Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of cimetidine on dextromethorphan O-demethylase activity of human liver microsomes and recombinant… Madeira, Maria 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0269.pdf [ 4.98MB ]
Metadata
JSON: 831-1.0091077.json
JSON-LD: 831-1.0091077-ld.json
RDF/XML (Pretty): 831-1.0091077-rdf.xml
RDF/JSON: 831-1.0091077-rdf.json
Turtle: 831-1.0091077-turtle.txt
N-Triples: 831-1.0091077-rdf-ntriples.txt
Original Record: 831-1.0091077-source.json
Full Text
831-1.0091077-fulltext.txt
Citation
831-1.0091077.ris

Full Text

T H E E F F E C T O F C I M E T I D I N E O N D E X T R O M E T H O R P H A N O - D E M E T H Y L A S E A C T I V I T Y O F H U M A N L I V E R M I C R O S O M E S A N D R E C O M B I N A N T C Y P 2 D 6 B y M A R I A M A D E I R A B.Sc. (Honours, Toxicology) , Universi ty o f Gu e l p h , 2000 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E i n T H E F A C U L T Y O F G R A D U A T E S T U D I E S Faculty o f Pharmaceutical Sciences D i v i s i o n o f Pharmacology and Toxico logy W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A M a r c h 2003 © Mar ia Madeira, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. DejjartmefrT'of -^gtcjJJw Jr 1v\^jv^aca<^rkc^A ^ c A e ^ c e ^ The University of British Columbia Vancouver, Canada Date }^>v\\ 7003 DE-6 (2/88) ABSTRACT Cimetidine is a liistamine H2-receptor antagonist used in the treatment o f peptic ulcer disorders. There have been numerous clinical observations o f cimetidine drug-drug interactions, in some cases involving known substrates o f cytochrome P450 (P450) 2D6. One explanation for these observations is impaired hepatic metabolism via inhibition o f P450s, the primary mediators of hepatic phase I metabolism of drugs. Cimetidine has been demonstrated to inhibit various P450 enzymes in vitro, including C Y P 2 D 6 , but the inhibition constants are generally higher than the plasma concentrations measured following in vivo administration. In rats, the in vitro selectivity and potency o f cimetidine can be increased by preincubating the inhibitor and N A P D H with hepatic microsomes, and the formation o f a metabolite-intermediate complex has been implicated. The effect o f preincubation on inhibition o f C Y P 2 D 6 by cimetidine has not been investigated. The objective o f this project was to further characterize the effect o f cimetidine on C Y P 2 D 6 in vitro using human liver microsomes and recombinant C Y P 2 D 6 . Dextromethorphan O-demethylase activity was used as a probe o f C Y P 2 D 6 activity. Metabolite formation was quantified using a high performance liquid chromatography assay with fluorescence detection. The Vmax and K m o f this reaction were 0.82+0.06 n m o l / m i n / n m o l P450 and 4.1 ±0.1 u.M, respectively, in microsomes; and 15.9±0.8 n m o l / m i n / n m o l P450 and 1.4±0.6 [iM, respectively, with recombinant C Y P 2 D 6 . W i t h human liver microsomes, cimetidine was a competitive inhibitor o f C Y P 2 D 6 with a K , o f 38.0±5.3 u M ; the observed inhibition was not effected by preincubation. W i t h the recombinant enzyme, cimetidine acted as a mixed inhibitor with a K i o f 102.5±16.8 [j.M. Preincubation with the inhibitor and N A D P H led to a further decrease in activity o f 11% to 18% i n a manner that was consistent with mechanism-based inactivation. The effect o f preincubation was attenuated by including quinidine in the preincubation reaction. The k;nact and K i were estimated to be 0.03 min- 1 and 19.2 \LM, respectively, while the ti/2 o f inactivation was 25 minutes. In the present study, cimetidine acted as an inhibitor o f dextromethorphan O-demethylation in both in vitro systems. W i t h recombinant C Y P 2 D 6 , loss o f activity was enhanced with preincubation. In both cases, the effect o f cimetidine as an inhibitor o f C Y P 2 D 6 was modest and does not fully explain the clinical findings. ii TABLE OF CONTENTS Page Abstract i i Table o f contents i i i L i s t o f tables v i L is t o f figures v i i i L i s t o f abbreviations x i i 1. Introduction 1 1.1 Cytochromes P450 2 1.2 Cytochrome P450 2 D 6 5 1.2.1 Pharmacogenetics o f C Y P 2 D 6 5 1.2.2 C Y P 2 D 6 * 1 0 7 1.3 In vitro methods used to study drug metabolism 8 1.3.1 H u m a n liver microsomes and recombinant enzymes 8 1.3.2 Use o f probe substrates 9 1.3.3 Dextromethorphan O-demethylase activity o f C Y P 2 D 6 10 1.4 Inhibi t ion o f cytochrome P450 enzymes 13 1.4.1 Competi t ive, non-competitive, and mixed inhibi t ion 13 1.4.2 Reversible inhibi t ion 14 1.4.2.1 Quinid ine 15 1.4.3 Quasi-irreversible and irreversible inhibi t ion 15 iii 1.4.3.1 Triacetyloleandomycin and 1-aminobenzottiazole 17 1.4.4 Cl in ica l consequences o f P450 inhibi t ion 17 1.5 Cimetidine 18 1.5.1 Drug-drug interactions wi th cimetidine 20 1.5.2 Inhibi t ion o f C Y P 2 D 6 by cimetidine 21 1.6 Hypothesis and Objectives 22 2. Materials and methods 25 2.1 Chemicals 25 2.2 E n z y m e incubations 25 2.2.1 Incubation conditions for human liver microsomes 25 2.2.2 Incubation conditions for recombinant C Y P 2 D 6 26 2.2.3 Preincubation experiments 27 2.3 H P L C assay methodology 28 2.3.1 Assay conditions 28 2.3.2 Calibrat ion curves 29 2.3.3 Assay validations 29 2.4 Data analysis 30 3. Results 32 3.1 Assay validation 32 3.2 Opt imiza t ion o f incubation conditions 43 3.3 Studies wi th human liver microsomes 49 3.3.1 Kine t i c constants o f dextromethorphan O-demethylation 49 3.3.2 Effect o f quinidine and cimetidine on dextromethorphan 0 - 52 demethylase activity iv - ' V 3.3.3 Preincubation studies wi th cimetidine 57 3.3.4 Mechanism-based inhibi t ion wi th triacetyloleandomycin and 70 1-aminobenzotriazole 3.3.5 Summary o f studies wi th human liver microsomes 70 3.4 Studies wi th recombinant C Y P 2 D 6 73 3.4.1 Kine t i c constants o f dextromethorphan O-demethylation 73 3.4.2 Effect o f quktidine and cimetidine on dextromethorphan O- 73 demethylase activity 3.4.3 Preincubation studies wi th cimetidine 79 3.4.4 Preincubation o f recombinant C Y P 2 D 6 wi th 1- 85 aminobenzotriazole and quinidine 3.4.5 Effect o f preincubation o f the K ; o f inhibi t ion by cimetidine 88 3.4.6 Effect o f quinidine on inactivation o f recombinant C Y P 2 D 6 95 by cimetidine 3.4.7 Determinat ion o f k, n a c t o f cimetidine 97 3.4.8 Summary o f studies wi th recombinant C Y P 2 D 6 97 4. Discuss ion 99 5. Future studies 111 5.1 Studies wi th human liver microsomes 111 5.2 Studies w i th recombinant C Y P 2 D 6 112 5.3 Studies wi th recombinant C Y P 2 D 6 * 1 0 114 6. Summary & Conclusions 115 7. References 117 v LIST OF TABLES Table 1: Intra-assay precision and accuracy o f quality control (QC) samples prepared with human liver microsomal matrix. Table 2: Inter-assay precision and accuracy o f quality control (QC) samples prepared with human liver microsomal protein matrix. Table 3: Intra-assay precision and accuracy o f quality control (QC) samples prepared with inactivated recombinant human C Y P 2 D 6 . Table 4: Inter-assay precision and accuracy o f quality control (QC) samples prepared with inactivated recombinant human C Y P 2 D 6 . Table 5: Summary o f the kinetic constants for dextromethorphan O -demethylation by pooled human liver microsomes determined in three separate experiments. Table 6: Summary o f estimated K i values for inhibition o f dextromethorphan O -demethylation by quinidine and cimetidine with pooled human liver microsomes, determined in three separate experiments. Table 7: Preincubation o f human liver microsomes with cimetidine (25 u.M) for up to 30 minutes. Table 8: Effect o f adding N A D P H to the preincubation mixture every 60 minutes versus only once at the beginning o f the preincubation reaction. Table 9: Summary o f the kinetic constants for dextromethorphan O -demethylation by recombinant C Y P 2 D 6 determined in three separate experiments. page 39 40 41 42 51 56 58 61 75 V I T a b l e 10: Summary o f estimated K i values for inhibition o f dextromethorphan O- 78 demethylation by quinidine and cimetidine with recombinant C Y P 2 D 6 , determined in three separate experiments. T a b l e 11: Summary o f K i values for the inhibition of dextromethorphan O- 91 demethylase activity o f recombinant C Y P 2 D 6 by cimetidine. vii L I S T O F F I G U R E S Page Figure 1: The P450 catalytic cycle 4 Figure 2: Metabolic pathways o f dextromethorphan demethylation 10 Figure 3: Chemical structure o f cimetidine and its major metabolites 19 Figure 4: Chromatographic separation o f pure standards o f dextromethorphan 34 and its metabolites. Figure 5A: Representative chromatogram o f dextrorphan and 3- 35 methoxymorphinan produced in a microsomal incubation. Figure 5B: Representative chromatogram o f dextrorphan and 3- 36 methoxymorphinan produced i n an incubation with recombinant C Y P 2 D 6 . Figure 6: Representative calibration curves for (A) dextrorphan ( D O R ) , (B) 3- 37 hydroxymorphinan (3-HM), and (C) 3-methoxymorphinan (3-MM). Figure 7: Optimization o f microsomal protein concentration 44 Figure 8: Optimization o f incubation time with pooled human liver 45 microsomes. Figure 9: Optimization o f N A D P H concentration with human liver 46 microsomes. Figure 10: Optimization o f recombinant C Y P 2 D 6 concentration. 47 Figure 11: Optimization o f incubation time with recombinant C Y P 2 D 6 . 47 viii Figure 12: Optimization o f N A D P H concentration with recombinant C Y P 2 D 6 . 48 Figure 13: Rate o f formation o f dextrorphan by human liver microsomes. 50 Figure 14: Inhibition o f dextromethorphan O-demethylase activity in human liver 53 microsomes by quinidine. Figure 15: Effect o f cimetidine free base and hydrochloride salt on 54 dextromethorphan O-demethylase activity in human liver microsomes. Figure 16: Inhibition o f dextromethorphan O-demethylase activity in human liver 55 microsomes by cimetidine. Figure 17: Effect o f addition o f N A D P H to the preincubation mixture every 30 60 minutes. Figure 18: Preincubation o f pooled human liver microsomes with cimetidine for 63 up to 120 minutes. Figure 19: Preincubation o f human liver microsomes from individual H G 95 66 with cimetidine for up to 120 minutes. Figure 20: Preincubation o f human liver microsomes from individual H G 64 67 with cimetidine for up to 120 minutes. Figure 21: Effect o f preincubation o f human liver microsomes with cimetidine 68 for 90 minutes, followed by incubation with 50 U . M dextromethorphan, on dextromethorphan O-demethylase activity in human liver microsomes. Figure 22: Effect o f preincubation o f human liver microsomes with cimetidine 69 for 90 minutes, followed by incubation with 50 u.M or 300 u.M dextromethorphan, on dextromethorphan N-demethylase activity o f human liver microsomes. Figure 23: Inactivation o f dextromethorphan N-demethylase activity in human 71 liver microsomes by triacetyloleandomycin ( T A O ) . ix Figure 24: Inactivation o f dextromethorphan O-demethylase activity in human 72 liver microsomes by 1-aminobenzotriazole (ABT) . Figure 25: Rate o f formation o f dextrorphan by recombinant C Y P 2 D 6 . 74 Figure 26: Inhibition o f dextromethorphan O-demethylase activity o f 76 recombinant C Y P 2 D 6 by quinidine. Figure 27: Inhibition o f dextromethorphan O-demethylase activity o f 77 recombinant C Y P 2 D 6 by cimetidine. Figure 28: Preincubation o f recombinant C Y P 2 D 6 with cimetidine for 5, 10, and 80 20 minutes, followed by incubation with 1 fxM dextromethorphan. Figure 29: Preincubation o f C Y P 2 D 6 with cimetidine (50 u,M) for 20 minutes, 81 followed by incubation with 1 u.M dextromethorphan. Figure 30: Preincubation o f recombinant C Y P 2 D 6 with cimetidine for 5, 10, and 83 20 minutes, followed by incubation with a saturating concentration o f dextromethorphan. Figure 31: Preincubation o f C Y P 2 D 6 with cimetidine for 20 minutes, followed 84 by incubation with a saturating concentration o f dextromethorphan. Figure 32: Mechanism-based inactivation o f recombinant C Y P 2 D 6 by 1- 86 aminobenzotriazole (ABT) . Figure 33: Lack o f mechanism-based inactivation o f recombinant C Y P 2 D 6 by 87 quinidine. Figure 34: Inhibition o f recombinant C Y P 2 D 6 by cimetidine following 20 89. minutes preincubation using the dilution method. Figure 35: D i x o n plot for the inhibition o f recombinant C Y P 2 D 6 by cimetidine 90 following 20 minutes preincubation. x Figure 36A: D i x o n plot for the inhibition o f recombinant C Y P 2 D 6 by cimetidine following 20 minutes preincubation using saturating substrate concentrations. 93 Figure Lineweaver-Burk transformation o f the data presented in figure 33A. 94 36B: Figure 37: Effect o f quinidine on inactivation o f recombinant C Y P 2 D 6 by 96 cimetidine. Figure 38: Inactivation o f recombinant C Y P 2 D 6 by cimetidine. 98 xi LIST OF ABBREVIATIONS u.L microlitre U . M micromolar A B T 1-aminobenzotriazole A U C area under the plasma concentration versus time curve C degrees Celsius C l i n t in vitro intrinsic clearance, equivalent to V m a x / K m C V coefficient o f variation C Y P 2 D 6 cytochrome P450 2D6 dex dextromethorphan D O R dextrorphan E M extensive metabolizer 3 - H M 3-hydroxymorphinan H P L C high performance liquid chromatography K i inhibition constant; concentration o f inhibitor required for half-maximal inhibition K i concentration o f inactivator required for half-maximal inactivation kinact rate constant for inactivation K m Michaelis-Menten constant; concentration o f substrate required to achieve half-maximal activity L O Q limit o f quantitation mg milligram min minute mL millilitre m M millimolar xii 3 - M M 3-methoxymorphinan N A D P H p-nicotinamide adenine dinucleotide phosph n M nanomolar n m o l nanomole P450 cytochrome P450 P M poor metabolizer p m o l picomole P V D F polyvinylidene fluoride Q C quality control tl/2 half-life T A O triacetyloleandomycin "Vmax maximal velocity o f reaction xiii Chapter 1 INTRODUCTION The effects o f drug therapy can vary markedly among individuals and one o f the major factors contributing to this inter-individual variation is altered capacity o f the enzymes involved in drug metabolism. T w o o f the ways in which enzyme activity can be altered are the expression o f polymorphic forms o f the enzyme, or by drug-drug interactions resulting in enzyme inhibition. Variant alleles can arise from a number o f mutational events and can result in increased, diminished, or absent enzyme activity. Such polymorphisms are known to exist for over 20 phase I and phase II drug metabolizing enzymes, including several cytochrome P450 (P450) enzymes (Ingelman-Sundberg 1999). P450s, which are the predominant mediators o f phase I metabolism o f xenobiotics, are also susceptible to inhibition by a large number o f drugs. Therefore, there is a potential for a compound effect caused by the inhibition o f polymorphically expressed enzymes, which may result in unexpected responses. Cimetidine is an example o f a drug that has been implicated in a number o f clinical interactions, some o f which are thought to involve C Y P 2 D 6 . C Y P 2 D 6 is one o f the most extensively characterized polymorphic P450s and over 75 alleles, conferring varying degrees o f metabolic activity, have been identified (Ingelman-Sundberg and Evans 2001). The frequency o f a given allele can vary gready between ethnic populations, as is seen with the CYP2D6*10 allele, which is almost absent in Caucasians (Marez et al. 1997), but has a frequency o f over 50% in some Asian populations (Garcia-Barcelo et al. 2000). Expression o f the C Y P 2 D 6 * 1 0 protein results in an enzyme with decreased activity. It is possible that the effect o f cimetidine as an inhibitor o f C Y P 2 D 6 may be dependent on the phenotype 1 o f the expressed enzyme, and this could explain some o f the clinical observations that have previously not had a clear mechanism. In vitro models can be used to predict where drug-drug interactions are likely to occur, and an individual's genotype for polymorphically expressed drug metabolizing enzymes, such as C Y P 2 D 6 , can be accurately determined. Information regarding how an inhibitor affects the activity o f an enzyme, or any o f its genetic variants, could therefore be used to determine where adjustments in dosing are needed to reduce the potential for toxicity in individuals with altered metabolic capacity. 1.1 Cytochromes P450 P450s are heme-thiolate proteins that catalyze mono-oxygenase reactions in the presence o f molecular oxygen, N A D P H , and the accessory enzyme N A D P H - c y t o c h r o m e P450 reductase (Gonzalez 1989). P450s are expressed in all eukaryotes and some prokaryotes; however, the expression o f specific enzymes is highly species-specific and, in some cases, sex- and age-dependent (Omura 1999). Individual enzymes are divided into families (denoted by arabic numerals) and subfamilies (denoted by letters) based on the similarity o f their amino acid sequences. Enzymes within the same family demonstrate less than 40% homology with proteins in other families, while enzymes within the same subfamily share at least 59% homology in their amino acid sequences (Gonzalez and Nebert 1990). Individual genes are distinguished by a second arabic numeral. In humans, at least 14 different P450 families exist (Nelson et al. 1996). Families 1 through 3 are predominandy involved in the metabolism of xenobiotics, while the other families catalyze endogenous reactions, such as those involved in the synthesis o f steroid hormones, leukotrienes, prostaglandins and bile acids (Smith et al. 1998). P450s differ from most enzymes in that they have broad and overlapping substrate specificity. A s a result o f this, a large number o f structurally diverse chemicals can be metabolized and several enzymes may be involved in the metabolism o f a single substrate. In humans, P450s are responsible for the majority o f phase I reactions involved in the 2 metabolism o f drugs and other xenobiotics. This step functions to render generally hydrophobic compounds more susceptible to phase II conjugation, which results in the addition o f a hydrophilic moiety to the compound. The end result is the formation o f a more water-soluble species which can be readily excreted. The majority o f the xenobiotic metabolizing P450s are expressed in the liver, but can also be found in extrahepatic tissues such as the gastrointestinal tract, kidneys, and lungs (Smith et al. 1998). A t a subcellular level, these P450s are embedded within the membrane o f the endoplasmic reticulum. Dur ing the isolation o f these enzymes from cells, the endoplasmic reticular membrane is disrupted and forms small vesicles called microsomes, which contain the P450 enzymes. The catalytic cycle o f P450s can be generally described by 8 steps (Guengerich 2001), which are depicted in figure 1. The substrate binds to the enzyme in step one, while the iron is in the ferric state. In step 2, the i ron is reduced by the transfer o f a single electron from N A P D H - P 4 5 0 reductase. Molecular oxygen then binds to the ferrous iron (step 3), resulting in an Fe 2 + , G"2 complex. A second electron is then transferred (step 4), followed by the addition o f a proton (step 5). The nature o f the iron-oxygen complex in these, and subsequent, steps is unclear. In step 6, the O - O bond is cleaved and, by addition o f another proton, water is released. The product is then formed (step 7) and released from the active site o f the enzyme (step 8). 3 R O H R H Figure 1: The P450 catalytic cycle. See text for a description of each step. Adapted from Guengerich (2001). 4 1.2 Cy toch rome P450 2D6 Only a single enzyme in the C Y P 2 D subfamily has been detected in the human liver: C Y P 2 D 6 (Nelson et al. 1996). The human C Y P 2 D 6 enzyme was first purified by Distierath et al. (1985) and was later cloned by Kimura et al. (1989). Located on the long arm of chromosome 22, the CYP2D gene locus contains three genes: CYP2D6, which codes for the functional enzyme, and CYP2D7P and CYP2D8P. The latter two genes are not transcribed; CYP2D7P contains an insertion mutation which disrupts the reading frame, while CYP2D8P contains multiple nucleotide deletions and insertions (Kimura et al. 1989). Although C Y P 2 D 6 represents only 2% of total hepatic C Y P content (Shimada et al. 1994), it plays a major role in drug metabolism and is involved in the metabolism of over 50 clinically important drugs (Hasler et al. 1999). C Y P 2 D 6 substrates, which include various antiarrhythmics, (3-blockers, antihypertensives, antidepressants, neuroleptics, and opioid derivatives (Cholerton et al. 1992), have litde in common other than the presence o f a nitrogen which can be protonated at physiological p H and is thought to form an ion pair with the negatively charged aspartate residue located in position 301 o f the C Y P 2 D 6 amino acid sequence (Ellis et al. 1995). This nitrogen atom is usually found 5 to 7 A from the site o f hydroxylation (Lewis et al. 1999). 1.2.1 Pharmacogenetics of CYP2D6 It has been recognized for over 30 years that genetic factors play a major role in inter-individual differences in response to drug therapy (Kalow 1968, 1971). Genetic polymorphisms o f the genes encoding drug transporters, receptors, or metabolizing enzymes can lead to the expression o f proteins with altered activity towards drugs. M u c h research has been focused on characterizing and determining the functional consequences o f mutations within the genes o f drug metabolizing enzymes. These variant alleles can arise from a number o f mutational events including gene deletions, single nucleotide polymorphisms (which can in turn result in frameshift, missense, nonsense, or splice-site mutations), or gene duplications (Ingelman-Sundberg et al. 1999) and can result in absent, decreased, or 5 increased enzyme activity or expression. Clinically, polymorphisms can have several effects on the outcome o f pharmacotherapy, including therapeutic failure or an increased susceptibility to toxicity. These effects are generally most noticeable with drugs that are frequendy prescribed or have a narrow therapeutic index (Bronsen and G r a m 1989). The frequency o f a polymorphic allele can vary greatiy between ethnic populations so that the responses o f different ethnic groups to the same drug may be quite different. A s a result o f this, dosages o f drugs primarily metabolized by polymorphic enzymes may have to be determined for individual populations to reduce the negative influence o f these polymorphisms on effective therapy (Tanaka 1999). The C Y P 2 D 6 polymorphism was first identified by Mahgoub et al. (1977). Using debrisoquine (a drug which is predominandy metabolized by C Y P 2 D 6 [Dalen et al. 1998]) as a probe, metabolic ratios were determined by measuring the amount o f the parent drug relative to the metabolite 4-hydroxydebrisoquine in the subjects' urine. T w o groups o f individuals, classified as non-metabolizers and extensive metabolizers, were identified based on the bimodal distribution o f the calculated metabolic ratios and the non-metabolizer phenotype was found to be associated with the inheritance o f a single autosomal recessive gene. Further studies, however, have shown that the classification o f C Y P 2 D 6 phenotypes is more complex and four phenotypic groups are now recognized: poor, intermediate, extensive, and ultrarapid metabolizers, with extensive metabolizers (EMs) having "normal" enzyme activity and poor metabolizers (PMs) having no C Y P 2 D 6 activity (Van der Weide and Steijns 1999). Certain alleles are associated with specific phenotypic groups. However, the divisions between these groups are not well defined and the large number o f possible combinations o f C Y P 2 D 6 alleles adds an additional complexity to the issue. Over 75 variant alleles, in addition to the 'wild-type' allele (denoted CYP2D6*1 [Marez et al. 1997]), have been identified and the functional effects o f all o f these polymorphisms have not been determined (Ingelman-Sundberg and Evans 2001). Mos t o f these alleles, however, are rare and an individual's phenotype can be accurately predicted by screening for a small group o f alleles known to be prevalent in a given population. For example, in the 6 German population, screening for four genes provided an accurate assessment o f an individual's phenotype (Sachse et al. 1998). 1.2.2 CYP2D6*10 O f the many C Y P 2 D 6 variants that have been identified, C Y P 2 D 6 * 1 0 , which results in an intermediate metabolizer phenotype, is o f particular importance due to its prevalence in Asian populations. The occurrence o f the poor metabolizer phenotype in Asians is rare. However, impaired metabolism of C Y P 2 D 6 substrates was observed in studies with Chinese patients being treated with various neuroleptics agents, compared to Caucasians being treated with the same drugs (Potkin et al. 1984, Rudorfer etal. 1984). Further research lead to the identification o f the CYP2D6*10 allele, which is characterized by the presence o f three nucleotide substitutions: C 1 8 8—>T, G 1 7 4 9 —>C, and G 4 2 6 8 —>C (Marez et al. 1997). Several other forms o f the allele have been identified, which differ based on the presence o f additional nucleotide substitutions, but code for enzymes that display the same catalytic activity (Marez etal. 1997). Estimates o f the CYP2D6*10 allele frequency range from 0.41 to 0.68 in Japanese populations, where this variant represents the most common C Y P 2 D 6 allele (Tateishi et al. 1999, Kubota et al. 2000), and 0.64 in the Chinese population (Garcia-Barcelo et al. 2000). The proportion o f homozygous carriers o f this allele has been estimated to be from 0.17 among Japanese (Tateishi et al. 1999) to 0.45 and 0.41 in Korean and Chinese populations, respectively (Yoon et al. 1997, Garcia-Barcelo et al. 2000). In contrast, the frequency o f the CYP2D6*10 allele in Caucasians is less than 0.02 (Sachse et al. 1997, Grise et al. 1998). Various clinical studies have demonstrated that individuals who have the CYP2D6*10 allele experience a reduced ability to metabolize C Y P 2 D 6 substrates (Wang et al. 1993, Fukada et al. 1999, Y o o n et al. 2000) which is direcdy related to the number o f CYP2D6*10 alleles present (Johansson et al. 1994). This reduction in clearance is due to two factors: a decreased expression o f the enzyme and a reduced affinity o f the mutant enzyme for substrates. Johansson et al. 7 (1994) demonstrated that the presence o f the C 1 8 8—>T mutation, which results in a Pro—»Ser amino acid substitution at position 34 o f the amino acid sequence, leads to the expression o f an unstable gene product, thus resulting in lower enzyme levels. In studies o f the catalytic activity o f C Y P 2 D 6 * 1 0 , Fukada et al. (2000) found that while the V m a x is unaffected, the K m o f both bufuralol l '-hydroxylation and venlafaxine O-demethylation is increased by 5- to 10-fold compared to C Y P 2 D 6 * 1 . 1.3 In vitro methods used to study drug metabolism 1.3.1 Human liver microsomes and recombinant enzymes There are a number o f different in vitro models available for the study o f drug metabolism, which include cultured hepatocytes, liver tissue slices, subcellular fractions such as microsomes, and heterologously expressed drug metabolizing enzymes (Rodrigues 1994). The most commonly used models in studies' o f cytochrome P450-related drug metabolism are hepatic microsomes and heterologously expressed enzymes. These two in vitro systems each have their own advantages and disadvantages. Microsomes offer the advantage o f having a full complement o f P450s, with each enzyme present in ratios and concentrations typically seen in vivo (Crespi 1995). Levels o f enzyme expression and activity can vary greatly between individuals, however, and the source o f human microsomes used can influence experimental results (Crespi and Penman 1997). This influence can be minimized by using pooled microsomes from a number o f donors, which has the effect o f averaging out any extreme differences in enzyme expression. A n additional advantage found with microsomes is that the enzymes are present in their "natural" environment in terms o f the lipid membrane in which they are embedded and the number o f accessory proteins present, such as N A D P H - c y t o c h r o m e P450 reductase. Changes in either o f these factors can lead to altered enzyme activity (Venkatakrishnan et al. 2000). Some o f the disadvantages associated with microsomes are the absence o f phase II metabolizing enzymes, which results in an incomplete representation o f metabolic pathways seen in vivo, and the fact that inhibitory 8 antibodies or chemicals (both o f which are rarely specific for an single enzyme) are usually required to study the activity o f a single enzyme (Crespi 1995). The study o f single enzymes is facilitated by the use o f heterologously expressed, or recombinant, P450s. Expression systems that are commonly used include bacteria, yeast, insect, and human cells (Crespi and Miller 1999). Experiments carried out using this type o f model system can provide qualitative information regarding the enzymes involved in the metabolism of drugs, but give no indication as to how much an individual enzyme contributes to metabolism in vivo (Crespi and Penman 1997). A s a result o f this, it is difficult to relate data derived from recombinant P450s to what might be seen in vivo. A number o f different methods o f extrapolating this data from recombinant to microsomal models have been developed, including the use o f relative activity scaling factors (Nakajima etal. 1999, Shimada etal. 1999, Venkatakrishnan etal. 2000). 1.3.2 Use of probe substrates The activity o f P450 enzymes can be studied using probe substrates: compounds that are metabolized to a known product by the enzyme o f interest. Preferably, only a single enzyme is involved in the production o f the metabolite; with P450s, however, this is rarely the case. Due to their overlapping substrate specificity, several enzymes may be involved in the metabolism of a single compound, and even in the production o f a single metabolite. However, i f the different enzymes exhibit sufficiendy different affinities for the substrate, the specificity o f the reactions may be increased under defined experimental conditions. There are several experimental approaches used to determine which P450s are involved in the metabolism o f a compound, which include correlating the enzyme activity o f interest to that o f a known marker activity o f a specific P450 enzyme; assessing the effect o f chemical inhibitors or inhibitory antibodies; or by using individual recombinant enzymes to determine which P450s are capable o f catalyzing the reaction (Halpert et al. 1994). Often, a combination o f these methods is used to provide a more accurate estimate o f the specificity o f the reaction. 9 1.3.3 Dextromethorphan O-demethylase activity of CYP2D6 Dextromethorphan is a commonly used in vivo and in vitro probe o f C Y P 2 D 6 activity. The metabolic pathways o f dextromethorphan have been well characterized in both human liver microsomes (Dayer et al. 1989, von Moltke et al. 1998) and with recombinant P450 enzymes (Yu and Haining 2001), and its use as an in vivo marker has been demonstrated (Capon et al. 1996, Jones et al. 1996). Dextromethorphan undergoes O- or AT-demethylation to form dextrorphan or 3-methoxymorphinan, respectively, and these two metabolites are both further demethylated to form the secondary metabolite, 3-hydroxymorphinan (figure 2). HO 3-hydroxymorphinan Figure 2: Metabolic pathways of dextromethorphan demethylation. The major P450 enzyme responsible for the formation of each metabolite is shown in bold lettering 1 0 The formation o f dextrorphan, the O-demethylated metabolite, has been well documented in vitro and the conditions under which this pathway is specific for C Y P 2 D 6 activity are known. Studies with human liver microsomes have shown the formation o f dextrorphan to be biphasic, with the high affinity pathway having a K m in the range o f 2.3 to 5.3 u.M and the low affinity pathway having a K m approximately ten-fold higher (Dayer et al. 1989, Kerry et al. 1994), suggesting that at least two different enzymes contribute to the formation o f this metabolite. While this pattern is typically seen in microsomes from extensive metabolizers o f debrisoquine, one-site kinetics are observed with microsomes from poor metabolizers, with a K m similar to that o f the low affinity pathway in extensive metabolizers (Kerry et al. 1994). The absence o f a high affinity enzyme pathway in poor metabolizers, who demonstrate litde or no C Y P 2 D 6 activity, suggests that this pathway is primarily mediated by C Y P 2 D 6 . The effect o f selective inhibitors o f P450 enzymes on dextromethorphan O-demethylase activity was investigated by Schmider et al. (1997) and confirms the role o f C Y P 2 D 6 as the high affinity enzyme involved i n dextrorphan formation. Quinidine, a C Y P 2 D 6 inhibitor, reduced the formation o f dextrorphan by 80%, while other C Y P 2 D 6 substrates, such as bufuralol, debrisoquine, and sparteine competitively inhibited dextromethorphan O-demethylation in microsomes from extensive metabolizers with K i values which ranged from 7.5 U . M for bufuralol to 45 [xM for sparteine, when the concentrations o f dextromethorphan used were well below the k m for the low affinity pathway (Dayer et al. 1989). In contrast, inhibitors o f C Y P s 1 A 1 / 2 (oc-naphthaflavone), 2C9 (sulfaphenazole), and 3A (triacetyloleandomycin and ketoconazole) had no effect (Dayer et al. 1989). In vivo administration o f quinidine has been shown to inhibit dextrorphan formation in extensive metabolizers, but has no effect in poor metabolizers (Capon et al. 1996). Studies with individual P450 enzymes (von Moltke et al. 1998, Y u and Haining 2001) have shown that recombinant C Y P s 2D6, 3A4, 2C9, 2C19, and 2B6 are all capable o f metabolizing dextromethorphan to dextrorphan, but C Y P 2 D 6 has the highest intrinsic clearance (Clint , which is represented by the ratio o f V m i x / K m for the reaction). Using in vitro relative activity factor scaling to 11 determine the relative contribution o f each enzyme, it was estimated that at substrate concentrations up to two-fold higher than the K m for the high affinity pathway, C Y P 2 D 6 is responsible for approximately 70% o f dextrorphan formation, while at higher substrate concentrations, the involvement o f C Y P 2 C 9 increases (von Moltke et al. 1998). The iV-demethylation o f dextromethorphan has been proposed as a probe o f C Y P 3 A activity, although its suitability is questionable. Using recombinant P450 enzymes, Y u and Haining (2001) showed that the formation o f 3-methoxymorphinan was mediated by C Y P s 2B6, 3A4, 2C19, 2D6, 2C9, and 2C19. C Y P 2 B 6 and 3A4 demonstrated the highest intrinsic clearance, but the overall contribution o f C Y P 2 B 6 in human liver microsomes or in vivo was predicted to be minor due to the low hepatic expression o f this enzyme, and C Y P 3 A 4 was estimated to be responsible for greater than 90% o f dextromethorphan IV-demethylation activity. In a similar study, however, C Y P 3 A 4 was predicted to be responsible for only 50% o f N-demethylation activity, with C Y P s 2C9 and 2C19 equally responsible for the residual activity (von Moltke et al. 1998). This second estimate is in agreement with the results from several studies with human liver microsomes, where maximal inhibition o f dextromethorphan N -demethylase activity by triacetyloleandomycin and indinivar, both o f which were shown to completely inhibit recombinant C Y P 3 A 4 / 5 activity, was only 60% with human liver microsomes (Wang and Unadkat 1999). Similarly, an t i -CYP3A antibodies resulted in an inhibition o f 40-50% (Gorski et al. 1994, Schmider et al. 1997). The formation o f the secondary metabolite, 3-hydroxymorphinan, is a result o f the IV-demethylation o f dextrorphan or the O-demethylation o f 3-methoxymorphinan. The O-demethylation pathway, mediated by C Y P 2 D 6 , is thought to be the major route o f formation. The formation o f 3-hydroxymorphinan by the N-demethylation pathway is mediated by C Y P 3 A 4 (Yu and Haining 2001). 12 1.4 I n h i b i t i o n o f cytochrome P 4 5 0 enzymes A large number o f compounds, including many drugs, are known to inhibit P450 enzymes. Impairment o f any o f the steps o f the P450 catalytic cycle can lead to loss o f enzyme activity; however, the steps involving the binding o f the substrate, binding o f molecular oxygen, and oxidation o f the substrate (steps 1, 3, and 7, respectively, in figure 1) are most susceptible (Oritz de Montellano and Correia 1995). The resulting inhibition may be relatively selective, and affect only one or two closely related P450 enzymes, or may be more general, and depends on several factors, including the chemical properties o f the inhibitor and its mechanism o f action (Lin and L u 1998). There are two systems o f classification used to categorize enzyme inhibitors. The first system describes inhibitors based on their effect on the kinetic parameters o f the catalyzed reaction and includes competitive, non-competitive, and mixed inhibitors. The second system addresses the mechanism by which the inhibitor acts. In this case, inhibitors can be classified as reversible, quasi-irreversible, or irreversible. The latter method o f classification is often more useful when discussing inhibitors o f P450s because it is more specific (for example, non-competitive inhibition can be caused by both reversible and irreversible inhibitors). 1.4.1 Competitive, non-competitive, and mixed inhibition Competitive inhibitors act by binding to the active site o f the enzyme, thus preventing the substrate from binding and being metabolized. They are also often, but not always, capable o f acting as substrates for the enzymes they inhibit. Competitive inhibitors are characterized by reducing the apparent affinity (or increasing the K m , the concentration o f substrate required to reach half-maximal activity) o f the enzyme for its substrate, but have no effect on the maximal rate o f the reaction (Vmax). Increasing concentrations o f the substrate wi l l displace the inhibitor from the active site and reverse its effect on enzyme activity. Non-competitive inhibitors, on the other hand, bind to a site distinct from the substrate binding site. A s a result, these inhibitors do not affect the enzyme's affinity for the 13 substrate. Inhibition cannot be overcome by increasing the concentration o f the substrate and the V m a x o f the enzyme wil l be decreased. Mixed inhibitors act by a combination o f competitive and non-competitive inhibition and can affect both the K m and the Vmax (Copeland 1996, L i n and L u 1998). 1.4.2 Reversible inhibition Reversible inhibitors can bind to P450s by one o f two modes: substrate (or productive) binding and ligand binding (Testa 1990). Inhibition by the first mode occurs when two compounds compete for binding to the active site o f the same enzyme. Maximal inhibition is seen when one substrate binds with a greater affinity, but has a much lower catalytic turnover, than the other (Oritz de Montellano and Correia 1995). Enzyme activity is restored when the inhibitor is metabolized or diffuses away from the active site (Murray 1987). Ligand binding usually involves nitrogen-containing compounds, such as quinolones, imidazoles, and pyridines. The lone pair o f electrons on the nitrogen forms a strong coordinate bond with the ferric form of the prosthetic heme, resulting in a type II difference spectrum with a soret band in the region o f 420-425 n m (Halpert 1995). Inhibition due to ligand binding prevents substrate turnover by increasing the redox potential o f the enzyme, thus making the electron transfer step more difficult, and by occupying the binding site required for molecular oxygen (Oritz de Montellano and Correia 1995). Compounds rarely form only one type o f bond, and those which are able to bind to both the heme iron and the protein moiety through lipophilic interactions are inherendy more potent (Oritz de Montellano and Correia 1995). The selectivity o f these inhibitors is determined by how tightly they fit into the active site o f the enzyme, and whether the inhibitor's orientation in the active site allows for the formation o f a coordinate bond with the ferric heme (Halpert 1995). The relative potency o f reversible inhibitors is defined by the inhibition constant (Ki) which is a measure o f the inhibitor's binding capacity for the enzyme (Copeland 1996). 14 1.4.2.1 Quinidine Quinidine, an example o f a reversible P450 inhibitor, is a potent and selective inhibitor o f C Y P 2 D 6 in vitro. Studies with human liver microsomes have shown that quinidine is a competitive inhibitor o f dextromethorphan O-demethylase activity with a K ; ranging from 0.028 to 0.4 U . M (Kobayashi et al. 1989, Ching et al. 1995, Bourrie et al. 1996). Furthermore, concentrations o f quinidine which result in more than 80% inhibition o f C Y P 2 D 6 activity inhibit other P450 enzymes less than 10% (Newton et al. 1995). The potency o f quinidine as an inhibitor is related to its ability to bind to both the substrate-binding site and to co-ordinate with heme iron (Oritz de Montellano and Correia 1995). The binding o f quinidine to rat hepatic microsomes results in a type II binding spectrum, suggesting that the inhibitor is interacting with the ferric i ron (Murray 1984). Unlike the majority o f reversible, competitive inhibitors, quinidine does not undergo metabolic biotransformation by the enzyme it inhibits; instead, its metabolism is primarily mediated by C Y P 3 A (Guengerich etal. 1986). 1.43 Quasi-irreversible and irreversible inhibition These types o f inhibitors require catalytic activation to a reactive species by the target enzyme; for this reason, they are also referred to as being "mechanism-based". In this thesis, inhibition o f enzyme activity by a mechanism-based process wi l l be referred to as "inactivation" to distinguish it from loss o f activity due to reversible inhibition. There are three steps involved in mechanism-based inactivation: the inactivator reversibly binds to the active site o f the enzyme, undergoes catalytic conversion, and the reactive species binds to the enzyme leading to loss o f activity (Halpert et al. 1994). Quasi-irreversible and irreversible inactivators differ in the way in which they bind to the target enzyme. W i t h quasi-irreversible inactivators, a metabolite intermediate (MI) complex is formed, in which the product o f P450-dependent metabolism binds tighdy, but not covalentiy, to the heme group (Oritz de Montellano and Correia 1995). These inactivators are described as quasi-irreversible because although the M I complex can be dissociated under certain conditions in vitro, such as incubation with a 15 highly Hpophilic compound or oxidation o f the heme group with potassium ferricyanide, these conditions are rarely achieved in vivo and enzyme activity is only restored on synthesis o f new protein. Irreversible inactivators, on the other hand, lead to loss o f enzyme activity through covalent modification o f the enzyme. This can occur by several mechanisms, including binding to the prosthetic heme group o f the P450 enzyme, the apoprotein moiety, or through activation o f the heme group to products that subsequently covalently bind to the apoprotein (Halpert 1995). Mechanism-based inactivators are generally more selective and potent than reversible inhibitors because the compound must be able to bind to the active site o f a specific P450 enzyme, must have the properties required o f a substrate in order to be catalytically activated, and then alter the enzyme structure, resulting in its permanent removal from the catalytic pool (Oritz de Montellano and Correia 1995). Several criteria must be met for an inactivator to be classified as mechanism-based (Silverman 1995). Inactivation is time-dependent, as it is limited by the rate o f conversion o f the inactivator to the reactive species, and is saturable, with maximal inactivation seen when a sufficient concentration o f the inhibitor is present to occupy all o f the enzyme molecules. The presence o f a substrate o f the target enzyme wi l l slow the rate o f inactivation by restricting access o f the inactivator to the active site, through a process called substrate protection. Enzyme inactivation is irreversible, and follows a 1:1 stoichiometry: one inhibitor molecule wi l l inactivate one enzyme molecule. Finally, enzyme inactivation must occur prior to the release o f the reactive species from the active site o f the enzyme. While reversible inhibitors are defined by their K ; value, mechanism-based inactivation is more appropriately described by the rate constant for inactivation (k;nact) and the concentration o f inactivator required to achieve a half-maximal rate o f inactivation (Ki). The kinact value is a complex expression which combines the rate constants that define binding o f the inhibitor, its catalytic conversion to a reactive species, and the rate at which this reactive species is either released from the active site as a product or binds to and inactivates the enzyme (Silverman 1995). 16 1.4.3.1 Triacetyloleandomycin and 1-aminoben^ptria^ple Triacetyloleandomycin ( T A O ) and 1-arninobenzotriazole (ABT) are two compounds which act as mechanism-based inactivators o f P450s. T A O is a highly selective inactivator o f C Y P 3 A (Chang et al. 1994) which acts by the formation o f a metabolite-intermediate complex with the ferrous heme o f the enzyme, a complex which can be readily dissociated with potassium ferricyanide (Pessayre et al. 1982, Lindstrom et al. 1993). The parent compound is first N-delakylated to a primary amine, which then undergoes further metabolism to a nitroso intermediate. This species then coordinates the heme iron through the lone pair o f electrons on the nitroso functional group (Oritz de Montellano and Correia 1995). Unlike T A O , A B T causes irreversible inactivation o f P450 enzymes through alkylation o f the heme group by a reactive species, which has been demonstrated through the isolation o f abnormal porphyrin products following incubation o f hepatic microsomes with A B T (Oritz de Montellano and Mathews 1981). The destruction o f microsomal P450s can result in a loss o f almost 80% o f total P450 following administration o f A B T to rats, and is both time- and concentration-dependent (Colby et al. 1995). In comparison to T A O , the selectivity o f A B T for specific P450 enzymes is not as great. A t lower concentrations o f the inactivator, both human recombinant C Y P s 3 A and 2 D 6 are inactivated with IC50 values less than 10 p.M, while C Y P s 1A2, 2B6, 2C9, and 2C19 are all inactivated to a lesser extent, with IC50 values between 50 and 250 ( i M (Balani etal. 2002). 1.4.4 Clinical consequences ofP450 inhibition Drug-drug interactions can occur when one drug inhibits the hepatic clearance o f another. Impairment o f the activity o f drug-metabolizing enzymes can have several clinical consequences, including an exaggerated pharmacological effect or increased incidence o f adverse effects due to elevated level o f the parent drug, or therapeutic failure in the case o f prodrugs that require activation by metabolic enzymes (Benetti and Bani 1998). Enzyme inhibition is most likely to have clinical consequences when the affected drug has a low therapeutic index and small changes in plasma 17 concentrations can lead to significant alteration in the toxicity profile o f the drug. The duration and severity o f the effect is dependent on the type o f inhibition and the degree to which the inhibited enzyme is involved in the metabolism of concurrendy administered drugs. Interactions due to competitive inhibition last only as long as the inhibitor is present in the liver at sufficient concentrations; enzyme activity is restored as the inhibitor gradually diffuses away from the active site. W i t h mechanism-based inactivation, however, recovery is dependent on the synthesis o f new enzyme and therefore requires a longer period o f time (Murray 1997). Inhibition o f the P450 enzymes involved in the metabolism o f a drug wi l l lead to a decrease in the drug's clearance, with the magnitude o f the effect related to the role the inhibited enzyme plays in the overall clearance o f the drug. Changes in clearance are reflected by changes in the area under the plasma concentration time curve (AUC) : a decrease in clearance wi l l result in an increased A U C (Lin and L u 1998). The pharmacological response or toxicity o f a drug can also be related to the A U C (Ito et al. 1998), therefore this pharmacokinetic parameter provides a good assessment o f the effect o f in vivo enzyme inhibition. 1.5 Cimetidine Cimetidine is an H2-receptor antagonist used in the treatment o f gastric acid disorders (Feldman and Burton 1990). Structurally, cimetidine is a substituted imidazole (figure 3), which is slightly basic. The pharmacokinetic properties o f cimetidine have been well documented. Fol lowing oral administration, the drug is rapidly absorbed, with a maximal plasma concentration o f 6 to 9 \iM achieved 60 to 90 minutes after administration o f a 300 mg dose (Brogden et al. 1978, Somogyi and Gugler 1983). Plasma concentrations are proportional to dose when the drug is administered in single doses o f up to 800 mg (Bodemar et al. 1981). The major route o f elimination o f cimetidine is renal excretion. Wi th in 24 hours o f dosing, 50 to 80% of the drug is recovered unchanged in the urine 18 (Somogyi and Gugler 1983). The remainder o f the drug is metabolized to four products: cimetidine-N ' -glucuronide , cimetidine sulphoxide, hydroxymethyl cimetidine, and guanylurea cimetidine (Somogyi and Gugler 1983). Severe adverse effects are rarely seen with cimetidine treatment, however, the drug has been implicated in a number o f drug-drug interactions. cimetidine N'-glucuronide cimetidine guanylurea Figure 3: Chemical structure of cimetidine and its major metabolites 19 1.5.1 Drug-drug interactions with cimetidine The first study o f a drug interaction with cimetidine was reported by Serlin et al. in 1979, where concurrent treatment with warfarin was found to potentiate the effect o f the anticoagulant. This effect was associated with an increased plasma concentration and decreased clearance o f warfarin. Since this initial report, interactions with over 40 drugs have been reported (reviewed in Somogyi and Gugler 1982, Feldman and Burton 1990), including benzodiazepines, theophylline, phenytoin, and propranolol. Cimetidine's ability to impair the clearance o f a wide variety o f drugs led to the initial assumption that it was a general inhibitor o f P450s (Puurunen et al. 1980, Pelkonen and Puurunen 1980). This was supported by early in vitro studies where it was demonstrated that cimetidine was able to reversibly bind to the ferric iron o f rat microsomal P450, resulting in a typical type II binding spectrum, and two different binding affinities were detected, indicating that at least two separate forms o f P450s were interacting with cimetidine (Rendic et al. 1979). Further in vivo and in vitro studies were conducted, which addressed specific P450 activities, and it was found that some enzymes are refractory to inhibition by cimetidine. In rats, in vivo administration o f cimetidine inhibited activities specific to C Y P s 2C11 and 2C6, but not C Y P s 2 A 1 , 2 B 1 / 2 , or 3 A 1 / 2 (Chang et al. 1992A, Levine et al. 1998). When the same enzyme activities were examined in vitro, cimetidine was a weak inhibitor o f C Y P s 2 B 1 / 2 and 3 A 1 / 2 in addition to C Y P 2 C 1 1 ; however, when rat hepatic microsomes were preincubated with cimetidine and N A D P H , the potency o f inhibition o f C Y P 2 C 1 1 approximated that seen in vivo (Chang et al. 1992B). A t high concentrations, cimetidine appears to act as a weak, general P450 inhibitor; following preincubation both the potency and specificity o f inhibition is increased, suggesting that cimetidine may be metabolized to an inhibitory species. Spectral evidence has shown that a metabolite-intermediate complex is formed when cimetidine is incubated with rat hepatic microsomes and N A D P H , and the formation o f this complex can be blocked by anti-rat C Y P 2 C 1 1 antibodies (Levine and Bellward 1995). Preincubation o f cimetidine with human liver microsomes and N A D P H also resulted in an inhibition o f dapsone metabolism, with increasing inhibition seen with longer 20 preincubation times (Tingle et al. 1991). Mechanism-based inactivation o f P450 enzymes by cimetidine is further supported by a delay in the onset o f the cimetidine-drug interaction. For example, with propranolol, clearance was reduced by 33% after one day o f treatment with cimetidine; following five days, however, a 51% decrease was observed (Reiman et al. 1981). 1.5.2 Inhibition of CYP2D6 by cimetidine Based on clinical and in vitro data, C Y P 2 D 6 appears to be one o f the P450 enzymes inhibited by cimetidine in humans. The clearance o f several C Y P 2 D 6 substrates in vivo is decreased by approximately 30% to 50% by cimetidine, including propranolol (Reimann et al. 1981), imipramine (Miller and Mackl in 1983, Abernethy et al. 1984), desipramine (Amsterdam et al. 1984), and sparteine (Schellens et al. 1989). However, with other C Y P 2 D 6 substrates, such as nortriptyline, the effect o f cimetidine is inconsistent. While cimetidine was found to increase plasma concentrations o f nortriptyline in one reported case (Miller et al. 1983), a subsequent controlled clinical trial involving six subjects failed to observe the same effect (Henauer et al. 1984). Similar results were observed with desipramine, where the A U C of this drug was increased in some, but not all subjects (Amsterdam et al. 1984, Steiner and Spina 1987); however, this effect was explained by the individuals' C Y P 2 D 6 phenotype. In the E M s , the A U C o f desipramine was increased by an average o f 50%, whereas P M s , who do not express functional C Y P 2 D 6 , demonstrated no change in the A U C . C Y P 2 D 6 phenotype, therefore, appears to play a significant role in determining the effect o f cimetidine in vivo. Al though this effect is not large enough to effectively convert the E M s to a P M phenotype (Philip et al. 1989), the lack of effect in P M s provides strong evidence that cimetidine is an effective inhibitor o f C Y P 2 D 6 in vivo. In vitro inhibition studies have determined the K ; o f inhibition by cimetidine on several C Y P 2 D 6 substrates. Bufuralol hydroxylase activity was competitively inhibited with a K j o f 50 to 55 [0.M (Knodell et al. 1991, Furuta et al. 2001), while dextromethorphan O-demethylation was inhibited with an IC50 o f 200 [xM when a substrate concentration o f 25 u.M was used (Martinez et al. 1999). 21 Compared to maximum plasma concentrations o f approximately 10 u M , however, these values are too high to fully explain the clinical observations. The effect o f preincubation on inhibition o f C Y P 2 D 6 by cimetidine has not been investigated. 1.6 Hypothesis and objectives The in vivo and in vitro studies described above have shown that cimetidine is a P450 inhibitor. In vitro treatment with a high concentration o f cimetidine results in general inhibition o f P450 enzymes; however, cimetidine is a more selective and more potent inhibitor at the lower concentrations seen in vivo. The effect seen after in vivo administration o f cimetidine can be emulated in vitro when hepatic microsomes are preincubated with the inhibitor, as was demonstrated with inhibition o f rat C Y P 2 C 1 1 , which leads to the formation o f a metabolite-intermediate complex. In humans, cimetidine has been shown to decrease the clearance o f many therapeutically important drugs, some o f which are known substrates o f C Y P 2 D 6 . Inhibition o f C Y P 2 D 6 by cimetidine has also been observed in vitro, but the estimated K i values are approximately 50 u.M, which is above the maximum plasma concentration o f approximately 10 [xM attained during therapy. Clinical observation has also shown that drug interactions with C Y P 2 D 6 substrates are not consistent among individuals. The C Y P 2 D 6 enzyme displays a large degree o f genotypic and phenotypic variation, which impacts the metabolism o f C Y P 2 D 6 substrates, and which may explain these observations. Al though both in vitro and clinical data indicate that C Y P 2 D 6 is inhibited by cimetidine, the mechanism of this inhibition is unclear. Based on the effect o f cimetidine on rat C Y P 2 C 1 1 and the discrepancy between the in vitro inhibitory potency relative to plasma levels o f cimetidine, it is possible that mechanism-based inhibition may be occurring. The inhibitory effect o f cimetidine in vivo is also dependent on C Y P 2 D 6 phenotype. One allelic variant o f this enzyme, C Y P 2 D 6 * 1 0 , has been shown to be less susceptible to in vitro competitive inhibition by compounds such as sparteine, debrisoquine, and norfluoxetine, compared to C Y P 2 D 6 * 1 (Ramamoorthy et al. 2001). Therefore, it could be 22 hypothesized that the CYP2D6*10 variant is less susceptible to inhibition by cimetidine than CYP2D6*1. Were this found to be the case, this may provide an explanation for the inconsistency in clinically observed interactions. In order to begin testing this hypothesis, the first premise to be addressed is the mechanism by which cimetidine acts as an inhibitor o f C Y P 2 D 6 . The specific hypothesis tested in this thesis is: cimetidine is a mechanism-based inactivator of CYP2D6. The specific objectives o f this project were: 1. to develop and validate a quantitative assay for the measurement o f the metabolites o f dextromethorphan; 2. using human liver microsomes, to determine the effect o f quinidine and cimetidine on dextromethorphan O-demethylase activity, with and without preincubation; 3. using recombinant C Y P 2 D 6 , to determine the effect o f quinidine and cimetidine on dextromethorphan O-demethylase activity, with and without preincubation. I f cimetidine is acting as a mechanism-based inactivator o f C Y P 2 D 6 , preincubation o f the enzyme with cimetidine should lead to an increase in the potency o f the inhibitor, as seen by a decrease in the estimated K i value. Quinidine wi l l be utilized to show that an increase in inhibition with preincubation is not observed with reversible inhibitors. A s a competitive inhibitor o f C Y P 2 D 6 , quinidine's effect should be similar both with and without preincubation. Finally, two different in vitro model systems, human liver microsomes and recombinant C Y P 2 D 6 , wi l l be used. I f a metabolite o f cimetidine is involved in the inhibition o f C Y P 2 D 6 , this metabolite may be formed by a P450 enzyme other than C Y P 2 D 6 . A lack o f effect with preincubation with the individual recombinant enzyme could have various possible explanations: cimetidine may not be acting as a mechanism-based inactivator or, alternatively, another P450 enzyme could be involved in the production o f the reactive 23 species. Therefore, studies wi l l initially be done with microsomes, in which all hepatic P450 enzymes are present. 24 Chapter 2 M A T E R I A L S & M E T H O D S 2.1 C h e m i c a l s Dextromethorphan hydrobromide, cimetidine, quinidine, 1-aminobenzotriazole (ABT) , and triacetyloleandomycin ( T A O ) were purchased from Sigma Chemical Co . (St. Louis, M O ) . Dextrorphan ( D O R ) and 3-methoxymorphinan hydrochloride (3-MM) were obtained from Gentest Corp. (Woburn, M A ) , while 3-hydroxymorphinan (3-HM) was from R B I (Oakville, O N ) . A l so purchased from Sigma Chemical Co . were P-nicotinamide adenine dinucleotide phosphate, reduced form ( N A D P H ) , and reagent grade magnesium chloride. H P L C grade potassium phosphate monobasic and methanol, and perchloric acid were bought from Fisher Scientific (Toronto, O N ) . Reagent grade potassium phosphate monobasic and dibasic were from Fisher Scientific and Anachemica (Montreal, P Q ) , respectively. 2.2 E n z y m e incubat ions 2.2.11ncubation conditions for human liver microsomes Pooled human liver microsomes and microsomes from two individual donors with high C Y P 2 D 6 activity (catalog numbers H G 64 and H G 95) were purchased from Gentest Corp. (Woburn 25 M A ) . The protein and P450 concentration for each lot o f microsomes was provided by the manufacturer. Preliminary experiments were performed to ensure that the protein concentration and incubation time used were within the linear range o f metabolite formation, and that the concentration o f N A D P H was in excess (see section 3.2). The incubation mixtures consisted o f microsomal protein (0.5 m g / m L ) , dextromethorphan, 50 m M potassium phosphate (pH 7.4), and 1 m M magnesium chloride. Reaction tubes were placed in a 37 C shaking water bath for 2 minutes to allow for temperature equilibration prior to initiation o f the reaction with 1 m M N A D P H dissolved in 50 m M potassium phosphate (pH 7.4). For inhibition studies, cimetidine or quinidine was added to the incubation mixture before initiation o f the reaction with N A D P H . A l l reactions were carried out in a final volume o f 200 p L in glass tubes. Reactions were terminated after 20 minutes by precipitation o f the protein with 20 p L ice-cold perchloric acid and mixing with a vortex. Samples were then centrifuged at 3000 xg for 10 minutes in a Beckman model J -6B centrifuge. The supernatant was removed with a 1 m L syringe, passed through a 0.22 u m P V D F membrane syringe driven filtration unit (Millipore L td . , Bedford M A ) , and transferred to autosampler vials fitted with 150 p L conical inserts (both from V W R , Edmonton A B ) for H P L C analysis. Solutions o f dextromethorphan, cimetidine, and quinidine were made in 50 m M potassium phosphate (pH 7.4), with minimal amounts o f methanol used to aid in solubility. In all cases, the final volume o f methanol in the microsomal incubation was less than 0.1% of the total incubation volume, which is below the concentration found to inhibit C Y P activity (Busby et al. 1998). 2.2.2 Incubation conditions for recombinant CYP2D6 Baculovirus-insect-cell-expressed human C Y P 2 D 6 supersomes (recombinant C Y P 2 D 6 ) and control supersomes, which consisted o f microsomes produced from the expression system transfected with the wild-type vector, were purchased from Gentest Corp. The protein and P450 concentrations 26 were provided by the manufacturer. A s with the human liver microsomes, preliminary experiments were conducted to determine the optimal incubation conditions with respect to C Y P 2 D 6 and N A D P H concentration, and incubation time (see section 3.2). N o formation o f dextrorphan was observed when control microsomes were incubated with dextromethorphan and N A D P H . Incubation mixtures consisted o f 2.5 pmol P 4 5 0 / m L , dextromethorphan, 100 m M potassium phosphate (pH 7.4), and 3.3 m M magnesium chloride. For inhibition studies, cimetidine or quinidine was added to the incubation mixture. The incubation mixtures were preincubated for 2 minutes at 37 C in a shaking water bath before initiation o f the reaction with 1 m M N A D P H . A l l reactions were carried out in a final volume o f 200 u L in glass tubes. After 15 minutes, reactions were terminated with 20 uL ice-cold perchloric acid. The samples were then centrifuged and filtered, as described for the human liver microsome samples, prior to analysis by H P L C . 2.2.3 Preincubation experiments The following procedure was used for the majority o f the preincubation experiments; any variations in this method are noted in the figure legends where appropriate. For experiments with human liver microsomes, the preincubation mixture consisted o f microsomal protein (2 m g / m L ) , inhibitor, 50 m M potassium phosphate (pH 7.4), and 1 m M magnesium chloride. Where the recombinant enzyme was used, the preincubation mixture contained C Y P 2 D 6 (10 p m o l / m L ) , inhibitor, 100 m M potassium phosphate (pH 7.4), and 3.3 m M magnesium chloride. The preincubation mixture was warmed i n a 37 C shaking water bath for 2 minutes before the reaction was initiated by the addition o f 1 m M N A D P H . A t various time points following the start o f the reaction, 50 uL aliquots were removed from the preincubation mixture and added to reaction tubes containing dextromethorphan, 1 m M N A D P H , potassium phosphate, and magnesium chloride in a volume o f 150 U.L. The reactions were then incubated for a further 20 or 15 minutes, for samples containing human liver microsomes or recombinant C Y P 2 D 6 , respectively, and were terminated with 27 20 U.L perchloric acid. In the final incubation step, the concentration o f microsomal protein was 0.5 m g / m L , and the concentration o f recombinant C Y P 2 D 6 was 2.5 p m o l / m L . Control preincubation reactions, in which microsomal protein (2 m g / m L ) or recombinant C Y P 2 D 6 (10 p m o l / m L ) were preincubated with 1 m M N A D P H in the absence o f inhibitor, were assayed at the same time. The activity in the inhibitor-treated groups was compared to the control activity at each time point. 2.3 H P L C Assay M e t h o d o l o g y 23.1 Assay conditions The chromatographic system consisted o f a 1525 binary pump, 717plus autosampler, and 474 scanning fluorescence detector (all from Waters, Mil l i ford M A ) . Operation o f all equipment, as well as data acquisition and peak integration, was managed with Waters' Breeze software. Separation was achieved on a Waters SymmetryShield RPis column (4.6 m m x 150 m m , 5 p m particle size) equipped with a SymmetryShield RPi8 Sentry guard column (3.9 m m x 20 m m , 5 p m particle size). A mobile phase o f 60% 50 m M potassium phosphate (adjusted to p H 4.75 with 50 m M potassium phosphate dibasic) and 40% methanol was pumped through the column at 1 mL/minu te at ambient temperature. A l l mobile phase components were H P L C grade, including the water used to prepare the 50 m M potassium phosphate, which was produced on site by reverse osmosis and subsequent purification using a M i l l i - O filtration system (Millipore Ltd . , Bedford M A ) . The mobile phase was degassed by vacuum filtration through 0.45 u m filters (Millipore Ltd.) prior to use. A n injection volume o f 25 u L was used and the fluorescence excitation and emission wavelengths were set at 280 n m and 310 nm, respectively. 28 2.3.2 Calibration curves Stock solutions o f the metabolite standards (200 U.M) were prepared in distilled water and stored in 1 m L aliquots at -20 C in glass screw-top vials ( V W R Scientific, Edmonton A B ) . Working solutions o f 20, 2, and 0.2 U . M were prepared on the day o f each experiment by dilution o f freshly-thawed 200 uM stock solutions in potassium phosphate (50 or 100 m M , for microsomal or recombinant C Y P 2 D 6 , respectively). Standard (n=3) and quality control (QC) samples (n=5) were prepared using a method which paralleled the incubation conditions. Mixtures containing either microsomal protein (0.5 m g / m L ) and 50 m M potassium phosphate, or recombinant C Y P 2 D 6 (2.5 p m o l / m L ) and 100 m M potassium phosphate, were incubated for 2 minutes at 37 C in a shaking water bath. The protein was subsequently precipitated by the addition o f 20 uL perchloric acid, and the samples were vortex mixed and placed on ice for 10 minutes. Appropriate volumes o f the working solutions were then added to give metabolite concentrations o f 0.1, 0.25. 0.5, 0.75, 1, 2.5, and 5 uM, in a volume o f 200 U .L . The samples were then centrifuged at 3000 x g for 10 minutes, and the supernatant was filtered through a M i l l e x - G V syringe driven filter unit with a 0.22 p m P V D F membrane (Millipore Ltd.) prior to analysis by H P L C . Binding o f the analytes to the filter membrane was less than 5%. Calibration curves for each o f the three metabolites were constructed by plotting the concentration o f the standard against the integrated peak area. The equations o f the curves were determined by unweighted linear regression using Microsoft Exce l (1995). 2.3.3 Assay Validation The inter- and intra-assay accuracy and precision were evaluated by determining the percent bias (% bias) and percent coefficient o f variation (% C V ) , respectively, o f the high, mid, and low Q C samples. Inter-assay variability was assessed by analyzing a set o f standards (n=3) and Q C samples (n=5) on five separate days, while the intra-assay variability was assessed by analyzing a set o f standards 29 (n=3) and Q C samples (n=5) in a single run. Q C samples were quantitated by back calculation o f the concentration using the equation o f the calibration curve. The % bias and % C V were calculated using the following equations: % bias = mean concentration o f 5 O C samples - nominal concentration x 100% nominal concentration % C V = standard deviation o f 5 O C samples x 100% mean concentration 2.4 D a t a Ana lys i s A l l kinetic modeling was done with the SigmaPlot enzyme kinetics module, version 1.1 (SPSS Inc., Chicago I L , 2001). The K m and V m a x values were determined using the Michaelis-Menten equation, where V represents the measured enzyme velocity at substrate concentration S: V = Vmax • [S] K m + [S] K ; values were estimated using the following equations: for competitive inhibition: v = Vmax (I + ( K m / S ) ( I + I / K ) ) for noncompetitive inhibition: v = V m a x ((I + I /K;)(I + K m / S ) ) for mixed inhibition: v = Vmax ( (K m /S ) ( I + I /Kj) I + I / ( a / K m ) where S and I represent the concentration o f substrate and inhibitor, respectively, and a is the factor change to the K m when the inhibitor is bound to the enzyme-substrate complex. The mode o f inhibition which best described the data was determined by comparing R 2 values for the fit o f the various models. 30 Measurements described as "duplicate" or "triplicate" refer to samples analyzed within the same experiment; separate experiments are described as such. A l l data are expressed as means + standard deviation. Two-way analysis o f variance ( A N O V A ) was used to determine differences between the inhibitor-treated and control groups in preincubation studies. Significant interactions between preincubation time and presence o f inhibitor were used to demonstrate that the loss in enzyme activity following preincubation was greater than would be expected due to either o f these factors alone. T o compare K i values obtained by several different methods, a one-way A N O V A was used. Statistical analyses were performed using SigmaStat for Windows, version 1.0 (Jandel Inc., San Rafael C A , 1994). The level o f significance was set, a priori, at p<0.05. 31 C h a p t e r 3 R E S U L T S 3 . 1 Assay V a l i d a t i o n H P L C is a commonly used technique for the detection o f dextromethorphan and its metabolites. The H P L C assay used in this thesis was adapted from previously published methods. The chromatographic conditions o f von Moltke et al. (1998) were used, with the organic component o f the mobile phase changed from acetonitrile to methanol to achieve better separation o f the analytes. The use o f fluorescence detection was established by L a m and Rodriguez (1993). Before the assay was used quantitatively, it was validated with respect to its selectivity, linearity o f response, limits o f quantitation, accuracy, and precision. The selectivity o f an assay defines whether the response from a single analyte can be distinguished from that o f other analytes and the biological matrix (Karnes et al.. 1991). Peaks corresponding to pure standards o f dextrorphan, 3-hydroxymorphinan, dextromethorphan, and 3-methoxymorphinan were eluted with retention times o f 4, 5, 12, and 16 minutes, respectively (figure 4, chromatogram 1). Blank samples containing either microsomal protein (0.5 m g / m L ) or recombinant C Y P 2 D 6 (2.5 pmol P450 /mL) produced no response within the elution windows o f any o f the four analytes (figure 4, chromatograms 2 and 3). Incubation o f either human liver microsomes or recombinant C Y P 2 D 6 with dextromethorphan resulted in the generation o f peaks corresponding to 32 dextrorphan and 3-methoxymorphinan, which had similar retention characteristics to the pure metabolite standards (figures 5A and 5B). W i t h recombinant C Y P 2 D 6 , 3-methoxymorphinan was only detected in incubations with high concentrations o f dextromethorphan or, as in figure 5B, enzyme. W i t h the range o f substrate concentrations and incubation conditions used experimentally, 3-hydroxymorphinan was not produced. In order to use an assay quantitatively, a predictable, linear response must be established between the amount o f analyte present in the samples and the instrument response (Causon 1997). The calibration curves for each metabolite were constructed by plotting analyte concentration against the integrated peak area (figure 6), and the equation o f the line o f best fit was determined using unweighted linear regression. The response was found to be linear over the range o f 0.1 to 5 U.M, with R 2 values greater than 0.99 for all three metabolites. 33 2.00 4.00 6.00 8.00 Minutes 10.00 12.00 14.00 16.00 Figure 4: Chromatographic separation of pure standards of dextromethorphan and its metabolites. Chromatogram 1: resolution of a mixture of the three metabolite standards and dextromethorphan. The peaks correspond to dextrorphan (DOR), 3-hydroxymorphinan (3.-HM), dextromethorphan (DEX) and 3-methoxymorphinan (3-MM). Chromatograms 2 and 3 represent blank microsomal and recombinant CYP2D6 matrix, respectively. Chromatographic conditions were the same as those described in the Materials and Methods section 2.3.1. 34 i.OO-Minutes 12.00 14.00 16.00 Figure 5A: Representative chromatogram of dextrorphan and 3-methoxymorphinan produced in a microsomal incubation. Pooled human liver microsomes (0.5 mg/mL) were incubated with 7.5 uM dextromethorphan and 1 mM N A D P H for 20 minutes. 35 8.00 Minutes 10.00 12.00 14.00 16.00 Figure 5B: Representative chromatogram of dextrorphan and 3-methoxymorphinan produced in an incubation with recombinant CYP2D6. Recombinant CYP2D6 (1.5 pmol/mL) was incubated with 10 uM dextromethorphan and 1 mM N A D P H for 15 minutes. 36 A 1 0 0 0 0 0 0 7 5 0 0 0 0 5 0 0 0 0 0 2 5 0 0 0 0 0 + 8 ^ 1 , 1 1 r 0 1 2 3 4 5 C o n c e n t a t i o n ((iM) B 1 000000 0 1 2 3 4 5 C o n c e n t r a t i o n (wM) Figure 6: Representative calibration curves for (A) dextrorphan (DOR), (B) 3-hydroxymorphinan (3-HM), and (C) 3-methoxymorphinan (3-MM). Each point is a mean of three measurements. The line of best fit was determined by unweighted linear regression. 37 The lowest concentration standard included in the calibration curves corresponded to the limit o f quantitation ( L O Q ) o f the assay. The L O Q is the lowest concentration o f analyte than can be measured with acceptable accuracy and precision. Conventionally, acceptable accuracy is defined as a % bias within 20% of the nominal concentration, while for acceptable precision, the coefficient o f variation should not be greater than 20% (Shah et al. 1992). Fo r all three metabolites, the lowest concentration which fulfilled these requirements was 0.1 U.M. The accuracy and precision o f the assay were determined by measuring the %bias and % C V , respectively, o f the Q C samples. Q C samples corresponding to the high, middle, and low range o f the calibration curve were prepared at 4 u.M, 2.5 u.M, and 0.1 u M , respectively, and included either inactivated microsomal protein or recombinant C Y P 2 D 6 . The concentrations o f the Q C samples were determined using calibration curves analyzed at the same time, which also included the appropriate matrix. The variability o f the assay is considered acceptable i f the % C V is less than 15% and the % bias is within 15% o f the nominal concentration; however, because the low Q C sample represented the L O Q , a % C V and % bias o f less than 20% was considered acceptable (Shah et al. 1992). Bo th the intra- and inter-assay variability were acceptable when either microsomal protein (tables 1 and 2) or recombinant C Y P 2 D 6 (tables 3 and 4) was used as the matrix. 38 6 o o o J3 I TJ OH "a. a CO u a c o u 3 >> o <s 3 u u « •a s fi o cu too a ,G -a » s S ^ s s 3 ^ H .S .S -a "d vj _Q OH U CL) W VJ CL) CL. T3 CL) •£ CL) O & -a <_) IT T t T3 d J3 > •-3 o > •a 3 u C O •a VJ JD =1 be G _o ca 3 -a o Xi OH s u cy CL) •a c 3 cr OL, e in u a X H CL) a -a CL) T3 G cj VJ CU cj s •p CL) s -a oi G O •a o CL) G « -a c a , •a ~<3 « -S3 K ^ E a> cn CO un CO Ol CO CO Tt Tt CO CO o o o o o CO o o o o LO CD CD LO LO c\i oi oi oi oi o o o o o o o o o o O CO 00 o h-- CD Tt CO Tt Tt Tt Tt 0 T - T - CO CO CD CD CD LO LO 01 oi oi oi oi CJ) O) O) o o O O O T- T-d d d c i d T- N (O f U) Ol Tt o CO o o Oj CO O o ui CO co o O Ol LO LO to o LO Ol O CD LO CO oi oi oi oi oi oi oi o Tt Tt CO IT) oi o o CO o Tt LO c re <D E e "o c o u 15 c I o c o o o o o o re > a> T3 re TJ c re o oo CO LO Tt Tt CO o LO o CO 00 o T T— T— o a CO CO CO Tt Tt Tt Tt Tt Tt o CO o LO Ol 00 If) o Ol oi o oi oi 00 o CD CO Tt o T- 00 If) Tt o CO T-" o CO CN O in o O If) oi CD CO o co CO "? §1 o o C3 o -a G 3 o o o o o o o G 3 O vj of o o o o o o o "0 CL) -a G 3 o vj m" o o © CL) ^3 39 -a a o co O u u e a J3 •a u u OH 1 cn u a c o o >-> u CS u 3 u o TI C rt a o u a U 3 "§0 -a a 6 oj -a a in -a cj a, I .3 & 3 " „, a. e u O 3 u C O •J3 ^ >, 3 G a « G u I) O CJ a o G O •a OJ J 3 oj > •a -o c of > u aj C/D 4_l J_l -J a o 13 a, a, -a 0J OJ -a OJ u OJ S5 -a CO 0) (1) Q£ £ O O U) n CD > < a a> a> m CO 0 0 T— CN in 0 CM CM 0 CM T t T t T t T t T t •«t T t O CO CD CD (—> O in CO s CD in LO CD CO in 0 O) CM c\i CM c\i oi oi 0 0 CN CN T— a • 10 ^ — t- T— ^ — 0 0 0 0 O 0 O 0 0 0 CO 0 5 T t c o i n T t O 1 - T t O CM T t T t T t : T t CD O CO ID CT> in CD in in in CM CM CM CN (N 00 T t O) LO CO o T - m o CM T t T t T t T t T t CD ai in CD o in in in in CD CN CM CM CM CM O T— O CM CJ> T - T - T - T - O 0 0 0 0 0 CM CO •>* IO o CM CM CM CM m CM c CQ CU E 3 c "o c o u ro c E o c o 0 c o cs '> a> •a XJ k_ cs XI c cs 00 CO CO CO 0 0 00 0 O O T t 0 T t in 0 CM CM m O 00 o i d 0 o i 0 T— T— T— CD 0 * — O O cn CM 00 T— d d d d d C) O d 06 If) O CM CD If) O CM CM T t T t O LO in 0 CM 00 T t m O t-- 00 oi O d oi 0 a T - CO 01 T t §3 40 Table 3: Intra-assay precision and accuracy of quality control (QC) samples prepared with inactivated recombinant human CYP2D6. Five Q C samples at each of the low (0.1 (xM), mid (2.5 and high (4 fiM) concentrations were analyzed in a single run. The Q C samples were quantitated using calibration curves for the respective metabolites, which were prepared in triplicate and analyzed at the same time. Precision (%CV) and accuracy (%bias) were assessed using the equations described in the Materials and Methods section 2.3.3. DOR (uM) Replicate QC low QC mid QC high 1 0 . 1 0 2 . 6 3 4 . 1 9 2 0 . 1 0 2 . 5 2 4 . 0 3 3 0 . 1 0 2 . 5 5 4 . 0 4 4 0 . 0 9 2 . 5 4 3 . 9 9 5 0 . 0 9 2 . 5 5 4 . 0 1 mean 0.09 2.56 4.05 nominal conc'n (uM) 0.10 2.50 4.00 standard deviation 0 . 0 0 A 0 . 0 4 0 . 0 8 %CV 4.39 1.67 2.00 %bias -4.13 2.33 1.28 a actual value 0.004, rounded to 0.00 41 Table 4: Inter-assay precision and accuracy of quality control (QC) samples prepared wi th inactivated recombinant human C Y P 2 D 6 . Five replicates of Q C samples at the low (0.1 U.M), mid (2.5 U.M), and high (4 p.M) concentrations were prepared and analyzed along with a calibration curve (in triplicate) each day for five days. Inter-assay precision and accuracy were determined by calculating the % C V and %bias, respectively, of the mean concentration measured on each of the five days. Average DOR (uM) Day QC low QC mid QC high 1 0.10 2.56 4.05 2 0.12 2.56 . 4.01 3 0.09 2.56 4.06 4 0.09 2.57 4.09 5 0.10 2.50 3.94 mean 0.10 2.55 4.03 nominal conc'n (uM) 0.10 2.50 4.00 standard deviation 0.01 0.03 0.06 %CV 12.64 1.19 1.42 %bias -2.26 1.96 0.75 42 3.2 Optimization of incubation conditions Incubation conditions for human liver microsomes were optimized with respect to protein concentration, incubation time, and N A D P H concentration; while the conditions used for the recombinant enzyme were optimized for P450 concentration, incubation time, and N A D P H concentration. Protein or P450 concentrations and incubation times were selected to be within the linear range o f product formation and the N A P D H concentration was chosen so as not to be a limiting factor in the reactions. The formation o f dextrorphan and 3-methoxymorphinan were linear with microsomal protein concentrations up to 1.0 m g / m L (figure 7). A protein concentration o f 0.5 m g / m L was selected, as it was the mid-point o f the linear range for dextrorphan formation, which is the metabolite o f interest i n this thesis. Incubation times o f up to 30 minutes produced a linear response for dextrorphan formation, while the formation o f 3-methoxymorphinan was linear for only 25 minutes (figure 8). The secondary metabolite, 3-hydroxymorphinan, was not detected with incubation times less than 25 minutes. In order to maximize the formation o f dextrorphan, while minimizing the formation o f the secondary metabolite, an incubation time o f 20 minutes was used. There was no difference in the amount o f dextrorphan or 3-methoxymorphinan formed with N A D P H concentrations from 0.1 to 2 m M (figure 9). A final N A D P H concentration o f 1 m M was used to ensure that N A D P H was i n excess. W i t h the recombinant enzyme, the formation o f dextrorphan was linear with C Y P 2 D 6 concentrations o f 0.25 to 1 p m o l / m L (figure 10) and incubation times up to 20 minutes (figure 11). There was no difference in the amount o f dextrorphan formed with N A D P H concentrations up to 2 m M (figure 12). A final P450 concentration o f 0.5 p m o l / m L , an incubation time o f 15 minutes, and an N A D P H concentration o f 1 m M were used in all subsequent incubations. Under these conditions, the formation o f 3-hydroxymorphinan was not detectable. 43 o E c tt O Q 0.0 0.2 0.4 0.6 0.8 Microsomal protein concentration (mg/mL) 0.4 H ^ 0.3 0.0 0.2 0.4 0.6 0.8 Microsomal protein concentration (mg/mL) Figure 7: Optimization of microsomal protein concentration. Pooled human liver microsomes (0.15 to 1 mg protein/mL) were incubated with 10 u.M dextromethorphan and 1 m M N A D P H for 20 minutes. The amount of dextrorphan (DOR) formed is shown in the top panel, while 3-methoxymorphinan (3-MM) is shown in the bottom panel. Each point represents the mean of two measurements. 44 40 60 80 Incubation time (minutes) 0.00 20 40 60 80 Incubation time (minutes) 100 120 Figure 8: Opt imizat ion of incubation time wi th pooled human liver microsomes. Microsomal protein (0.5 mg/mL) was incubated with dextromethorphan (10 \LM) and N A D P H (1 mM) for 2.5 to 120 minutes. The top panel shows the formation o f dextrorphan (DOR) with time, while the bottom shows the formation of 3-methoxymorphinan (3-MM). Each point represents the mean of duplicate measurements. 45 0.0 "I • • • • r-0 1 2 3 4 5 NADPH concentration (mM) ~ 0.4 -o E co 0.0 -I . • • • r-0 1 2 3 4 5 NADPH concentration (mM) Figure 9: Optimization of N A D P H concentration with human liver microsomes. Pooled human liver microsomes (0.5 mg prote in /mL) were incubated with 10 u M dextromethorphan and various concentrations of N A D P H (0.1 to 5 mM) for 20 minutes. The amount of dextrorphan (DOR, top panel) and 3-methoxymorphinan (3-MM, bottom panel) formed was measured in duplicate at each point. 46 o E c o Q 1.00 0.75 0.50 0.25 0.00 • —• / / / / 2 3 4 C Y P 2 D 6 c o n c e n t r a t i o n (pmol/mL) Figure 10: Optimization of recombinant CYP2D6 concentration. Recombinant C Y P 2 D 6 (0.25 to 5 pmol /mL) was incubated with 5 \iM dextromethorphan and 1 m M N A D P H for 15 minutes and the amount of dextrorphan (DOR) formed was measured. Each point represents the mean of two measurements. 0.0 -I 1 • • ' ' >— 0 20 40 60 80 100 120 Incubat ion t ime (minutes) Figure 11: Optimization of incubation time with recombinant CYP2D6. Recombinant C Y P 2 D 6 (0.5 pmol /mL) was incubated with 5 \iM dextromethorphan and 1 m M N A D P H for 5 to 120 minutes and the amount of dextrorphan (DOR) formed was measured. Each point represents the mean of two measurements 47 0.4 0.0 4 • . • • 1 0.0 0.5 1.0 1.5 2.0 NADPH concentration (mM) Figure 12: Optimization of N A D P H concentration with recombinant CYP2D6. Recombinant C Y P 2 D 6 (0.5 pmol /mL) was incubated with 5 u M dextromethorphan and N A D P H concentrations ranging from 0.1 to 2 m M for 15 minutes and the amount of dextrorphan (DOR) formed was measured. Each point is the mean of two measurements. 48 3.3 Studies w i t h h u m a n liver mic rosomes The effect o f cimetidine on C Y P 2 D 6 activity was first determined using human liver microsomes. This in vitro model is more representative o f the metabolic conditions seen in the liver in vivo, as the cytochrome P450 enzymes are present in their normal lipid environment and the accessory proteins, such as N A D P H - P 4 5 0 reductase, are present in normal proportions. A l so , a full complement o f hepatic P450 enzymes is present. While this leads to a more complex system, where several enzymes may be contributing to the probe activity, use o f appropriate substrate concentrations can increase the specificity o f the reaction o f interest. When investigating the potential formation o f a catalysis-dependent inhibitory species, the presence o f all P450 enzymes is beneficial because the inhibitory species could possibly be formed by an enzyme other than the one which is targeted. Therefore, by using a model in which all enzymes are present, the likelihood o f developing the conditions under which a catalysis-dependent increase in inhibitory potency can be observed is improved. 3.3.1 Kinetic constants of dextromethorphan O-demethylation The formation o f dextrorphan, the O-demethylated metabolite o f dextromethorphan, was used as a marker o f C Y P 2 D 6 activity. The apparent K m and Ym3X o f dextrorphan formation were estimated with pooled human liver microsomes by fitting data to the Michaelis-Menten equation (Figure 13). The apparent K m was determined to be 4.1 ± 0.1 (xM, while the Vmax was 0.82 ± 0.06 n m o l / m i n / n m o l P450 in three trials (Table 5). The K m value determined from this experiment was used to select appropriate substrate concentrations in subsequent inhibition studies. The Eadie-Hofstee plot was linear over the range o f substrate concentrations used to determine the kinetic constants (figure 13B), suggesting that the observed activity is predominandy representative o f a single enzyme. 49 0.0 -I 1 1 1 1 1 1 1 0 10 20 30 40 50 60 70 [dextromethorphan] (uM) B 0.00 0.05 0.10 0.15 0.20 0.25 v 0 (nmol/min/nmol P450 ) / [dextromethorphan] (uM) Figure 13: Rate of formation of dextrorphan by human liver microsomes. Pooled human liver microsomes (0.5 mg protein/mL) were incubated with dextromethorphan (0.5 to 60 uM) and N A D P H (1 mM) for 20 minutes. Panel A : data were fit to the Michaelis-Menten equation using non-linear regression. Panel B: Eadie-Hofstee transformation of the data presented in panel A . Each point is the mean of duplicate measurements. 50 Table 5: Summary of the kinetic constants for dextromethorphan O-demethylation by pooled human liver microsomes determined in three separate experiments. Tr ia l (nmol/min/nmol P450) (uM) 1 0.86 4.0 0.22 2 0.84 4.2 0.20 3 0.75 4.2 0.18 m e a n 0.82 4.1 0.20 s t a n d a r d dev ia t ion 0.06 0.1 0.02 51 33.2 Hffect of quinidine and cimetidine on dextromethorphan O-demethylase activity Quinidine is a known competitive inhibitor o f C Y P 2 D 6 (Kobayashi et al. 1989). Pooled human liver microsomes were incubated with various concentrations o f inhibitor, at three fixed concentrations o f dextromethorphan (approximately 0.5-, 1-, and 2-times the K m ) , and the K i value was estimated to be 52.6 ± 28.2 n M (figure 14, panel A ) . A competitive inhibition model provided the best fit for the data. A representative Lineweaver-Burk plot is shown in figure 14, panel B . T w o forms o f cimetidine, the hydrochloride salt and the free base, were available for use in the inhibition studies. Cimetidine hydrochloride was obtained from the manufacturer, SmithKline & French Canada (Mississauga O N ) , in 1988 and was used in previous studies in this laboratory which demonstrated that cimetidine acts as an in vitro mechanism-based inactivator o f rat C Y P 2 C 1 1 (Chang et al. 1992). However, because o f the age o f the product, there was a possibility that it may have undergone significant chemical degradation. Cimetidine has been shown to be stable when stored at room temperature in a dry environment for up to five years (Bavin et al. 1984), but there is no data to support its stability beyond this period o f time. The free base, on the other hand, was readily available from a commercial source. Dose response curves for the two forms o f cimetidine are shown in figure 15. The curve for the hydrochloride salt is shifted to the right compared to the curve for the free base, but the IC50 values are similar (approximately 100 \iM for the free base, 150 u.M for cimetidine hydrochloride, as determined by visual inspection o f the curves). Because there was litde difference between the two forms, the free base, which was more easily obtained, was used in all subsequent experiments. The K i for inhibition o f dextromethorphan O-demethylase activity by cimetidine was determined using a similar method to that used with quinidine, and was found to be 38.0 + 5.3 u.M. A s with quinidine, a competitive inhibition model provided the best fit for the data (figure 16). The individual K ; values for quinidine and cimetidine, determined in three separate trials, are summarized in table 6. 52 T I I T -0.1 0.0 0.1 0.2 [quindine] (uM) Figure 14: Inhibition of dextromethorphan O-demethylase activity in human liver microsomes by quinidine. A: Dixon plot. Pooled human liver microsomes (0.5 mg protein/mL) were incubated with dextromethorphan (dex, 2.5, 4.5, or 7.5 jxM), quinidine (0.025 to 0.2 jxM), and N A D P H (1 mM) for 20 minutes. Data were fit to a competitive inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of duplicate measurements. B: Lineweaver-Burk transformation of the data in panel A . 53 100 0 q , 1 10 100 1000 [cimetidine] (uM) Figure 15: Effect of cimetidine free base and hydrochloride salt on dextromethorphan O-demethylase activity in human liver microsomes. Pooled human liver microsomes (0.5 mg protein/mL) were incubated with dextromethorphan (10 |xM), cimetidine free base or hydrochloride salt (10 to 1000 \iM), and N A D P H (1 mM) for 20 minutes. Percent control activity was determined by comparing the activity in the samples treated with inhibitor to that of a group incubated with dextromethorphan (10 jxM) and N A D P H (1 mM) alone. The activity of the control groups were 0.458 and 0.493 nmol /min /nmol P450, for cimetidine free base and hydrochloride salt, respectively. Each point represents the mean of two experiments in which duplicate measurements were made. 54 -100 0 100 200 300 [cimetidine] (uM) -0.4 -0.2 0.0 0.2 0.4 0.6 1/[dextromethorphan] (pM) Figure 16: Inhibition of dextromethorphan O-demethylase activity in human liver microsomes by cimetidine. A : Dixon plot. Pooled human liver microsomes (0.5 mg protein/mL) were incubated with dextromethorphan (dex, 2.5, 4.5, or 7.5 uM), cimetidine (25 to 250 uM), and N A D P H (1 mM) for 20 minutes. Data were fit to a competitive inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of duplicate measurements. B: Lineweaver-Burk transformation of the data presented in panel A . 55 Table 6: Summary of estimated K i values for inhibition of dextromethorphan O -demethylation by quinidine and cimetidine with pooled human liver microsomes, determined in three separate experiments. The K i values for quinidine and cimetidine were estimated using a competitive inhibition model. a p p a r e n t K, Tr ia l q u i n i d i n e c imet id ine (nM) (uM) 1 77.6 32.2 2 22.0 42.6 3 58.2 39.2 m e a n 52.6 38.0 s t a n d a r d dev ia t ion 28.2 5.3 56 3.3.3 Preincubation studies with cimetidine In order to determine whether cimetidine was capable of acting as a mechanism-based inactivator of C Y P 2 D 6 , human liver microsomes were preincubated with cimetidine and N A P D H for various periods of time before measuring the dextromethorphan O-demethylase activity. The method used for the preincubation studies involved two steps: a preincubation reaction with the inhibitor and incubation reaction with the substrate. The microsomal protein concentration in the preincubation reaction was 2 m g / m L , four times higher than the optimized concentration for incubations with dextromethorphan. A t appropriate time points, an aliquot of the preincubation mixture was added to a reaction tube containing substrate and N A P D H with a four-fold dilution, resulting in a microsomal protein concentration of 0.5 m g / m L in the final incubation. This dilution step also had the effect of reducing the concentration of cimetidine in the incubation to one-fourth of what was used in the preincubation reaction. Because cimetidine was demonstrated to act as a competitive inhibitor (see section 3.3.2), any cimetidine carried over from the preincubation reaction to the incubation reaction could potentially cause reversible inhibition of C Y P 2 D 6 activity. However, the observation of mechanism-based inactivation, not competitive inhibition, was the goal of the following experiments. Using this method for the preincubation experiments miriimized the competitive inhibition of dextromethorphan O-demethylase activity by cimetidine during the incubation reaction. Initially, pooled human liver microsomes were preincubated with 25 u M cimetidine for up to 30 minutes. Without preincubation, there was a 25% inhibition of activity, which was likely due to competitive inhibition by the residual cimetidine in the incubation reaction. With 5 to 30 minutes of preincubation, there was no trend towards a further decrease in dextromethorphan O-demethylase activity (table 7) 57 Table 7: Preincubation of human liver microsomes with cimetidine (25 |xM) for up to 30 minutes. Pooled human liver microsomes (2 mg protein/mL) were preincubated with cimetidine (25 yM) and N A P D H (1 mM), or with N A P D H alone (control). A t 10, 20, and 30 minutes after initiation of the preincubation reaction, a 50 p.L aliquot was removed from the preincubation mixture and added to a new reaction tube containing dextromethorphan and N A D P H . Samples were incubated for a further 20 minutes. In the final incubation step, the concentration of dextromethorphan was 5 piM and the concentration of microsomal protein was 0.5 mg /mL. Data shown are the mean of duplicate measurements. Activity in the cimetidine-treated groups was normalized to the activity of the control group preincubated for the same period of time to give the % control activity. preincubation time dextromethorphan O-demethylation (nmol/min/nmol P450) (minutes) control cimetidine % control activity 0 0.42 0.32 75% 10 0.36 0.30 83% 20 0.35 0.30 86% 30 0.37 0.30 81% 58 T o investigate whether a trend toward increasing inhibition could be observed with longer preincubation times, the preincubation period was increased to 120 minutes. Over this longer period o f time, the preincubation mixture was supplemented with N A D P H every 30 minutes to ensure that the cofactor did not become a limiting factor in the preincubation reaction. Under these conditions, however, activity o f the groups preincubated with cimetidine and N A D P H , or N A D P H alone, decreased by 72.6% and 84.7%, respectively, after a period o f 120 minutes (figure 17). In contrast, activity in the group to which neither N A D P H nor cimetidine were added dropped by only 17.7% over the same period o f time. Because the drop in activity appeared to be related to the addition o f excess N A D P H , the conditions under which sufficient N A D P H was added so as not to become the limiting reagent, but also to not cause excessive loss o f microsomal activity, were investigated (table 8). Microsomes were incubated with either 25 \iM or 100 \JM cimetidine for 120 minutes. N A D P H was added to initiate the preincubation reaction, and was then added again after 60 minutes ( + N A D P H ) , or was only added at the beginning o f the preincubation period ( - N A D P H ) . In both cases, the loss o f control activity was approximately 30% after 120 minutes o f preincubation. The absolute activity was slightly, but significandy, higher in the groups to which N A D P H was added only once; however, the % control activities o f the cimetidine-treated groups were similar after 120 minutes o f preincubation whether extra N A D P H was added at 60 minutes or not. These results suggest that adding N A D P H to the preincubation mixture only once does not limit the reaction o f cimetidine with C Y P 2 D 6 and maintains a high level o f enzyme activity. 59 Figure 17: Effect of addition of N A D P H to the preincubation mixture every 30 minutes. Pooled human liver microsomes (2 mg protein/mL) were preincubated with N A D P H (1 mM), and 25 | i .M cimetidine (+cimetidine+NADPH) or an equal volume of buffer ( + N A D P H ) . Every 30 minutes, a 50 [iL aliquot was removed from the preincubation mixture and added to a reaction tube containing dextromethorphan and N A D P H in a volume of 150 \iL. In the final incubation, the concentration of microsomal protein was 0.5 m g / m L and dextromethorphan was 5 [J .M. Every 30 minutes, immediately following the removal of the 50 (xL aliquots, an amount of N A D P H equivalent to a final concentration of 0.5 m M was added to the preincubation mixture. A control preincubation mixture, which consisted of pooled human liver microsomes (2 mg/mL) and incubation buffer alone, and to which N A D P H was added only at the beginning of the preincubation period, was run in parallel. Data shown are the mean of four measurements ± standard deviation. # activity is significantly (p<0.05) lower than that of control group preincubated for die same period of time and groups preincubated with N A D P H alone for 0, 30, 60, and 90 minutes. * activity is significantly (p<0.05) lower than that of group preincubated with N A D P H alone for the same period of rime and groups preincubated with cimetidine and N A D P H for 0, 30, 60, and 90 minutes. 60 6D g '8 g 'So CD 43 O x . u y c o o o 3 a c o • iH J-» rt ja 3 o a u XI OJO g TJ W do CJ H o o c •a <u -a a -a o 4 3 3 e a s 3 X 0 O P H e o 4 3 3 O hfi a '3 a •5b 4 3 -a CL) -a -a P H P < z o IS cu 4-1 C « 43 3 o _g '33 O H <U 43 3 4 2 <+H O u B o a o a o •a OS -a a a 'a a 'So CL) 4 3 o 43 Q 3. a o •d c« 4 3 3 U .g u 4 3 43 a .s 3 cr cr •j3 'P -5 T3 0 O J 3 0 VJ ba in -a H U -a •j3 cr V c 'N •x 4 3 TJ -VJ LO jy cr '£ •a u 43 a, vj 0 4 3 a) 6 o •J3 vf T5 c o O J o © V .OH a e •J3 3 TJ CL) 43 4 3 3 U g '33 o •s 8 S t i " 33 hp O J 3 0 Go X P H Q <: z + 0-Q < CD 2 2 E o CD II LO £ CN E o I a. Q < CD 2 .E o a5 ? E o LO J . CN E O CU ro = 42 c I 1 00 CO o o o +1 CO 1^ -00 CO CN -— O CN 00 CN CD co -C O o o 00 CM CO • r t o <=> :° +1 ^ LO 00 00 CN CO o ° ^ CN CD CD CO CN CD ( O o o © +1 m CM co o o CD 00 +1 T t T— O T t CN CD * T— o o CD LO +1 CO LO o <tf CN —' CD CM O O ©' +1 o CO CM o „ o 5-o' °: +1 ^ LO CD T - CD CN ' CO O O O +1 h~ CO CN o o o +1 f-CM o CM 61 After optimization o f the conditions for the longer preincubation period, the experiment was repeated using pooled human liver microsomes. Microsomes were also obtained from two individuals who had relatively high C Y P 2 D 6 activity as assessed by bufuralol 1-hydroxylase activity (activity in the pooled microsomes was 120 pmol product /minute/mg protein, while for individual H G 95 and H G 64, activity was 160 and 220 pmol product /minute/mg protein, respectively), as measured by the supplier. T w o concentrations o f cimetidine, 25 |xM and 100 |xM, were used in each experiment. The 25 ( i M concentration represents an amount o f inhibitor that should produce little or no competitive inhibition in the incubation step. After the dilution step, the concentration o f cimetidine in the final incubation was 6.25 |xM, much lower than the K i for competitive inhibition (which was found to be 38.0 (xM, as discussed in section 3.3.2). A n y inactivation o f C Y P 2 D 6 should be more apparent under these conditions because the loss in activity due to competitive inhibition alone would potentially be small compared to the effect o f enzyme inactivation. The second concentration (100 jxM) was selected in order to increase the effect o f the inactivation reaction, i f such an effect were occurring. Because the concentration o f cimetidine in the incubation step approached the K i in this case, carrying out the preincubation reaction with more than 100 [xM cimetidine was impractical. The results from the pooled human liver microsomes are shown in figure 18. Preincubation for 120 minutes with N A P D H alone resulted in a decrease in activity o f approximately 30%. W i t h 25 \xM cimetidine, a 10% decrease in activity was seen without any preincubation, and with preincubation times o f 30 to 120 minutes no difference in activity was seen in the cimetidine-treated groups compared to the appropriate control. W i t h 100 \iM cimetidine, an inhibition o f 26.4% was seen without preincubation, which represents the competitive inhibition caused by the presence o f 25 \iM cimetidine in the final incubation step. When preincubated from 30 to 120 minutes, the activity in the 100 [xM cimetidine-treated group remained lower than that o f the control group at each time point, but there was no trend towards increasing loss o f activity with time, as shown in the lack o f change in % control activity. 62 0.6 0.2 A 0.0 — o — control • • • • v • • • 25 uM cimetidine — - a — 100 uM cimetidine * * cr— * — * preincubation time 0 % control 30 activity 60 90 120 25 |xM cim 100 LIM cim 90.7 73.6 100.3 82.2 97.9 85.6 101.4 89.5 98.8 86.8 0 i 30 1 60 i 90 1 120 Preincubation time (minutes) Figure 18: Preincubation of pooled human liver microsomes with cimetidine for up to 120 minutes. Pooled human liver microsomes (2 mg protein/mL) were preincubated with cimetidine (25 \xM or 100 uM) or incubation buffer (control) and N A D P H (1 mM). A t 30 minute intervals, 50 jxL of the preincubation mixture was removed and added to a reaction tube containing dextromethorphan and N A D P H , in a volume of 150 \iL, and incubated for a further 20 minutes. In the final incubation, the concentration of dextromethorphan was 5 u.M and the concentration of microsomal protein was 0.5 mg/mL. Each point represents the mean of triplicate measurements ± standard deviation. * significantly lower (p<0.05) than control group preincubated for the same period of time. 63 W i t h the microsomes from the individuals H G 95 (figure 19) and H G 64 (figure 20), the amount o f inhibition seen without preincubation was similar to that seen with the pooled human liver microsomes with both 25 and 100 jxM cimetidine. Again, there was no trend toward increasing inhibition with longer preincubation times at either concentration o f cimetidine. In addition, the activity o f the control groups decreased more rapidly than that o f the cimetidine-treated groups, thus masking any inhibitory effect cimetidine might have been having. Because preincubation with cimetidine did not result in an increase in inhibition under the conditions used in the experiments described above, the effect o f using a saturating concentration o f substrate in the incubation step was investigated. Microsomes from individual H G 64 were preincubated with 100 \iM cimetidine for 90 minutes and then incubated with 50 \LM dextromethorphan, a concentration roughly 10-times the K m . Under conditions where a saturating substrate concentration was used, the competitive inhibition by the cimetidine present in the incubation reaction should be minimized, therefore a decrease in activity seen following preincubation should be exclusively due to mechanism-based inactivation. Using a higher substrate concentration also provided information regarding the iV-demethylation pathway o f dextromethorphan metabolism, which has a lower affinity compared to the O-demethylation pathway. A second substrate concentration (300 [xM), which approximated the K m o f the N-demethylation pathway (von Moltke et al. 1998) was also included. Microsomes from individual H G 64 were used because this sample demonstrated the highest dextromethorphan O-demethylase activity. Additionally, in the experiment described in figure 20, there appeared to be a slight trend towards increasing inhibition o f 3-methoxymorphinan formation by cimetidine with preincubation, which was most apparent after 90 minutes o f preincubation; however, because o f the low substrate concentration used in that experiment, the formation o f 3-methoxymorphinan was too low to be accurately quantitated. A s expected, without preincubation the activity in the cimetidine-treated group was the same as that o f the control group (figure 21), indicating that at higher substrate concentrations the competitive 64 inhibition o f cimetidine was overcome. There was no increase in cimetidine's inhibitory effect on dextrorphan formation with preincubation. For the N-demethylase pathway, there was an 11.5% drop i n activity in the cimetidine-treated group preincubated for 90 minutes compared to the control group preincubated for the same period o f time, when the incubation reaction contained 50 [xM dextromethorphan (figure 22A). There was no observable inhibition without preincubation. A t the higher substrate concentration (figure 22B), there was also no observable inhibition, either with or without preincubation. Because the inhibitory effect seen at the lower substrate concentration was small and the N-demethylation pathway involves several C Y P enzymes, thus making it difficult to determine which enzyme is involved in the inhibition, it was not investigated further. 65 Figure 19: Preincubation of human liver microsomes from individual H G 95 with cimetidine for up to 120 minutes. Microsomal protein (2 mg/mL) was preincubated with cimetidine (25 [xM or 100 p.M), or incubation buffer (control), and N A D P H (1 mM). A t 30 minute intervals, 50 fxL of the preincubation mixture was removed and added to a reaction tube containing dextromethorphan and N A D P H , in a volume of 150 |xL, and incubated for a further 20 minutes. In the final incubation , the concentration o f dextromethorphan was 5 uL, and the concentration of microsomal protein was 0.5 mg/mL. Each point represents the mean of triplicate measurements + standard deviation. * significantly lower (p<0.05) than control group preincubated for the same period of time. # significantly lower than the control group preincubated for the same period of time and the group treated with 100 f iM cimetidine that was not preincubated. 66 Figure 20: Preincubation of human liver microsomes from individual H G 64 with cimetidine for up to 120 minutes. Microsomal protein (2 mg/mL) was preincubated with cimetidine (25 \xM or 100 \iM) or incubation buffer (control) and N A D P H (1 mM). A t 30 minute intervals, 50 uL of the preincubation mixture was removed and added to a reaction tube containing dextromethorphan and N A D P H , in a volume of 150 ftL, and incubated for a further 20 minutes. In the final incubation, the concentration of dextromethorphan was 5 [iM, and the concentration of microsomal protein was 0.5 mg/mL. Each point represents the mean of triplicate measurements ± standard deviation. * activity is significantly lower (p<0.05) than control group preincubated for the same period of time. 67 Figure 21: Effect of preincubation of human liver microsomes with cimetidine for 90 minutes, followed by incubation with 50 [ i M dextromethorphan, on dextromethorphan O-demethylase activity in human liver microsomes. Microsomal protein (2 mg/mL) from donor H G 6 4 was preincubated with cimetidine (100 uM) and N A D P H (1 mM). Before initiation of the preincubation reaction, and 90 minutes later, a 50 [LL aliquot was removed from the preincubation mixture and added to a reaction tube containing dextromethorphan and N A D P H , in a volume of 150 Samples were then incubated for a further 20 minutes. The final concentration of dextromethorphan in the incubation reaction was 50 u M . Each bar represents the mean of triplicate measurements + standard deviation. N o statistically significant inhibition was seen in either of the cimetidine-treated groups. 68 X TJ A: 50 uM dextromethorphan without preincubation with preincubation B: 300 uM dextromethorphan oj without preincubation with preincubation TJ Figure 22: Effect of preincubation of human liver microsomes with cimetidine for 90 minutes, followed by incubation with 50 tiM or 300 \tM dextromethorphan, on dextromethorphan iV-demethylase activity of human liver microsomes. Microsomal protein (2 mg/mL) from donor H G 6 4 was preincubated with cimetidine (100 \iM) and N A D P H (1 mM). Before initiation of the preincubation reaction, and 90 minutes later, a 50 uL aliquot was removed from the preincubation mixture and added to a reaction tube containing dextromethorphan and N A D P H , in a volume of 150 \iL. Samples were then incubated for a further 20 minutes. The final concentration of dextromethorphan in the incubation reaction was 50 u.M or 300 jxM. Each bar represents the mean of 3 measurements + standard deviation. * significantly different (p<0.05) from control group preincubated for 90 minutes and from cimetidine-treated group which was not preincubated. 69 3.3.4 Mechanism-based inactivation with triacetyloleandomycin and 1 -aminobenzotriazole T w o compounds known to act as mechanism-based inactivators, triacetyloleandomycin and 1-aminobenzotriazole, were used as positive controls to demonstrate that this type o f inhibition could be observed under the same conditions used for the cimetidine preincubation studies. Triacetyloleandomycin has been shown to be a specific inactivator o f C Y P 3 A (Chang et al. 1994). Without preincubation, triacetyloleandomycin inhibited the iV-demethylation o f dextromethorphan, a pathway partially mediated by C Y P 3 A , by 27.4%. Following preincubation with N A D P H for 30 minutes, a 45.7% decrease in activity was observed (figure 23). In contrast to triacetyloleandomycin, 1-aminobenzotriazole is a more general P450 inactivator and has been shown to inactivate a number o f enzymes, including C Y P 2 D 6 (Balani et al. 2002). A more specific C Y P 2 D 6 inactivator could not be found; however, using incubation conditions under which the formation o f dextrorphan is primarily mediated by C Y P 2 D 6 , mechanism-based inactivation was demonstrated. Fol lowing 30 minutes o f preincubation with N A D P H , 1-aminobenzotriazole decreased the formation o f dextrorphan by 34%, compared to no observable loss in activity without preincubation (figure 24). 3.3.5 Summary of studies with human liver microsomes Both cimetidine and quinidine were shown to be competitive inhibitors o f dextromethorphan O-demethylation. Preincubation with N A D P H and cimetidine for up to 120 minutes did not result in a further loss o f activity compared to the appropriate control. The lack o f observable mechanism-based inactivation by cimetidine under these conditions, however, is not due to any intrinsic properties o f the experimental system as positive results were observed with both triacetyloleandomycin and 1-aminobenzotriazole. 70 without preincubation with preincubation Figure 23: Inactivation of dextromethorphan JV-demethylase activity in human liver microsomes by triacetyloleandomycin (TAO). Pooled human liver microsomes (2 mg/mL) were preincubated with 5 U . M T A O , or an equal volume of buffer (control), and N A D P H (1 mM). A t 0 and 30 minutes after initiation of the preincubation reaction, a 50 u L aliquot was removed and added to a reaction tube containing dextromethorphan (300 uM) and N A D P H (1 mM), and incubated for a further 15 minutes. In the final incubation step, the concentration of microsomal protein was 0.5 m g / m L and the concentration of T A O was 1.25 u M . Each bar represents the mean of four measurements + standard deviation. * significantly different (p<0.05) from control group preincubated for the same period of time. There was a significant interaction (p<0.05) between preincubation time and T A O treatment. 71 without preincubation with preincubation Figure 24: Inactivation of dextromethorphan O-demethylase activity in human liver microsomes by 1-aminobenzotriazole (ABT). Pooled human liver microsomes (2 mg/mL) were preincubated with 20 u M A B T , or an equal volume of buffer (control), and N A D P H (1 mM). A t 0 and 30 minutes after initiation of the preincubation reaction, a 50 p L aliquot was removed and added to a reaction tube containing dextromethorphan (5 pM) and N A D P H (1 mM), and incubated for a further 15 minutes. In the final incubation step, the concentration of microsomal protein was 0.5 m g / m L and the concentration of A B T was 5 u M . Each bar represents the mean of four measurements + standard deviation. * significandy different (p<0.05) from the control group preincubated for the same period of time. There was a significant interaction (p<0.05) between preincubation time and A B T treatment. 72 3.4 Studies w i t h recombinan t h u m a n C Y P 2 D 6 3.4.1 Kinetic constants of dextromethorphan O-demethylation The V m a x and K m values for dextromethorphan O-demethylation were determined from Michaelis-Menten plots and were estimated to be 15.9 ± 0.8 n m o l / m i n / n m o l P450 and 1.4 ± 0.6 \xM, respectively (figure 25). The average ratio o f V m a x / K m with the recombinant enzyme was 13.1 (table 9). This same ratio with the human liver microsomes was 0.2 (table 5), indicating that the catalytic efficiency o f the recombinant system for this pathway is considerably higher than that o f the microsomal system. A s with the results from the microsomal experiment, the K m value determined in this experiment was used to select appropriate concentrations o f substrate for subsequent experiments. 3.4.2 Effect of quinidine and cimetidine on dextromethorphan O-demethylase activity The K i values for inhibition by quinidine and cimetidine were determined by incubating recombinant C Y P 2 D 6 with various concentrations o f inhibitor at three fixed concentrations o f substrate (approximately 0.5-, 1-, and 2-times the K m ) . F r o m D i x o n plots, (figure 26A) the JQ value for quinidine was estimated to be 1.3 ± 0.4 n M (n=3). The competitive inhibition model best fit the data, as was seen with the human liver microsomes, and a representative Lineweaver-Burk plot is shown in figure 26B. Fo r cimetidine, the model that provided the best fit for the data was that o f mixed inhibition, and the estimated JQ value was 102.5 ± 16.8 u.M (n=3). Representative D i x o n and Lineweaver-Burk plots for cimetidine are shown in figure 27. The individual K i values from three independent trials with both quinidine and cimetidine are summarized in table 10. 73 c ro JZ Q. ** O X LO II .2 c to E o 4> re 20 n 15 10 20 40 60 [dex t romethorphan] (pM) 80 100 Figure 25: Rate of formation of dextrorphan by recombinant CYP2D6. Recombinant C Y P 2 D 6 (2.5 pmol /mL) was incubated with dextromethorphan (0.25 to 80 uM) and N A D P H ( lmM) for 15 minutes. Data were fit to the Michaelis-Menten equation using non-linear regression. Each point represents the mean of duplicate measurements. 74 Table 9: Summary of the kinetic constants for dextromethorphan O-demethylation by recombinant CYP2D6 determined in three separate experiments. (nmol/min/nmol P450) (uM) 1 15.2 1.0 15.2 2 16.8 1.0 16.8 3 15.6 2.1 7.4 m e a n 15.9 1.4 13.1 s t a n d a r d dev ia t ion 0.8 0.6 5.1 75 -10 -5 0 5 10 [quinidine] (nM) - 5 - 4 - 3 - 2 - 1 0 1 2 3 ^[dextromethorphan] (LIM"1) Figure 26: Inhibition of dextromethorphan O-demethylase activity of recombinant CYP2D6 by quinidine. Recombinant C Y P 2 D 6 (2.5 pmol /mL) was incubated with dextromethorphan (dex, 0.5 to 2 uM), quinidine (0.5 to 7.5 nM), and N A D P H (1 mM) for 15 minutes. Data were Et to a competitive inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of duplicate measurements. A : Dixon plot. B: Lineweaver-Burk transformation of the data presented in panel A . 76 -200 -100 0 100 200 [cimetidine] (uM) - 2 - 1 0 1 2 3 [^dextromethorphan] (uM'1) Figure 27: Inhibition of dextromethorphan O-demethylase activity of recombinant CYP2D6 by cimetidine. Recombinant C Y P 2 D 6 (2.5 pmol /mL) was incubated with dextromethorphan (dex, 0.5 to 2 \iM), cimetidine (25 to 150 \xM), and N A D P H (1 mM) for 15 minutes. Data were fit to a mixed inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of duplicate measurements. A : Dixon plot. B: Lineweaver-Burk transformation of data in panel A . 77 Table 10: Summary of estimated Kj values for inhibition of dextromethorphan O-demethylation by quinidine and cimetidine with recombinant CYP2D6, determined in three separate experiments. The Kj values for qumidine were determined using a competitive inhibition model, while a mixed inhibition model was used for cimetidine. a p p a r e n t K, Tr ia l q u i n i d i n e c imet id ine (nM) (u.M) 1 1.4 117.6 2 1.6 105.5 3 0.9 84.3 m e a n 1.3 102.5 s t a n d a r d dev ia t ion 0.4 16.8 78 3.4.3 Preincubation studies with cimetidine T o determine whether cimetidine could act as a mechanism-based inactivator in the recombinant system, C Y P 2 D 6 was preincubated with this inhibitor for up to 20 minutes (figure 28). Fol lowing 20 minutes o f preincubation, the activity o f the control group, which included C Y P 2 D 6 and N A P D H in the absence o f cimetidine, had decreased by 54.3%. Due to the instability o f the enzyme activity, the preincubation period was not extended beyond this time. A cimetidine concentration o f 50 u M was used in the preincubation reaction, resulting in an inhibitor concentration o f 12.5 u M in the final incubation step. This concentration was selected to rninimize the amount o f competitive inhibition seen, based on the K ; value o f 102.5 [iM. In the cimetidine-treated group, there was a trend toward increasing loss o f activity with time from 7.2% without preincubation to 34.7% after 20 minutes o f preincubation, compared to the control group preincubated with N A D P H alone. In addition to being dependent on preincubation time, the loss o f activity observed in the cimetidine-treated group was also dependent on the presence o f N A P D H in the preincubation mixture. In groups that did not include N A P D H in the preincubation mixture, an initial drop in activity o f 6.4% was seen in the cimetidine treated group compared to the control without preincubation, however, this drop in activity did not increase with longer preincubation times. T o determine the consistency o f the effect o f preincubation on inhibition with cimetidine, three independent experiments were performed in which C Y P 2 D 6 was preincubated with 100 u M cimetidine for 20 minutes. The average activity o f the cimetidine-treated group was 35.6% (95% confidence interval, 33.7-37.5) lower than that o f the control group after preincubation (figure 29). 79 4-1 > o CO a> (A > o 7 i c E CO i= a . E E o X a> Q 12 10 H i 6 H 4 4 preincubation +NADPH +NADPH +CIM -NADPH -NAPDH +CIM time (minutes) 0 5 10 20 +NAPDH 92.8 83.1 70.9 65.3 -NADPH 93.6 95.3 92.0 94.9 i 0 i 5 i 10 i 15 *# 20 Preincubation time (minutes) Figure 25: Preincubation of recombinant CYP2D6 with cimetidine for 5, 10, and 20 minutes, followed by incubation with 1 u.M dextromethorphan. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with cimetidine (50 u.M) and N A D P H (1 mM), or N A D P H (1 mM) alone. Aliquots (50 p.L) were removed at 5, 10, and 20 minutes following initiation of the preincubation reaction and were added to reaction tubes containing dextromethorphan (1 u.M) and N A D P H (1 mM) and reactions were incubated for a further 15 minutes. In the final incubation step, the concentration of recombinant C Y P 2 D 6 was 2.5 p m o l / m L and the concentration of cimetidine was 12.5 [ iM in a volume of 200 (xL. A second set of samples were preincubated with or without cimetidine in the absence of N A D P H . Each point represents the mean of three measurements ± standard deviation. * significantly different from groups preincubated with N A P D H alone for the same period of time. # significantly different from groups preincubated with N A D P H and cimetidine for 0, 5, and 10 minutes. 80 > O ra o V) n got % i 61 s i J= o II ** OJ E o L . <-> X OJ Q 12 3H 1 control cimetidine (50 uM) * J l without preincubation with preincubation Figure 29: Preincubation of CYP2D6 with cimetidine (50 (JiM) for 20 minutes, followed by incubation with 1 [*M dextromethorphan. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with cimetidine (50 uM) and N A D P H (1 mM), or N A D P H alone (control), for 20 minutes. A 50 uL aliquot was then removed and added to a reaction tube containing dextromethorphan (1 uM) and N A D P H (1 mM), in a volume of 150 uL. Samples were then incubated for a further 15 minutes. In the final incubation step, the concentration of recombinant C Y P 2 D 6 was 2.5 p m o l / m L and the concentration of cimetidine was 12.5 u M . Data shown are the mean o f four separate experiments, in which activity was measured in triplicate in each group, + standard deviation. * significandy different from preincubated control and cimetidine-treated group without preincubation (p<0.05). 81 Despite the presence o f a relatively low concentration o f cimetidine (12.5 (xM) in the incubation step o f these experiments, a small degree o f competitive inhibition was still observed, as can be seen by the 7% inhibition in activity without preincubation in figure 28. T o eliminate the contribution o f competitive inhibition, and to ensure that enzyme inactivation due to preincubation with cimetidine is primarily responsible for the observed decrease in activity, a saturating concentration o f dextromethorphan was used in the incubation reaction. Fol lowing 20 minutes o f preincubation with 50 u.M cimetidine, dextromethorphan O-demethylase activity was decreased by 10.6% (95% confidence interval, 5.0-16.2; figure 30) when a saturating substrate concentration (equivalent to approximately 20 times the K m ) was used in the incubation reaction. Increasing the cimetidine concentration to 100 u.M resulted in a 17.8% (95% confidence interval, 13.4-22.2) decrease in activity with preincubation (figure 30). In both cases, without preincubation there was no difference i n the activity o f the cimetidine treated groups compared to that o f the control. In addition to being dependent on inhibitor concentration, the decrease in activity was also dependent on preincubation time and the presence o f N A P D H (figure 31). 82 X Q without preincubation with preincubation u ro o co TO ^ t § O Q_ I i si P O c 0) E o X 0) Q 18 15 12 £ ~ 3 • control =i cimetidine (100 uM) without preincubation with preincubation Figure 30: Preincubation of CYP2D6 with cimetidine for 20 minutes, followed by incubation with a saturating concentration of dextromethorphan. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with 50 u-M (n=3) or 100 |xM (n=4) cimetidine and N A D P H (1 mM), or N A D P H alone (control), for 20 minutes. A 50 [xL aliquot was then removed and added to a reaction tube containing dextromethorphan (20 (xM) and N A D P H (1 mM). Samples were then incubated for a further 15 minutes. In the final incubation step, the concentration o f recombinant C Y P 2 D 6 was 2.5 p m o l / m L and the concentration of cimetidine was 12.5 or 25 (xM, for samples preincubated with 50 or 100 p.M, respectively. Data shown are the mean of three or four separate experiments, in which activity was measured in triplicate in each group, + standard deviation. * significantly different from preincubated control (p<0.05). There was a significant interaction (p<0.05) between preincubation time and treatment with cimetidine. 83 Figure 31: Preincubation of recombinant CYP2D6 with cimetidine for 5, 10, and 20 minutes, followed by incubation with a saturating concentration of dextromethorphan. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with cimetidine (100 [iM) and N A D P H (1 mM), or N A D P H (1 mM) alone. AHquots (50 uL) were removed at 5, 10, and 20 minutes following initiation of the preincubation reaction and were added to reaction tubes containing dextromethorphan (20 uM) and N A D P H (1 mM). Reactions were incubated for a further 15 minutes. In the final incubation step, the concentration of recombinant C Y P 2 D 6 was 2.5 p m o l / m L and the concentration of cimetidine was 25 u M in a volume of 200 uL. A second set of samples were preincubated with or without cimeticiine in the absence of N A D P H . Each point represents the mean of three measurements + standard deviation. * significantiy different from groups preincubated with N A P D H alone for the same period of time. # significantiy different from groups preincubated with N A D P H and cimetidine for 0, 5, and 10 minutes. 84 3.4.4 Preincubation of recombinant CYP2D6 with 1 -aminobenzotriazole and quinidine Previously, 1-aminobenzotriazole was used as a positive control for mechanism-based inactivation o f C Y P 2 D 6 with human liver microsomes and was shown to inhibit dextromethorphan 0 -demethylase activity in a time-dependent manner (section 3.3.4). Preincubation o f recombinant C Y P 2 D 6 with 1-aminobenzotriazole and N A P D H resulted in a 51.5% decrease in activity compared to the control group preincubated for the same period o f time (figure 32). Without preincubation, no inhibition was observed. Quinidine, which acts as a competitive inhibitor o f C Y P 2 D 6 was used as a further control to demonstrate that the time-dependent decrease in activity seen with 1-aminobenzotriazole and cimetidine is not a general property o f all C Y P 2 D 6 inhibitors. Without preincubation, quinidine competitively inhibited dextrorphan formation by 58.3% (figure 33). Fol lowing 20 minutes o f preincubation with the inhibitor and N A D P H , an inhibition o f 41.2% was observed. The preincubation step, therefore, did not result in a further decrease in activity with a competitive inhibitor. The only experimental conditions under which a significant decrease in activity was demonstrated with preincubation occurred with the known mechanism-based inactivator, 1-aminobenzotriazole. 85 10 without preincubation with preincubation Figure 32: Mechanism-based inactivation of recombinant CYP2D6 by 1-aminobenzotriazole (ABT). Recombinant CYP2D6 (10 pmol/mL) was preincubated with 4 u.M A B T , or an equal volume of buffer (control), and N A D P H (1 mM). A t 0 and 15 minutes after initiation of the preincubation reaction, a 50 [iL aliquot was removed and added to a reaction tube containing dextromethorphan (1 u.M) and N A D P H (1 mM), and incubated for a further 15 minutes. Each bar represents the mean of four measurements + standard deviation. * significantly different (p<0.05) from preincubated control. There was a significant interaction (p<0.05) between preincubation time and treatment A B T . 86 12 without preincubation with preincubation Figure 33: Lack of mechanism-based inactivation of recombinant CYP2D6 by quinidine. Recombinant C Y P 2 D 6 (10 pmol/mL) was preincubated with 0.03 u M quinidine or an equal volume of buffer (control), and N A D P H (1 mM). At 0 and 20 minutes after initiation of the preincubation reaction, a 50 uL aliquot was removed and added to a reaction tube containing dextromethorphan (1 \xM) and N A D P H (1 mM), and incubated for a further 15 minutes. Each bar is the mean of four measurements + standard deviation. * significantly different (p<0.05) from control, # significantly different (p<0.05) from and quinidine-treated group that was not preincubated. 87 3 . 4 . 5 Effect of preincubation on the K of inhibition by cimetidine After the preincubation conditions for cimetidine had been established, the K i determination was repeated with the preincubation step included. Initially, this was carried out using the same preincubation method described in the Materials and Methods (section 2.2.3). Wi th this method, the enzyme and inhibitor concentrations in the preincubation reaction are diluted four-fold in the final incubation. D i x o n plots were constructed using the cimetidine concentration in the final incubation reaction (figure 34) and the JQ was estimated to be 23.6 ± 7.5 u.M (n=3). However, because the concentrations o f cimetidine and C Y P 2 D 6 in the preincubation and incubation steps were different, these results are difficult to interpret. T o overcome this issue, the experiment was repeated using a method that did not involve the dilution step, as described in the legend for figure 35. Briefly, recombinant C Y P 2 D 6 (2.5 p m o l / m L ) was preincubated with cimetidine and N A P D H in a volume o f 190 |xL. After 20 minutes, dextromethorphan was added in a volume o f 10 u.L, to give a final incubation volume o f 200 u.L. D i x o n plots were constructed, this time using the initial cimetidine concentrations, and the K was estimated to be 40.5 ± 3.5 ( iM (n=3). A representative D i x o n plot is shown i n figure 35. There was no significant difference between the K ; values determined with these two methods and addition o f the preincubation step lowered cimetidine's K i by 2.5- to 4-fold compared to the value obtained without preincubation (table 11). In both cases, a mixed inhibition model was used to estimate the JQ values, but, based on R 2 values, it was difficult to distinguish which inhibition model provided the best fit for the data. When the dilution method was used, the mixed and noncompetitive inhibition models both provided a good fit for the data, with R 2 values o f 0.945 and 0.940, respectively. W i t h the second preincubation method, however, the competitive and mixed models provided similar fits to the data with R 2 values o f 0.976 and 0.975, respectively. T o facilitate comparison o f the IQ values determined previously and with the two preincubation methods, the mixed inhibition model was used in both cases. 88 -80 -60 -40 -20 0 20 40 60 [cimetidine] (pM) - 2 - 1 0 1 "(/[dextromethorphan] (JJM1) Figure 34: Inhibition of recombinant CYP2D6 by cimetidine following 20 minutes preincubation using the dilution method. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with cimetidine (25 to 200 p.M) and N A D P H (1 mM). After 20 minutes, 50 uL aliquots were removed from the preincubation mixtures and added to reaction tubes containing dextromethorphan and N A D P H in incubation buffer in a volume of 150 uL. In the final incubation, the concentration of cimetidine ranged from 6.25 to 50 u,M, and the C Y P 2 D 6 concentration was 2.5 p m o l / m L . Data were fit to the mixed inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of two measurements. Panel A : Dixon plot. Panel B : Lineweaver-Burk transformation of the data presented in panel A . 89 —i 1 1 1 r 1 1 1 -80 -60 -40 -20 0 20 40 60 [cimetidine] (uM) B - 3 - 2 - 1 0 1 2 [^dextromethorphan] (pM) Figure 35: Dixon plot for the inhibition of recombinant CYP2D6 by cimetidine following 20 minutes preincubation. Recombinant C Y P 2 D 6 (2.5 pmol /mL) was preincubated with N A D P H (1 mM) and cimetidine (6.25 to 50 uM) in a volume of 190 p.L. After 20 minutes, 10 p.L dextromethorphan was added to each reaction tube and samples were incubated for a further 15 minutes. Data was fit to a mixed inhibition model using the SigmaPlot enzyme kinetics module. Each point represents the mean of duplicate measurements. A : Dixon plot. B: Lineweaver-Burk transformation of the data shown in panel A . 90 Table 11: Summary of K i values for the inhibition of dextromethorphan O-demethylase activity of recombinant CYP2D6 by cimetidine. K i values were determined without a preincubation step and with a 20 minute preincubation step. The preincubation experiments were conducted with or without a dilution step, as described in the legends for figures 34 and 35, respectively. K i values were estimated using the mixed inhibition model. * significandy different (p<0.05) than the average K i values determined with preincubation by either method. There was no statistically significant difference between the two groups of K ; values determined with preincubation. c imet id ine Kj (pM) Tr ia l W i t h o u t p r e i n c u b a t i o n Wi th p r e i n c u b a t i o n with d i lu t ion wi thout d i lu t ion 1 117.6 18.9 41.8 2 105.5 32.2 36.5 3 84.3 19.7 43.2 m e a n 102.5* 23.6 40.5 s t a n d a r d 16.8 7.5 3.5 dev ia t ion 91 T o further clarify which mode o f inhibition was occurring under conditions o f preincubation, higher concentrations o f dextromethorphan were included in the D i x o n plot (figure 36A). Under these conditions, the non-competitive component o f mixed inhibition should be more apparent, i f this type o f inhibition is occurring. However, in this case, both the mixed and competitive models o f inhibition fit the data with the same R 2 value (0.902) and the Lineweaver-Burk plots for each o f these models are identical (figure 36B). 92 —I 1 1 1 1 \ 1 1 1 -100 -80 -60 -40 -20 0 20 40 60 [cimetidine] (uM) Figure 36A: Dixon plot for the inhibition of recombinant CYP2D6 by cimetidine following 20 minutes preincubation using saturating substrate concentrations. Recombinant C Y P 2 D 6 (2.5 pmol /mL) was preincubated with N A D P H (1 mM) and cimetidine (6.25 to 50 uM) in a volume of 190 uL. After 20 minutes, 10 [tL dextromethorphan was added to each reaction tube and the samples were incubated for a further 15 minutes. Each point represents the mean of duplicate measurements. Solid lines represent fit of a mixed inhibition model; however, this model does not appear to provide an adequate fit to the data. 93 A 0.3 -o O E c | 0 . 2 -o E _c _ o • /// • //// [cimetidine] i i _> T— tStr s i ^ P -• o T V m a i 0 uM 6.25 LLM 12.5 LIM 25 LIM 37.5 LLM 50 LLM 1 i -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 1/[dextromethorphan] (uM 1) -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 1/[dextromethorphan] (MM'1) Figure 36B: Lineweaver-Burk transformation of the data presented in figure 36A. Data were fit to the mixed (panel A) or competitive (panel B) model of inhibition. 94 3.4.6 Effect of quinidine on inactivation of CYP2D6 by cimetidine Mechanism-based inactivators require catalytic activation to a reactive species which then binds to and inactivates the target enzyme. Like other enzyme-mediated reactions, this activation step is susceptible to competitive inhibition. T o determine whether the time- and NADPH-dependen t decrease in dextromethorphan O-demethylase activity by cimetidine could be attenuated by a competitive inhibitor o f C Y P 2 D 6 , increasing concentrations o f quinidine were included in the preincubation reaction. The concentrations o f quinidine used were equivalent to 0.5-, 1-, and 2-times the K i o f inhibition o f dextromethorphan O-demethylation; a saturating substrate concentration was used i n the incubation reaction to ensure that quinidine would not competitively inhibit the formation of dextrorphan. Without preincubation, there was no significant inhibition caused by either cimetidine or quinidine (figure 37). Preincubation with N A D P H alone for 20 minutes reduced the dextromethorphan O-demethylase activity to 8.65 n m o l / m i n / n m o l P450, a decrease o f approximately 50% compared to the control group which was not preincubated. In the presence o f 100 u M cimetidine, preincubation with N A P D H lowered the activity to 7.23 n m o l / m i n / n m o l P450), a decrease of 16.8%, which was consistent with the effect previously observed with this inhibitor concentration (section 3.4.3). W i t h the addition of increasing amounts o f quinidine to the preincubation mixture, the effect o f cimetidine was attenuated and, at the two highest concentrations o f quinidine, the activity in the cimetidine-treated groups was not significantly different from that o f the control group. The effect o f preincubation on inactivation o f C Y P 2 D 6 was blocked by the addition o f quinidine to the preincubation reaction. 95 I without preincubation ] with preincubation [quinidine] (nM) Figure 37: Effect of quinidine on inactivation of recombinant CYP2D6 by cimetidine. Recombinant C Y P 2 D 6 (10 pmol /mL) was preincubated with cimetidine (100 u.M) and varying concentrations of quinidine (0 to 2.6 nM), and N A P D H (1 mM) for 0 or 20 minutes. A 50 |j.L aliquot was removed and added to a reaction tube containing dextromethorphan and N A P D H in a volume of 150 u.L and was incubated for a further 15 minutes. The concentration of dextromethorphan in the final incubation was 20 u M . Activity in each group was normalized to the activity of the control group which was preincubated with N A P D H (1 mM) for 0 or 20 minutes in the absence of cimetidine or quinidine. Data are expressed as the mean of three experiments + standard deviation. * significantly lower (p<0.05) compared to group containing the same concentration of quinidine that was not preincubated. There was no difference in the activity of the groups that were not preincubated. 96 3.4.7 Determination of the kmaaimtion of cimetidine A s a final demonstration o f cimetidine's ability to act as a mechanism-based inactivator o f recombinant C Y P 2 D 6 , the rate constant for inactivation (kinact) was estimated. This was achieved by preincubating the recombinant enzyme with increasing concentrations o f cimetidine (50 to 400 uM) for 0 to 20 minutes (figure 38A). The observed decrease in activity due to preincubation with cimetidine was time- and concentration-dependent. Based on a Ki t z -Wi l son transformation o f the data (figure 38B), in which the half-life o f inactivation at each concentration o f cimetidine was plotted against the inverse concentration o f cimetidine in the incubation reaction, the kinact, or the rate o f maximal inactivation, was estimated to be 0.03 min- 1 and the time required to inactivate 50% o f the enzyme was estimated to be 25.3 minutes. The concentration o f cimetidine required for half-maximal inactivation (Ki) was estimated to be 19.2 u M . 3.4.8 Summary of studies with recombinant CYP2D6 Cimetidine acted as a mixed inhibitor o f C Y P 2 D 6 , with a K o f 102.5 u M , while quinidine was a competitive inhibitor with a K j o f 1.3 n M . Preincubation o f recombinant C Y P 2 D 6 with cimetidine resulted in an increase in inhibition, which was dependent on both preincubation time and the presence of N A P D H . The maximal effect, observed after 20 minutes o f preincubation, resulted in a further 18% decrease i n activity compared to the group treated with the same concentration cimetidine in the absence o f preincubation. The same type o f effect was seen when recombinant C Y P 2 D 6 was preincubated with 1-aminobenzotriazole and N A D P H ; in contrast, preincubation with quinidine did not have an effect on inhibition. Preincubation with cimetidine lowered the K o f the inhibitor by 2.5-to 4-fold, and was characterized by a k;nact o f 0.03 min- 1 and a K i o f 19.2 U .M. The inclusion o f quinidine in the preincubation reaction attenuated the degree o f inactivation o f C Y P 2 D 6 by cimetidine. 97 Figure 38: Inactivation of recombinant C Y P 2 D 6 by cimetidine. Recombinant C Y P 2 D 6 (10 pmol/mL)'was preincubated with buffer (control) or various concentrations of cimetidine, and N A P D H (1 mM). Immediately before initiating the preincubation reaction and 5, 10, and 20 minutes later, 50 uL aliquots of the preincubation mixture were added to reaction tubes containing dextromethorphan (20 u.M) and N A D P H (1 mM) and were incubated for a further 15 minutes. A : plot of log percent control activity versus preincubation time. Each point represents the mean of two separate experiments in which measurements were made in triplicate. The concentration of cimetidine reported is that in the final incubation. B: Ki tz -Wilson plot of half-life of inactivation versus inverse concentration of cimetidine. 98 Chapter 4 DISCUSSION In vitro studies have demonstrated that cimetidine is a weak, reversible inhibitor o f various P450 enzymes, including C Y P 2 D 6 . Cimetidine has been implicated in several drug-drug interactions involving known substrates o f this enzyme, suggesting that enzyme inhibition may be occurring at the relatively low plasma concentrations o f cimetidine seen in vivo. It has been previously shown with rat hepatic microsomes that both the potency and selectivity o f inhibition by cimetidine can be increased in vitro when experimental conditions allowing for mechanism-based inactivation are used. The purpose o f the present study was to further characterize the inhibition o f C Y P 2 D 6 activity by cimetidine and to determine whether mechanism-based inactivation could be observed under the appropriate experimental conditions. T o this end, two in vitro model systems were used: human liver microsomes and recombinant C Y P 2 D 6 . The results obtained from these two experimental models, however, were not in agreement. Wi th human liver microsomes, cimetidine was found to act as a competitive inhibitor o f dextromethorphan O-demethylase activity, while with the recombinant enzyme, both mixed inhibition and mechanism-based inactivation were observed, depending on the experimental conditions used. There are several possible explanations for this discrepancy, which relate to the intrinsic properties o f each o f the model systems used, such as their relative rates o f C Y P 2 D 6 activity and the presence or absence o f additional P450 enzymes. The findings from this study suggest that the choice o f model system can have a major impact on the types o f conclusions drawn from in vitro studies. 99 Using pooled human liver microsomes, the dextromethorphan O-demethylase activity was comparable to that reported by other laboratories using similar experimental conditions. In the present study, the K m and V m a x o f the reaction were estimated to be 4.1 u M and 0.82 n m o l / m i n / n m o l P450 (or 0.31 n m o l / m i n / m g protein), respectively. Previously reported values range from 2.8 \iM to 8.3 u M for the K m , and 0.14 to 0.26 n m o l / m i n / m g protein for the V m a x (Dayer et al. 1989, Kerry et al. 1994, Schmider et al. 1997, von Moltke et al. 1998). This indicates that the microsomes exhibited expected C Y P 2 D 6 expression and activity. The effect o f quinidine on dextromethorphan O-demethylation was also consistent with that reported by other investigators. The inhibitor was found to act competitively, with a K o f 53 n M , which is within the range o f 28 n M to 400 n M previously reported (Kobayashi et al. 1989, Ching et al. 1995, Bourrie et al. 1996). Inhibition o f C Y P 2 D 6 by cimetidine has also been investigated, to a lesser extent than quinidine, with human liver microsomes. Using bufuralol hydroxylase activity as a marker o f C Y P 2 D 6 activity, the K o f cimetidine was estimated to be 50 u M (Knodel l et al. 1991) and 55.5 p.M (Furuta et al. 2001). The former study also found that cimetidine was acting as a mixed inhibitor. Martinez et al. (1999) estimated that cimetidine inhibited dextromethorphan O-demethylase activity o f human liver microsomes with an IC50 o f 210 u M . In this experiment, a relatively high concentration o f dextromethorphan was used (25uM, or 2.5-times the K m ) , which may explain the high concentration o f cimetidine required to inhibit the activity by 50%. Competitive inhibition is dependent on both substrate and inhibitor concentrations, and the degree o f observed competitive inhibition is reduced at higher substrate concentrations. A l so , at this substrate concentration, there is likely a considerable contribution o f other P450 enzymes to the formation o f dextrorphan, which may be refractory to inhibition by cimetidine. In the present study, inhibition o f C Y P 2 D 6 by cimetidine was found to occur by competitive inhibition, with a K ; o f 38 u M . Despite the differences in the probe drugs used, this value is in agreement with the K i values obtained in the studies reported above. The difference in the mechanism o f inhibition may be a result o f differences i n methodology. The study which found cimetidine acted as a mixed inhibitor used substrate 100 concentrations which ranged from slightly below the K m o f the probe reaction to 5-times greater than the K m , whereas the substrate concentrations used here were limited to 0.5- to 2-times the K m . I f the inhibitor is in fact acting by a combination o f both competitive and non-competitive mechanisms, this would be more apparent at higher substrate concentrations. Preincubation o f human liver microsomes with cimetidine and N A P D H did not affect the potency o f the inhibitor, suggesting that cimetidine does not act as a mechanism-based inactivator o f C Y P 2 D 6 . However, an alternative explanation for this lack o f observable effect may be due to the use o f inappropriate experimental conditions. For example, the preincubation time may not have been long enough to allow for the production o f a sufficient amount o f the putative inhibitory species to result in a significant, measurable, degree o f enzyme inactivation. The preincubation reaction was carried out for a maximum of 120 minutes. A t this point, the activity o f the control groups had decreased by 25% to 40%, depending on which human liver microsomes were used, and it was not practical to further extend the duration o f preincubation for this reason. A significant degree o f l ipid peroxidation is known to occur during in vitro incubations with hepatic microsomes which leads to a loss o f enzyme activity and destruction o f the P450 heme (Levin et al. 1973, Renton et al. 1976). The N A P D H concentration-dependent formation o f superoxide and hydrogen peroxide, which can be converted to the l ipid peroxidation-inducing hydroxy radical in the presence o f iron, has been implicated (Bondy and Naderi 1994, Rashba-Step and Cederbaum 1994). Another possible explanation for the lack o f effect may relate to the inability to distinguish competitive inhibition from mechanism-based inactivation. I f there is a significant overlap in the concentration o f inhibitor that produces reversible and mechanism-based inhibition, it is possible that the reversible inhibition may be sufficient to mask the effect o f preincubation on a further decrease in enzyme activity. T o minimize the contribution o f competitive inhibition, a saturating substrate concentration can be used in the inhibition studies. Unfortunately, this methodology is problematic when using human liver microsomes because at concentrations above two times the K m , 101 dextromethorphan loses its selectivity as a probe o f C Y P 2 D 6 activity (von Moltke et al. 1998). However, the way in which the present studies were performed minimized the concentration o f cimetidine present in the final incubation reaction. The transfer o f an aliquot o f the preincubation mixture into a separate reaction tube for the incubation reaction resulted in a four-fold dilution o f the cimetidine concentration. A s a result, the decrease in activity attributable to competitive inhibition was only 10% at the highest concentration o f cimetidine used. Therefore, unless preincubation was having only a very small effect on the potency o f cimetidine as an inhibitor o f dextromethorphan O-demethylase activity, it is unlikely that the effect o f competitive inhibition would be sufficient to obscure any further inactivation o f the enzyme. When a saturating concentration o f dextromethorphan was used, preincubation still had no effect on inhibition by cimetidine. It is not possible to conclude whether this observation is due to the absence o f mechanism-based inactivation or due to the loss o f specificity o f dextrorphan formation as a marker o f C Y P 2 D 6 activity at this higher substrate concentration. The experimental method used for the preincubation studies may also have contributed to the lack o f effect o f preincubation on inhibition by cimetidine by reducing the free concentration o f inhibitor due to binding to the microsomal protein. The concentration o f protein in the preincubation reaction was 2 m g / m L , four times higher that that used in the competitive inhibition studies. The inhibitor may have been binding non-specifically to the protein, or to specific P450 enzymes other than C Y P 2 D 6 , thus reducing the amount o f cimetidine available to inactivate C Y P 2 D 6 . Spectral binding studies have shown that cimetidine does bind to at least two rat microsomal P450 enzymes (Rendic et al. 1979), and in vitro studies have demonstrated that activities of CYPs 1A2 and 3A can be inhibited by cimetidine, therefore the inhibitor may also be binding to these enzymes. Human liver microsomes provide a good in vitro model for hepatic metabolism because all P450 enzymes and accessory proteins are present in their appropriate proportions; however, this can be problematic when trying to isolate the effect o f an inhibitor on a single enzyme. Due to overlaps in 102 substrate specificity, a loss in activity in one enzyme can be compensated for by other enzymes which are capable o f mediating the same metabolic pathway. Although C Y P 2 D 6 displays the highest affinity for dextromethorphan O-demethylase activity, this pathway o f metabolism has a limited capacity due to the low expression o f C Y P 2 D 6 in the human liver compared to the other enzymes involved in the metabolism o f dextromethorphan. Specific inhibition o f C Y P 2 D 6 may therefore be difficult to observe using this marker activity because other enzymes may compensate for a loss o f activity and prevent a further decrease in dextrorphan formation. It could be argued that the lack o f effect after preincubation o f human liver microsomes with cimetidine was due to a flaw in the methodology used for these experiments. However, two positive controls for mechanism-based inactivation, triacetyloleandomycin and 1-aminobenzotriazole, both demonstrated that this type o f inhibition could be observed under the same experimental conditions used for the cimetidine preincubation experiments. O n the other hand, these two inhibitors are relatively potent, and the experimental conditions required to observe the same type o f inhibition with a potentially less potent inhibitor, such as cimetidine, may be quite different. The lack o f effect with 120 minutes o f preincubation suggests that, i f cimetidine is capable o f acting as a mechanism-based inactivator, the turnover o f the inhibitor to the inactivating species is very slow. This is consistent with in vivo observations, where the degree o f P450-mediated biotransformation o f cimetidine is a relatively minor pathway o f metabolism (Mitchell et al. 1982) and a period o f several days is required to achieve maximal inhibition following administration o f cimetidine to patients (Reimann et al. 1981). Therefore, it is possible that this in vitro system is not capable o f supporting the required enzyme activity for the duration o f time needed to produce a sufficient amount o f the inactivator to cause significant loss o f enzyme activity. Incongruously, preincubation with cimetidine for long periods o f time with microsomes from individuals with high C Y P 2 D 6 activity appeared to cause an increase in activity compared to the appropriate controls. After preincubation for 120 minutes, the activity in the cimetidine-treated groups 103 was 25% to 43% higher than the control groups (figures 19 and 20). Al though this could be interpreted as enzyme activation, the absolute activity in the cimetidine-treated groups is not actually increasing with time. Instead, this apparent increase in activity was due to a greater decrease in the control activity with than that o f the cimetidine-treated groups. The failure o f the enzyme activity to decrease in the inhibitor-treated group could have occurred because cimetidine was having a stabilizing effect on C Y P 2 D 6 or could have been due to consumption o f excess N A P D H by metabolism o f cimetidine, therefore minimizing the effect o f N A D P H on lipid peroxidation. This last point cannot be supported based on the current experimental results as the metabolism o f cimetidine was not monitored. However, it is unlikely to be a factor as P450-mediated metabolism is only a minor elimination pathway for cimetidine and the amount o f N A P D H consumed during this process would be very small. It is important to note that these experiments were conducted only once with microsomes from each individual, and this effect needs to be further characterized before conclusions can be drawn. When higher substrate concentrations were used, the formation o f the second metabolite, 3-methoxymorphinan, could be measured. This metabolite has been used as a probe o f C Y P 3 A activity, although its lack o f selectivity has been documented (Wang and Undakat 1999). It has been suggested that C Y P 3 A may be responsible for a maximum o f only 50% o f dextromethorphan N-demethylase activity and maximal inhibition by triacetyloleandomycin was 50% using concentrations that inhibit testosterone 6(->-hydroxylase activity almost completely (Schmider et al. 1997). When triacetyloleandomycin was used as a positive control for mechanism-based inhibition in this study, an effect o f similar magnitude was seen. Although preincubation with cimetidine appeared to have a slight, but significant, effect on inactivation o f dextromethorphan N-demethylase activity, it is difficult to conclude whether cimetidine is in fact acting as a mechanism-based inactivator o f C Y P 3 A due to both the magnitude o f the effect and the lack o f specificity o f the probe substrate. I f this effect is to be 104 further investigated, a more specific probe o f C Y P 3 A , such as testosterone 6p-hydroxylase activity, should be used, in order to be more confident that the monitored activity is primarily due to C Y P 3 A . The estimated K m for the recombinant enzyme was similar to that obtained with human liver microsomes, but the V m a x was much higher, leading to a greater than sixty-fold increase in the intrinsic clearance o f dextromethorphan by the recombinant enzyme compared to hepatic microsomes. This large difference in the maximal rate o f the reaction between the two model systems is expected for a number o f reasons. First, with human liver microsomes, the amount o f C Y P 2 D 6 present represents approximately 2% o f the total P450 content(Smith et al. 1998); when the recombinant enzyme is used, however, the enzyme concentration is much higher than would normally be found in the human liver. A l s o , i n the recombinant system, the ratio o f the accessory protein N A P D H P450-reductase to C Y P 2 D 6 is different than would be found under physiological conditions, and this can affect the activity o f a given P450 enzyme in different preparations (Remmel and Burchell 1993). In recombinant systems the proportion o f the reductase can be increased through co-expression o f this protein with the P450 and this has been found to result in an elevation o f the maximal activity o f the P450 enzyme (Nakajima et al. 2002, Yamazaki et al. 2002). Finally, differences in the osmolarity o f the buffer systems used with the two different enzyme preparations can affect maximal activity (Remmel and Burchell 1993). The activity o f enzymes produced using different expression systems can also vary considerably. Recombinant P450 enzymes expressed using human lymphoblast cells generally have activities which are more similar to human liver microsomal enzyme activities than are the baculovirus-insect cell-expressed enzymes used in the present study (Gonzalez and Korzekwa 1995). A s was observed with the human liver microsomes, quinidine acted as a competitive inhibitor o f C Y P 2 D 6 . W i t h the recombinant enzyme, however, the K i was an order o f magnitude lower, perhaps reflecting the absence o f contribution o f other enzymes to dextromethorphan O-demethylase 105 activity which are refractory to inhibition by quinidine. Cimetidine, in contrast, .demonstrated rnixed inhibition o f recombinant C Y P 2 D 6 . Preincubation o f the recombinant enzyme with cimetidine resulted in a consistent slight, but significant, increase in inhibition, the magnitude o f which depended on the concentration o f both the substrate and inhibitor used. A preincubation period o f 20 minutes resulted in a further 11% to 36% decrease in dextromethorphan O-demethylase activity. The greatest effect with preincubation was observed when 50 u.M cimetidine and 1 u M dextromethorphan were used in the preincubation and incubation reactions, respectively. A t this substrate concentration, which approximated the K m o f the probe reaction, a small degree o f competitive inhibition by cimetidine was apparent in the absence o f preincubation. The final drop in activity o f 36% was, therefore, due to a combination o f both reversible inhibition and mechanism-based inactivation. By using a higher substrate concentration, the contribution o f reversible inhibition was eliminated. When a saturating substrate concentration was used, the resulting 18% or 11% decrease in activity following preincubation with 100 u.M or 50 u M cimetidine, respectively, for 20 minutes can be attributed to mechanism-based inactivation o f C Y P 2 D 6 . Regardless o f the substrate or inhibitor concentrations used, this further decrease in activity was only seen when the enzyme was preincubated in the presence N A P D H and was dependent on preincubation time, supporting the role o f a reactive metabolite or intermediate species in the inactivation o f C Y P 2 D 6 . The effect o f preincubation on inhibition by cimetidine was also consistent with that seen with 1-aminobenzotriazole, a known mechanism-based inactivator. In contrast, inhibition o f C Y P 2 D 6 by quinidine, a known reversible inhibitor, was not affected by preincubation. In addition to being time- and NAPDH-dependen t , cimetidine demonstrated several other properties o f mechanism-based inactivation, including substrate protection. Dur ing the preincubation reaction, a mechanism-based inactivator is catalytically converted to the reactive species that acts as the enzyme inactivator and, like all enzymatic reactions, this process is susceptible to competitive inhibition. In the present study, quinidine was used as the 'substrate', or competitive inhibitor o f 106 C Y P 2 D 6 , to reduce the formation o f the inactivator species. By increasing the concentration o f quinidine from 0.5- to 2-times the K i value for this inhibitor, the effect o f preincubation with cimetidine was attenuated in a dose-dependent manner. The inactivation o f C Y P 2 D 6 was reduced from 17% in the absence o f quinidine, to 12% in the presence o f the lowest concentration o f quinidine. A t the two highest concentrations o f quinidine, the effect o f preincubation on inactivation o f C Y P 2 D 6 was completely abolished. This further supports the hypothesis that cimetidine is acting as a mechanism-based inactivator o f C Y P 2 D 6 , using this model system. Preincubation o f the recombinant enzyme with cimetidine increased the potency o f the inhibitor, as was reflected in a 2.5- to 4-fold decrease in K . Without preincubation, cimetidine appeared to be acting as a mixed inhibitor o f recombinant C Y P 2 D 6 ; following preincubation, it was more difficult to determine which model o f inhibition provided the best fit for the data. When lower dextromethorphan concentrations were used (approximately 0.5-, 1-, and 2-times the K m ) , there was little difference between the fit o f the competitive, mixed and non-competitive models. The effect o f a mechanism-based inactivator on enzyme activity would be expected to be similar to that o f non-competitive, reversible inhibitors, which cause a decrease in the maximal rate o f the inhibited reaction. Fol lowing binding o f the inactivator species, the affected enzyme is rendered catalytically inactive and the total number o f enzyme molecules available to catalyze the reaction o f interest is decreased, thus resulting in a lower V m a x . The enzyme molecules which are not inactivated still function normally so that the binding o f the substrate is not affected and the K m is unaltered. In this case, the decrease in enzyme activity should be apparent even at saturating substrate concentrations. When the K» determination was repeated under these conditions, the kinetics module was unable to find an appropriate fit for the data, as was demonstrated in figure 36A. This suggests that the effect o f cimetidine on recombinant C Y P 2 D 6 is too complex to be fit to a simple model o f inhibition. This could be due to an overlap between the concentrations o f cimetidine which cause competitive 107 inhibition and mechanism-based inactivation. Activity o f C Y P 2 D 6 could, therefore, be affected by more than one inhibitory mechanism simultaneously. Inhibition and inactivation, however, are two distinct process, and a Ki value does not provide the best description o f the latter process. The effectiveness o f a mechanism-based inactivator is more appropriately defined by the maximal rate o f formation o f the inactivator species, or ki n a ct, and the concentration o f inhibitor required to produce a half-maximal rate o f inactivation (Ki). The k;nact and K i o f cimetidine were estimated to be 0.03 min- 1 and 19.2 u.M, respectively. Compared to other well characterized mechanism-based inactivators, cimetidine is relatively weak. Fo r example, furafylline, a selective and potent mechanism-based inactivator o f C Y P 1 A 2 (Newton et al. 1995), is characterized by a kmactof 0.3 min- 1 and a K i o f 18 [iM (Clarke etal. 1994). The modest effect o f preincubation on dextromethorphan O-demethylase activity can be partially explained by its long half-life o f inactivation. F r o m the Ki t z -Wi l son plot in figure 38, the time required to inactivate 50% o f the enzyme is approximately 25 minutes, which is longer than the preincubation periods used in these studies. A s was seen with human liver microsomes, control activity decreased with increasing preincubation times. W i t h the recombinant enzyme, this effect was much more pronounced and occurred more rapidly: by 20 minutes, approximately 50% o f control activity was lost. Fo r this reason, the preincubation period could not be prolonged. Presumably, i f the preincubation period could be extended, a greater effect would be observed. The most striking difference between the two model systems used in the present study was the elevated catalytic efficiency o f the recombinant enzyme compared to human liver microsomes for dextromethorphan O-demethylation. This observation may partially explain the discrepancy between the mechanisms o f inhibition o f cimetidine observed in the two model systems. In order to act as a mechanism-based inactivator, a compound must be catalytically converted to a reactive species by the enzyme that is subsequently inactivated. Using the recombinant enzyme, the present study provided 108 several key pieces o f evidence to indicate that cimetidine was converted into a reactive species which, in turn, was acting as a mechanism-based inactivator o f C Y P 2 D 6 : preincubation with the inhibitor and N A D P H for a finite period o f time resulted in a decrease in enzyme activity that was apparent even in the presence o f high substrate concentrations capable o f blocking any competitive inhibition; competitive inhibition o f C Y P 2 D 6 during the preincubation reaction blocked the effect o f preincubation on inactivation by cimetidine; and the rate o f inactivation o f C Y P 2 D 6 could be estimated. This final piece o f information indicated that a preincubation period o f approximately 25 minutes was required to inactivate 50% of the enzyme. Thus, the preincubation conditions used were insufficient to observe the maximal effect o f cimetidine on recombinant C Y P 2 D 6 activity, despite the relatively high rate o f turnover o f the recombinant enzyme. It is not surprising that the same effect was not seen when cimetidine was preincubated with human liver microsomes. In this case, the relative intrinsic clearance o f dextromethorphan was approximately 60-fold smaller than that seen with the recombinant enzyme; therefore, the rate o f turnover o f cimetidine would also be expected to be considerably slower. G iven that optimal preincubation conditions could not be achieved using the recombinant system, it is even less likely that these conditions could be established with the microsomal model. The second major difference between the two in vitro model' systems involves the specificity o f the probe reaction used to evaluate enzyme activity. In the case o f the recombinant enzyme, the formation o f dextrorphan is obviously exclusively due to CYP2D6-mediated dextromethorphan O-demethylation. However, with human liver microsomes, several P450 enzymes have been shown to be involved in this reaction. A s a result, the measured dextromethorphan O-demethylase activity may reflect the activity o f two or more P450 enzymes, with the contribution o f each enzyme dependent on the substrate and inhibitor concentrations used. The effect o f an inhibitor on a single enzyme is therefore difficult to isolate using human liver microsomes and is dependent on the experimental conditions. The inability to observe the same effect o f preincubation with cimetidine on recombinant 109 and microsomal C Y P 2 D 6 activity may have been a result o f this. W i t h human liver microsomes, other enzymes may have been compensating for loss in activity o f C Y P 2 D 6 . Additionally, in order to maintain maximal selectivity o f the probe reaction for C Y P 2 D 6 , low substrate concentrations had to be used. However, mechanism-based inactivation is best observed when saturating substrate concentrations are used, as was seen with the recombinant enzyme. The conditions under which the microsomal studies were conducted were a balance between maximizing the specificity o f the probe reaction and optimizing the conditions under which mechanism-based inactivation could be observed and the optimal experimental conditions for observing each o f these conditions were compromised. A s a result, it is unlikely that the effect o f preincubation with cimetidine on the activity o f the recombinant enzyme could be mimicked using the microsomal system. The in vitro experiments carried out in this study indicate that cimetidine is a weak reversible inhibitor, and potentially a mechanism-based inactivator, o f C Y P 2 D 6 . The results described here are inadequate to fully explain the in vivo effect o f cimetidine on the clearance o f CYP2D6-metabol ized drugs noted in a number o f clinical studies, based on a maximal plasma concentration o f cimetidine o f 10 u M . The concentration o f the inhibitor at the active site(s) o f the affected enzyme(s), however, is unknown and it is possible that cimetidine may be concentrated within the liver compared to plasma. Furthermore, the clinical observation o f inhibited drug metabolism is dependent not only on the concentration o f the inhibitor, but also that o f the substrate whose biotransformation is impaired. I f the substrate is present at a low concentration relative to the K m o f the reaction pathway, it w i l l be more susceptible to inhibition. Further in vivo experimentation is therefore required in order to fully understand the mechanism(s) underlying the clinical observations. 110 Chapter 5 F U T U R E S T U D I E S 5.1 Studies w i t h h u m a n liver mic rosomes One possible explanation for the lack o f effect o f preincubation on inhibition by cimetidine is that the contribution o f P450 enzymes other than C Y P 2 D 6 to the formation o f dextrorphan is compensating for a loss in activity o f C Y P 2 D 6 . In order to overcome this, the preincubation studies could be repeated with the inclusion o f inhibitors (chemical or antibodies) o f these other enzymes, such as C Y P 2 C 9 , which is the major low-affinity enzyme involved in dextrorphan formation (von Moltke et al. 1998). Assuming that optimal conditions o f inhibition could be achieved, resulting in complete inhibition o f these other enzymes, the observed dextromethorphan O-demethylase activity should be exclusively due to C Y P 2 D 6 . A l so , these experimental conditions would allow the use o f higher substrate concentrations, thus isolating the effect o f enzyme inactivation from that o f competitive inhibition Alternatively, a different substrate could be used to measure C Y P 2 D 6 activity. Fo r example, the clearance o f both desipramine and nortriptyline, two C Y P 2 D 6 substrates, has been shown to be affected by cimetidine in vivo (Miller et al. 1983, Steiner and Spina 1987). Although dextromethorphan was initially selected as the probe substrate in this study due to its relatively high selectivity for C Y P 2 D 6 , perhaps using a substrate with which cimetidine had already been demonstrated to interact would have been a more appropriate choice. In the present study, there was some evidence to suggest that C Y P 3 A may be inhibited by cimetidine. Cimetidine has been shown to inhibit C Y P 3 A 4 in vitro (Martinez et al. 1999), but again, the 111 effect o f preincubation has not been studied. In order to more fully investigate this, a more selective probe o f C Y P 3 A activity should be used. The 6p-hydroxylation o f testosterone has been demonstrated to be a selective marker o f C Y P 3 A activity. The same types o f experiments conducted in this study, including determination o f the K i o f cimetidine and the effect o f preincubation on this value, could be repeated for this activity, using both human liver microsomes and the recombinant enzyme. These experiments would also provide information on the specificity o f cimetidine as a mechanism-based inactivator. 5.2 Studies w i t h recombinan t C Y P 2 D 6 Al though there were various indications that cimetidine was acting as a mechanism-based inactivator o f recombinant C Y P 2 D 6 , the effect o f preincubation on the observed model o f inhibition was inconsistent with this type o f inhibition. Wi th preincubation, one would expect a shift from competitive to non-competitive inhibition; however, with cimetidine, the kinetics modelling software had difficulty fitting any model to the data, but the competitive and mixed inhibition models were found to fit better than the non-competitive model. It would be interesting to determine the effect o f preincubation on the observed model o f inhibition o f quinidine. Preincubation was shown to have no effect on the potency o f this inhibitor, therefore, after preincubation, the competitive model should still provide the best fit. A l so , the same comparison could be made with 1-aminobenzotriazole, which demonstrated mechanism-based inactivation o f C Y P 2 D 6 . In this case, non-competitive inhibition should be observed. The results from the studies described in this thesis suggest that the experimental conditions were not optimal for observing mechanism-based inactivation using the recombinant model. Based on the estimated half-life o f inactivation o f 25.3 minutes, the maximal preincubation time used did not cover even one half-life. However, due to the instability o f the recombinant enzyme, the preincubation period cannot be prolonged using the experimental conditions described in this study. However, it has 112 been shown that the use o f antioxidants, such as a-methyldopa, in the preincubation reaction results in an attenuation o f the rapid loss in enzyme activity with increasing preincubation times (Veres et al. 2001). Al though this study was conducted using rat hepatic microsomes, the same type o f effect would be expected with recombinant enzymes. Before including an additional compound in the incubation reaction, a series o f preliminary experiments would have to be conducted to ensure that the activity o f the enzyme was not affected. I f the conditions under which mechanism-based inactivation occurs could be optimized through the use o f an antioxidant, it would then be worthwhile to repeat the determination o f the kinact, because use o f a longer preincubation time would presumably provide a more accurate prediction o f this value. Furthermore, the other characteristics o f a mechanism-based inactivator that were not addressed in this study could be investigated. The major properties that were not demonstrated include the irreversible nature o f the inactivation, the type o f complex formed between the reactive species and the enzyme, and inactivation o f the enzyme prior to the release o f the reactive species from the active site. The first point can be investigated by treating the enzyme with potassium ferricyanide or by determining whether dialysis o f the complexed enzyme restores activity. This would also provide evidence as to whether the reactive species covalently or tightly binds to the enzyme. I f activity can be restored following treatment with either o f the two methods described above, this would suggest that a metabolite-intermediate complex is being formed, while a lack o f effect with these agents would indicate covalent modification o f the enzyme. The formation o f a metaboHte-intermediate complex could also be monitored spectrophotometrically as a time-dependent increase in a peak near 420 nm. T o evaluate whether the reactive species inactivates the enzyme before being released from the active site, trapping agents such as glutathione or N-acetylcysteine can be used in the preincubation reaction. These compounds can effectively inactivate reactive electrophilic species, and would therefore be able to decrease the effect o f preincubation on enzyme inactivation in a concentration-dependent manner i f the reactive species were released before binding to the enzyme. 113 5.3 Studies with recombinant CYP2D6*10 Finally, to address the initial hypothesis presented in section 1.6 regarding the susceptibility o f C Y P 2 D 6 * 1 0 compared to the wild-type enzyme, the experiments conducted with recombinant C Y P 2 D 6 presented in this thesis could be repeated using the variant enzyme. Briefly, the kinetic constants o f dextromethorphan O-demethylation would have to be determined to allow for the rational selection o f substrate concentrations for subsequent experiments. The effect o f both quinidine and cimetidine on the activity o f C Y P 2 D 6 * 1 0 could then be determined and estimates o f the K i values for each o f these inhibitors could be obtained and compared to the K ; values already determined for the wild-type enzyme. The effect o f preincubation could then be investigated to determine whether the mechanism-based inactivation observed with the wild-type enzyme could also be observed with C Y P 2 D 6 * 1 0 . Under the present experimental conditions, however, it is unlikely that such inactivation would be observed with the C Y P 2 D 6 * 1 0 variant. The rate o f turnover o f the variant enzyme is lower than that o f the wild-type enzyme used in the present study and, given the small decrease in activity seen following preincubation with the wild-type enzyme, being able to reproducibly measure the even smaller decrease expected with C Y P 2 D 6 * 1 0 would be difficult. The use o f appropriate controls in the preincubation experiments, such as 1-aminobenzotriazole and quinidine, would be important in such a situation. 114 Chapter 6 SUMMARY & CONCLUSIONS 1. Under the experimental conditions used, cimetidine acted as a competitive inhibitor, but not a mechanism-based inactivator, of dextromethorphan O-demethylation in human liver microsomes. The estimated K i for competitive inhibition was higher than the peak plasma concentration o f cimetidine observed following clinical administration o f the drug. Therefore, competitive inhibition alone cannot explain the drug-drug interactions observed with cimetidine and C Y P 2 D 6 substrates. It is possible that the appropriate conditions could not be achieved to observed mechanism-based inhibition in these hepatic microsome samples. 2. Using recombinant CYP2D6, cimetidine was shown to act as both a mixed inhibitor and a mechanism-based inactivator. Following preincubation of the recombinant enzyme with cimetidine and N A D P H , the potency of the inhibitor was increased. This effect was relatively weak, resulting in a further 15% loss of activity or a 2.5- to 4-fold decrease in the K;, but could potentially be increased i f the preincubation period could be extended beyond 20 minutes. 3. The type of inhibition seen is dependent on the model system and experimental conditions used. The large difference in absolute C Y P 2 D 6 activity between the human liver microsomes and recombinant system may explain the differences in the observed mechanisms of inhibition of 115 cimetidine. The presence or absence of a preincubation step and the concentration of substrate used can also affect the observed inhibition. Based on the present study, an appropriate in vitro model was not found to explain the effect of cimetidine on C Y P 2 D 6 activity in vivo. 116 Chapter 7 R E F E R E N C E S Abernethy D R , Greenblatt D J , Shader RI. 1984. Imipramine-cimetidine interaction: impairment of clearance and enhanced bioavailability. Journal of Pharmacology and Experimental Therapeutics 229:702-705. Amsterdam J D , Brunswick D J , Potter L , Kaplan MJ . 1984. Cimetidine-induced alterations in desipramine plasma concentration. Psychopharmacology 83:373-375. Balani S K , Zhu T, Yang TJ , L iu Z , He B , Lee FW. 2002. Effective dosing regimen of 1-aminobenzotriazole for inhibition of antipyrine clearance in rats, dogs, and monkeys. Drug Metabolism and Disposition 30:1059-1062. Bavin P M G , Post A , Zarembo J E . 1984. Cimetidine. Analytical Profiles of Drug Substances 13:127-183. Benedetti M S , Bani M . 1998. Design of in vitro studies to predict in vivo inhibitory drug-drug interactions. Pharmacological Research 38:81-88. Bodemar G , Norlander B , Walan A . 1981. Pharmacokinetics of cimetidine after single doses and during continuous treatment. Clinical Pharmacokinetics 6:306-315. Bondy SC, Naderi S. 1994. Contribution of hepatic cytochrome P450 systems to the generation of reactive oxygen species, biochemical Pharmacology 48:155-159. Brosen K , Gram L F . 1989. Clinical significance of the sparteine/debrisoquine oxidation polymorphism. European Journal of Clinical Pharmacology 36:537-547. Bourrie M , Meunier V , Berger Y , Fabre G . 1996. Cytochrome P450 isoform inhibitors as a tool for the investigation of metabolic reactions catalyzed by human liver microsomes. Journal of Pharmacology and Experimental Therapeutics 277:321-332. Brogden R N , Heel R C , Speight T M , Avery GS. 1978. Cimetidine: a review of its pharmacological properties and therapeutic efficacy in peptic ulcer disease. Drugs 15:93-131. Busby W F , Ackermann J M , Crespi C L . 1998. Effect of methanol, ethanol, dimethyl sulfoxide, and acetonitrile on in vitro activities of cDNA-expressed human cytochromes P450. Drug Metabolism and Disposition 27:246-249. 117 Capon D A , Bochner F, Kerry N , Mikus G , Danz C, Somogyi A A . 1996. The influence of C Y P 2 D 6 polymorphism and quinidine on the disposition and antitussive effect of dextromethorphan in humans. Clinical Pharmacology and Therapeutics 60:295-306. Causon R. 1997. Validation of chromatographic methods in biomedical analysis: viewpoint and discussion. Journal of Chromatography B 689:175-180. Chang T, Levine M , Bandiera S M , Bellward G D . 1992A. Selective inhibition of rat hepatic microsomal cytochrome P450 I. Effect of the in vivo administration of cimetidine. Journal of Pharmacology and Experimental Therapeutics 260:1441-1449. Chang T, Levine M , Bellward G D . 1992B. Selective inhibition of rat hepatic microsomal cytochrome P450 II. Effect of the in vitro aciministration of cimetidine. Journal of Pharmacology and Experimental Therapeutics 260:1450-1455. Chang T K H , Gonzalez FJ, Waxman DJ . 1994. Evaluation of triacetyloleandomycin, a-naphthaflavone, and diethyldithiocarbamate as selective chemical probes for inhibition of human cytochromes P450. Archives of Biochemistry and Biophysics 311:437-442. Ching M A , Blake C K , Ghabrial H , Ellis SW, Lennard MS, Tucker G T , Smallwood R A . 1995. Potent inhibition of yeast-expressed C Y P 2 D 6 by dihydroquinidine, quinidine, and its metabolites. Biochemical Pharmacology 50:833-837. Cholerton S, Daly A K , Idle JR. 1992. The role of individual human cytochromes P450 in drug metabolism and clinical response. Trends in Pharmacological Science 13:434-439. Clarke SE , Ayrton A D , Chenery RJ. 1994. Characterization of the inhibition of P4501A2 by furafylline. Xenobiotica 24:517-526. Colby ELD, Abbott B , Cachovic M , Debolt K M , Mico B A . 1995. Inactivation of adrenal cytochromes P450 by 1-aminobenzotriazole. BiochemicalPharmacology 49:1057'-1062. Copeland R A . 1996. Enzymes: A practical introduction to structure, mechanism, and data analysis. V C H Publishers Inc, New York. Crespi C L . 1995. Xenobiotic-metabolizing human cells as tools for pharmacological and toxicological research. Advances in Drug Research 26:179-235. Crespi C L , Penman B W . 1997. Use of cDNA-expressed human cytochrome P450 enzymes to study potential drug-drug interactions. Advances Pharmacology 43:171-188. Crespi C L , M i l l e r V P . 1999. The use of heterologously expressed drug metabolizing enzymes - state of the art and prospects for the future. Pharmacology and Therapeutics 84:121-131. Dalen P, Dahl M L , Ruiz M L B , Nordin J , Bertilsson L . 1998. 10-hydroxylation of nortriptyline in white persons with 0, 1, 2, 3, and 13 functional CYP2D6 genes. Clinical Pharmacology and Therapeutics 63:444-452. Dayer P, Leemann T, Striberni R. 1989. Dextromethorphan O-demethylation in liver microsomes as a prototype reaction to monitor cytochrome P-450 dbi activity. Clinical Pharmacology and Therapeutics 45:34-40. 118 Distlerath L M , Reilly P E B , Martin M V , Davis G G , Wilkinson G R . 1985. Purification and characterization of the human liver cytochromes P-450 involved in debrisoquine 4-hydroxylation and phenacetin 0-demethylation, two prototypes for genetic polymorphism in oxidative drug metabolism. Journal of Biological Chemistry 260:9057-9067. Ducharme J , Abdullah S, Wainer IW. 1996. Dextromethorphan as an in vivo probe for the simultaneous determination of C Y P 2 D 6 and C Y P 3 A activity. Journal of Chromatography B 678:113-128. Ellis SW, Hayhurst G P , Smith G , Lightfoot T, Wong M M S . 1995. Evidence that aspartic acid 301 is a critical substrate-contact residue in the active site of cytochrome P450 2D6. Journal of Biological Chemistry 270:29055-29058. Feldman M , Burton M E . 1990. I-hstamine2-receptor antagonists: standard therapy for acid-peptic diseases (first of two parts). The New England Journal of Medicine 323:1672-1680. Fukada T, Yamamoto I, Nishida Y , Zhou Q, Ohno M . 1999. Effect of the CYP2D6*10 genotype on venlafaxine pharmacokinetics in healthy adult volunteers. British Journal of Clinical Pharmacology 47:450-453. Fukuda T, Nishida Y , Imaoka S, Euroi T, Naihara M . 2000. The decreased in vivo clearance of C Y P 2 D 6 substrates by CYP2D6*10 might be caused not only by the low-expression but also by low affinity of C Y P 2 D 6 . Archives of Biochemistry and Biophysics 380:303-308. Furuta S, Kamada E , Suzuki T, Sugimoto T, Kawabata Y , Shinozaki Y , Sano H . 2001. Inhibition of drug metabolism in human liver microsomes by nizatidine, cimetidine and omeprazole. Xenobiotica 31-1-10. Garcia-Barcelo M , Chow L Y , Chiu H F K , Wing Y K , Lee D T S . 2000. Genetic analysis of the CYP2D6 locus in a Hong K o n g Chinese population. Clinical Chemistry 46:18-23. Gonzalez FJ. 1989. The molecular biology of cytochrome P450s. Pharmacological Previews 40:243-288. Gonzalez FJ, Korzekwa K R . 1995. Cytochromes P450 expression systems. Annual Reviews of Pharmacology . and Toxicology 35:369-390. Gonzalez FJ, Nebert D W . 1990. Evolution of the P450 superfamily. Trends in Genetics 6:182-186 Gorski J C , Jones D R , Wrighton SA, Hall SD. 1994. Characterization of dextromethorphan N-demethylation by human liver microsomes. Biochemical Pharmacology 48:173-182. Griese E U , Zanger U M , Brudermanns U , Gaedigk A , Mikus G . 1998. Assessment of the predictive power of genotypes for the in vivo catalytic function of C Y P 2 D 6 in a German population. Pharmacogenetics 8:15-26. Guengerich FP. 2001. Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicology. Chemical Research in Toxicology 14:611-650. Guengerich FP , Muller-Enoch D , Blair IA. 1986. Oxidation of quinidine by human liver cytochrome P450. Molecular Pharmacology 30:287-295. HalpertJR. 1995. Structural basis of selective cytochrome P450 inhibition. Annual Review of Pharmacology and Toxicology. 35:29-53. 119 Halpert JR, Guengerich FP , Bend JR, Correia M A . 1994. Selective inhibition of cytochromes P450. Toxicology and Applied Pharmacology 125:163-175. Hasler J A , Estabrook R, Murray M , Pikuleva I, Waterman M . 1999. Human cytochromes P450. Molecular Aspects of Medicine 20:1-137. Henhauer SA, Hollister L E . 1984. Cimetidine interaction with imipramine and nortriptyline. Clinical Pharmacology and Therapeutics 35:183-187. Ingelman-Sundberg M , Oscarson M , McLeilan R A . 1999. Polymorphic human cytochrome P450 enzymes: an opportunity for individualized drug treatment. Trends in Pharmacological Science 20:342-349. Ingelman-Sundberg M , Evans W E . 2001. Unravelling the functional genomics of the human C Y P 2 D 6 gene locus. Pharmacogenetics 11:553-554. Ito K , Iwatsubo T, Kananitsu S, Nakajima Y , Sugiyama Y . 1998. Quantitative prediction of in vivo drug clearance and drug interactions from in vitro data on metabolism, together with binding and transport. Annual Reviews of Pharmacology and Toxicology 38:461-499. Johansson I, Oscarson M , Yue Q Y , Bertilsson L , Sjoqvist F. 1994. Genetic analysis of the Chinese cytochrome P450 locus: characterization of variant CYP2D6 genes present in subjects with (diminished capacity for debrisoquine hydroxylation. Molecular Pharmacology 46:452-459. Jones D R , Gorski J C , Hamman M A , Hal l SD. 1996. Quantification of dextromethorphan and metabolites: a dual phenotypic marker for cytochrome P450 3A4/5 and 2D6 activity. Journal of Chromatography B 678:105-111. Kalow W. 1968. Pharmacogenetics in animals and man. Annals of the NY Academy of Science 151:694-698. Kalow W. 1971. Topics in pharmacogenetics. Annals of the NY Academy of Science 179:654-659. Karnes H T , Shiu G , Shah V P . 1991. Validation of bioanalytical methods. Pharmaceutical Research 8:421-426. Kerry N L , Somogyi A A , Bochner F, Mikus G . 1994. The role of C Y P 2 D 6 in primary and secondary oxidative metabolism of dextromethorphan: in vitro studies using human liver microsomes. British Journal of Clinical Pharmacology 38:243-248. Kimura S, Umeno M , Skoda R C , Meyer U A , Gonzalez FJ. 1989. The human debrisoquine 4-hydroxylase (CYP2D) gene locus: sequence and identification of the polymorphic CYP2D6 gene, a related gene, and a pseudogene. American Journal of Human Genetics 45:889-904. Knodel l R G , Browne D G , Gwozdz G P , Brian W R , Guengerich FP. 1991. Differential inhibition of individual human liver cytochromes P450 by cimetidine. Gastroenterology 101:1680-1691. Kobayashi S, Murray S, Watson D , Sesardic D , Davis D S , Boobis A R . 1989. The specificity of inhibition of debrisoquine 4-hydroxylase activity by quinidine and quinine in the rat is the inverse of that in man. Biochemical Pharmacology 38:2795-2799. Kubota T, Yamaura Y , Ohkawa N , Hara H , Chiba K . 2000. Frequencies of CYP2D6 mutant alleles in a normal Japanese population and metabolic activity of dextromethorphan O-demethylation in different CYP2D6 genotypes. British Journal of Clinical Pharmacology 50:31-34. 120 Lam Y W F , Rodriguez SY. 1993. High-performance liquid chromatography determination of dextromethorphan and dextrorphan for oxidation phenotyping by fluorescence and ultraviolet detection. Therapeutic Drug Monitoring 15:300-304 Levin W, L u A Y H , Jacobson M , Kuntzman R. 1973. Lip id peroxidation and the degradation of cytochrome P450 heme. Archives of Biochemistry and Biophysics 158:842-852. Levine M , Bellward G D . 1995. Effect of cimetidine on hepatic cytochrome P450: evidence for formation of a metabolite-intermediate complex. Drug Metabolism and Disposition 23:1407-1411. Levine M , Law E Y W , Bandiera S M , Chang T K H , Bellward G D . 1998. In vivo cimetidine inhibits hepatic C Y P 2 C 6 and CYP2C11 but not C Y P 1 A 1 in adult male rats. Journal of Pharmacology and Experimental Therapeutics 284:493-499. Lewis D F V , Dickins M , Eddershaw PJ, Tarbit M H , Goldfarb PS. 1999. Cytochrome P450 substrate specificities, substrate structural templates and enzyme active site geometries. Drug Metabolism & Drug Interactions 15:1-49. L in J H , L u A Y H . 1998. Inhibition and induction of cytochrome P450 and the clinical implications. Clinical Pharmacokinetics 35:361-390. Lindstrom T D , Hanssen B R , Wrighton SA. 1993. Cytochrome P450 complex formation by dirithromycin and other macrolides in rat and human livers. Anitmicrobial Agents and Chemotherapy 37:265-269. Mahgoub A , Idle JR, Dring L G , Lancaster R, Smith R L . 1977. Polymorphic hydroxylation of debrisoquine in man. Lancet 2:584-586. Marez D , Legrand M , Sabbagh N , L o Guidice J M , Spire C. 1997. Polymorphism of the cytochrome P450 C Y P 2 D 6 gene in a European population: characterization of 48 mutations and 53 alleles, their frequencies and evolution. Pharmacogenetics 7:193-202. Martinez C, Albet C , Agundez J A G , Herrero E , Carrillo J A , Marquez M , Benitz J , Oritz J A . 1999. Comparative in vitro and in vivo inhibition of cytochrome P450 C Y P 1 A 2 , C Y P 2 D 6 , and C Y P 3 A by H2-receptor antagonists. Clinical Pharmacology and Therapeutics 65:369-376. Miller D D , Macklin M . 1983. Cimetiaine-imipramine interaction: a case report. American Journal of Psychiatry 140:351-352. Miller D D , Sawyer JB , Duffy JP. 1983. Cimetidine's effect on steady-state serum nortriptyline concentrations. Drug Intelligence and Clinical Pharmacy 17:904-905. Mitchell SC, Idle JR, Smith R L . 1982. The metabolism of [1 4C] cimetidine in man. Xenobiotica 12:283-292. Murray M . 1984. In vitro effects of quinoline derivatives on cytochrome P450 and aminopyrine N-demethylase activity in rat hepatic microsomes. Biochemical Pharmacology 33:3277-3281. Murray M . 1987. Mechanisms of the inhibition of cytochrome P450-mediated drug oxidation by therapeutic agents. Drug Metabolism Reviews 18:55-81. Murray M . 1997. Drug-mediated inactivation of cytochrome P450. Clinical and experimental pharmacology and physiology 24:465-470. 121 Nakajima M , Nakamura S, Tokudome S, Shimada N , Yamazaki H . 1999. Azelastine N-demethylation by cytochrome P450 (CYP)3A4, C Y P 2 D 6 , and C Y P 1 A 2 in human liver microsomes: evaluation of approach to predict the contribution o f multiple CYPs . Drug Metabolism and Disposition 27:1381-1391. Nakajima M , Tane K , Nakamura S, Shimada N , Yamazaki H , Yoko i T. 2002. Evaluation of approach to predict the contribution of multiple cytochrome P450s in drug metabolism using relative activity factor: effects of the differences in expression levels of NAPDH-dytochromeP450 reductase and cytochrome bs in the expression system and the differences in the marker activities. Journal of Pharmaceutical Sciences 91:952-963. Nelson D R , Koymans L , Kamataki T, Stegeman JJ, Feyereisen R, Waxman D J , Gotoh O, Coon M J , Estabrook R W , Gunsalus E C , Nebert D W . 1996. P450 superfamily: update on new sequences, gene mapping accession numbers and nomenclature. Pharmacogentics 6:1-42. Newton D J , Wang R W , L u A Y H . 1995. Evaluation of specificities in the in vitro metabolism of therapeutic agents by human liver microsomes. Drug Metabolism and Disposition 23:154-158. Omura T. 1999. Forty years of cytochrome P450. Biochemical and Biophysical Research Communications 266: 690-698. Oritz de Montellano PR, Mathews J M . 1981. Autocatalytic alkylation of the cytochrome P450 prosthetic haem group by 1-aminobenzotriazole. Biochemistry Journal 195:761-764. Oritz de Motellano PR, Correia M A . 1995. Inhibition of Cytochrome P450 enzymes. From Oritz de Monellano, P R (ed). Cytochrome P450: structure, Mechanism, and Biochemistry (Second Edition), Plenum Press, N e w York. Pelkonen O , Puurunen J. 1980. The effect of cimetidine on in vitro and in vivo microsomal drug metabolism in the rat. Biochemical Pharmacology 29:3075-3080. Pessayre D , Larrey D , Vitaux J , Breil P, Belghiti J , Benhamou JP. 1982. Formation of an inactive cytochrome P450 Fe(II)-metabolite complex after administration of troleandomycin in humans. Biochemical Pharmacology 31:1699-1704. Phillip P A , James C A , Rogers HJ . 1989. The influence of cimetidine on debrisoquine 4-hydroxylation in extensive metabolizers. European Journal of Clinical Pharmacology 36:319-321. Potkin S G , Shen Y , Pardes H , Phelps B H , Zho D , 1984. Haloperidol concentrations elevated in Chinese pateints. Psychiatry Research 12:167-172. Puurunen J , Sotaniemi E , Pelkonen O. 1979. Effect of cimetidine on microsomal drug metabolism in man. European Journal of Clinical Pharmacology 18:185-187. Ramamoorthy Y , Tyndale R F , Sellers E M . 2001. Cytochrome P450 2D6.1 and cytochrome P450 2D6.10 differ in catalytic activity for multiple substrates. Pharmacogenetics 11:477-487. Rashba-Step J , Cederbaum A L 1994. Generation of reactive oxygen intermediates by human liver microsomes in the presence of N A P D H or N A D H . Molecular Pharmacology 45:150-157. Reimann IW, Klotz U , Siems B, Frolich JC. 1981. Cimetidine increases steady state plasma levels of propranolol. British Journal of Clinical Pharmacology 12:785-790. 122 Remmel R P , Burchell B . 1993. Validation and use of cloned, expressed human drug-metabolizing enzymes in heterologous cells for analysis of drug metabolism and drug-drug interactions. Biochemical Pharmacology 46:559-566. Rendic S, Sunjic V , Kajfez F, Ruf H H . 1979. Interactions of cimetidine with liver microsomes. Xenobiotica 9:555-564. Ren ton K W , Aranda J V , Eade N R . 1976. NADPH-dependent lipid peroxidation and its effects on aminopyrine N-demethylation in subcellular fractions of human neonatal liver. Canadian Journal of Physiology and Pharmacology 56:838-843. Rodrigues A D . 1994. Use of in vitro human metabolism studies in drug development. Biochemical Pharmacology 48:2147-2156. Rudorfer M V , Lane E A , Chang W H , Zhang M , Potter W Z . 1984. Desipramine pharmacokinetics in Chinese and Caucasian volunteers. British Journal of Clinical Pharmacology 17:433-440. Sachse C, Brockmoller J , Bauer S, Roots I. 1997. Cytochrome P450 2D6 variants in a Caucasian population: allele frequencies and phenotypic consequences. American Journal of Human Genetics 60:284-295. Sachse C, Brockmoller J , Hildebrand M , Muller K , Roots I. 1998. Correctness of prediction of the C Y P 2 D 6 phenotype confirmed by genotyping 47 intermediate and poor metabolizers of debrisoquine. Pharmacogenetics 8:181-185. Schellens J H M , van der Wart J H F , Brugman M , Breimer D D . 1989. Influence of enzyme induction and inhibition of the oxidation of nifedipine, sparteine, mephenytoin and antipyrine in humans as assessed by a "cocktail" study design. Journal of Pharmacology and Experimental Therapeutics 249:638-645. Schmider J , Greenblatt D J , Fogelman S M , von Moltke L L , Shader R L 1997. Metabolism of dextromethorphan in vitro: involvement of cytochromes P450 2D6 and 3A3/4 , with a possible role of 2E1. Biopharmaceutics & Drug Disposition 18:227-240. Serlin M J , Sibeon R G , Mossman S, Breckenridge A M . 1979. Cimetidine: interactions with oral anticoagulants in man. Eancet 2:317-319. Shah V P , Midha K K , Dighe S, McGilveray IJ, Skelly JP, Yacobi A , Layloff T, Viswanathan C T , Cook C E , McDowel l E D , Pittman K A , Spector S. 1992. Analytical methods validation: bioavailability, bioequivalence, and pharmacokinetic studies. Pharmaceutical Research 9:588-592. Shimada T, Yamazaki H , Mimura M , Inui Y , Guengerich FP. 1994. Interindividual variations in human liver cytochrome P450 enzymes involved in the oxidation of drugs, carcinogens and toxic chemicals: studies with liver microsomes of 30 Japanese and 30 Caucasians. Journal of Pharmacology and Experimental Therapeutics 270:414-423 Shimada T, Tsumura F, Yamazaki H . 1999. Prediction of human liver microsomal oxidations of 7-ethoxycoumarin and chlorzoxazone with kinetic parameters of recombinant cytochrome P450 enzymes. Drug Metabolism and Disposition 27:1274-1280. Silverman R B . 1995. Mechanism-based enzyme inactivators. Methods in En^ymology 249:240-283. Smith G , Stubbins M J , Harries L W , Wol f CR. 1998. Molecular genetics of the human cytochrome P450 monooxygenase superfamily. Xenobiotica 28:1129-1165 123 Somogyi A , Gugler R. 1982. Drug interactions with cimetidine. Clinical Pharmacokinetics 7:23-41. Somogyi A , Gugler R. 1983. Clinical Pharmacokinetics o f cimetidine. Clinical Pharmacokinetics 8:463-495. Steiner E , Spina E . 1987. Differences in the inhibitory effect of cimetidine on desipramine metabolism between rapid and slow debrisoquin hydroxylators. Clinical Pharmacology and Therapeutics 42:278-282. Tanaka E . 1999. Update: genetic polymorphism of drug metabolizing enzymes in humans. Journal of Clinical Pharmacy and Therapeutics 24:323-329. Tateishi T, Chida M , Ariyoshi N , Mizorogi Y , Kamataki T. 1999. Analysis of the CYP2D6 gene in relation to dextromethorphan O-demethylation capacity in a Japanese population. Clinical Pharmacology and Therapeutics 65:570-575. Testa B. 1990. Mechanisms of inhibition of xenobiotic-metabolizing enzymes. Xenobiotica 20:1129-37. Tingle M D , Coleman M D , Park B K . 1991. The effect of preincubation with cimetidine on the N-hydroxylation of dapsone by human liver microsomes. British Journal of Clinical Pharmacology 32:120-123. Van der Weide J , Steijns L S W . 1999. Cytochrome P450 enzyme system: genetic polymorphisms and impact on clinical pharmacology. Annals of Clinical Biochemistry 36:722-729. Venkatakrishnan K , von Moltke L L , Court M H , Harmatz JS, Crespi C L . 2000. Comparison between cytochrome P450 (CYP) content and relative activity approaches to scaling from cDNA-expressed CYPs to human liver microsomes: ratios of accessory proteins as sources of discrepancies between the approaches. Drug Metabolism and Disposition 28:1493-1504. Veres A , Szakacs T, Vereczkey L . 2001. Effect of a-methyldopa on pentoxyresorufin O-delakylation in liver microsomes from rats treated with phenobarbital. Archives of Biochemistry and Biophysics 302:59-64. V o n Moltke L L , Greenblatt D J , Grassi J M , Granada B W , Venkatakrishnan K , Schmider J , Harmatz JS, Shader R l . 1998. Multiple human cytochromes contribute to biotransformation of dextromethorphan in vitro: role of CYP2C9 , CYP2C19, C Y P 2 D 6 , and C Y P 3 A . Journal of Pharmacy and Pharmacology 50:997-1004. Wang SL, Huang J D , La i M G , L iu B H , Lai M L . 1993. Molecular basis of genetic variation in debrisoquine hydroxylation in Chinese subjects: polymorphism in R F L P and D N A sequence of CYP2D6. Clinical Pharmacology and Therapeutics 53:410-418. Wang Y , Unadkat J D . 1999. Enzymes in addition to C Y P 3 A 4 and 3A5 mediate N-demethylation of dextromethorphan inhuman liver microsomes. Biopharmaceutics & Drug Disposition 20:341-346. Yamazaki H , Nakamura M , Komatsu T, Ohyama K , Hatanaka N , Asahi S, Shimada N , Guengerich FP , Shimada T, Nakajima M , Yokoi T. 2002. Roles of N A P D H - P 4 5 0 reductase and apo-and holo-cytochrome bs on xenobiotic oxidations catalyzed by 12 recombinant human cytochrome P450s expressed in membranes of Escherichia coli. Protein Expression and Purification 24:329-337. Y o o n Y R , Cha IJ, Shon J H , K i m K A , Cha Y N . 2000. Relationship o f paroxetine disposition to metoprolol metabolic ration and CYP2D6*10 genotype of Korean subjects. Clinical Pharmacology Therapeutics 67:567-576. 124 Y u A , Haining R L . 2001. Comparative contribution to dextromethorphan metabolism by cytochrome P450 isoforms in vitro: can dextromethorphan be used as a dual probe for both C Y P 2 D 6 and C Y P 3 A activities? Drug Metabolism and Disposition 29:1514-1520. 125 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0091077/manifest

Comment

Related Items