Citation
Synthesis and in vitro Evaluation of Cytochrome P450 Isoform-Selective Substrate Analogs as Exhaled-Breath Markers of Enzyme Competency

Material Information

Title:
Synthesis and in vitro Evaluation of Cytochrome P450 Isoform-Selective Substrate Analogs as Exhaled-Breath Markers of Enzyme Competency
Creator:
WANG XIAOLI ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Aldehydes ( jstor )
Cytochromes ( jstor )
Enzymes ( jstor )
In vitro fertilization ( jstor )
Incubation ( jstor )
Mass spectra ( jstor )
Metabolism ( jstor )
Metabolites ( jstor )
Phenotypes ( jstor )
Urine ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Xiaoli Wang. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/31/2007

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

SYNTHESIS AND IN VITRO EVALUATION OF CYTOCHROME P450 ISOFORMSELECTIVE SUBSTRATE ANALOGS AS EXHALED-B REATH MARKERS OF ENZYME COMPETENCY By XIAOLI WANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Xiaoli Wang

PAGE 3

This document is dedicated to my parents and my family.

PAGE 4

iv ACKNOWLEDGMENTS I would like to express my a ppreciation and grateful thanks to Dr. Laszlo Prokai for his intelligent guidance and generous suppor t throughout my Ph.D. program. I also would like to thank the members of my superv isory committee, Dr. Kenneth Sloan, Dr. Margaret O. James, and Dr. Richard J. Me lker, for their valuable advice and time throughout my doctoral research. I would like to express my gratitude to Dr. Katalin Ta trai-Prokai and Dr. Alevtina Zharikova for their technical support and expert opinion. I would like to extend my thanks to the secretaries of both the Department of Medicina l Chemistry at the University of Florida and the Department of Molecu lar Biology and Immunology at the University of North Texas Health Science Center for th eir support. I also would like to thank lab members and graduate students of the depa rtment for their help and friendship. Most of all, I give special thanks to my family for their support and encouragement in many ways.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABBREVIATIONS.............................................................................................................x ABSTRACT.......................................................................................................................xi CHAPTER 1 INTRODUCTION........................................................................................................1 Drug Metabolism..........................................................................................................1 CYP Enzymes...............................................................................................................3 Phenotyping..................................................................................................................7 2 RESEARCH DESIGN AND PRELIMINARY STUDIES........................................11 Current Phenotyping Methods....................................................................................11 CYP 2D6.............................................................................................................11 CYP 3A4.............................................................................................................14 Research Design.........................................................................................................15 Preliminar y Studies.....................................................................................................18 Detection of Cyclopropane Ca rboxaldehyde by Breath Assays..........................18 Introduction..................................................................................................18 Experimental................................................................................................18 Results and discussion..................................................................................20 Detection of Fluorinated Aldehyde from Metabolism of a Fluoroalkylcontaining Drug...............................................................................................23 Introduction..................................................................................................23 Experimental................................................................................................24 Results and discussion..................................................................................27 Hypothesis..................................................................................................................31 3 SYNTHESIS AND METABOLIC STUDI ES OF CYP 2D6 SUSTRATE ANALOGUE: TRIFLUOROETHYLDEXTRORPHAN...........................................32

PAGE 6

vi Introduction.................................................................................................................32 Materials and Methods...............................................................................................34 Materials..............................................................................................................34 Synthesis of O-Trifl uoroethyldextrorphan..........................................................35 CYP 2D6 Inhibition Assays................................................................................38 Human Liver Microsome Assays........................................................................38 Instrument............................................................................................................40 Quantitation.........................................................................................................40 Results........................................................................................................................ .41 Inhibition of CYP 2D6 by Dextromethor phan and Trifluoroethyldextrorphan..41 O-Dealkylation of Dextromethorphan and Trifluoroethyldextrorphan...............41 Determination of Trifluoroacetaldehyde.............................................................44 Discussion...................................................................................................................45 4 SYNTHESIS AND METABOLIC STUDI ES OF CYP 3A4 SUSTRATE ANALOGUE: TRIFLUOROETHYLNORVERAPAMIL.........................................52 Introduction.................................................................................................................52 Materials and Methods...............................................................................................54 Materials..............................................................................................................54 Synthesis of N-Trifluoroethylnorverapamil........................................................54 CYP 3A4 Inhibition Assays................................................................................55 Human Liver Microsome Assays........................................................................55 Instrument............................................................................................................57 Quantitation.........................................................................................................57 Results........................................................................................................................ .58 Inhibition of CYP 3A4 by Verapam il and Trifluoroethylnorverapamil..............58 N-Dealkylation of Verapamil a nd Trifluoroethylnorverapamil..........................58 Determination of Trifluoroacetaldehyde.............................................................61 Discussion...................................................................................................................61 5 CONCLUSION...........................................................................................................67 APPENDIX A MS AND 1H-NMR SPECTRA...................................................................................68 B CALIBRATION CURVES........................................................................................75 LIST OF REFERENCES...................................................................................................76 BIOGRAPHICAL SKETCH.............................................................................................90

PAGE 7

vii LIST OF TABLES Table page 3-1 Heat of reaction of dextromethorphan and trifluoroalkyldextrorphan analogues....51 4-1 Heat of reaction of verapamil a nd trifluoroalkylnorverapamil analogues...............66

PAGE 8

viii LIST OF FIGURES Figure page 1-1 General catalytic cycle for P450 catalyzed reactions.................................................4 2-1 O-Demethylation of de xtromethorphan by CYP2D6...............................................12 2-2 N-Demethylation of verapamil by CYP3A4............................................................18 2-3 Metabolism of naltr exone produces CPCA..............................................................19 2-4 Reaction of a carbonyl compound with DNP H to form the hydrazone derivative..20 2-5 Chromatograms from LC/MS/MS analys is (left) and product ion mass spectra (right) of A) CPCA standard, B) patie nt’s breath before naltrexone and C) patient’s breath after naltre xone after DNPH derivatization....................................22 2-6 Flecainide and its two main metabolites MODF and MODLF................................24 2-7 Synthesis of trifluoroacetaldehyde DNPH adduct...................................................28 2-8 Apparatus for trapping trifluoroacetal dehyde. A: incubation mixture; B: DNPH cartridge; C: syringe.................................................................................................28 2-9 Synthesis of trifluoroacetaldehyde [15N4] DNPH. All Ns are 15N labeled...............28 2-10 SIM chromatograms (A and B) and fu ll scan mass spectra (C and D) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubati on of flecainide........................................................................30 3-1 Metabolic pathways for dextromethorphan..............................................................33 3-2 Synthesis of O-trif luoroethyldextrophan..................................................................34 3-3 Inhibition of CYP 2D6 activities of AMMC by increasing concentration of dextromethorphan and trifluor oethyldextrorphan from 10-10 M to 10-5 M. Activities are given as percen t of control rates (no inhi bitor). Data are given as means of duplicate determinations...........................................................................42 3-4 Chromatograms (A and B) and full scan mass spectra (C and D) for dextrorphan and ethylmorphine, respectively..............................................................................43

PAGE 9

ix 3-5 Proposed electron impact mass fragmentation of dextrorphan................................44 3-6 Time-dependent dextrorphan forma tion from both dextromethorphan and its analogue trifluoroethyldextrorphan. Results are expressed as means SD (n=3)...44 3-7 SIM chromatograms (A and B) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubation of trifluoroethyldextrorphan. For full scan mass spectra, see Figure 2-10 (C and D)..45 3-8 Proposed mechanism for O-demethylation..............................................................49 4-1 Metabolism scheme of verapamil. Th e asymmetric carbons are marked with asterisks....................................................................................................................53 4-2 Synthesis of N-trif luoroethylnorverapamil..............................................................53 4-3 Inhibition of CYP 3A4 activities of DBF by increasing concentration of verapamil and trifluoroethylnorverapamil from 10-10 M to 10-5 M. Activities are given as percent of control rates (no i nhibitor). Data are given as means of duplicate determinations..........................................................................................59 4-4 Chromatograms (A and B) and fu ll scan mass spectra (C and D) for norverapamil and d -propranolol, respectively.........................................................60 4-5 SIM chromatograms (A and B) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubation of trifluoroethylnorverapamil. For full s can mass spectra, see Figure 2-10 (C and D)............................................................................................................................. .62 4-6 Proposed mechanism for N-demethylation..............................................................65 A-1 MS and 1H NMR of dextrorphan.............................................................................68 A-2 MS and 1H NMR of N,O-bisVOC morphinan........................................................69 A-3 MS and 1H NMR of 3-hydroxyN-VOC morphinan................................................70 A-4 MS and 1H NMR of 3-(2,2,2-trifluoroe thyl)-N-VOC morphinan............................71 A-5 MS and 1H NMR of O-trifluoroe thyl dextrorphan...................................................72 A-6 1H NMR of 2,2,2-trifluoroethyl-p-t oluenesulfonate (upper) and 2,2,2trifluoroethyl triflate (lower)....................................................................................73 A-7 MS and 1H NMR of trifluoroethyl norverapamil.....................................................74 B-1 Calibration curve for dextrorphan............................................................................75 B-2 Calibration curve for norverapamil..........................................................................75

PAGE 10

x ABBREVIATIONS AMHC: 3-[2-(N, N-diethyl-N-methylammoni um)ethyl]-7-hydroxy-4-methylcoumarin AMMC: 3-[2-(N, N-diethyl-N-methylammoni um)ethyl]-7-methoxy-4-methylcoumarin APCI: Atmospheric pressure chemical ionization CI: Chemical ionization CPCA: Cyclopropane carboxaldehyde DBF: Dibenzylfluorescein DMF: N,N-Dimethylformamide DNPH: 2,4-Dinitrophenylhydrazine EI: Electron impact ionization GC/MS: Gas chromatography mass spectrometry 1H NMR: Proton nuclear magnetic resonance HPLC: High-performance liquid chromatogrpahy. LC/MS: Liquid chromatography mass spectrometry MRM: Multiple reaction monitoring NADP+: Nicotinamide adenine dinucleotide phosphate SIM: Selected ion monitoring TLC: Thin-layer chromatography UV: Ultraviolet VOC: Vinyloxycarbonyl group

PAGE 11

xi Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHESIS AND IN VITRO EVALUATION OF CYTOCHROME P450 ISOFORMSELECTIVE SUBSTRATE ANALOGS AS EXHALED-B REATH MARKERS OF ENZYME COMPETENCY By Xiaoli Wang August 2006 Chair: Laszlo Prokai Major Department: Medicinal Chemistry People respond differently to drugs. Some (poor metabolizers) could develop serious adverse effects at normal doses, wh ile others (extensive metabolizers) may require a higher dose to achieve the expected therapeutic response. Inherited differences in drug metabolism play an important role in the variable response to drug therapies. Identification of individuals’ phenotype (extensive metabolizers versus poor metabolizers) will allow physicians to i ndividualize drug therapy to maximize drug efficacy and minimize drug toxicity. This knowle dge is also valuable to drug discovery and development. Direct phenotyping with a probe drug gives the end result of pharmacogenetic differences between people. However, current phenotyping techniques are not easily accessible to clinical settings because they are labor intensive and time consuming. Our objective is to find a si mple and convenient method to phenotype patients with respect to enzyme competency.

PAGE 12

xii Cytochrome P450 enzymes are important Phase I drug metabolizing enzymes. Among them, CYP 2D6 is the most polymorphic enzyme which accounts for 25% of drug metabolism; CYP3A4 is the most abunda nt enzyme involved in the metabolism of more than 50% of currently used drugs. We will target these two enzymes as examples. Based on metabolism of dextromethorphan (CYP 2D6 substrate) and verapamil (CYP 3A4 substrate), we designed new anal ogues by introducing the trifluoroethyl group to dextrorphan and norverapamil (demethyl ated metabolites for dextromethorphan and verapamil, respectively). Our hypothesis wa s that these new analogues would produce a trifluorinated aldehyde which c ould be excreted into the brea th. Since breath sampling is noninvasive and convenient compared to trad itional urine and blood sampling methods, it will facilitate phenotyping patie nts about their enzyme activity in the clinic. In this proof-of-concept study, we synthesized the s ubstrate analogues trifluoroethyldextrorphan (for CYP 2D6) a nd trifluoroethylnorver apamil (for CYP 3A4), and investigated their in vitro metabolism using human liver microsomes. Analysis was conducted by GC(EI)/MS for the metabolite s dextrorphan and norverapmil, and by GC(CI)/MS for trifluoroacetaldehyde. Due to the electron-withdrawing effects of fluorine atoms, both analogues showed more potent inhib itory effect than the original substrate drugs and were less metabolized by enzyme s. However, trifluoroacetaldehyde was detectable in both cases, as we hypothesized. To overcome this slow metabolism of the designed analogues and generate more trifluorinated aldehyde as a breath marker, elongation of the alkyl chain containing the terminal trifluoromethyl group was proposed to make them better enzyme substrates.

PAGE 13

1 CHAPTER 1 INTRODUCTION There are often large differences among i ndividuals in the way they respond to drugs, varying from potentially life-threatening adverse drug reactions at one end of the spectrum to a lack of desired therapeutic eff ect at the other end. Ma ny factors contribute to the interpatient variability, such as the nature of the disease, the individual’s age and race, organ function, concomitant therapy, drug interactions, and concomitant illnesses. However, an individual’s genetic makeup can have an even greater influence on the efficacy and toxicity of drugs (Evans and Johnson, 2001). This pharmacogenetic variation can be either pharmac okinetic (variation in the leve l of the active substance at the site of its action), or pharmacodynamic (qua ntitative/qualitative va riation in response of the target) in nature. The former is caused by genetic polymorphism in drug metabolizing enzymes (DMEs) or drug transp orters, while the latter is by genetic polymorphism in drug targets: receptors, enzy mes, etc. (Obach et al., 1997; Vesell, 1997). In this dissertation, emphasis will be put on pharmacogenetics associated with DMEs. Drug Metabolism Metabolism is the biochemical modificat ion of chemical compounds in living organisms and cells. This includes the bi osynthesis of complex organic molecules (anabolism) and their breakdown (catabolism). Metabolism usually consists of sequences of enzymatic steps called metabolic pathways (Wikimedia Foundation, 2006). Drug metabolism often involves a metabolic pathwa y by which a lipophilic drug is converted

PAGE 14

2 to a more hydrophilic metabolite which is readily excreted by the kidney. This biotransformation is important since it ca n affect other pharmacokinetic processes (absorption, distribution and excretion) by altering the physicochemical properties of drugs, including lipophilicity, solubility, binding to proteins and excr etion rates (Leahy et al., 1989; Seydel and Schaper, 1986). It also a ffects pharmacological properties of drugs. Although in most cases, metabolism leads to in activation of drugs, sometimes it leads to the generation of active metabolites, which can be solely or partially responsible for pharmacological response. For example, acetaminophen, an O -deethylated metabolite of phenacetin, shows superior analgesic activity when compared with phenacetin (Lin and Lu, 1997). In some instances, metabolic tr ansformation can cause toxicity through formation of haptens that can initiate systemic hypersensitivity or immune responses, production of reactive intermediates that either adduct to tissue macromolecules or initiate destructive free radical chain r eactions, and depletion of endogenous compounds that act as intracellular and extracellular antioxidants (Leyland-Jones, 2004). Another important aspect of drug metabolism is the po tential for drug-drug in teractions resulting from enzyme inhibition or i nduction because DMEs have br oad substrate specificities (Kumar and Surapaneni, 2001). There are two categories of reactions cat alyzed by DMEs: Phase I and Phase II. Phase I reactions are functional reactions which usually introduce a polar functional group into a parent molecule to form a meta bolite. The most common Phase I functional reactions include, but are not limited to, oxidation, re duction, hydrolysis, hydroxylation and deamination. Phase II reactions are usua lly conjugation reactions that conjugate a polar moiety to the parent compound or its ph ase I metabolite. Therefore, the metabolites

PAGE 15

3 resulting from Phase II are usually much more polar than the parent molecule, and are readily excreted from urine or/and bile. Th e common conjugation reactions for drugs are glucuronidation, sulphation, glutathione conj ugation, methylation and acetylation (Kumar and Surapaneni, 2001). Drug metabolizing enzymes have been classified according to the type of reaction they catalyze and the cofactors involved. The major classes of Phase I enzymes include cytochrome P450 (CYP) enzymes, flavin-con taining monooxygenases (FMO), esterases, and amidases. The major classes of Phase II enzymes include UDPglucuronyltransferases (UGT), sulphotransfe rases (SUT), glutathione-S-transferases (GST) and N-acetyltransferases (NAT). DMEs are mainly located in the liver, although other organs (such as lung, kidney and intes tine) also contain metabolizing enzymes (Lin et al., 2003). CYP Enzymes Among these DMEs, CYP enzymes are the most important for numerous xenobiotics (e.g., drugs, pollutants, and di etary components) as well as endogenous substrates (e.g., bile acids, steroids, and cholesterol). They mediate several biotransformations including hydroxylation of aliphatic and aromatic carbon, epoxidation of aromatic or olefinic double bond, heteroatom oxidation, dealkylation, and dehydrogenation (Wrighton and Stevens, 1992; Rendic and Di Carlo, 1997). Their catalysis is completed through reduction of cytochrome P450 by NADPH cytochrome P450 reductase. Figure 1-1 shows the gene ral catalytic cycle for P450 catalyzed reactions. NADPH cytochrome P450 reductase takes two electrons from NADPH. These electrons go to reduce two flavins sequentia lly, FAD and FMN, contained in separate domains in P450 reductase. FMN then tran sfers the reduction potential to the Fe3+ in the

PAGE 16

4 heme ring of cytochrome P450. The oxygen molecule binds to a site adjacent to the iron, creating an unstable state. To increase stability, the oxygen is autooxidized to O2 and released as activated oxygen. Next, the transf er of the second elec tron to oxygen creates O2 2-, which attracts protons in solution to form water. The activated oxygen created with the first electron transfer can react with a drug (substrate) to a dd a hydroxyl group, thus completing monooxygenation (Guengerich, 2004). Figure 1-1. General catalytic cy cle for P450 catalyzed reactions. CYP superfamily members are divided into various families based on amino acid sequence homology (more than 40 % identical amino acid sequen ces in the same family). The drug metabolizing CYP enzymes are confined to families 1, 2, 3, and 4. These families are further divided into subfamilies according to the similarity of amino acid sequences of the encoded CYP isoforms (mor e than 55% identical amino acid sequences in the same subfamily)(Chang and Kam, 1999; Nelson et al., 1993). Major isoforms

PAGE 17

5 involved in the metabolism of drugs in humans are CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and CYP3A4. They are primarily located in the endoplasmic reticulum of hepatocytes in the liver a nd account for more than 90% of oxidative reactions in drug metabolism. Several CYP isoforms are associated with genetic polymorphism. The reason behind the polymorphism is the alteration in nucleotide sequence of the genes coding the DMEs. It could be single nuc leotide polymorphism (SNP), in sertion or deletions in the nucleotide sequence, or amplif ication/rearrangement of gene s. SNP is the most common process (Srivastava, 2003). CYP 2D6 was the first CYP enzyme to demonstrate polymorphic expression in humans. Although it only constitutes about 2% of the total CYP 450 enzymes, it participates in about 25% of drug metabolis m (Michalets, 1998; Shimada et al., 1994). There are three metabolic phenotypes: poor (PM), extensive (EM) and ultraextensive (UEM). PM is caused by the homozygous pres ence of defect allele s leading to a total absence of active enzyme and an impaired ab ility to metabolize pr obe drugs specific for this drug-metabolizing enzyme (Gaedigk et al., 1991; Kagimoto et al., 1990). UEM is caused by occurrence of duplicate d, multiduplicated or amplified CYP2D6 genes. At present, alleles with two, three, four, five and 13 gene copies in tandem have been reported and the number of individuals carrying multiple CYP2D6 gene copies is highest in Ethiopia and Saudi Arabia, where up to one third of the population displays this phenotype. A total of more th an 30 different defective CYP2D6 and 55 CYP2D6 variations have been identified to date (Ingelman-Sundberg et al., 1999). Examples of alleles with normal, wild type func tion are CYP2D6*1, CYP2D6*1, CYP2D6*2A and

PAGE 18

6 CYP2D6*2B. Alleles resulting in an absence of function are CYP2D6*3, CYP2D6*4A, CYP2D6*4B, CYP2D6*5, CYP2D6*6A, CYP2D6*6B, CYP2D6*7, CYP2D6*8, CYP2D6*11, CYP2D6*12. Alleles resulting in a reduced function are CYP2D6*9, CYP2D6*10A, CYP2D6*10B. Raci al and ethnic studies of drug metabolism have shown substantial inter-population differences in the poly morphic distribution of CYP2D6 activity. This polymorphism has been extensiv ely studied in Caucasians and Orientals with results consistently showing a prev alence of PMs of 5-10% in Caucasians (Europeans and white North Americans) and 1% in Orientals (Chinese, Japanese and Koreans) (Lou, 1990; Kalow, 1997). The ultraex tensive metabolisers are reported with a prevalence of 1.5-29% in different ethnic groups. The frequency of the CYP2D6 gene duplication was found to be 2-3% among most European populations and a proportion of 12% in Turkish subjects. The carriers of gene duplication in Saudi Arabia and Ethiopia are 21% and 29% respectively (McLella n et al., 1997; Akillu et al., 1996). CYP3A4 is the most abundant enzyme in the human liver and accounts for 30% of total CYP protein. CYP3A4 has broad substr ate specificity and is estimated to be involved in the metabolism of more than 50% of drugs used in humans (Rendic and Di Carlo, 1997; De Wildt et al., 1999). To date, thre e mutant alleles have been identified for the CYP 3A4 gene ( CYP3A4*1B , CYP3A4*2 , and CYP3A4*3 ). The allelic frequency for the CYP3A4*1B allele, which contains an A (-290) G substitution in the promoter region of CYP3A4 , ranges from 0% in Chinese and Japa nese Americans to >54% in African Americans. American and European Caucas ians were reported to have an allelic frequency of ~4%. The CYP3A4*2 allele, which encodes a Ser222Pro change, has an allelic frequency of 2.7% in the white (Finni sh) population (Ball et al ., 1999; Sata et al.,

PAGE 19

7 2000). Because variant alleles that are found in >1% of th e population are defined as genetic polymorphisms, both the CYP3A4*1B and the CYP3A4*2 allele are considered to be genetic polymorphisms of CYP3A4 . The CYP3A4*3 allele, which has a Thr1473Cys change that produces a Met445Thr substi tution in exon 12, was detected in only 1 Chinese subject and has been thought of as a rare allele. Recent studies found that CYP3A4*3 has an allelic frequency of 1.1% in Caucasians. This implies that the variant CYP3A4*3 allele is not a rare allele, but instead represents a geneti c polymorphism that can be found in a substantial part of the population (Van Schaik et al., 2001). Phenotyping The existence of enzymatic polymorphism contributes to mark ed inter-individual and inter-ethnic variability in human dr ug metabolism which accounts for different responses to drug therapy. Slow metabolizers have high plas ma drug concentration which may lead to significant side effects at nor mal dose, while fast metabolizers have low plasma concentration and may have no eff ect at normal dose. For drugs with narrow therapeutic indices or drugs which require bi oactivation (e.g. codeine, prodrugs), this polymorphism can be critical (Nebert, 1997) . Therefore, individualized drug therapy based on each patient’s inherited ability to me tabolize, eliminate, a nd respond to specific medications is necessary to maximize dr ug efficacy and minimize drug toxicity. The knowledge of an individual’s phenotype allows a clinician to determine a safe and therapeutically effective drug treatment re gimen because the phenotype is correlated to an individual’s susceptibility to toxic ch emicals and diseases. This knowledge is also valuable to the drug development process. Many drugs are eliminated in clinic trials or withdrawn from the market due to toxicities that only affect a segment of the individuals in a target group. However, these drugs may be reconsidered for others who can safely

PAGE 20

8 metabolize them and have good therapeutic eff ects. Screening individuals with safe and effective drug metabolism will lead to the decrease in the number of individuals needed to participate in a dr ug treatment testing trial and, thus, decrease cost and allow the drug to reach the market faster (Leyland-Jones, 2004). Currently, the determination of an indi vidual’s phenotype for a given metabolic enzyme can be performed either via di rect metabolic phenotyping or indirect extrapolation of an individual’s ge notype to a given phenotype (Gonzales and Idle, 1994). Phenotyping involves the use of a probe substrate known to be metabolized by a given enzyme. After administering the probe to an individual to be phenotyped, a biological sample (e.g., blood, urine) is co llected and analyzed for metabolites corresponding to the probe. The rate of its metabolism is calculated and is used to determine a metabolic phenotype (Linder et al., 1997). Genotyping involves identificati on of defined genetic mutati on that gives rise to the specific drug metabolism phenotype. These muta tions include genetic alterations that lead to overexpression (gene amplification) , absence of an active protein product (null allele), or production of a mutant protein with diminished catalytic capacity (inactivating allele). DNA isolated from peripheral lymphocytes can be used for genotyping. Two commonly used methods in genotyping are the PCR-RFLP method and allele-specific PCR. In the former technique, a specific regi on of the gene of inte rest is amplified by PCR followed by digestion of the amplifie d DNA product with restriction endonucleases. The size of the digestion products is easily evaluated by agarose gel electrophoresis with ethidium bromide staining and UV tran sillumination. In allele specific PCR amplification, oligonucleotides specific fo r hybridizing with the common or variant

PAGE 21

9 alleles are used for parallel amplification r eactions. Analysis for the presence or absence of the appropriate amplified product is accomplished by agarose gel electrophoresis (Heim and Meyer, 1990; Sachse et al., 1997; Daly et al., 1999). Indirect phenotyping may be limited by seve ral factors that can result in an alteration in the theoretical phenotype. For example it has been well established that genotype does not always correlate with phe notype. Likewise gene expression does not always correlate with protein expressi on, and protein expression does not always correlate with protein func tion. Indirect phenotyping fails to account for many factors that affect protein function including but not limited to post-translational protein modification, polypharmacy, and exposure to inducers or inhibitors (Leyland-Jones, 2004). Furthermore, other limitations include the potential complexity of performing a complete genotyping. The mutation sequence must first be identified before it can be examined in a genotyping assay. Subsequent to identification, th e mutation must be linked to a definitive effect on phenotype. For so me enzymes, there appear to be very few mutations and those found have been well characterized, while for other enzymes multiple mutations are present with new mutations found regularly (e.g., CYP2D6 has over 53 mutations and 48 allelic variants). Therefore, while genotyping for CYP2C19 might be performed with relatively few measurements, a complete and accurate genotyping of CYP2D6 would be comple x and require multiple measurements (Weinshilboum and Wang, 2004). Phenotyping that determines the presence and activity of a particular metabolic enzyme is often referred to as functio nal phenotyping. It gives the end result of pharmacogenetic differences between people. However, current phenotyping techniques

PAGE 22

10 are not easily accessible to clinical settings because they are labor intensive and time consuming. Our objective is to find a convenien t method to identify a patient’s phenotype and, thus, to facilitate indivi dualized drug therapy: the right drug for the right patient at the right dose.

PAGE 23

11 CHAPTER 2 RESEARCH DESIGN AND PRELIMINARY STUDIES CYP 2D6 and CYP3A4 are the two most im portant CYP isoforms. CYP 2D6 is the most polymorphic isoform, and about 25% drug s are substrates for the CYP2D6 isoform. Over 50 widely used drugs have been identi fied in which CYP 2D6-mediated metabolism is involved in the overall elimination, in cluding many antidepressants, neuroleptics, antiarrhythmics, beta-adrenergic blockers , serotonin 5-HT3-antagonists and opioids (Dalen et al., 1999). CYP 3A4 is the most abundant isoform and is responsible for approximately 30 to 40% of the total CYP c ontent in human liver and small intestine. CYP3A4 is estimated to metabolize more th an 50% of the currently used drugs and contribute to the disposition of more than 60 therapeutically important drugs. Substrates of CYP3A4 can be subject to high first-pass metabolism as it is expressed in abundance in both liver and intestine (Paine et al., 1997; Wrighton et al., 2000). Although the activity of CYP3A4 metabolism is distributed in a unimoda l fashion, a large degree of inter-individual variability has been observ ed. PMs may have an increased risk of developing concentration-related adverse eff ects or reduced therapeutic effect in the situation where an active metabolite is form ed. The reverse situation applies to UEMs. CYP 2D6 and 3A4 enzymes will be targ eted as examples in our design. Current Phenotyping Methods CYP 2D6 Phenotyping individuals for CYP 2D6 ha s been extensively studied since the discovery of the genetic polymorphism of the oxidative metabolism of debrisoquine and

PAGE 24

12 sparteine in the 1970’s (Vlncent-Viry et al ., 1991). Debrisoquine, an antihypertensive drug, was initially used as the standard probe to assess the activity of CYP2D6 and to distinguish between EMs and PMs. However, it is not availabl e for human medicine in all countries and sometimes causes severe adve rse effects (e.g. orthos tatic hypotension in PMs) (zdemir et al., 2004). Today, other CY P2D6 probe drugs have been validated to study CYP 2D6 phenotype in populations and in dividuals. These incl ude sparteine (an oxytocic and an antiarrhythmia agent), codeine (an opioid an algesic), metoprolol (a beta1selective adrenoceptor blocki ng agent) and dextromethor phan (a nonopioid antitussive agent). Metabolic ratio (MR) of a probe dr ug is calculated and used to determine the phenotype of an individual. Different probe s have different MRs. For example, to identify PMs, a urinary MR of 0.3 is used for dextromethorphan (Schmid et al., 1985); 12.6 for debrisoquine (Evans et al., 1980); 20 for sparteine (Eichelbaum and Woolhouse, 1985). In a study comparing three CYP2D6 probe substrates, dextromethorphan was found to be more sensitive than debrisoquine and sparteine in reve aling variant CYP2D6 catalytic activity (Droll e al., 1998). Also dext romethorphan (Figure 2-1) is the substrate used widely for CYP2D6 phenot ypic study in vivo since it is a readily available over-thecounter drug and relatively safe. H3CO NCH3HO NCH3+ HCHOCYP2D6 Figure 2-1. O-Demethylation of dextromethorphan by CYP2D6.

PAGE 25

13 A traditional study was performed by collec ting urine over 8 h to 12 h after a single dose of dextromethorphan (DM, 20-30 mg). DM and its O-demethylated metabolite dextrorphan (DEX) are measured and their c oncentration ratio DM/D EX is calculated. An antimode of 0.3 separates EM and PM (Hou et al., 1996). This phenotyping method, however, is time-consuming because of the long sampling period. Moreover, measurements in urine are susceptible to impaired renal function. The ratio calculated from concentrations in urine over 8 h is a function of both intrinsic clearance of the precursor to products and the renal clear ance of the precursor and the products, respectively. It could, therefore, provide a flawed index of hepatic enzyme activity in individuals with renal impairm ent (Kevorkian et al., 1996). Kohler et al. (1997) found there was lack of correlation (r = 0.364) between urine MR and serum MR when using DM to identify phenotypes of healthy controls and patient s under psychotropic medication. Deconjugation of the urine sample (37C for 18 h) using -glucuronidase is required because DEX is mainly excreted as its glucuronide in urine. Omission of deconjugation resulted in approximately ten times lower recovery of DEX in urine as shown in French, South Indian, and Turk ish populations (Chladek et al., 2000). Therefore, urine assay is time-consuming and complicated. Then blood sample was used as an alternative way for phenotying. Blood was obtained from the antecubital vein a few hours after administration of DM. Serum or heparinized plasma was prepared by centr ifugation of blood at 3000 for 10 min before analysis (Kohler et al. 1997). Compared with the standard urine procedure, serum/plasma assay is faster and more accurate, and the resu lts are likely to come closer to initial velocities of hepatic CYP2D6 activity. However this method is invasive.

PAGE 26

14 Hou et al. (1991) tried salivary sample s to determine CYP 2D6 phenotype for patients with renal failure re quiring hemodialysis. Although sa livary analysis can be used to identify PMs, it is not likely to replace urine collection in phenotyping subjects with normal renal function. It required a large dos e of DM (50 mg) in attempt to ensure detectable salivary concentra tions, which would cause more side effects. In their study, 22 out of 61 complained about discomfort. T echnically, it is more time-consuming and difficult than the urinary assay. DM was wei ghed and placed into empty gelatin capsules. Care has to be taken to prevent contamina tion of the outer surface of the capsule by DM powder. When taken by subjects, capsules mu st not be broken or opened to prevent contamination of saliva by drug. To collect saliva, all particip ants were asked to chew a piece of parafilm, and also some people dis like being asked to collect saliva samples. CYP 3A4 A number of drugs metabolized by CYP 3A4 have been studied in determining CYP3A4 phenotype in vivo. These include midazolam, erythromycin, omeprazole, dextromethorphan, cortisol, lidocaine, nifedi pine, dapsone and al fentanil. However, identification of the ideal probe has been difficult. The most widely accepted and tested CYP 3A4 probes are midazolam and erythromyc in. Others usually lack correlation with these established probe drugs (Streetman et al., 2000). Concentration of midazolam (a reagent for intravenous sedation and the induc tion of anesthesia) and its metabolite 1’hydroxymidazolam in plasma are determined to identify CYP 3A4 phenotype. A single blood sample has not been shown to reliab ly predict CYP 3A4 ac tivity in non-induced subjects (Thummel et al., 1994). The pro cedure requires multiple intravenous blood samples. It is invasive and not convenie nt, which limits this method's large-scale application. The use of urine samples to pr edict in vivo CYP 3A4 activity has not yet

PAGE 27

15 been fully assessed due to the impact of renal CYP 3A5 on various urinary measures (Streetman et al., 2000). The intravenous erythromycin breath test (ERMBTIV) is another method to phenotype CYP 3A4. Erythromycin undergoes N-demethylation to produce formaldehyde, and formaldehyde is further enzymatically convert ed to bicarbonate, which then equilibrates with alveol ar carbon dioxide. Because the ERMBTIV uses [14CN methyl] erythromycin, 14CO2 is liberated in breath, which is measured by liquid scintillation counting (Rivory et al., 2001). In addition to being highly expressed in the liver, CYP3A4 is highly expressed in the small intestine. The ERMBTIV result would, therefore, appear to refl ect only hepatic CYP3A4 activity because intravenous administration should avoid significant metabolism by the gut. The problem for ERMBTIV is the use of radioactive compounds, whic h has safety issues and more care is needed for both patients and analyzers when dealing with them. Accordingly, Paine et al. (2002) developed an oral stable isotope 13C-labeled formulation of the ERMBT (ERMBToral). 13CO2 was determined by use of a Europa Scientific 20/20 gas isotope ratio mass spectrometer. The ratio of 13CO2 to 12CO2 was measured. Because 12CO2 was influenced by food and exercise, large valu es for interday variation were observed. Therefore search for ideal probes is still ongoing. Research Design Our goal is to find an easy and convenient way to collect samples, making enzymes phenotyping accessible in clinical practice. Oxidative dealkylation of drug substrates containing an ether or ami no functional group produces th e corresponding aldehyde in addition to the dealkylated drug (McMa hon, 1966). Volatile compounds can be easily excreted into the breath. Some are already us ed as biochemical markers in the clinic. For example, exhaled ethane is a marker of lipid peroxidation which reflects the damage of

PAGE 28

16 cell membranes caused by reactive oxygen sp ecies; NO and CO are markers of airway inflammation and oxidative stre ss in the study of pathogenesi s and progression of asthma, chronic obstructive respiratory disease and cyst ic fibrosis (Paredi et al., 2002). The lung receives all of the blood fl ow from the right side of the heart and it has been demonstrated that measurements from breat h correlate with blood concentration. Small molecules can penetrate the blood-lung barri er, mix, and appear in the exhaled breath. Specifically, due to the reversibility of the biochemical process involved in the conversion of ammonia and ammonium ions to urea, a reliable correlation between breath ammonia and blood urea nitrogen has b een established (Narasimhn et al., 2001). There are several advantages of using exhaled breath as a sample. Compared to blood sample, it is noninvasive and contains mu ch less cells and proteins. Compared to urine sample, it does not need a long samp ling period and complicated procedure, reducing time for sample preparation. Compared to salivary sample, it is easy to obtain and there is no concern about sample cont amination. Upon analyzing the metabolism of DM, besides the O-demethylated DEX, there is another product, formaldehyde (HCHO). This compound is highly volatile with boiling point -20 C. Our body also produces a variety of sma ll molecule carbonyl compounds in the normal breath, such as formaldehyde, acetaldehyde, acetone, propanal, 2-butanone, butanal, pentanal, hexanal etc. (Lin et al ., 1995). To increase the selectivity and specificity of the breath assay, a product whic h keeps the volatility of formaldehyde and at the same time is not interfered with by existing carbonyl compounds is needed. Trifluorinated aldehydes meet that requireme nt. Their boiling points are essentially the same as that of unfluorinated aldehydes. Trifluoroacetaldehyde is even 40C lower than

PAGE 29

17 that of acetaldehyde (21C) and highly volatile (Henne et al., 1950). Moreover, endogenous carbonyl compounds present in the br eath do not contain fluorine, and they will not interfere w ith the determination of fluorine-containing aldehydes. Since trifluorinated drugs have already been used in the clinic and they can undergo detrifluoroalkylation after me tabolism. For example, flecainide, an antiarrhythmic, is meta-O-detrifluoroethylated by CYP 2D6 (McQ uinn et al., 1984; Tenneze et al., 2002), halazepam and quazepam, sedatives, are meta bolized through N-detrifluoroethylation (Hilbert et al., 1984; Lu et al., 1991). Based on DM, we design a new CYP 2D6 substrate analogue which could produce a trifluorinated aldehyde after metabolism. Therefore, we could identify CYP2D6 phenotype in individuals by using exhaled breath. Verapamil was selected as the CYP 3A4 phenotyping probe in our design. Verapamil is a calcium channel blocker widely used in the treatment of angina pectoris, coronary artery disease, car diac arrhythmias, and hypertensi on. This calcium antagonist is effective in young and old, black and white hypertensive patients and is free of metabolic side effects (Wang et al., 2004). ODemethylation and Ndealkylation are the two major metabolic pathways of verapamil. CYP3As are the primary enzymes involved in Ndealkylation, which produces two major meta bolites, norverapamil (Figure 2-2) and Ndesalkylverapamil (D617) (Kroemer et al., 1992) . Due to the advantages of breath test we mentioned above and generation of the al dehyde after verapam il demethylation, the same strategy of introducing a trifluorinated group will be employed to design a new substrate analogue which could produce trif luorinated aldehyde after metabolism for CYP3A4 phenotyping.

PAGE 30

18 CH3N CN OCH3OCH3H3CO H3CO H N CN OCH3OCH3H3CO H3CO + HCHO CYP 3A4 Figure 2-2. N-Demethylati on of verapamil by CYP3A4. Preliminary Studies Detection of Cyclopropane Carboxaldehyde by Breath Assays Introduction To test the feasibility of our design usi ng exhaled breath to id entify activity of CYP enzymes, we collected breath samples from a patient who was taking naltrexone to treat alcoholism. Naltrexone is a -opioid receptor antagonist used for the treatment of opioid and alcohol dependence in th e clinic. It undergoes extens ive metabolism after oral administration. The major human metabolite 6 -naltrexol is formed by aldo-keto reductase enzymes (AKR1C1, AKR1C2, and AKR1C4) that catalyse stereospecific reduction of naltrexone (Ohara et al., 1995; Breyer-Pfaff and Nill, 2004). Other metabolites include 2-hydroxy-3-O-met hylnaltrexone and 2-hydroxy-3-O-methylnaltrexol. Both naltrexone and metabolites ar e excreted as glucur onide conjugates (Wall et al., 1981). N-dealkylation was also reported which could produce volatile cyclopropane carboxaldehyde (CPCA) (Misra et al., 1976) (Figure 2-3). Detection of CPCA was performed by LC/MS/MS (APCI negative mode) after 2,4dinitrophenylhydrazine(DNPH) deriva tization of breath samples. Experimental Materials. Naltrexone (ReVias) was from Dupont Pharma (Wilmington, DE). Cyclopropane carboxaldehyde and 2,4-dinitr ophenylhydrazine were purchased from Sigma (St. Louis, MO). All solvents were HPLC grade and were obtained from Fisher Scientific (Atlanta, GA).

PAGE 31

19 O O H O OH N H O O H O OH N O O H OH N OH O O OH N O H MeO O OH N O H MeO OH OHC + Figure 2-3. Metabolism of naltrexone produces CPCA. Sample preparation. Breath samples were collected from the patient through breath storage bags before and 30 min afte r taking naltrexone. After adding 50 mL DNPH solution (3.1g/L in 3.6M HCl) to ea ch bag and shaking for 15 min, the DNPH adducts were extracted with the same vol ume of hexane twice. Then hexane was evaporated and sample was dried under N2 for analysis. Synthesis of DNPH derivatized CPCA. DNPH derivatized CPCA standard was synthesized by the following procedure. To th e clear solution obtained by warming 500 mg DNPH, 1 mL HCl and 9 ml ethanol, was added 200 L CPCA and heated just to boiling. The mix was allowed to cool to room temperature, the product was filtered and recrystallised it from ethanol. Instrumentation. A quadrupole ion trap system (L CQ, Thermofinnigan, San Jose, CA) was used in the APCI mode. Negative ions were detected. The s can range of the MS was from 100 to 275 for the determination of m/z values. MS/MS product-ion scans were obtained after collisioninduced dissociation with helium as the target gas. In full-scan

PAGE 32

20 MS/MS mode, m/z 249 was used as the precursor ion for the measurement of CPCA. In multiple reaction monitoring (MRM) 249 163,173,179 and 191 were used to monitor CPCA. The DNPH derivative was separated on a Zorbax SB C-18 column (3 cm 2.1 mm) with acetonitrile/water (70:30) as a m obile phase. The eluent flow rate was 0.2 mL/min. The sample residues were dissolved in 50 L mobile phase, and 5 L was injected for analysis. Results and discussion Because of the high volatility and reactivit y, aldehydes are usually first derivatized before determination (In the absence of deri vatization, sample treatment and storage may introduce very large errors in the results.). Th is not only improves the analytical process, but also fixes the aldehyde concentration at a given time (Nagy et al., 2004). The most common method is derivatization with 2,4dinitrophenylhydrazine (DNPH), since the reaction with DNPH (Figure 24) is typical of carbonyl compounds, very sensitive and rapid. RR' O NO2NO2HN NH2NO2NO2HN N R' R + H2O+ [ H+ ] Figure 2-4. Reaction of a carbonyl compound with DNPH to form the hydrazone derivative. Analysis for aldehyde derivatives is of ten performed using gas chromatography, with flame ionization (Priego-Lopez and Luque de Casto, 2002), mass spectrometric

PAGE 33

21 (Park et al., 1998), and electr on capture detection (Tomita et al., 1990; Ohata et al., 1997). DNPH treated samples were also analyzed by HPLC with a UV detector due to the strong UV absorbance of DNPH derivatized al dehydes at 307 and 356 nm (Lucas et al., 1986; Shara et al., 1992; Cordis et al., 1994). In recent year s, HPLC methods utilizing atmospheric pressure ionization coupled to ma ss spectrometric detection have also been published. Mass spectrometry is more specific than UV, and detection limits may reach the low picogram range (Kolliker et al., 1998; Kempter et al., 1999; Zurek and Karst, 1999). To further increase the sensitivity a nd selectivity, HPLC coupled with tandem mass spectrometry (MS/MS) has been st udied (Nagy et al., 2004). In our study, LC/MS/MS was employed to improve detecti on limits and simplify sample preparation procedure as well. Figure 2-5 shows the chromatograms (left) of DNPH derivatized CPCA standard (A), breath samples before (B) and 30 min after naltrexone (C). We can clearly see the DNPH derivatized CPCA from the patient’s breath, which has a RT 1.21 min. There are four characteristic peaks in the MS/MS of the CPCA derivative: 163, 173, 179 and 191. The breath sample (C) has the same product ion spectra (Figure 2-5, right). This suggests that breath samples could be used to dete ct the compound that we are interested in. Tandem mass spectrometric technique is more selective in identific ation of compounds through comparison of product ions spectra. In this example, detection of CPCA was slightly interfered with by the other components in breath. The method was also sensitive, with the detection limit of CPCA derivative in the level of 10-7 g. Naltrexone has been shown to be effective in the treatment of alcohol dependence. However, a major limitation of its clinical util ity has been poor patient adherence to the

PAGE 34

22 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 173 191 160 120 179 152 203 117 147 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 173 191 160 120 152 203 117 147 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 160 120 152 203 117 147 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 150 172 178 152 163 231 202 216 119 249 133 104 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 150 172 178 152 163 231 202 216 191 119 249 133 104 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 173 191 160 120 179 152 203 117 147 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 173 191 160 120 179 152 203 117 147 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.21 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 173 191 160 120 152 203 117 147 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 160 120 152 203 117 147 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 150 172 178 152 163 231 202 216 119 249 133 104 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 150 172 178 152 163 231 202 216 191 119 249 133 104 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 0 1 2 3 4 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.40 1.211.211.21 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 173 179 151 120 202 132 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 173 151 120 202 132 1.211.211.21 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 0 1 2 3 4 5 Time (min) 0 10 20 30 40 50 60 70 80 90 100 1.41 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 173 179 151 120 202 132 100 120 140 160 180 200 220 240 m/z 100 120 140 160 180 200 220 240 m/z 0 10 20 30 40 50 60 70 80 90 100 163 191 173 151 120 202 132 Figure 2-5. Chromatograms from LC/MS/MS analysis (left) and product ion mass spectra (right) of A) CPCA standard, B) patient’s breath before naltrexone and C) patient’s breath after naltre xone after DNPH derivatization. daily dosing schedule (Volpicelli et al., 1997). Reasons for nonadherence include poor motivation, cognitive impairment (Rinn et al ., 2002), and other adverse effects of the medication which may result in interrupted therapy or premature discontinuation (Croop et al., 1997; Rohsenow et al ., 2000; Oncken et al., 2001). N onadherence may also result from the ability of alcohol to disrupt behavi oral control and an individual’s capacity to recognize that he/she has an illness requi ring treatment (Leshner, 2003). Thus, alcohol A B C

PAGE 35

23 dependence itself contributes to the adhere nce difficulties encountered by individuals who suffer from the disease. Patient adherenc e to the oral naltrexone regimen has been found to be a crucial factor in the pharm acological treatment of alcohol dependence (Litten et al., 2005). The traditional way to dete ct if patients are taking the medication is to measure the naltrexone and 6 -naltrexol in the blood or urin e. The result of this study also suggests that breath assay would pr ovide an easy and convenient way to tell physicians about patients’ compliance to th e medication by detecti ng the presence of CPCA in patients’ breath. Detection of Fluorinated Aldehyde from Metabolism of a Fluoroalkyl-containing Drug Introduction To test if trifluorinated aldehyde will be generated and detectable after metabolism, an existing fluoroalkyl-containing drug was utiliz ed as an example to verify our design through in vitro studies. Flecainide is a cla ss Ic antiarrhythmic agent causing a decreased intracardiac conduction velocity an d therefore is used for the treatment of ve ntricular and supraventricular arrhythmias (Levine et al., 199 0). It has two trifluoroethyl groups in the molecular structure. The major metabolites of flecainide in humans are m -O-dealkylated flecainide (MODF) and m -O-dealkylated lactam flecainide (MODL F) (Figure 2-6). The latter is formed by further oxidation of MODF at the piperidine ring. Both metabolites exist in urine primarily as glucuronide conjugates (Conard et al., 1984; McQuinn et al., 1984). We will focus on the detection of trifluor oacetaldehyde, a concomitant product of dealkylation.

PAGE 36

24 CF3CH2O OCH2CF3 O N H N H O H OCH2CF3 O N H N H O H OCH2CF3 O N H N H O CF3CHO + Figure 2-6. Flecainide and its two ma in metabolites MODF and MODLF. The determination involves the capture of trifluoroacetaldehyde by DNPH followed by the analysis of the hydrazone by GC/MS with negative chemical ionization (NCI). Trifluoroacetaldehyde was generated from in vitro metabolism of flecainide using human liver microsomes as a model. Experimental Materials. Flecainide acetate was purchased from Tocris (Ellisville, MO). Pooled human liver microsomes and NADPH regenera ting system were obtained from Gentest (Woburn, MA) and stored at -80C. Lp DNPH S10 cartridge was from Supelco (Bellefonte, PA). 15N2-labeled hydrazine sulfate and [15N] KNO3 were obtained from Cambridge Isotope Laboratories (Andove r, MA). Trifluor oacetaldehyde 2,4dinitrophenylhydrazone and 15N labeled internal standard were synthesized in our lab. All reagents were purchased from Sigma (St. Loui s, MO). All solvents were HPLC grade and were obtained from Fisher Scien tific (Atlanta, GA). Synthesis of 15N4-labeled DNPH. To a mixture of 136 mg (1.2 mmol) chlorobenzene in 0.8 mL carbon tetrachloride and 1.2 mL H2SO4, 292 mg (2.9 mmol) MODLF MODF CYP 2D6

PAGE 37

25 [15N]KNO3 was added with stirring for 1 hour, while the temperature was maintained below 5 oC. Then the solution was stirred at 45 oC for 2.5 hours. The mixture was poured onto ice and extracted with 4 10 mL ether, washed with water and dried. The residue ([15N2] 2,4-dinitrochlorobenzene) was recr ystallized from methanol. [15N2] 2,4-Dinitrochlorobenzene 50 mg (0.25 mmol) was diss olved in 2 mL ethanol followed by addition of 35 mg (0.27mmol) of [15N2] hydrazine (the free base was extracted with ethanol from mixture of sul phate and potassium acetate water solution.) which was refluxed for an hour. The reaction mi xture was cooled to room temperature, the product was filtered and recrysta llized from dioxane to yield [15N4] DNPH as an orange powder. Synthesis of trifluoroacetaldehyde 2,4dinitrophenylhydrozone and stableisotope labeled internal standard. Trifluoroacetaldehyde ethyl hemiacetal (310 L) was dissolved in 15 mL toluene followed by addition of 480 mg 2,4-DNPH and 15 mg p toluenesulfonic acid as a catalyst. Molecular sieves 4A (1 g) were then added, and the mixture was stirred for 4 hours at 100 oC. After the addition of 10 mL diethyl ether, the mixture was washed with 1% sodium bicar bonate and brine and dried over sodium sulfate. Solvent was evaporated under reduced pressure to give a yellow solid. TLC and GC showed the presence of one co mpound. The procedure for synthesis of 15N labeled trifluoroacetaldehyde 2,4-dinitrophenyl hydrazone was the same as above, except 15N4 labeled DNPH was used. Testing of aldehyde trappi ng with DNPH cartridge. The mixture of trifluoroacetaldehyde ethyl hemiacetal (15 L) and concentrated sulfuric acid (350 L) was heated at 90 oC for 2.5 hours in a 2 mL vial conn ected to an LpDNPH S10 cartridge

PAGE 38

26 through a needle. The other end of the cartridge was connected to a 5 mL syringe through which a reduced pressure was generated by withdrawing the plunger. After the reaction was done, the cartridge was washed with 3 mL acetonitrile. Acetonitrile was removed under a nitrogen stream and the residue was dissolved in 20 L dichloromethane, and 2 L was used for GC analysis. Microsomal incubation. The microsomal incubation mixture included 2 mg/mL human liver microsome, 5.2 mM NADP+, 13.2 mM glucose-6-phosphate, 1.6 U/mL glucose-6-phosphate dehydrogenase, 13.2 mM ma gnesium chloride and 1 mM flecainide in 100 mM potassium phosphate buffer (pH 7.5) . The total volume was 1.5 mL. LpDNPH S10 cartridges were employed to collect trifluoroacetaldehyde released from the incubation mixture for 1.5 hours through a suction generated by a 5 mL syringe. Incubations were performed in triplicate. After 1.5 hours of inc ubation, the cartridges were washed with 3 ml acetonitrile and 20 l [15N] trifluoroacetalde hyde DNPH (internal standard, 50 g/mL in acetonitrile) was added. Th e solvent was removed at room temperature under nitrogen stream, the residue was dissolved in 20 L dichloromethane and 2 L was injected for GC-MS analysis. Instrumentation. A Polaris Q mass spectrometer system interfaced to a TRACE GC and controlled by an Xcalibur data syst em (all from Thermo Electron Corporation, Trace Chemical Analysis, Austin, TX) was used in the study. Separations were done on a 30 m 0.25 mm i.d., df = 0.25 m Rtx-5MS fused silica column (Restec, Bellefonte, PA). The injector temperature was 220C. The carri er gas was helium at a flow rate of 1 mL/min. Column head pressure was 7.8 psi. The initial oven temperature was 40C for 1 min, then increased to 300C at 25C/min and maintained at 300C for 3 min. All

PAGE 39

27 injections were carried out using the splitless mode. Conditions for mass spectrometry were as follows: ion source temperature, 200C; interface temperature, 300C; ionizing voltage, 70eV; NCI mode with methane as a reagent. Quantitation. Trifluoroacetaldehyde DNPH generated by in vitro metabolism was identified by comparison of its retenti on time and mass spectra with the synthetic reference compound. Quantitation was performe d by selected ion monitoring (SIM) of the most intensive fragment ion of the anal yte and the internal st andard, respectively. [15N] Trifluoroacetaldehyde DNPH was used as an internal standard. The amount of trifluoroacetaldehyde DNPH captured by th e LpDNPH S10 cartridge was calculated by multiplying the ratio of the analyte to internal standard peak areas of the SIM chromatograms with the known quantity (1 g ) of the internal standard added. The reported value was obtained from three expe riments; error was given as standard deviation. Results and discussion LC/MS/MS (APCI negative mode) was used to analyze CPCA DNPH derivatives in the CAPA breath samples. Later, we f ound DNPH adduct could also be easily detected under our GC/MS (NCI mode) conditions . To confirm the structure of trifluoroacetaldehyde DNPH, we synthe sized the standard compound by treating trifluoroacetaldehyde ethyl hemiacetal with DNPH in toluene using p -toluenesulfonic acid as a catalyst at reflux for 4 hours (Fi gure 2-7) (Abouabdellah et al., 1997). Before using the DNPH cartridge to trap trifl uoroacetaldehyde released from microsomal incubation, the apparatus was tested by collec ting trifluoroacetaldehyde generated from hydrolysis of trifluoroacetaldehyde ethyl he miacetal with sulfuric acid (Figure 2-8) (Kumadaki et al., 1999). GC/MS showed that the compound from the cartridge was

PAGE 40

28 trifluoroacetaldehyde DNPH. Its retention time and mass spectra matched those of the standard trifluoroacetaldehyde DNPH. Therefore, the apparatus could be used in our metabolic studies that made sample prep aration more convenient and easier than traditional methods where DNPH solution was added to incubation mixture followed by organic solvent extrac tion. We also found that this met hod was more efficient than the traditional one. The amount of aldehyde co llected through cartridge was 20 times that extracted from the incubation mixture in the in vitro metabolism of CYP 2D6 enzyme substrate analogue. CF3CH(OEt)OH N H2N H NO2O2N NN H NO2O2N CF3CH Toluene p-TsOH + Figure 2-7. Synthesis of trif luoroacetaldehyde DNPH adduct. Figure 2-8. Apparatus for trapping trifluor oacetaldehyde. A: inc ubation mixture; B: DNPH cartridge; C: syringe. Cl Cl NO2NO2 NO2NO2N H NH2 KNO3H2SO4 NH2NH2 Figure 2-9. Synthesis of trifluoroacetaldehyde [15N4] DNPH. All Ns are 15N labeled. B C A

PAGE 41

29 We have also synthesized 15N labeled trifluoroacetalde hyde DNPH as an internal standard for quantitation by isotope dilution method. 15N4-labeled DNPH was obtained in a customary manner from 15N2-labeled hydrazine sulfate and 15N2-labeled 2,4dinitrochlorobenzene (Figure 2-9). Figure 2-10 shows GC/MS analysis of the analyte and its stable isotope labeled analogue. As e xpected, these two compounds had identical retention times and 4 u differe nce in molecular ions. Frag mentation between the two N atoms of the hydrazone moiety yielded the base peaks of the mass sp ectra for the analyte and internal standard; therefore, their m/z values differed by 3 u ( m/z 182 versus 185). SIM based quantitation was performed by using the latter ions. The limit of detection for trifluoroacetaldehyde DNPH was 0.02 g/mL. According to our GC/MS assay that employed the isotope dilution method, 24 6 ng trifluoroacetaldehyde DNPH (equivalent to 0.09 0.02 nmol trifluoroace taldehyde) was captured from the headspace of the vial during the microsomal incubation of flecainide. Flecainide has a narrow serum therapeu tic range, 0.2-1.0 mg/L, and its adverse effects have been found to re late to its plasma concentr ations (Boriani et al., 1993). Therefore, serum concentration monitoring ha s been a standard cl inical practice for patients (Valdes et al., 1998). Since flecaini de metabolism affects its clearance and consequently the blood concentration, flecainide metabolites have been included in drug monitoring (McQuinn et al., 1988). Our in vitro studies suggest that trifluoroacetaldehyde could be detected and may be used as a breath marker to indicate meta bolic clearance of flecainide, which will make routine flecainide monitoring much more conveni ent in the clinic. We expect that breath tests would not only reflect dr ug amount in the circulation, bu t also provide information

PAGE 42

30 5 6 7 8 9 10 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 10.33 10.32 5 6 7 8 9 10 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 10.33 5 6 7 8 9 10 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 10.33 10.32 182 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 185 265 169 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 261 166 182 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 185 265 169 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 261 166 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 185 265 169 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 261 166 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 185 265 169 50 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 261 166 Figure 2-10. SIM chromatograms (A and B) and full scan mass spectra (C and D) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubation of flecainide. A B C D CF3CH N N H NO2O2N 182

PAGE 43

31 on the change of drug concentration in th e blood upon repeated sampling and substitute or complement blood or urine assays. Sin ce flecainide and its metabolites are mainly excreted in urine, renal function affect s the urinary eliminat ion of the drug and, consequently, its serum concen tration (Conard and Ober, 1984). Therefore, this proposed method may not be able to reflect the flecainide clearance for patients with renal disease. Hypothesis Aldehyde released into breat h could be detected by anal yzing breath samples. Our preliminary studies demonstrated that tr ifluoroalkoxy-containing drugs could produce trifluorinated aldehyde after dealkylation. He re, the hypothesis is that if we introduce trifluoro-containing groups to CYP 2D6 and CYP 3A4 substrates to make new analogues, trifluorinated aldehyde will be produced after metabolism. They could be used as breath markers to reflect enzyme activity and, thus , to identify individuals’ phenotype, which will provide a simple and convenient method in the clinic to facilitate individualized drug therapy. To prove the concept, we modified dextrorphan and norverapamil by introducing trifluoroethyl groups as CYP 2D6 and CYP 3A4 substr ate analogues, respectively. In vitro metabolism studies using human liver micros omes were performed for evaluation of the designed analogues.

PAGE 44

32 CHAPTER 3 SYNTHESIS AND METABOLIC STUDIES OF CYP 2D6 SUSTRATE ANALOGUE: TRIFLUOROETHYLDEXTRORPHAN Introduction Dextromethorphan is a widely used antitu ssive ingredient in many over-the-counter cough formulations. It is primarily metabo lized by CYP 2D6 to dextrorphan through Odemethylation, which accounts for about 5060% of total dextromethorphan clearance (Schmider et al., 1997). It is also me tabolized to 3-methoxymorphinan and 3hydroxymorphinan, but these appear to be mi nor pathways mediated by CYP 2D6 and CYP 3A, respectively (Jacqz-Aigran et al ., 1993) (Figure 3-1). Dextromethorphan has been used as a probe substrate for CYP 2D6 in in vivo (Guttendorf et al.,1988; Baumann et al., 1992; Jacqz-Aigran et al., 1993), as well as in vitro studies (Wu et al., 1993; Kerry et al., 1994). Our preliminary studies showed the feasibility of our de sign. In this chapter, we report the synthesis of trifl uoroethyldextrorphan and its co mparison with current probe substrate dextromethorphan regarding their in vitro metabolism. Trifluoroacetaldehyde was also measured to confirm the form ation of the expected breath marker. Synthesis of O-trifluoroethyldextrorpha n was performed by modifying the method of Senderoff et al. (2000) and Olofson et al. (1977). It is a five-step procedure. First, dextromethorphan hydrobromide was O-demeth ylated by treatment with 48% aqueous hydrobromic acid at reflux and the resu lting O-desmethyldextromethorphan

PAGE 45

33 Figure 3-1. Metabolic path ways for dextromethorphan. hydrobromide was converted to the free base. Ne xt, this free base was treated with 2.25 eq of vinylchloroformate and Proton Sponge in 1, 2 dichloroethane to give bis-VOC derivative. Then, bis-VOC-morphinan was sele ctively O-deprotected by hydrolysis with dioxane/aqueous sodium hydroxide (3:1) at 60 C. In the fourth step, the resulting NVOC phenol was treated with 2,2,2-trifluoroethyl p -toluenesulfonate (made by tosylation of 2,2,2-trifluoroethanol using p -toluenesulfonyl chloride) an d sodium hydride in N,Ndimethylformamide to give 3-trifluoroet hoxy-N-VOC morphinan (Sonesson et al., 1994). Last, this trifluoroethyl derivative wa s reduced by lithium aluminum hydride in tetrahydrofuran to give O-tr ifluoroethyldextrorphan free base and the final compound

PAGE 46

34 was obtained by treating the free base w ith 48% hydrobromic acid. The scheme of synthesis is shown in Figure 3-2. H3CO NCH3HBr 1. 48%HBr, reflux O N O F3CH2CO NCH3HBr HO NCH3F3CH2CO NCH3VOC-CI, CICH2CH2CI Proton sponge2. K2CO3O O O O 3:1v/v NaOH: water (1.2eq) Dioxane, 60oC. 3hr O N HO O O TsOCH2CF3, NaH DMF O N F3CH2CO O O LiAIH4, THF 48%HBr Et2O 1 2 3 4 6 7 SO2CI SO2OCH2CF3CF3CH2OH, TEA DCM 5 Figure 3-2. Synthesis of Otrifluoroethyldextrophan. Materials and Methods Materials All reagents were purchased from Sigma (S t. Louis, MO). All solvents used were reagent grade and obtained from Fisher Scientific (Atlanta, GA). Column chromatography was carried out on Fisher silica gel (230-400 mesh). Thin-layer

PAGE 47

35 chromatography (TLC) analyses were performed on Fisher silica gel 60-F254 plates and visualized using UV light (254 nm). 1H NMR (300 MHz) spectra were recorded on a Varian Unity 300 spectrometer. Chemical shif ts are given in parts per million (ppm). Only diagnostic peaks are reported. APCI mass spectra we re obtained using a Thermo Finnigan LCQ mass spectrometer (San Jose, CA ). Elemental combustion analyses were performed by Atlantic Micr olab, Inc. (Norcross, GA). Synthesis of O-Trifluoroethyldextrorphan 3-(O-Desmethyl)dextromethorphan 2. Dextromethorphan hydrobromide 1 (10 g, 28.4 mmol) was dissolved in 48% aqueous hydr obromic acid (50 mL). The solution was heated at reflux for 18 hours. The mixture wa s poured on crushed ice, and treated with K2CO3 until pH = 10. The mixture was extracte d with chloroform (3 100 mL). The combined organic extracts were washed with brine, dried over Na2SO4, filtered, and the solvent was removed at reduced pressure to give a solid (6 g, 81%). TLC showed one spot, Rf = 0.2 (95:5 CH2Cl2/MeOH), the material was used without further purification. 1H NMR (CDCl3): 6.97 (d, J = 8.1Hz, 1H, C1-H), 6.72 (d, J = 2.4Hz, 1H, C4-H), 6.61 (dd, J = 2.7, 8.1Hz, 1H, C2-H), 2.40 (s, 3H, N-CH3); MS (APCI): m/z 258 [M+1]+. N, O-bis-VOC morphinan 3. 2 (6 g, 23.4 mmol) and Proton Sponge (1,8-bis (dimethylamino) naphthalene, 6 g, 28.2 mmol) were dissolved in 1,2-dichloroethane (50 mL) at 60C under N2. After adding vinyl chloroformate (6 g, 53 mmol), the solution was heated at reflux overnight. TLC revealed no starting material remaining. The mixture was filtered and concentrated, and the residue was purified by column chromatography eluting with CH2Cl2. Combination of the desired fractions followed by solvent removal gave a yellow oil (5 g, 56 %), Rf = 0.67(CH2Cl2). 1H NMR (CDCl3): 7.28 (d, J = 8.1Hz, 1H, C1-H), 7.20 (m, 2H, H2C=CH O-COR, R=N and O), 7.11 (s, 1H, C4-H), 7.01 (dd, J =

PAGE 48

36 2.4, 8.1Hz, 1H, C2-H), 5.04 (dd, J = 2.1, 13.8Hz, 1H, trans-H 2 C=CHO-COO), 4.77 (d, J = 14.4Hz, 1H, trans-H 2 C=CHO-CON), 4.68 (dd, J = 2.1, 6.3Hz, 1H, cis-H 2 C=CHOCOO), 4.45 (d, J = 6.9Hz, 1H, cis-H 2 C=CHO-CON); MS (APCI): m/z 384 [M+1]+. 3-Hydroxy-N-VOC morphinan 4. 3 (3.27 g, 8.5 mmol) was dissolved in dioxane (36 mL) and water (12 mL) containing 408 mg (10.2 mmol) of NaOH. The solution was heated at 60C for 3 hours. TLC revealed no starting material pres ent. The mixture was cooled to room temperature, poured into brin e, and extracted with ether (3 50 mL). The combined ether extracts were dried over Na2SO4, filtered, and solvent was evaporated in vacuo to leave a residue. The residue was purified by column chromatography, eluting with 10-50% EtOAc in hexane to yield an oil (2.5 g, 94%), Rf = 0.29 (CH2Cl2). 1H NMR (CDCl3): 7.18 (m, 1H, NCOOCH =CH2), 6.93 (d, J = 6.6Hz, 1H, C1-H), 6.76 (d, J = 1.8Hz, 1H, C4-H), 6.64 (dd, J = 2.1, 6.3Hz, 1H, C2-H), 4.69 (dd, J = 0.9, 10.5Hz, 1H, trans-H 2 C=CHO-CON), 4.37 (dd, J = 0.9, 4.8Hz, 1H, cis-H 2 C=CHO-CON); MS (APCI): m/z 314 [M+1]+. 2,2,2-Trifluoroethyl p-toluenesulfonate 5. p -Toluenesulfonyl chloride (4.5 g, 24 mmol) dissolved in CH2Cl2 (10 mL) was added dropwise under N2 to a solution of 2,2,2trifluoroethanol (1.6 g, 16 mmol) in 10 mL CH2Cl2, followed by 4.5 mL triethylamine at 0C. After completion of adding TEA, the reac tion was stirred at room temperature for 16 hours. The mixture was concentrated, and the residue was dissolved in EtOAc (150 mL), washed with NaHCO3 (100 mL), 0.5 M Citric acid (100 mL), distilled water (100 mL), then dried over Na2SO4. Solvent was evaporated in vacuo to leave a residue, which was purified by column chromatography eluting with 10% EtOAc in hexane to yield product 5 (3.2 g, 80%), Rf = 0.33 (4:1 hexane/CH2Cl2). 1H NMR (CDCl3): 7.75 (d, J = 6.9Hz,

PAGE 49

37 2H, aromatic H), 7.33 (d, J = 6.9Hz, 2H, aromatic H), 4.32 (q, J = 7.8Hz, 2H, CH2CF3), 2.38 (s, 3H, CH3). 3-(2,2,2-Trifluoroethyl) N-VOC morphinan 6. To a rapidly stirred solution of 4 (2 g, 6.4 mmol) in dry DMF (25 mL) was added NaH (60% oil dispersion, 400 mg, 10 mmol) under N2. After stirring for 1 hour, 5 (2.3 g, 9.1 mmol) in 10 mL of DMF was added dropwise. The mixture wa s stirred at room temperatur e overnight and then poured into 100 mL brine, extracted with ether (3 40 mL). The combined organic extracts were dried over Na2SO4, filtered and concentrated in vacuo. The residue was purified by column chromatography using CH2Cl2/MeOH as eluent to yield an oil (1 g, 40%), Rf = 0.22 (4:1:0.1 hexane/CH2Cl2/MeOH). 1H NMR (CDCl3): 7.18 (m, 1H, NCOOCH =CH2), 6.93 (d, J = 6.6Hz, 1H, C1-H), 6.76 (d, J = 1.8Hz, 1H, C4-H), 6.64 (dd, J = 2.1, 6.3Hz, 1H, C2-H), 4.69 (dd, J = 0.9, 10.5Hz, 1H, trans-NCOOCH=CH 2 ), 4.37 (dd, J = 0.9, 4.8Hz, 1H, cis-NCOOCH=CH 2 ); MS (APCI): m/z 396 [M+1]+. O-Trifluoroethyldextrorphan 7. 6 (730 mg, 1.84 mmol) was dissolved in dry THF (25 mL) and stirred in ice bath for 30 minutes under N2. LiAlH4 (300 mg, 7.9 mmol) was added in small portions with rapid stirri ng at 0C. The mixture was stirred overnight at room temperature. To the mixture was added 0.3 mL water and 0.3 mL of 15% aqueous NaOH. The mixture was poured into 75 mL of water and extr acted with ether (4 30 mL). The combined organic extracts were dried over Na2SO4, filtered and concentrated in vacuo. The residue was purif ied by column chromatography using 5-20% MeOH in CH2Cl2 as eluent to yield an oil (360 mg, 58%), Rf = 0.2 (95:5 CH2Cl2/MeOH). 1H NMR (CDCl3): 7.05 (d, J = 8.1Hz, 1H, C1-H), 6.85 (d, J = 2.7Hz, 1H, C4-H), 6.70 (dd, J = 2.7, 8.1Hz, 1H, C2-H), 4.32 (q, J = 8.1Hz, 2H, CF3CH2), 2.39 (s, 3H, NCH3);

PAGE 50

38 MS (APCI): m/z 340 [M+1]+. The oil (60 mg) was dissolved in diethyl ether (1 mL) and treated with 48% aqueous HBr (200 L), then evaporated in v acuo and dried to give the white solid hydrobromide. Analysis (C19H24F3NO.5HBr.6H2O): C, H, N. CYP 2D6 Inhibition Assays CYP 2D6 inhibition assays were perfor med by Novascreen (Hanover, MD). In brief, after 100 L inhibitors (dextromethorphan or trifluoroethyldextrorphan, concentration ranged from 10-10 M to 10-5 M) and 50 L cofactors were preincubated in 100 mM potassium phosphate buffer (pH = 7.4) for 10 min at 37C, 50 L of freshly mixed solution of human recombinant CYP2 D6 (1.5 pmole) and AMMC (3-[2-(N, Ndiethyl-N-methylammonium)ethyl ]-7-methoxy-4-methylcoumarin, 1.5 M) was added to each well. The final cofactor co ncentrations were 0.0081 mM NADP+, 0.41 mM glucose6-phosphate, 0.4 U/mL glucose-6-phospha te dehydrogenase. (The lower NADP+ concentration was to avoid interference fr om the fluorescence of NADPH which is readily apparent with a 390 nm excitation wavelength.) Samples were incubated for 45 minutes and then the reaction was stoped with 75 L 20% 0.5 M Tris/80% acetonitrile stop solution. Quinidine was used as the reference compound. The AMMC metabolite, AMHC (3-[2-(N, N-diethyl-N-methylammoni um)ethyl]-7-hydroxy-4-methylcoumarin), was measured using a fluorescent plate scanne r at an excitation wa velength of 390 nm and emission wavelength of 460 nm. Human Liver Microsome Assays Dextrorphan tartrate, dextromethorpha n hydrobromide and ethylmorphine were purchased from Sigma (St. Louis, MO). Pooled human liver microsomes and NADPH regenerating system were obtained from Gentest (Woburn, MA) and stored at -80C. LpDNPH S10 cartridges were from Supelco (Bellefonte, PA). Trifluoroethyldextrorphan

PAGE 51

39 hydrobromide and [15N] trifluoroacetaldehyde 2, 4-dinitrophenylhydrazone were synthesized in our lab. All solvents were HP LC grade and were purchased from Fisher Scientific (Atlanta, GA). Microsomal incubations were conducted in 1.5 mL polypropylene vials containing 1 mg/mL human liver microsomes, 2.6 mM NADP+, 6.6 mM glucose-6-phosphate, 0.8 U/ml glucose-6-phosphate dehydrogenase, 6.6 mM magnesium chloride and 0.1 mM dextromethorphan in 100 mM potassium phos phate buffer (pH 7.5) at 37C. The total volume was 1 mL. These incubations were performed in triplicate. Aliquots (100 L) were taken at 5, 15, 25, 45 and 75 min, and 10 L 70% perchloric acid was added to stop the reaction. After adding 20 L ethylmorphine (internal standard, 0.1 mM), the incubation mixture was basified by adding 600 L saturated sodium carbonate solution, followed by ethyl acetate (3 1 mL). Each tube was vortexed and centrifuged for 10 min. The organic layer was removed and dried under nitrogen stream . The residue was dissolved in 20 L dichloromethane and 1 L was used for GC analysis. Metabolism of trifluoroethyldextrorphan was performed unde r the same conditions as dextromethorphan in triplicate. Trifluoroacetaldehyde trapping experiment s were similar as described earlier (Figure 2-8). The final concentration of tr ifluoroethyldextrorphan was 1 mM, while the microsome and NADPH generating system were doubled. The total volume was 1.5 mL. LpDNPH S10 cartridges were employed to co llect trifluoroacetaldehyde produced from incubation mixture for 1.5 hours through a vacc um generated by the suction of a 5 mL syringe. The incubations were performed in triplicate. After 1.5 hours incubation, cartridges were washed with 3 mL acetonitrile and 20 L [15N] trifluoroacetaldehyde

PAGE 52

40 DNPH (internal standard, 50 g/mL) was added. Acetonitrile was removed under nitrogen stream. The residue was dissolved in 20 L dichloromethane, and 2 L was used for GC analysis. Instrument A Thermo Finnigan Polaris Q Mass Spectrometer (MS) system (consists of a Polaris Q MS, a TRACE GC, and the Xcalibur Data System) equipped with a Rtx-5MS silica capillary column (30 m 0.25 mm i.d., 0.25 m) was used. GC conditions were as follows. The injector temperatur e was 220C. The carrier gas wa s helium at a flow rate of 1 mL/min. Column head pressure was 7.8 psi. The initial oven te mperature was 40C for 1 min, increased to 300C at 25C/min and mainta ined at 300C for 3 min. All injections were carried out using the sp litless mode. MS conditions we re: ion source temperature, 200C; interface temperature, 300C; ionizing voltage, 70eV. GC/MS positive electron impact (EI) mode was used for identification and quantitation of dextrorphan. GC/MS negative ch emical ionization (CI) mode was used for identification and quantitation of trifluoroacetaldehyde DNPH. Methane was used as the reagent gas at a flow rate of 1.2 mL/min in the negative CI mode. Quantitation Ethylmorphine was used as an internal st andard for dextrorphan. A standard curve was constructed in the range of 0-50 M by using known standards, which were extracted from the incubation mixtures containing heat denatured microsomes (100C, 10 min). All unknown determinations were within the con centration range of the standard curves. [15N] Trifluoroacetaldehyde DNPH was us ed as an internal standard for trifluoroacetaldehyde DNPH. Quantitation wa s performed by selected ion monitoring

PAGE 53

41 (SIM) of most intensive fragment ions of dextrorphan and trif luoroacetaldehyde DNPH against the internal standard. Results Inhibition of CYP 2D6 by Dextrometh orphan and Trifluoroethyldextrorphan Dextromethorphan is a well known CYP 2D 6 substrate. Metabolism of AMMC was inhibited in the presence of dextromethor phan. To test the substrate properties of the synthesized analogue, both dextromethorpha n and trifluoroethyldextrorphan were coincubated with AMMC. Quinidine was used as a reference inhibitor which had an IC50 and Ki of 0.02 M and 0.01 M, respectively, in our assay. Results were means from duplicate experiments. Compared to dextromethorphan, whose IC50 and Ki values were 1.82 M and 1.06 M, respectively, trifluoroethyldext rorphan was about 5 times more potent with IC50 value of 0.43 M and Ki of 0.24 M. Figure 3-3 shows inhibition of CYP 2D6 activities of AMMC by increasing concentration of dextromethorphan and trifluoroethyldextrorphan. As we can see, at low concentration (<10-7 M), both has similar inhibitory effect; how ever, at high concentration ( 10-6 M), trifluoroethyldextrorphan is a much more potent inhibitor than dextromethorphan. O-Dealkylation of Dextromethorphan and Trifluoroethyldextrorphan The O-demethylated product of dextrome thorphan, dextrorphan, is often measured to evaluate CYP 2D6 activity. Our designe d analogue is also supposed to produce dextrorphan after metabolism. To compare th e two compounds, both were incubated with human liver microsomes under the same conditions. Experiments were conducted in triplicate for each. Dextrorpha n was quantitated by using GC (EI)/MS and selected ion monitoring (SIM) mode with et hylmorphine as an internal standard. Confirmation of the metabolite structure was achieved by matching GC retention time and mass spectra to

PAGE 54

42 -20 0 20 40 60 80 100 120 -10-9-8-7-6-5 Log [inhibitor] (M)% enzyme activity Dextromethorphan Trifluoroethyl dextrorphan Figure 3-3. Inhibition of CYP 2D6 activities of AMMC by increasi ng concentration of dextromethorphan and trifluor oethyldextrorphan from 10-10 M to 10-5 M. Activities are given as percen t of control rates (no inhi bitor). Data are given as means of duplicate determinations. that of authentic dextrorphan. Figure 3-4 s hows the chromatograms and mass spectra for dextrorphan and ethylmorphine. Proposed fragmentation of dextrorphan by EI (Wu et al., 2003) is given in Figure 3-5. Fragment ions m/z 157 and 313 were monitored for dextrorphan and ethylmorphine, respectively, in the quantitative analysis. Peak area ratios of the analyte and internal standard ( y ) versus concentration of the analyte ( x ) were linear over the range 0-50 M. The regression equation of the standard curve was y = 0.0384 x 0.0337, R2 = 0.998. The limit of detection for de xtrorphan was <0.002 nM. Figure 3-6 shows time-dependent dextrorphan forma tion from both dextromethorphan and its analogue trifluoroethyldextrorphan. Compar ed to dextromethorphan, metabolism of trifluoroethyldextrorphan was slow. Especi ally after 25 minutes, dextromethorphan was metabolized much faster than trifluoroethyldextrorphan. At 75 minutes, the amount of

PAGE 55

43 dextrorphan produced by dextromethorphan was 6 times lower by trifluoroethyldextrorphan. 9.5 10.0 10.5 11.0 11.5 12.0 12.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 11.78 0 10 20 30 40 50 60 70 80 90 100 11.04 9.5 10.0 10.5 11.0 11.5 12.0 12.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 11.78 0 10 20 30 40 50 60 70 80 90 100 11.04 9.5 10.0 10.5 11.0 11.5 12.0 12.5 Time (min) 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 11.78 0 10 20 30 40 50 60 70 80 90 100 11.04 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 313 162 243 214 284 256 115 100 150 200 250 m/z 0 10 20 30 40 50 60 70 80 90 100 157 200 257 150 160 128 77 189 91214Relative Abundance 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 313 162 243 214 284 256 115 100 150 200 250 m/z 0 10 20 30 40 50 60 70 80 90 100 157 200 257 150 160 128 77 189 91214Relative Abundance 100 150 200 250 300 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 313 162 243 214 284 256 115 100 150 200 250 m/z 0 10 20 30 40 50 60 70 80 90 100 157 200 257 150 160 128 77 189 91214Relative Abundance Figure 3-4. Chromatograms (A and B) a nd full scan mass spectra (C and D) for dextrorphan and ethylmorphine, respectively. A B C D

PAGE 56

44 NCH3 O H Figure 3-5. Proposed electron impact mass fragmentation of dextrorphan. 0 5 10 15 20 25 30 020406080time (minutes)formation of dextrorphan nmoles per ml Dextromethorphan Trifluoroethyl dextrorphan Figure 3-6. Time-dependent dextrorphan fo rmation from both dextromethorphan and its analogue trifluoroethyldextrorphan. Re sults are expressed as means SD (n=3). Determination of Trifluoroacetaldehyde According to our hypothesis, trifluoroacetaldehyde should be produced after detrifluoroethylation of trif luoroethyldextrorphan. The a bove experiment showed less formation of dextrorphan by trifluor oethyldextrorphan, so we collected NCH3 CH2NCH3 CH2CH2NCH3 CH2PhOH m/z 228 m/z 214 m/z 200 m/z 150 m/z 157 m/z 257 +

PAGE 57

45 trifluoroacetaldehyde after 90 minutes incuba tion instead of starti ng from 5 minutes. The incubation conditions were the same as above . Trifluoroacetaldehyde was derivatized as its DNPH adduct before determination under negative CI mode. Figure 3-7 shows the chromatograms of trifluoroacetaldehyde DNPH and the internal standard. Their mass spectra were identical with those in Figure 2-10 (C and D). The amount of trifluoroacetaldehyde DNPH was calculated by isotope dilution method. A total of 120.33 ng 1.53 trifluoroacetaldehyde DNPH (equivale nt to 0.43 nmol trifluoroacetaldehyde) was generated from the incubation mixture. Time (min) 4 5 6 7 8 9 10 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 9.62 0 10 20 30 40 50 60 70 80 90 100 9.62 Time (min) 4 5 6 7 8 9 10 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 9.62 0 10 20 30 40 50 60 70 80 90 100 9.62 4 5 6 7 8 9 10 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 9.62 0 10 20 30 40 50 60 70 80 90 100 9.62 Figure 3-7. SIM chromatograms (A and B) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubation of trifluoroethyldextrorphan. For full scan mass spectra, see Figure 2-10 (C and D). Discussion Dextromethorphan is a widely used pr obe drug for CYP 2D6. To facilitate phenotyping patients about their drug metabo lizing enzyme activity, we designed a new A B

PAGE 58

46 substrate analogue based on dextromethorphan in an attempt to ex ploit exhaled breath which is more convenient to sample. In th e proof-of-concept studies, we synthesized trifluoroethyldextrorpha n and investigated its in vitro metabolism using human liver microsomes. For the synthesis of trifluor oethyldextrorpan, due to the existence of competitive Nalkylation which leads to qua ternary ammonium salt forma tion, direct alkylation of 3OH-N-methylmorphinan is not preferred (Se nderoff et al., 2000). Ol ofson et al. (1977) reported an effective way to only alkylate th e hydroxyl group without effect on the amine group, which is protection of N and O w ith vinyloxycarbonyl (VOC) group followed by selective O-deprotection. The phenolic VOC function is easily removed by mild basic hydrolysis, while N-VOC function is refrac tory to cleavage under these conditions. Although the N-methyl group would be removed by carbamate formation in the generation of key intermediate (N,O-bis-VOC morphinan), the literature suggested that the carbamate could be convert ed efficiently back to the desired N-methyl moiety by lithium aluminum hydride reduction, obviat ing the need for realkylation. The fourth step, introduction of the trifluoroethoxy gr oup to 3-OH-N-VOC morphinan, was a key step in the whole pr ocess because the presence of trifluorosubstituents highly decreased SN2 reactivity on the -carbon atom of the ethyl group (Camps et al., 1980). First, we tried 3-br omo-1,1,1-trifluoropropane as an alkylating agent to introduce the trifluoropropoxy group. It was reported that the trifluoropropyl group was much easier to introduce compared to the trifluoroethyl group due to the reduced electron-withdrawing effect of tr ifluoromethyl group on th e electrophilic carbon. Hence SN2 reactivity should be increased (Singh et al., 2003). However, reaction with

PAGE 59

47 BrCH2CH2CF3 (3 eq) and NaH in dry DMF under N2 at 50C overnight did not work (Michellys et al., 2003). Furt her attempts using prolonged reaction time (3 days) and excess BrCH2CH2CF3 (10 eq) also failed. Then, we tried with 2,2,2-trifluoroethyl p toluenesulfonate that has a good leaving group (tosyl group). Reaction was carried out in EtOH with EtONa as a base to deprotonate ph enol (Bergeron et al., 2005). However, an ethoxy group was introduced instead of tr ifluoroethoxy. The NMR spectrum showed a quartet peak at 4 ppm that is methylene sp lit by an adjacent methyl group and not by the trifluoromethyl group. Otherwise, the chemical shift is 4.3 ppm as a result of the electronwithdrawing effect. Mass spectra also conf irmed it was the ethylat ed product. Finally, when 2,2,2-trifluoroethyl p -toluenesulfonate was used with NaH in dry DMF, we obtained the 3-trifluoroe thoxy N-VOC morphinan. Several methods have been developed for determination of dextrorphan, including HPLC (Stavchansky et al., 1995; Vielnascher et al., 1996; Mistry et al., 1998), LC/MS/MS (Eichhold et al., 1997; McCauley-Myers et al., 2000; Vengurlekar et al., 2002), GC with selective nitrogen detection (Kintz et al., 1989), electron capture dectection (Salsali et al., 1999), and mass spectrometric dete ction (Xu et al., 1993). It is usually necessary to carry out a pre-column derivatization reaction for GC analysis of dextrorphan. Wu et al. (2003) developed a method without derivatization. We also found dextrorphan could be directly detected under our GC/MS conditions, and the assay was rapid and simple. Human liver microsomal incubati on of both dextromethorphan and trifluoroethyldextrorphan showed that th e latter was metabolized more slowly. The generally accepted mechanism of CYP medi ated metabolism is that CYP mediated

PAGE 60

48 oxidations occur via ironoxo/porphyrin cation radical [F e(III)-Por] (Dawson and Sono, 1987; Dawson, 1988; Blake and Coon, 1989; Larroque et al ., 1990). The catalytic cycle for CYP starts with the binding of the substr ate to a pentacoordinate high-spin Fe(III) state that initiates the transf er of an electron from NADPH to give an Fe(II) complex. The reduced enzyme-substrate complex then binds molecular oxygen. Transfer of a second electron to this complex is followed by the uptake of two protons and cleavage of O-O bond to give rise to the ac tivated enzyme O=Fe(IV)-Por+ and water. The oxygen of the activated enzyme is then inserted into the substrate. For the mechanism of demethylation, according to some investigators (Watan abe et al., 1982, Karki and Dinnocenzo, 1995, Walles et al., 2001), first hydr ogen abstraction yield a ne utral radical, then hydroxy transfer from the enzyme gives rise to hemi acetal which is not stable and collapse to the corresponding phenol and formaldehyde (Figur e 3-8). So the possible reason for low metabolism of trifluoroethyl dextrorphan is that highly electronegative fluorine atoms decrease the electron charge density on the -carbon atom of the tr ifluoroethoxy group by spreading the charge across th e entire side chain, which reduces hydroxyl group attack on the -carbon. The heat of reaction for deal kylation of both dextromethorphan and trifluoroethyldextrorphan was calculated by a semiempirical quantum chemical method (PM3). Trifluoroethyldext rorphan showed a much hi gher reaction energy of 96.8 kcal/mol compared with -41.2 kcal/mol for dextromethorphan. Therefore, formation of the intermediate hemiacetal will be harder fo r trifluoroethyldextrorphan; consequently, less dealkylated product will be produced. On the other hand, trifluoroethyldextror phan demonstrated a potent inhibitory effect on CYP 2D6, suggesting th at trifluoroethyldextrorpan binds more tightly than

PAGE 61

49 O CH3 O=Fe(IV) P O CH2 Fe(IV)P O H O C H2 OH Fe(III)P O H H2C=OH-abstraction + + + + Figure 3-8. Proposed mechan ism for O-demethylation. dextromethorphan to the enzyme. Possibl y, these two compounds have different inhibitory mechanism for CYP 2D6. Dextrome thorphan is a competitive inhibitor which has the same binding site as the substrate A MMC, whereas trifluoroe thyldextrorpan is a noncompetitive inhibitor which can bind to some other site on the enzyme thus reducing enzyme’s activity. This difference could also be attributed to th e presence of highly electronegative fluorine atoms. It was reported that C-F bond in a ligand can interact with C=O bond in a protein to form C-F C=O th at leads to significan tly increased binding affinity (Olsen et al., 2003). The molecular conformation could be changed as well. The volume of a trifluoromethyl group is r oughly twice compared to a methyl group. Spectroscopic studies and hi gh-level quantum-mechanical calculations show that preference for the planar arrang ement in anisole is inverted to the orthogonal orientation in trifluoromethylanisole (Klocker et al., 2003; Bohm et al., 2004). Therefore, trifluoroethyldextrorphan could possibly bind to a site different from dextromethorphan due to the changed conformation and increase d binding affinity. Furt her studies will be needed to clarify their exact inhibitory mechanisms. Nevertheless, trifluoroacetald ehyde was produced from the trifluoroethyldextrorphan in cubation and was detectable as we hypothesized. However, due to the presence of electron-withdrawing fluorines which affect relative reactivity, binding affinity and orientation, the metabolis m of trifluoroethylde xtrorphan was less

PAGE 62

50 than that of dextromethorphan. Thus, less tr ifluoroacetaldehyde was released. Janzowski et al. (1982) reported that generation of trifluoroa cetaldehyde from N-nitrosotrifluoroethyl-ethylamine was only detect able at high substrate concentrations. Dealkylation of the et hyl group predominated over that of the fluorinated ethyl group. Baker et al. (1984) has illu strated the remote effects of fluorine substituents on microsomally mediated hydroxylation. Under conditions where the relative percentages of hydroxylation of hexane at positions 1, 2 and 3 were 3.4, 41.5, and 9.6, respectively (Kramer et al., 1974), 1,1,1,2,2,-pentafluorohexa ne gave only the 5-hydroxy derivative; no evidence was obtained for hydroxylation at positions 3 and 4. Likewise, 1,1difluorocyclohexane was not hydroxylated at position 2 and ratio for positions 3 and 4 was ~1:5.5. Based on previous investigator s’ work, introduction of a longer chain containing the trifluorogroup to dextrorphan could increase the metabolic properties of the new analogue. The longer chain can decrea se the strong electron-withdrawing effect of the trifluorogroup on the carbon of the substrate to be hydroxylated. And the introduction of longer chain is much easier from the perspective of synthesis. Analysis of the heat of reaction for a series of trifl uoro-containing analogues using semiempirical quantum chemical models showed that intr oduction of trifluoropropyl or trifluorobutyl group to dextrorphan instead of trifluoroeth yl group would not in crease the heat of reaction (Table 3-1), which would make dealkylation reactivity similar to dextromethorphan. A longer chain would also not increase the torsional potential of the molecule, so molecular conf ormation would be less affected. More trifluorinated aldehyde could be produced in analogues with longer trifluoroalkyl chains and could be considered as an improved exhaled breath marker.

PAGE 63

51 In summary, we designed a CYP 2D6 substrate analogue, trifluoroethyldextrorphan, and evaluated its in vitro metabolism. This study proved our concept that trifluorinated aldehyde can be generated after metabolism and may be considered as a breath marker for enzyme ac tivity. Modification of this analogue is still needed to make it a better substrate. Table 3-1. Heat of reaction of dextro methorphan and trifluoroalkyldextrorphan analogues. Heat of formation Heat of formation Heat of reaction (kcal/mol) (kcal/mol) (kcal/mol) Reactant Products Dextromethorphan HCHO Dextrorphan 33.3 -34.1 26.2 -41.2 Trifluoroethyldextrorphan CF3CHO Dextrorphan -123.7 -53.1 26.2 96.8 Trifluoropropyldextrorphan CF3CH2CHO Dextrorphan -130.4 -199.7 26.2 -43.1 Trifluorobutyldextrorphan CF3(CH2)2CHO Dextrorphan -135 -205.8 26.2 -44. 6 Heat of formation was calculated by using semiempirical quantum chemical models (PM3) with CAChe software. Heat of reaction = sum of heats of formati on (products) – sum of heats of formation (reactants).

PAGE 64

52 CHAPTER 4 SYNTHESIS AND METABOLIC STUDIES OF CYP 3A4 SUSTRATE ANALOGUE: TRIFLUOROETHYLNORVERAPAMIL Introduction Verapamil, a phenylalkylamine calcium cha nnel antagonist, is widely used in the treatment of various cardiovascular disord ers including supraventricular tachycardia, angina pectoris, and hyperte nsion. It has been reported that CYP 3A4 and 1A2 are responsible for verapamil N-dealkylation and that the CYP 2C subfamily are responsible for verapamil O-dealkylation (Eichelbaum et al., 1979; Levy et al., 2000 ). At therapeutic concentrations, CYP 1A2 is not involved in verapamil metabolism (Tracy et al., 1999). The N-dealkylated products norverapamil and D-617, mainly formed by CYP 3A4, are the major metabolites in humans, and th ey undergo further breakdown via the CYP system to form other metabolites (Fi gure 4-1) (Kroemer et al., 1993). According to our design, verapamil will be modified by introducing the trifluoroalkyl group into norverapamil whic h is expected to pr oduce trifluorinated aldehyde as an exhaled breath marker after me tabolism. In this chapter, we report the synthesis of trifluoroethylnorverapamil and compare it with verapamil regarding their in vitro metabolism. Trifluoroacetaldehyde was also measured to evaluate it as a breath marker. Synthesis of N-trifluoroe thylnorverapamil was performed by modifying the method of Johnstrom et al. (1995) and Lawal et al. (2004). First th e alkylating agent 2,2,2trifluoroethyltriflate was synthesized by using 2,2,2trifluoroe thanol and triflic

PAGE 65

53 Figure 4-1. Metabolism scheme of verapamil. The asymmetric carbons are marked with asterisks. anhydride in the presence of diisopropylethylamine. Norverapamil hydrochloride was then converted to the free base and treated with 2,2,2-trifluoroethyl triflate and sodium hydride in toluene to give N-trifluoroethyl norverapamil, which was converted back to hydrochloride to get the final compound. The sc heme of synthesis is shown in Figure 4-2. CF3CH2OH + CF3SO2OSO2CF3DIPEACF3SO2OCH2CF3 Figure 4-2. Synthesis of Ntrifluoroethylnorverapamil. H N OCH3OCH3H3CO H3CO CN NOCH3OCH3H3CO H3CO CN CH2CF3CF3SO2OCH2CF3, NaHToluene

PAGE 66

54 Materials and Methods Materials All reagents were purchased from Sigma (S t. Louis, MO). All solvents used were reagent grade and obtained from Fisher Scientific (Atlanta, GA). Thin-layer chromatography (TLC) analyses were performed on Fisher silica gel 60-F254 plates and visualized using UV light (254 nm). Prepara tive TLC was carried out on Analtech silica gel gf (2000 microns).1H NMR (300 MHz) spectra were recorded on a Varian Unity 300 spectrometer. Chemical shifts are given in parts per million (ppm). Only diagnostic peaks are reported. APCI mass spectra were obt ained using a Thermo Finnigan LCQ mass spectrometer (San Jose, CA). Elemental com bustion analysis was performed by Atlantic Microlab, Inc. (Norcross, GA). Synthesis of N-Trifluoroethylnorverapamil 2,2,2-Trifluoroethyltriflate. 2,2,2-Trifluoroethanol ( 1.77 g, 17.7 mmol, 1.27 mL) was trapped in diisopropylethylam ine (2.30 g, 17.8 mmol, 3.11 mL) under N2. After triflic anhydride (5 g, 17.7 mmol, 3 mL) was added dropwis e at 0C, the mixture was allowed to warm to room temperature and stirred for 3 hours at reflux. The product (3.2 g, 78%) was obtained by distilling reac tion mixture at 120C in oil bath. 1H NMR (CDCl3): 4.70 (q, J = 7.6Hz, 2H, OCH2CF3). N-Trifluoroethylnorverapamil. Norverapamil hydrochloride (40 mg, 0.084 mmol) was dissolved in water (8 mL) and treated with K2CO3 to give pH=10. The mixture was extracted with CH2Cl2 (3 8 mL). The combined organic extracts were dried over Na2SO4, filtered, and solvent was removed in vacuo to give norverapamil free base. This free base was dissolved in toluen e (4 mL). NaH (excess) was added to the mixture and stirred for 4 hours at 70C. Trifluoroethyltriflate (100 L) was added to the

PAGE 67

55 sodium salt of norverapamil and stirred overnight at reflux. The mixture was poured into water, extracted with ether (3 10 mL). The combined organi c extracts were dried over Na2SO4, filtered and concentrated in vacuo. The crude was pu rified by preparative TLC using 1% MeOH in CH2Cl2 to yield an oil (17 mg, 39%), Rf = 0.38 (99:1 CH2Cl2/MeOH). 1H NMR (CDCl3): 6.70-6.83 (m, 4H, aromatic H), 6.58 (s, 1H, aromatic H), 6.60 (s, 1H, aromatic H), 3.81 (s, 3H, OCH3), 3.80 (s, 3H, OCH3), 3.79 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 2.94 (q, J = 7.2Hz, 2H, NCH2CF3), 1.11 (d, J = 4.8Hz , 3H, CH(CH 3 )2), 0.72 (d, J = 5.1Hz , 3H, CH(CH 3 )2); MS (APCI): m/z 523 [M+1]+. The oil was dissolved in ethanol and treated with concentrated HCl, then evaporated in vacuo and dried to get the white so lid hydrochloride. Analysis (C28H37F3N2O4.7HCl): C, H, N. CYP 3A4 Inhibition Assays CYP 3A4 inhibition assays were perfor med by Novascreen (Hanover, MD). In brief, after 100 L inhibitors (verapamil or trifluoronorverapamil, concentration ranged from 10-10 M to 10-5 M) and cofactors were preincuba ted in 200 mM potassium phosphate buffer (pH = 7.4) for 10 min at 37C, 100 L of freshly mixed solution of human recombinant CYP3A4 (0.5 pmole) and DBF (dibenzylfluorescein, 1 M) was added to each well. The final cofactor co ncentrations were 1.3 mM NADP+, 3.3 mM glucose-6phosphate, 0.4 U/mL glucose-6-phosphate de hydrogenase. The samples were incubated for 10 minutes and then the reaction was stopped with 75 L 2M NaOH. Ketoconazole was used as the reference inhibitor. Th e DBF metabolite, fluorescein, was measured using a fluorescent plate scanner at an ex citation wavelength of 485 nm and emission wavelength of 538 nm. Human Liver Microsome Assays

PAGE 68

56 Norverapamil hydrochloride and verapamil hydrochloride were purchased from Sigma (St. Louis, MO). d -Propranolol was from Ayerst La boratories Incorporated (New York, NY). Pooled human liver microsomes and NADPH regenerating system were obtained from Gentest (Woburn, MA) and stored at -80C. LpDNPH S10 cartridges were from Supelco (Bellefonte, PA). Trifl uoroethylnorverapamil hydrochloride and [15N] trifluoroacetaldehyde 2,4-dinitrophenylhydr azone were synthesized in our lab. All solvents were HPLC grade and were purchas ed from Fisher Scientific (Atlanta, GA). Microsomal incubations were conducted in 1.5 mL polypropylene vials containing 1 mg/mL human liver microsome, 2.6 mM NADP+, 6.6 mM glucose-6-phosphate, 0.8 U/mL glucose-6-phosphate dehydrogenase, 6.6 mM magnesium chloride and 0.5 mM verapamil in 100 mM potassium phosphate buffer (pH 7.5) at 37C. The total volume was 1 mL. These incubations were pe rformed in triplicate. Aliquots (100 L) were taken at 5, 15, 25, 45 and 75 min, and 50 L acetonitrile was added to stop the reaction. After adding 10 L d -propranolol (internal standard, 0.1mM), the incubation mixture was basified by adding 50 L saturated sodium carbonate solu tion, followed by diethyl ether (3 1 mL). Each tube was vortexed and centrifuged for 10 min. The organic layer was removed and dried under nitrogen stream. The residue was dissolved in 20 L dichloromethane and 5 L was used for GC analysis. Metabolism of trifluoroethylnorverapamil was performed under the same condition as verapamil in triplicate. Trifluoroacetaldehyde trapping experiment s were similar as described earlier (Figure 2-8). The final concentration of trifluoroethylnorverapamil was 1 mM, microsome and NADPH generating system were doubled. The total volume was 1.5 mL.

PAGE 69

57 LpDNPH S10 cartridges were employed to co llect trifluoroacetaldehyde produced from incubation mixture for 1.5 hours through a vacc um generated by the suction of a 5ml syringe. The incubations were performed in triplicate. After 1.5 hours incubation, cartridges were washed with 3 mL acetonitrile and 20 L [15N] trifluoroacetaldehyde DNPH (50 g/mL) was added. Acetonitrile was re moved under nitrogen stream. The residue was dissolved in 20 L dichloromethane, and 2 L was used for GC analysis. Instrument A Thermo Finnigan Polaris Q Mass Spectrometer (MS) system (consists of a Polaris Q MS, a TRACE GC, and the Xcalibur Data System) equipped with a Rtx-5MS fused silica capillary column (30 m0.25 mm ID, 0.25 m) was used. GC conditions were as follows. The injector temperature was 220C. The carrier gas was helium at a flow rate of 1 mL/min. Column head pressu re was 7.8 psi. The initial oven temperature was 40C for 1 min, increased to 300C at 25 C/min and maintained at 300C for 10 min. All injections were carried out using the splitless mode. MS cond itions were: ion source temperature, 200C; interface temperat ure, 300C; ionizing voltage, 70 eV. GC/MS positive electron impact (EI) mode was used for identification and quantitation of norverapamil. GC/MS negativ e chemical ionization (CI) mode was used for identification and quantita tion of trifluoroacetaldehyde DNPH. Methane was used as the reagent gas at a flow rate of 1.2 mL/min in the negative CI mode. Quantitation d -Propranolol was used as an internal standard for norverapamil. The standard curve was constructed in the range of 5-50 M by using known standards, which were extracted from the incubation mixtures contai ning heat denatured microsomes (100C, 10 min). All unknown determinations were within the concentration range of the standard

PAGE 70

58 curves. [15N] Trifluoroacetaldehyde DNPH was used as an internal standard for trifluoroacetaldehyde DNPH. Quantitation wa s performed by selected ion monitoring (SIM) of most intensive fragment ions of norverapamil and trifluoroacetaldehyde DNPH against the internal standard. Results Inhibition of CYP 3A4 by Verapa mil and Trifluoroethylnorverapamil Verapamil is a CYP 3A4 substrate. Me tabolism of DBF was inhibited in the presence of verapamil. To test the substrat e properties of the s ynthesized analogue, both verapamil and trifluoroethylnorverapamil were coincubated with DBF. Ketoconazole was used as a reference inhibitor which had an IC50 and Ki of 0.06 M and 0.03 M, respectively, in our assay. Results were means from duplicate experiments. Figure 4-3 shows inhibition of CYP 3A4 activities to ward DBF by increasing concentration of verapamil and trifluoroethyl norvera pamil. At low concentration (<10-8 M), about 7% of DBF metabolism was inhibited by verapamil. When verapamil increased to 10-7-10-6 M, 17% metabolism was inhibited. Trifluoroe thylnorverapamil could produce the same inhibitory effect at 10-9 M. At high concentration (10-5 M), 85% metabolism was inhibited in trifluoroethylnorverapamil inc ubation, whereas only 30% inhibition occurred in verapamil incubation. With an IC50 of 1.93 M and Ki of 1.01 M, respectively, trifluoroethylnorverapamil was much more potent than verapamil (IC50: NA, Ki : NA). N-Dealkylation of Verapamil and Trifluoroethylnorverapamil Norverapamil is one of the major metabol ites of verapamil, mainly formed by CYP3A4. Our designed analogue is supposed to produce norverapamil as well after dealkylation. To compare the two compounds , both were incubated with human liver

PAGE 71

59 microsomes under the same conditions. Experi ments were conducted in triplicate for each. Norverapamil was quantitated by using GC(EI)/MS selected ion monitoring (SIM) 0 20 40 60 80 100 -10-9-8-7-6-5 Log [inhibitor] (M)% enzyme activity verapamil trifluoroethyl norverapamil Figure 4-3. Inhibition of CYP 3A4 activities of DBF by increasing concentration of verapamil and trifluoroethylnorverapamil from 10-10 M to 10-5 M. Activities are given as percent of control rates (no inhibitor). Data are given as means of duplicate determinations. mode with d-propranolol as an internal standard. Confirmation of the metabolite structure was achieved by matching GC retention time and mass spectra to that of authentic norverapamil. Figure 4-4 shows the chromat ograms and mass spectra for norverapamil and d-propranolol. Fragment ions 289 and 112 were monitored for norverapamil and dpropranolol, respectively, in the quantitative analysis. Peak area ratios of the analyte and internal standard (y) versus c oncentration of the analyte (x) were linear over the range 550 M. The regression equation of the standard curve was y = 0.1122x 0.3772, R2 =0.988. The limit of detection for norverapamil was 2.5 M. Although formation of norverapamil from verapamil was clearly detect able and measurable (> 20 nmol/mL after 75 min of incubation), formation of norver apamil from trifluoroethylnorverapamil was detectable but did not reach the level that allo wed for reliable quantif ication (5 nmol/mL).

PAGE 72

60 6 8 10 12 14 16 18 Time (min) 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100 17.83 11.29 6 8 10 12 14 16 18 Time (min) 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100 17.83 11.29 50 100 150 200 250 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 112 127 183 207 144 84 50 100 150 200 250 300 350 400 450 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 289 207 281 326 151 404 428 50 100 150 200 250 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 112 127 183 207 144 84 50 100 150 200 250 300 350 400 450 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 289 207 281 326 151 404 428 Figure 4-4. Chromatograms (A and B) a nd full scan mass spectra (C and D) for norverapamil and d -propranolol, respectively. B A C D N H MeO OMe OMe OMe CH3 C H3 CN 289 151

PAGE 73

61 Determination of Trifluoroacetaldehyde Trifluoroacetaldehyde was collected over 90 minutes incubation of trifluoroethylnorverapamil with human liver microsomes. The incubation conditions were the same as above. Trifluoroacetaldehyde was derivatized as it s DNPH adduct before determination under negative CI mode. Figure 4-5 shows the chromatograms of trifluoroacetaldehyde DNPH and the internal standard. Their mass spectra were identical with those in Figure 2-10 (C and D). Th e amount of trifluor oacetaldehyde DNPH was calculated by isotope dilution method. A to tal of 8.33 ng 1.15 trifluoroacetaldehyde DNPH, equivalent to 0.03 nmol trifluoroace taldehyde was generated from the incubation mixture. Discussion CYP 3A4 is an important CYP enzyme wh ich mediates the metabolism of more than 50% currently used drugs. Analyzing meta bolites in breath samples would facilitate phenotyping patients about their CYP 3A4 activ ity. Based on metabolism of verapamil, a new substrate analogue, trifluor oethylnorverapamil, was designed which was supposed to generate trifluorinated aldehyde in breath to be used as a breath marker. Synthesis and in vitro evaluation of this analogue was i nvestigated in this chapter. The new analogue was obtained by N-trif luoroalkylation of norverapamil. NAlkylation is usually performed by using alkyl halide or sulfate. Since the presence of fluorines decrease the activity towards nucleophi lic attack, N-trifluoroalkylation is more difficult than normal N-alkylation. Steinman et al. (1973) report ed the yield of 1trifluoroethylbenzodiazepin-2-one system was low when alkyl ated with trifluoroethyl iodide. The reaction could not be substa ntially improved by altering the reaction conditions or by the use of trif luoroethylbenzenesulfonate as the alkylating reagent. The

PAGE 74

62 5 6 7 8 9 10 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 10.30 10.30 5 6 7 8 9 10 Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 10.30 10.30 Figure 4-5. SIM chromatograms (A and B) of trifluoroacetaldehyde DNPH and labeled internal standard, respectively, after microsomal incubation of trifluoroethylnorverapamil. For full s can mass spectra, see Figure 2-10 (C and D). trifluoromethanesulfonate (triflate) group is the most activated functional groups for nucleophilic substitution reacti ons. It is much more active than benzenesulfonate or tosylate (Netscher and Bohrer, 1996). There is a factor of 40,000 di fference in reactivity of tosylates and triflates (Lawal et al. 2004). Both CF3CH2I and CF3CH2OTs did not work in our synthesis. Therefore, we chose trifluorosulfonyl as a better leaving group, which would allow alkylation with trifl uoroethyl on the secondary amine moiety. Because triflate is often used for the alkyl ation of particularly unreactive substrates, product distribution may be affected by comp eting reactions with other more reactive components in the media (base, water, solven t etc.). When the alkylation was attempted in aprotic, dipolar solvents at high temperature, polar byproducts were exclusively A B

PAGE 75

63 formed which presumably arose from the hydrolysis of CF3CH2OTf or from its reaction with the solvent (Johnstrom and Stonee-Elande r, 1995). Therefore, toluene was used as solvent in the synthesis. Verapamil contains an asymmetric carbon and, thus, exists in two enantiomeric forms: Rand Sverapamil. Clinically ad ministered verapamil is a racemic mixture, which has different pharmacokinetic and pha rmacological properties. S-Verapamil has been shown to have a negative dromotropic e ffect on atrioventricula r conduction that is 10 to 18 times greater than Rverapamil in man (Echizen et al., 1985). S-Verapamil is preferentially eliminated during first-pass me tabolism, and as a consequence, the plasma concentration ratio of Rto Sverapamil is around 5:1 after oral administration and is approximately 2:1 after intravenous administ ration (Vogelgesang et al., 1984). Verapamil undergoes extensive presystemic and systemic metabolism, in which only 3% to 4% of an oral dose is excreted unchanged in urine (Eichelbaum et al., 1979). The only metabolite with pharmacological activity is norverapam il which is primarily formed by CYP 3A4 (Kroemer et al., 1993; Tracy et al., 1999). Determination of norverapamil has been performed by several analytical methods including spectrofluorometry (McAllister and Howell, 1976), HPLC (Cole and Flanagan, 1981; Stagni and Gillespie, 1995; Garcia et al., 1997), GC (Ahnoff and Persson, 1990; Shin et al., 1996), and GC/MS (Mikus et al ., 1990). Hedeland et al. (2004) reported quantitation of the two norverapamil enantio mers in human plasma using LC/MS/MS. Since we were trying to find a correlation between norverapamil a nd dealkylation caused formation of aldehyde, which is the common byproduct of both norverapamil enantiomers, total norverapamil was determin ed in our case. Norverapamil gave good

PAGE 76

64 separation in our GC(EI)/MS method. Vasiliades et al. (1982) reported that norverapamil, a secondary amine, is not as readily exracted as the parent compound, which is a tertiary amine. This agreed with our observation that norverapamil had a very high detection limit of 2.5M. Due to the introduction of electron-withdr awing trifluoroethyl group, compared to verapamil, the synthesized analogue showed decreased metabolism, which is similar to trifluoroethyldextrorphan di scussed in Chapter 3. Accord ing to the demethylation mechanism (Figure 4-6) (Watanabe et al ., 1982, Karki and Dinnocenzo, 1995, Walles et al., 2001), presence of trifluoroethyl gr oup may impede hydrogen abstraction and the further step of hydroxy transfer from the enzyme. Semiempirical quantum chemical calculations showed Hf of hydrogen abstraction from verapamil was 29.5 kcal/mol, while for trifluoroethylnorverapamil, Hf was 33.6 kcal/mol. The heat of reaction for dealkylation of verapamil and trifluoroe thylnorverapamil was -32.5 and 105.8 kcal/mol, respectively. The results s upported the proposed rationale . This suggests that the introduction of trifluoroethyl group has an electronic effect on the metabolism of verapamil by decreasing subs trate reactivity toward enzy mes (Higgins et al., 2001). Actually, one of the strategies to improve th e metabolic stability in medicinal chemistry is to block the metabolically labile site wi th fluorine substituents (Smith et al., 2001). From our results, trifluoroethyl group seems to behave that way because the short alkyl chain could not reduce the effect of fluorines. The inhibition of CYP 3A4 experime nt showed a similar result as trifluoroethyldextrorphan did on CYP 2D6. Th e trifluorinated analogue had more potent inhibitory effect than the substrate drug. Sin ce it was less metabolized itself, yet inhibited

PAGE 77

65 NCH3 O=Fe(IV) P N CH2 Fe(IV) P O H NC H2 OH Fe(III) P NH H2C=OH-abstraction + + + + Figure 4-6. Proposed mechan ism for N-demethylation. enzymes that metabolize other substrates, it might act as a noncompetitive inhibitor to reduce the enzymes’ activity through increa sed binding affinity and changed molecular conformation caused by th e presence of electronegative fluorines. Nevertheless, trifluoroace taldehyde was also detected from gas phase of trifluoroethylnorverapamil incuba tion mixture as we expected . As discussed in Chapter 3, introduction of a long chain could decrease the electron-withdrawing effect of trifluorines on N-dealkylation, and is also preferable for s ynthesis. Analysis of the heat of reaction for a series of trifluoro-containing anal ogues using semiempirical quantum chemical models showed that introduction of trif luoropropyl or trifluorobutyl group to norverapamil would not increase the heat of reaction (Table 4-1), which would make its dealkylation reactivity similar to verapamil. Hf of hydrogen abstraction also showed trifluoropropyl (27.8 kcal/mol) or trifluorobut yl (29.3 kcal/mol) norverapamil had similar values as verapamil (29.5 kcal/mol). Ther efore, the metabolism of trifluorinated analogues would be affected less, and more tr ifluorinated aldehyde would be produced as an exhaled breath marker.

PAGE 78

66 Table 4-1. Heat of reaction of verapam il and trifluoroalkylno rverapamil analogues. Heat of formation Heat of formation Heat of reaction (kcal/mol) (kcal/mol) (kcal/mol) Reactant Products Verapamil HCHO Norverapamil -89.6 -34.1 -88 -32.5 Trifluoroethylnorverapamil CF3CHO Norverapamil -246.9 -53.1 -88 105.8 Trifluoropropylnorverapamil CF3CH2CHO Norverapamil -251.9 -199.7 -88 -35.8 Trifluorobutylnorverapamil CF3(CH2)2CHO Norverapamil -258.4 -205.8 -88 -35 .4 Heat of formation was calculated by using semiempirical quantum chemical models (PM3) with CAChe software. Heat of reaction = sum of heats of formati on (products) – sum of heats of formation (reactants).

PAGE 79

67 CHAPTER 5 CONCLUSION To exploit the advantages of breath sa mples and facilitate phenotyping patients about their enzyme activity in the clinic, new analogues trifluoroethyldextrorphan and trifluoroethylnorverapamil were designed base d on the metabolism of two substrate drugs dextromethorphan and verapamil. They were hypothesized to generate trifluoroacetaldehyde after metabolism which c ould be excreted in the breath and may be used as a breath marker for CYP 2D6 a nd CYP 3A4, respectively. Our preliminary evaluation showed the feasibility of this concept. Both analogues were synthesized by introducing trifluoroethyl group to demethylated substance drugs. In vitro metabolism studies using human liver microsomes showed the designed analogues were less metabo lized than the original substrate drugs and were more potent in inhibiting enzymes activities due to the electron-withdrawing effect of trifluoromethyl group which could a ffect relative reactivity, binding affinity and orientation of substrate analogue s. However, trifluoroacetald ehyde was detected from the gas phase of the incubation mixtures, as we hypothesized. In order to improve the substrate properties of the designed analogues, a longer trifluoroalk yl group could be considered in the future studies. Results fr om quantum chemical modeling showed that a long chain would not decrease the reactivities of analogues compared to the substrate drugs.

PAGE 80

68 APPENDIX A MS AND 1H-NMR SPECTRA 200 250 300 350 400 450 500 550 600 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 258.27 199.20 157.20 293.07 209.13 554.93 513.40 387.20 355.07 446.13 579.40 200 250 300 350 400 450 500 550 600 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 258.27 199.20 157.20 293.07 209.13 554.93 513.40 387.20 355.07 446.13 579.40 Figure A-1. MS and 1H NMR of dextrorphan.

PAGE 81

69 200 300 400 500 600 700 800 900 1000 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 384.00 215.07 232.07 633.00 339.93 789.60 699.20 455.07 856.53 554.87966.67 200 300 400 500 600 700 800 900 1000 m/z 200 300 400 500 600 700 800 900 1000 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 384.00 215.07 232.07 633.00 339.93 789.60 699.20 455.07 856.53 554.87966.67 Figure A-2. MS and 1H NMR of N,O-bi s-VOC morphinan.

PAGE 82

70 200 300 400 500 600 700 800 900 1000 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 314.13 270.07 199.13 327.93 649.07 503.67 350.07 784.67 725.20 457.47 961.80 609.07 835.60 200 300 400 500 600 700 800 900 1000 m/z 200 300 400 500 600 700 800 900 1000 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 314.13 270.07 199.13 327.93 649.07 503.67 350.07 784.67 725.20 457.47 961.80 609.07 835.60 Figure A-3. MS and 1H NMR of 3-hydroxyN-VOC morphinan.

PAGE 83

71 200 300 400 500 600 700 800 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 569.26 694.77 396.03 481.89 521.97 367.10711.65 242.07 190.71 326.15 757.63 200 300 400 500 600 700 800 m/z 0 10 20 200 300 400 500 600 700 800 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 30 40 50 60 70 80 90 100Relative Abundance 569.26 694.77 396.03 481.89 521.97 367.10711.65 242.07 190.71 326.15 757.63 Figure A-4. MS and 1H NMR of 3-(2,2,2-trifluor oethyl)-N-VOC morphinan.

PAGE 84

72 200 300 400 500 600 700 800 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 340.20 162.73 281.13 446.33 222.73 370.40 503.73 653.87 595.00714.93 748.00 200 300 400 500 600 700 800 m/z 0 10 20 30 200 300 400 500 600 700 800 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 340.20 162.73 281.13 446.33 222.73 370.40 503.73 653.87 595.00714.93 748.00 Figure A-5. MS and 1H NMR of O-trifluoroe thyl dextrorphan.

PAGE 85

73 Figure A-6. 1H NMR of 2,2,2-trifluoroethyl-p-t oluenesulfonate (upper) and 2,2,2trifluoroethyl triflate (lower).

PAGE 86

74 : 150 200 250 300 350 400 450 500 550 600 650 700 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 523.11 165.12 289.13 441.23 371.14 260.15 469.21616.95 562.05 218.13302.10 660.36 : 150 200 250 300 350 400 450 500 550 600 650 700 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance : 150 200 250 300 350 400 450 500 550 600 650 700 m/z 0 10 20 30 40 50 60 70 80 90 100Relative Abundance 523.11 165.12 289.13 441.23 371.14 260.15 469.21616.95 562.05 218.13302.10 660.36 Figure A-7. MS and 1H NMR of trifluoroethyl norverapamil.

PAGE 87

75 APPENDIX B CALIBRATION CURVES y = 0.0384x 0.0337 R2 = 0.99750 0.5 1 1.5 2 0204060Concentration (uM)Ratio (area of analyte / area of IS ) Figure B-1. Calibration curve for dextrorphan. y = 0.1122x 0.3772 R2 = 0.9875 0 1 2 3 4 5 6 0102030405060Concentration (uM)Ratio (area of analyte / area of IS) Figure B-2. Calibration curve for norverapamil.

PAGE 88

76 LIST OF REFERENCES Abouabdellah A, Begue JP, Bonnet-Delpon D a nd Nga TTT (1997) Diastereoselective synthesis of the nonracemic methyl syn-(3-fluoroalkyl)isoserinates. J Org Chem 62 : 8826-8833. Ahnoff M and Persson B (1990) Chromat ography of calcium channel blockers. J Chromatogr 531 : 181. Akillu E, Persson I, Bertilsson L, Johansson I, Rodrigues F and Ingelman-Sundberg M (1996) Frequent contribution of ultra ra pid metabolizers of debrisoquine in an Ethiopian population carrying duplicat ed and multiduplicated functional CYP2D6 alleles. J Pharmacol Exp Ther 278 : 441-446. Baker MH, Foster AB, Hegedus L, Jarman M, Rowlands MG, Coe PL and Troth J (1984) Metabolism of 1,1,1,2,2-pentafluorohexane a nd 1,1-difluorocyclohexane by rat liver microsomes in vitro. Biomed Mass Spectrom 11 : 512-521. Ball SE, Scatina J, Kao J, Ferron GM, Fr uncillo R and Mayer P (1999) Population distribution and effects on drug metabolis m of a genetic variant in the 59 promoter region of CYP3A4. Clin Pharmacol Ther 66 : 288. Baumann P, Meyer JW, Amey M, Baettig D, Bryois C, Jonzier-Perey M, Koeb L, Monney C and Woggon B (1992) Dext romethorphan and mephenytoin phenotyping of patients treated with thioridazine or amitriptyline. Ther Drug Monit 14 : 1-8. Bergeron RJ, Bharti N, Wiegand J, McManis JS, Yao H and Prokai L (2005) Polyaminevectored iron chelators: the role of charge. J Med Chem 48 : 4120-4137. Blake RC III and Coon MJ (1989) On the mech anism of action of cytochrome P-450. J Biol Chem 264 : 3694-3701. Bohm HJ, Banner D, Bendels S, Kansy M, Kuhn B, Muller K, Obst-Sander U and Stahl M (2004) Fluorine in medicinal chemistry. Chem Bio Chem 5 : 637-643. Boriani G, Capucci A, Strocchi E, Calliva R, Santarelli A, Biffi M and Magnani B (1993) Flecainide acetate: concentration-respons e relationships for antiarrhythmic and electrocardiographic effects. Int J Clin Pharmacol Res 13 : 211-219.

PAGE 89

77 Breyer-Pfaff U and Nill K (2004) Carbonyl re duction of naltrexone and dolasetron by oxidoreductases isolated from human liver cytosol. J Pharm Pharmacol 56 : 1601. Camps F, Coll J, Messeguer A and Pericas MA (1980) A simple method for preparation of aryl 2,2,2-trifl uoroethyl ethers, Synthesis 9 : 727-728. Chang GWM and Kam PCA (1999) The physiol ogical and pharmacological roles of cytochrome P450 isoenzymes. Anaesthesia 54 : 42. Chladek J, Zimova G, Beranek M and Mart inkova J (2000) In vivo indices of CYP 2D6 activity: comparison of dextromethorphan me tabolic ratios in 4-h urine and 3-h plasma. Eur J Pharmacol 56 : 651-657. Cole SC and Flanagan RJ (1981) Rapi d high-performance liquid chromatographic method for the measurement of verapamil and norverapamil in blood plasma or serum. J Chromatogr 218 : 621. Conard GJ and Ober RE (1984) Metabolism of flecainide. Am J Cardiol 53 : 41B-51B. Cordis GA, Bagchi D, Maulik N an d Das DK (1994) High-performance liquid chromatographic method for the simultaneous detection of malonaldehyde, acetaldehyde, formaldehyde, acetone a nd propionaldehyde to monitor the oxidative stress in heart. J Chromatogr A 661 : 181-191. Croop RS, Faulkner EB and Labriola DF (1997) The safety profile of naltrexone in the treatment of alcoholism. Results from a multicenter usage study. The naltrexone usage study group. Arch Gen Psychiatry 54 : 1130. Dalen P, Dahl M, Eichelbaum M, Bertilss on L and Wilkinson GR (1999) Disposition of debrisoquine in Caucasians with diffe rent CYP 2D6 genotypes including those with multiple genes. Pharmacogenetics 9 : 697-706. Daly AK, Steen VM, Fairbrot her KS and Idle JR (1999) CYP2D6 multiallelism. Methods Enzymol 272 : 199-210. Dawson JH (1988) Probing stru cture-function relations in heme-containing oxygenases and peroxidases. Science 240 : 433-439. Dawson JH and Sono M (1987) Cytochrome P-4 50 and chloroperoxidase: thiolate-ligand heme enzymes. Spectroscopic determinati on of their active site structures and mechanistic implications of thiolate ligation. Chem Rev 87 : 1255-1276. De Wildt SN, Kearns GL, Leeder JS and va n den Anker JN (1999) Cytochrome P450 3A: ontogeny and drug disposition. Clin Pharmacokinet 7 : 485.

PAGE 90

78 Droll K, Bruce-Mensah K, Otton SV, Gaed igk A, Sellers, EM and Tyndale RF (1998) Comparison of three CYP 2D6 probe subs trates and genotyping in Ghanaians, Chinese and Caucasians. Pharmacogenetics 8 : 325-333. Echizen H, Brecht T, Niedergesass S, V ogelgesang B and Eichelbaum M (1985) The effect of dextro-, levo-, and racemic ve rapamil on atrioventricular conduction in humans. Am Heart J 109 : 210-217. Eichelbaum M and Woolhouse NM (1985) Inter-ethnic difference in SP oxidation among Ghanaians and Germans. Eur J Clin Pharmacol 28 : 79-83. Eichelbaum M, Ende M, Remberg G, Schomerus M and Dengler HJ (1979) The metabolism of DL-[14C]verapamil in man. Drug Metab Dispos 7 : 145. Eichhold TH, Greenfield LJ, Hoke SH 2nd and Wehmeyer KR (1997) Determination of dextromethorphan and dextrorpha n in human plasma by liquid chromatography/tandem mass spectrometry. J Mass Spectrom 32 : 1205-1211. Evans DAP, Mahgoub A, Sloan TP, Idle JR and Smith RL (1980) A family and population study of the genetic polymorphism of DB oxidation in a white British population. J Med Genet 17 : 102-105. Evans WE and Johnson JA (2001) Pharm acogenomics: the inherited basis for interindividual differenc es in drug response. Annu Rev Genomics Hum Genet 2 : 939. Gaedigk A, Blum M, Gaedigk R, Eichelba um M and Mayer UA (1991) Deletion of the entire cytochrome P450 CYP2D6 gene as a cause of impaired drug metabolism in poor metaboliser of the debris oquine/sparteine polymorphism. Am J Hum Genet 48 : 943-950. Garcia MA, Aramayona JJ, Bregante MA, Fr aile LJ and Solans C (1997) Simultaneous determination of verapamil and norve rapamil in biological samples by highperformance liqiud chromatogra phy using ultraviolet detection. J Chromatogr 693 : 377. Gonzales FJ and Idle JR (1994) Pharm acogenetic phenotyping and genotyping. Present status and future potential. Clin Pharmacokinetics 26 : 59-70. Guanti G, Banfi L, Narisano E, Scolas tico C and Bosone E (1985) Monobactams: stereoselective synthesis of trans-3-am inoand 3-acylamino-4-trifluoromethyl-2azetidinones. Synthesis 6/7 : 609-611. Guengerich FP (2004) Cytochrome P450: what have we learned and what are the future issues. Drug Metab Rev 36 : 159.

PAGE 91

79 Guttendorf RJ, Wedlund PJ, Blake J and Cha ng SL (1988) Simplified phenotyping with dextromethorphan by thin-layer chroma tography: application to clinical laboratory screening for deficienci es in oxidative drug metabolism. Ther Drug Monit 10 : 490-498. Hedeland M, Fredriksson E, Lennernas H and Bondesson U (2004) Simultaneous quantification of the enantiomers of verapamil and its N-demethylated metabolite in human plasma using liquid chroma tography-tandem mass spectrometry. J Chromatogr B 804 : 303-311. Heim M and Meyer UA (1990) Genotyping of poor metabolisers of debrisoquine by allel-specific PCR amplification. Lancet 336 : 529-32. Henne AL, Pelley RL and Alm RM (1950) The preparation of trifl uorinated aldehydes. J Am Chem Soc 72 : 3370-3371. Higgins L, Korzekwa KR, Rao S, Shou M and Jones JP (2001) An assessment of the reaction energetics for cytochrome P450mediated reactions. Arch Biochem Biophys 385 : 220-230. Hilbert J, Pramanik B, Symchowicz S a nd Zampaglione N (1984) The disposition and metabolism of a hypnotic benzodiazepine, quazepam, in the hamster and mouse. Drug Metab Dispos 12 : 452-459. Hou ZY, Chen CP, Yang WC, Lai MD, Buchert ET, Chung HM, Pickle LW and Woosley RL (1996) Dertermination of de xtromethorphan metabolic phenotype by salivary analysis with a reference to ge notype in Chinese patients receiving renal hemodialysis. Clin Pharmacol Ther 59 : 411-417. Hou ZY, Pickle LM, Meyer PS and Woosley RL (1991) Salivary analysis for determination of dextromethorphan metabolic phenothype. Clin Pharmacol Ther 49 : 410-419. Ingelman-Sundberg M, Oscarson M and Mc Lellan RA (1999) Polymorphic human cytochrome P450 enzymes: an opportunity for individualized drug treatment. Trends Pharmacol Sci 20 : 342-9. Jacqz-Aigran E, Funck-Brentano C a nd Cresteil T (1993) CYP2D6 and CYP3A dependent metabolism of dextromethorphan in humans. Pharmacogenetics 3 : 197-204. Janzowski C, Pool BL, Preussmann R and Eisenbrand G (1982) Fluoro-substituted Nnitrosamines. 2. Metabolism of N-nitrosodi ethylamine and of fluorinated analogs in liver microsomal fractions. Carcinogenesis 3 : 155-159.

PAGE 92

80 Johnstrom P and Stonee-Elander S (1995) The 18F-labelled alkylating agent 2,2,2trifluoroethyl triflate: synthesis and specific activity. J Labelled Comp Radiopharm 36 : 537-547. Kagimoto M, Heim M, Kagimoto K, Zeugi n T and Mayer UA (1990) Multiple mutations of the human cytochrome P450IID6 gene ( CYP2D6 ) in poor metabolizers of debrisoquine. J Biol Chem 265 : 17209-17214. Kalow W. (1997) Pharmacogenetics in biological perspective. Pharmacol Rev 49 : 369379. Karki SB and Dinnocenzo JP (1995) On th e mechanism of amine oxidations by P450. Xenobiotica 25 : 711-724. Kempter C, Zurek G and Karst U (1999) Determination of carbonyls using liquid chromatography-mass spectrometry with atmospheric pressure chemical ionization. J Environ Monit 1 : 307-311. Kerry NL, Somogyi AA, Bochner F and Mikus G (1994) The role of CYP2D6 in primary and secondary oxidative metabolism of dext romethorphan: in vitro studies using human liver microsomes. Br J Clin Pharmacol 38 : 243-248. Kevorkian JP, Michel C, Hofmann U, J acqz Aigrain E, Kroemer HK, Peraldi MN Eichelbaum M, Jaillon P and Funck Bren tano C (1996) Assessment of individual CYP 2D6 activity in extensive metabolizer s with renal failure: comparison of spateine and dextromethorphan. Clin Pharmacol Ther 59 : 583-592. Kintz P, Mangin P, Lugnier A and Ch aumont AJ (1989) Determination of dextromethorphan and its major metabol ite dextrorphan by gas chromatography. Ann Biol Clin (Paris) 47 : 193. Klocker J, Karpfen A and Wolschann P (2003) On the structure and torsional potential of trifluoromethoxybenzene: an ab in itio and density functional study. Chem Phys Lett 367 : 566-575. Kohler D, Hartter S, Fuchs K, Sieghart W and Hiemke C (1997) CYP 2D6 genotype and phenotyping by dertermination of dextrome thorphan and metabolites in serum of healthy controls and of patie nts under psychotropic medication. Pharmacogenetics 7 : 453-461. Kolliker S, Oehme M and Dye C (1998) Structure elucidation of 2, 4dinitrophenylhydrazone derivatives of carbonyl compounds in ambient air by HPLC/MS and multiple MS/MS using atmo spheric chemical ionization in the negative ion mode. Anal Chem 70 : 1979-1985.

PAGE 93

81 Kramer A, Staudinger H and Ullrich V (1974) Effect of n-hexane inhalation on the monooxygenase system in mice liver microsomes. Chem Biol Interact 8 : 11-18. Kroemer HK, Echizen H, Heidemann H and Ei chelbaum M (1992) Predictability of the in vivo metabolism of verapamil from in vitro data: contribution of individual metabolic pathways and stereoselective aspects. J Pharmacol Exp Ther 260 : 1052. Kroemer HK, Gautier JC, Beaune P, Henders on C, Wolf CR and Ei chelbaum M (1993) Identification of P450 enzymes involved in metabolism of verapamil in humans. Naunyn Schmiedebergs Arch Pharmacol 348 : 332. Kumadaki I, Jonoshita S, Harada A, Omote M and Ando A (1999) Synthesis of organofluorine compounds using ene t ype reaction of N -(p-toluenesulfonyl) trifluoroacetaldehyde imine. J Fluor Chem 97 : 61-63. Kumar GN and Surapaneni S (2001) Role of drug metabolism in drug discovery and development. Med Res Rev 21 : 397411. Larroque C, Lange R, Maurin L, Bienvenue A and van Lier JE (1990) On the nature of the cytochrome P450scc "u ltimate oxidant": characterization of a productive radical intermediate. Arch Biochem Biophys 282 : 198-201. Lawal L, Nguyen T, Mayo JT, Tinnerman WN and Malloy TB Jr. (2004) Synthesis of Ofluoroalkylhydroxylamine hydrochlorides. Re gional American Chemical Society Meeting, Fort Worth, TX. Leahy DE, Lynch J and Taylor DD (1989) Mechan isms of absorption of small molecules, in Novel Drug Delivery (Prescott LF and Nimmo WS eds) pp 33, John Wiley & Sons, New York. Leshner AI (2003) Understanding drug addi ction: insights from the research, in Principles of Addiction Medicine (Graham AW, et al. eds) pp 47, American Society of Addiction Medicine, Chevy Chase, MD. Levine B, Chute D and Caplan YH (1990) Flecainide intoxication. J Anal Toxicol 14 : 335-336. Levy RH, Thummel KE, Trager WF, Hansten PD and Eichelbaum M (2000) Metabolic Drug Interaction . Lippincott William & Wilkins, Philadelphia, PA. Leyland-Jones B (2004) US Patent Pub. No. 2004/0101477 A1. Washington, DC: US Patent and Trademark Office.

PAGE 94

82 Lin J, Sahakian DC, Morais SMF, Xu JJ, Po lzer RJ and Winter SM (2003) The role of absorption, distribution, metabolism, excr etion and toxicity in drug discovery. Curr Top Med Chem 3 : 1125-1154. Lin JH and Lu AYH (1997) Role of pharmac okinetics and metabolism in drug discovery and development. Pharmacol Rev 49 : 403-449. Lin Y, Dueker SR, Jones AD, Ebeler SE and Clifford AJ (1995) Protocol for collection and HPLC analysis of volatile carbonyl compounds in breath. Clin Chem 41 : 1028-1032. Linder MW, Prough RA and Valdes R Jr. (1997) Pharmacogenetics: a laboratory tool for optimizing therapeutic efficiency. Clin Chem 43 : 254-266. Litten RZ, Fertig J, Mattson M and Egli M (2005) Development of medications for alcohol use disorders: recent advances and ongoing challenges. Expert Opin Emerg Drugs 10 : 323. Lou YC (1990) Differences in drug meta bolism polymorphism between Orientals and Caucasians. Drug Metab Rev 22 : 451-475. Lu XL, Guengerich FP and Yang SK (1991) St ereoselective metabo lism of prazepam and halazepam by human liver microsomes. Drug Metab Dispos 19 : 637-642. Lucas D, Menez JF, Berthou F, Pennec Y a nd Floch HH (1986) Determination of free acetaldehyde in blood as the dinitrophenylhydrazone derivative by highperformance liquid chromatography. J Chromatogr 382 : 57-66 McAllister RG and Howell SM (1976) Fluorometric assay of verapamil in biological fluids and tissues. J Pharm Sci 65 : 431-432 McCauley-Myers DL, Eichhold TH, Bailey RE , Dobrozsi DJ, Best KJ, Hayes JW 2nd and Hoke SH 2nd (2000) Rapid bioanalytical determ ination of dextromethorphan in canine plasma by dilute-and-shoot prepar ation combined with one minute per sample LC-MS/MS analysis to optimize formulations for drug delivery. J Pharm Biomed Anal 23 : 825-835. McLellan RA, Oscarson M, Seidegard J, Evans DA and Ingelman-Sundberg M (1997) Frequent occurrence of CYP2D6 gene duplication in Saudi Arabians. Pharmacogenetics 7 : 187-191. McMahon RE (1966) Microsomal dealkylati on of drugs, substrate specificity and mechanism. J Pharm Sci 55 : 457-466.

PAGE 95

83 McQuinn RL, Pentikainen PJ, Chang SF and Co nard GJ (1988) Pharmacokinetics of flecainide in patients with cirrhosis of the liver. Clin Pharmacol Ther 44 : 566572. McQuinn RL, Quarfoth GJ, Johnson JD, Banitt EH, Pathre SV, Chang SF, Ober RE and Conard GJ (1984) Biotransfo rmation and elimination of 14C-flecainide acetate in humans. Drug Metab Dispos 12 : 414-420. Michalets EL (1998) Review of therapeutics, update: clinically sign ificant cytochrome P450 drug interactions. Pharmacotherapy 18 : 84-112. Michellys PY, Ardecky RJ, Chen JH, D’Arri go J, Grese TA, Karanewsky DS, Leibowitz MD, Liu S, Mais DA, Mapes CM, Montrose-Rafizadeh C, Ogilvie KM, ReifelMiller A, Rungta D, Thompson AW, Tyhonas JS and Boehm MF (2003) Design, synthesis, and structure-activity relati onship studies of novel 6,7-locked-[7-(2alkoxy-3,5-dialkylbenzene)-3-methy locta]-2,4,6-trienoic acids. J Med Chem 46 : 4087-4103. Mikus G, Eichelbaum M, Fischer C, Gu mulka S, Klotz U and Kroemer H (1990) Interaction of verapamil and cimetidin c: stereochemical aspects of drug metabolism, drug disposition and drug action. J Pharmacol Exp Ther 253 : 1042– 1048. Misra AL, Bloch R, Vardy J, Mule SJ and Ve rebely K (1976) Disposition of [l5, l6-3H] naltrexone in the central nervous system of the rat. Drug Metab Dispos 4 : 276280. Mistry B, Leslie J and Eddington NE (1998) A sensitive assay of metoprolol and its major metabolite alpha-hydroxymetoprolol in human plasma and determination of dextromethorphan and its metabolite dextrorphan in urine with high performance liquid chromatography and fluorometric detection. J Pharm Biomed Anal 16 : 1041. Nagy K, Pollreisz F, Takats Z and Vekey K (2004) Atmospheric pressure chemical ionization mass spectrometry of aldehydes in biological matrices. Rapid Commun Mass Spectrom 18 : 2473-2478. Narasimhan LR, Goodman W and Patel CK ( 2001) Correlation of breath ammonia with blood urea nitrogen and crea tinine during hemodialysis. Proc Natl Acad Sci U S A 98 : 4617-4621. Nebert DW (1997) Polymorphisms in drug-meta bolizing enzymes: what is their clinical relevance and why do they exist. Am J Hum Genetics 60 : 265-271.

PAGE 96

84 Nelson DR, Kamataki T, Waxman DJ, Guendegrich FP, Estabrook RW, Feyeriesen R, Gonzalez FJ, Coon MJ, Gunsalus IC, Gotoh O, Okuda K and Nebert DW (1993) The P450 superfamily: update on new sequences, gene mapping, accession number, early trivial names of enzyme, and nomenclature. DNA Cell Biol 12 : 151. Netscher T and Bohrer P (1996) Formation of sulf inate esters in the synthesis of triflates. Tetrahedron Lett 37 : 8359-8362. Obach RS, Baxter JG, Liston TE, Silber BM , Jones BC, MacIntyre F, Rance DJ and Wastall P (1997) The prediction of hu man pharmacokinetic parameters from preclinical and in vitro metabolism data. J Pharmacol Exp Ther 283 : 46-58. Ohara H, Miyabe Y, DeyashikiY, Matsuur a K and Hara A (1995) Reduction of drug ketones by dihydrodiol dehydrogenases, carbonyl reductase and aldehyde reductase of human liver. Biochem Pharmacol 50 : 221. Ohata H, Otsuka M and Ohmo ri S (1997) Determination of acetaldehyde in biological samples by gas chromatography with electron-capture detection. J Chromatogr B 693 : 297-305. Olofson RA and Schnur RC (1977) Value of the vinyloxycarbonyl unit in hydroxyl protection: application to th e synthesis of nalorphine. Tetrahedron Lett 18 : 15711574. Olsen JA, Banner DW, Seiler P, Sander UO, D’Arcy A, Stihle M, Muller K and Diederich F (2003) A fluorine scan of thrombin inhibitors to map the fluorophilicity/fluorophobicity of an enzy me active site: Evidence for C-F center dot center dot center dot C=O interactions. Angew Chem Intl Ed 42 : 2507-2511. Oncken C, Van Kirk J and Kranzler HR (2001) Adverse effects of oral naltrexone: analysis of data from two clinical trials. Psychopharmacology (Berl) 154 : 397– 402. zdemir M, Crewe KH, Tucker GT and Rost ami-Hodjegan A (2004) Assessment of in vivo CYP2D6 activity: differential sensit ivity of commonly used probes to urine pH J Clin Pharmacol 44 : 1398-1404. Paine MF, Khalighi M, Fisher JM, Shen DD, Kunze KL, Marsh CL, Perkins JD and Thummel KE (1997) Characteri zation of intera nd intra-intestinal differences in human CYP3A-dependent metabolism. J Pharmacol Exp Ther 283 : 1552-1562. Paine MF, Wagner DA, Hoffmaster KA and Wa tkins PB (2002) Cytochrome P450 3A4 and P-glycoprotein mediate the interacti on between an oral erythromycin breath test and rifampin. Clin Pharmacol Ther 72 : 524-535.

PAGE 97

85 Paredi P, Kharitonov SA and Barnes PJ ( 2002) Analysis of expired air for oxidation products. Am J Respir Crit Care Med 166 : 531-537. Park HM, Eo YW, Cha KS, Kim YM and Lee KB (1998) Determination of free acetaldehyde in total blood for investigati ng the effect of aspartate on metabolism of alcohol in mice. J Chromatogr B 719 : 217-221. Priego-Lopez E and Luque de Casto MD (2002) Pervaporation-gas chromatography coupling for slurry samples. Determin ation of acetaldehyde and acetone in food. J Chromatogr A 976 : 399-407. Rendic S and Di Carlo FJ (1997) Human Cyto chrome P450 enzymes: a status report summarizing their reactions, subs trates, inducers and inhibitors. Drug Metab Rev 29 : 413-580. Rinn W, Desai N, Rosenblatt H and Gastfrie nd DR (2002) Addiction denial and cognitive dysfunction: a preliminary investigation. J Neuropsychiatry Clin Neurosci 14 : 5257. Rivory LP, Slaviero KA, Hoskins JM and Clarke SJ (2001) The erythr omycin breath test for the prediction of drug clearance. Clin Pharmacokinet 40 : 151-158. Rohsenow DJ, Colby SM, Monti PM, Swift RM, Martin RA, Mueller TI, Gordon A and Eaton CA (2000) Predictors of compliance with naltrexone among alcoholics. Alcohol Clin Exp Res 24 : 1542. Sachse C, Brockmoller J, Bauer S and Roots I (1997) Cytochrome P 450 2D6 variants in a Caucasian population: allele freque ncies and phenotypic consequences. Am J Hum Genet 60 : 284-295. Salsali M, Coutts RT and Baker GB (1999) Analysis of dextrorphan, a metabolite of dextromethorphan, using gas chromatogr aphy with electron-capture detection. J Pharmacol Toxicol 41 : 143-146. Sata F, Sapone A, Elizondo G, Stocker P, Miller VP and Zh eng W (2000) CYP3A4 allelic variants with amino acid substituti ons in exons 7 and 12: evidence for an allelic variant with altered catalytic activity. Clin Pharmacol Ther 67 : 48-56. Schmid B, Bircher J, Preisig R and Kupfer A (1985) Polymorphic DEX metabolism: cosegregation of oxidative O-demethyla tion with debrisoquin hydroxylation. Clin Pharmacol Ther 38 : 618-624. Schmider J, Greenblatt DJ, Fogelman SM, Moltke LLV and Shader RI (1997) Metabolism of dextromethorphan in vitro: involvement of cytochromes P450 2D6 and 3A3/4, with a possible role of 2E1. Biopharm Drug Dispos 18 : 227-240.

PAGE 98

86 Senderoff SG, Landvatter SW and Heys JR (2000) A novel synthesis of [O-methyl-14C] dextromethorphan: utility of vinyloxycarbonyl group. J Labelled Cpd Radiopharm 43 : 1283-1288. Seydel JK and Schaper KJ (1986) Quantitati ve structure-pharmacokinetic relationships and drug design, in Pharmacokinetics: Theory and Methodology (Rowland M and Tucker G eds) pp 311, Pergamon Press, New York. Shara MA, Dickson PH, Bagchi D and Stohs SJ (1992) Excretion of formaldehyde, malondialdehyde, acetaldehyde and acetone in the urine of rats in response to 2,3,7,8-tetrachlorodibenzo-p-dioxin, para quat, endrin and carbon tetrachloride. J Chromatogr B 576 : 221-233. Shimada T, Yamazaki H and Mimura M (1994) In terindividual variations in human liver cytochrome P450 enzymes involved in the oxidation of drugs, carcinogen, and toxic chemicals. Studies with liver micr osomes of 30 Japanese and 30 Caucasians. J Pharmacol Exp Ther 270 : 414-423. Shin HS, Shin YS, Kim HJ and Kang YK (1996) Sensitive assay for verapamil in plasma using gas-liquid chromatography w ith nitrogen-phosphorus detection. J Chromatogr 677 : 369. Singh RP, Manandhar S and Shreeve (2003) Monoand disubstituted polyfluoroalkylimidazolium quate rnary salts and ionic liquids. Synthesis 10 : 1579-1585. Smith DH, van de Waterbeemd H and Walker DK (2001) Pharmacokinetics and Metabolism in Drug Design, Methods and Pr inciples in Medicinal Chemistry . Vol. 13, Wiley-VCH, Weinheim. Sonesson C, Lin CH, Hansson L, Waters N, Svensson K, Carlsson A, Smith MW and Wikstrom H. (1994) Substituted (S)-phe nylpiperidines and rigid congeners as preferential dopamine autoreceptor anta gonists: synthesis an d structure-activity relationships. J Med Chem 37 : 2735-2753. Srivastava P (2003) Drug Metabo lism and Individualized Medicine. Curr Drug Metab 4 : 33-44. Stagni G and Gillespie WR (1995) Simultaneous analysis of verapamil and norverapamil enatiomers in human plasma by high-performance liquid chromatography. J Chromatogr 667 : 349. Stavchansky S, Demirbas S, Reyderman L and Chai CK (1995) Simultaneous determination of dextrorphan and gua ifenesin in human plasma by liquid chromatography with fluorescence detection. J Pharm Biomed Anal 13 : 919.

PAGE 99

87 Steinman M, Topliss JP, Alekel R, Wong YS and York EE (1973) 1Polyfluoroalkylbenzodizepines 1. J Med Chem 16 : 1354-1360. Streetman DS, Bertino JSJ and Nafziger AN (2000) Phenotyping of drug-metabolizing enzymes in adults: a review of in-v ivo cytochrome P450 phenotyping probes. Pharmacogenet Genomics 10 : 187-216. Tenneze L, Tarral E, Ducloux N and Funck-Br entano C (2002) Pharmacokenitics and electrocardiographic effect s of a new controlled-rele ase form of flecainide acetate: comparison with the standard form and influence of the CYP 2D6 polymorphism. Clin Pharmacol Ther 72 : 112-122. Thummel KE, Shen DD, Podoll TD, Kunze KL , Trager WF and Bacchi CE (1994) Use of midazolam as a human cytochrome P450 3A probe: II. Characterization of interand intraindivi dual hepatic CYP3A variability after liver tr ansplantation. J Pharmacol Exp Ther 271 : 557-566. Tomita M, Okuyama T, Hatta Y and Kawai S (1990) Determination of free malonaldehyde by gas chromatography w ith an electron-capture detector. J Chromatogr B 526 : 174-179. Tracy TS, Korzekwa KR, Gonzalez FJ and Wainer IW (1999) Cytochrome P450 isoforms involved in metabolism of the enantiomers of verapamil and norverapamil. Br J Clin Pharmacol 47 : 545-552. Valdes R, Jortani SA and Gheorghiade M (1998) Standards of laboratory practice: cardiac drug monitoring. Clin Chem 44 : 1096-1109. Van Schaik RHN, de Wildt SN, Brosens R, van Fessem M, van den Anker JN and Lindemans J (2001) The CYP3A4*3 Allele: Is It Really Rare. Clin Chem 47 : 1104-1106. Vasiliades J, Wilkerson K, Ellul D, Anticoli M and Rocchini AP (1982) Gaschromatographic determination of verapamil and norverapamil, with a nitrogenselective detector. Clin Chem 28 : 638-641. Vengurlekar SS, Heitkamp J, McCush F, Vela galeti PR, Brisson JH and Bramer SL (2002) A sensitive LC-MS/MS assay for the determination of dextromethorphan and metabolites in human urine-applicati on for drug interaction studies assessing potential CYP 3A a nd CYP 2D6 inhibition. J Pharm Biomed Anal 30 : 113-124. Vesell ES (1997) Therapeutic le ssons from pharmacogenetics. Ann Intern Med 126 : 653655.

PAGE 100

88 Vielnascher E, Spatzenegger M, Mayrhofer A, Klinger P and Jager W (1996) Metabolism of dextromethorphan in human liver micr osomes: a rapid HPLC assay to monitor cytochrome P450 2D6 activity. Pharmazie 51 : 586. Vlncent-Viry M, Muller J, Fournler B, Ga lteau MM and Siest G ( 1991) Relation between debrisoquine oxidation phenotype a nd morphological, biological, and pathological variables in a large population. Clin Chem 37 : 327-332. Vogelgesang B, Echizen H, Schmidt E and Eichelbaum M (1984). Stereoselective firstpass metabolism of highly cleared drugs: stud ies of the bioavailability of Land D-verapamil examined with a stable isotope technique. Br J Clin Pharmacol 18 : 733-740. Volpicelli JR, Rhines KC, Rhines JS, Volpicel li LA, Alterman AI and O’Brien CP (1997) Naltrexone and alcohol dependence. Role of subject compliance. Arch Gen Psychiatry 54 : 737. Wall ME, Brine DR and Perez-Reyes M (1981) Metabolism and disposition of naltrexone in man after oral and intravenous administration. Drug Metab Dispos 9 : 369. Walles M, Thum T, Levsen K and Borlak J (2001) Verapamil metabolism in distinct regions of the heart and in cultures of cardiomyocytes of adult rats. Drug Metab Dispos 29 : 761-768. Wang Y, Jones DR and Hall SD (2004) Predic tion of cytochrome P450 3A inhibition by verapamil enantiomers and their metabolites. Drug Metab Disp 32 : 259-266. Watanabe Y, Oae S and Iyanahi T (1982) M echanisms of enzymatic S-oxygenation of thioanisole derivatives and O-demethylat ion of anisole derivatives promoted by both microsomes and a reconstitutes sy stem with purified cytochrome P-450. Chem Soc Jpn 55 : 188-195. Weinshilboum R and Wang L (2004) Phar macogenomics: Bench to bedside. Nat Rev Drug Discov 3 : 739-748. Wikimedia Foundation, Inc.(2006) http://en.wikipedia.org/wiki/Metabolism . Last accessed: July 20, 2006. Wrighton SA and Stevens JC (1992) The human hepatic cytochrome P450 involved in drug metabolism, Crit Rev Toxicol 22 : 1-21. Wrighton SA, Schuetz EG, Thummel KE, Shen DD, Korzekwa KR and Watkins PB (2000) The human CYP3A subfamily practical considerations. Drug Metab Rev 32 : 339-361.

PAGE 101

89 Wu D, Otton SV, Sproule BA, Busto U, Inaba T, Kalow W and Sellers EM (1993). Inhibition of human cytochrome P450 2D6 (CYP2D6) by methadone. Br J Clin Pharmacol 35 : 30-34. Wu YJ, Cheng YY, Zeng S and Ma MM (2003) Determination of dextromethorphan and its metabolite dextrorphan in human urine by capillary gas chromatography without derivatization J Chromatogr B 784 : 219-224. Xu YX, Shen L and Zhang CJ (1993) Analysis of dextromethorphan and its metabolites in human urine by using gas chromatography-mass spectrometry. Acta Pharmaceutica Sinica 28 : 156-159. Zurek G and Karst U (1999) Liquid chroma tography-mass spectrometry method for the determination of aldehydes deriva tized by the Hantzsch reaction. J Chromatogr A 864 : 191-197.

PAGE 102

90 BIOGRAPHICAL SKETCH Xiaoli Wang was born and raised in Shaanxi, P. R. China. She received her master’s degree in pharmaceutical sciences fr om Xi’an Jiaotong University in June 2001. After graduation, she worked in an industrial departme nt of drug discovery and development for a year. In August 2002, she entered the Ph.D. program at the Department of Medicinal Chemistry, College of Pharmacy, University of Florida, working under the supervision of Dr. Laszlo Prokai. Xiaoli received her Doctor of Philosophy degree in pharmaceutic al sciences in August 2006.