UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

In vitro studies of allelic variation and function of aldo-keto reductases 1A1 and 1C2 Takahashi, Ryan Hiro 2009

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

Item Metadata

Download

Media
24-ubc_2009_fall_takahashi_ryan.pdf [ 6.42MB ]
Metadata
JSON: 24-1.0070830.json
JSON-LD: 24-1.0070830-ld.json
RDF/XML (Pretty): 24-1.0070830-rdf.xml
RDF/JSON: 24-1.0070830-rdf.json
Turtle: 24-1.0070830-turtle.txt
N-Triples: 24-1.0070830-rdf-ntriples.txt
Original Record: 24-1.0070830-source.json
Full Text
24-1.0070830-fulltext.txt
Citation
24-1.0070830.ris

Full Text

IN VITRO STUDIES OF ALLELIC VARIATION AND FUNCTION OF ALDO-KETO REDUCTASES 1A1 AND 1C2  by  Ryan Hiro Takahashi B.Sc. (Chemistry), Simon Fraser University, 2001  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Pharmaceutical Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  June 2009  © Ryan Hiro Takahashi, 2009  ABSTRACT The aldo-keto reductases (AKR) are a superfamily of enzymes involved in metabolizing  a  variety  of  endogenous  and  xenobiotic  chemicals.  These  biotransformation reactions are being recognized as important bioactivation or deactivation steps in humans that may determine the efficacy or toxicity of drug therapy or disease development. Consequently, the factors that affect the function of AKRs may be critical determinants for interpatient differences in therapeutic response and likelihood for disease occurrence. The metabolism of anthracycline drugs and androgen steroids are proposed to be involved in drug-induced cardiotoxicity and prostate cancer development, respectively. In the former case, the metabolism of anthracyclines can generate reactive species or toxic metabolites that are damaging to heart tissue. In the latter case, metabolic pathways that synthesize, interconvert, and deactivate androgen steroids regulate androgen receptor activation, and imbalance of these processes may promote abnormal prostate growth. The occurrence of both disorders is highly unpredictable, which may be, in part, due to interpatient variability in metabolism.  The anthracyclines doxorubicin (DOX) and  daunorubicin (DAUN) and the androgen steroid dihydrotestosterone (DHT) are substrates for two human isoforms of AKR protein, namely AKR1A1 and AKR1C2. Thus, factors affecting the function of these enzymes may contribute to altered metabolism and development of these conditions. The goal of this project is to characterize the function of AKR1A1 and AKR1C2 in metabolizing the anthracyclines DOX and DAUN, and the androgen steroid DHT using in vitro techniques. The naturally-occurring allelic variations in the coding regions  ii  of the AKR genes are compared to wild-type to identify their association with impaired reductase function. In addition, allosteric activation of AKR1C2, which has not been confirmed since its first report in the literature, is investigated for its relevance to DHT metabolism. The results of these studies demonstrate that genetic variation and chemical modulation can significantly alter the metabolism of DOX, DAUN, or DHT and, therefore, potentially contribute to the variability in patient response to these chemicals. These data can form a foundation of biochemical evidence to design in vivo studies that will elucidate the role of altered metabolism in developing anthracycline-associated cardiotoxicity and prostate cancer.  iii  TABLE OF CONTENTS Abstract...........................................................................................................................ii Table of Contents...........................................................................................................iv List of Tables ................................................................................................................vii List of Figures..............................................................................................................viii List of Abbreviations....................................................................................................... x Acknowledgements ......................................................................................................xiii Co-authorship Statement .............................................................................................. xiv CHAPTER 1 Background .............................................................................................. 1 1.1. Aldo-Keto Reductases ....................................................................................... 1 1.1.1. Overview .................................................................................................... 1 1.1.2. Human aldo-keto reductases........................................................................ 3 1.1.3. Mechanism of reduction.............................................................................. 5 1.1.4. Polymorphism in AKRs .............................................................................. 7 1.1.5. Aldo-keto reductase 1A1............................................................................. 7 1.1.6. Aldo-keto reductase 1C2........................................................................... 11 1.2. Anthracycline Chemotherapeutics and Their Cardiac Toxicity......................... 16 1.2.1. Anthracycline drugs.................................................................................. 16 1.2.2. Cardiotoxicity ........................................................................................... 19 1.2.3. Role of reductases in anthracycline metabolism ........................................ 24 1.2.4. Implications of altered anthracycline metabolism...................................... 24 1.3. Androgens and Prostate Cancer ....................................................................... 26 1.3.1. Prostate cancer.......................................................................................... 26 1.3.2. Androgen receptor activation and prostate cancer...................................... 26 1.3.3. Pharmacological role of DHT ................................................................... 29 1.3.4. Androgen synthesis and degradation ......................................................... 30 1.3.5. Role of aldo-keto reductase 1C2 in androgen metabolism ......................... 34 1.3.6. Pharmacological role of 3!-diol ................................................................ 35 1.3.7. Implications of altered androgen metabolism ............................................ 35 1.4. Rationale ......................................................................................................... 36 1.5. Research Overview and Hypotheses ................................................................ 37 1.6. Specific Objectives .......................................................................................... 38 1.7. References ....................................................................................................... 40 CHAPTER 2 Two Allelic Variants of Aldo-Keto Reductase 1A1 Exhibit Reduced In Vitro Metabolism of Daunorubicin, ............................................................................... 48 2.1. Preface............................................................................................................. 48 2.2. Materials and Methods..................................................................................... 50 2.2.1. Chemicals and enzymes ............................................................................ 50 2.2.2. Molecular cloning of the human AKR1A1 gene ......................................... 51 2.2.3. Expression of recombinant human AKR1A1............................................. 52 2.2.4. Purification of recombinant human AKR1A1............................................ 53 2.2.5. Total protein staining ................................................................................ 54 2.2.6. Western blotting........................................................................................ 54 2.2.7. AKR1A1 enzymatic activity assays........................................................... 55 iv  2.2.8. Statistical analysis..................................................................................... 57 2.3. Results............................................................................................................. 57 2.3.1. Expression and purification....................................................................... 57 2.3.2. AKR1A1 enzymatic activities with test substrates..................................... 58 2.3.3. AKR1A1 enzymatic activities with anthracyclines.................................... 61 2.4. Discussion ....................................................................................................... 64 2.5. References ....................................................................................................... 70 CHAPTER 3 Aldo-Keto Reductase 1C2 Fails to Metabolize Doxorubicin and Daunorubicin In Vitro ................................................................................................... 73 3.1. Preface............................................................................................................. 73 3.2. Materials and Methods..................................................................................... 74 3.2.1. Chemicals and enzymes. ........................................................................... 74 3.2.2. Cloning of human AKR1A1 and 1C2........................................................ 74 3.2.3. Expression of recombinant enzymes. ........................................................ 75 3.2.4. Measurement of AKR enzyme activities. .................................................. 76 3.2.5. Measurement of anthracycline reduction................................................... 76 3.3. Results and Discussion .................................................................................... 77 3.4. References ....................................................................................................... 82 CHAPTER 4 Quantitation of DHT and its Reduction Metabolites by LC/MS/MS ........ 84 4.1. Preface............................................................................................................. 84 4.2. Materials and Methods..................................................................................... 86 4.2.1. Chemicals ................................................................................................. 86 4.2.2. Extraction and derivatization of DHT and metabolites .............................. 86 4.2.3. Chromatographic separation and mass spectrometric analysis of derivatized DHT and metabolites............................................................................................. 87 4.2.4. Assay validation........................................................................................ 88 4.3. Results............................................................................................................. 89 4.4. Discussion ....................................................................................................... 94 4.5. References ....................................................................................................... 97 CHAPTER 5 Allosteric Activation of AKR1C2 Reduction of Dihydrotestosterone by Antidepressant Drugs and Its In Vivo Significance ........................................................ 99 5.1. Preface............................................................................................................. 99 5.2. Materials and Methods................................................................................... 102 5.2.1. Chemicals ............................................................................................... 102 5.2.2. AKR1C2 protein..................................................................................... 102 5.2.3. In vitro reduction of DHT by AKR1C2 ................................................... 103 5.2.4. Kinetics of DHT reduction...................................................................... 103 5.2.5. Effect of modulators on AKR1C2 reduction of DHT............................... 104 5.2.6. Effect of bovine serum albumin on AKR1C2 reduction of DHT.............. 105 5.3. Results........................................................................................................... 105 5.4. Discussion ..................................................................................................... 111 5.5. References ..................................................................................................... 119  v  CHAPTER 6 The Effect of Allelic Variation in Aldo-Keto Reductase 1C2 on the In Vitro Metabolism of Dihydrotestosterone.................................................................... 122 6.1. Preface........................................................................................................... 122 6.2. Materials and Methods................................................................................... 124 6.2.1. Materials................................................................................................. 124 6.2.2. Construction of expression plasmids ....................................................... 125 6.2.3. Expression of AKR1C2 in insect cells..................................................... 128 6.2.4. Western blotting of AKR1C2.................................................................. 129 6.2.5. Kinetic analysis of AKR1C2................................................................... 130 6.2.6. Data analysis........................................................................................... 132 6.3. Results........................................................................................................... 133 6.3.1. Expression of AKR1C2 in insect cells..................................................... 133 6.3.2. Probe substrate and DHT reduction by wild-type and variant alleles of AKR1C2 ............................................................................................................. 135 6.4. Discussion ..................................................................................................... 140 6.5. References ..................................................................................................... 148 CHAPTER 7 Summary .............................................................................................. 151 7.1. Overall Conclusions....................................................................................... 151 7.2. Implications of Findings ................................................................................ 155 7.3. Scope and Limitations of Research ................................................................ 159 7.4. Suggested Future Research Directions ........................................................... 164 7.4.1. Application of the developed technologies .............................................. 164 7.4.2. Biochemical investigations...................................................................... 166 7.4.3. Patient genotype-phenotype correlation studies....................................... 168 7.5. References ..................................................................................................... 171  vi  LIST OF TABLES Table 1.1 Human AKR isoforms identified to-date.......................................................... 4 Table 1.2 Non-synonymous single nucleotide polymorphisms naturally occurring in the AKR1A1 gene compiled from the National Centre for Biotechnology Information (NCBI) SNP database.............................................................................................. 9 Table 1.3 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1A1 compiled from the National Centre for Biotechnology Information (NCBI) SNP database......................................................................... 10 Table 1.4 Non-synonymous single nucleotide polymorphisms naturally occurring in the AKR1C2 gene compiled from the National Centre for Biotechnology Information (NCBI) SNP database............................................................................................ 14 Table 1.5 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1C2 compiled from the National Centre for Biotechnology Information (NCBI) SNP database......................................................................... 15 Table 2.1 Kinetic constants for DAUN reduction by recombinant AKR1A1 wildtype and variant allele proteins............................................................................................. 64 Table 4.1 Accuracy and precision, intra-day and inter-day, for the LC/MS/MS determinations of DHT, 3!-diol, and 3"-diol in in vitro samples. .......................... 93 Table 5.1 Summary of AKR1C2 catalyzed reduction of DHT in absence and presence of modulating compounds........................................................................................ 107 Table 6.1 PCR primers used for site-directed mutagenesis to create AKR1C2 variant alleles.. ................................................................................................................ 127 Table 6.2 Kinetic parameters determined for the reduction of the fluorogenic probe by wild-type and variant AKR1C2 expressed in T. ni insect cells. ............................ 138 Table 6.3 Kinetic parameters determined for the reduction of DHT by wild-type and variant AKR1C2 expressed in Sf9 insect cells. .................................................... 139 Table 6.2 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1C2 compiled from the National Centre for Biotechnology Information (NCBI) SNP database....................................................................... 146  vii  LIST OF FIGURES Figure 1.1 AKR protein structure showing conserved (!/")8 barrel structure.................... 2 Figure 1.2 Proposed catalytic mechanism for AKR-mediated reduction and oxidation reactions. ................................................................................................................. 6 Figure 1.3 Chemical structures of doxorubicin, daunorubicin, and epirubicin. A high level of structural similarity exists between these anthracyclines............................ 16 Figure 1.4 Metabolic scheme for anthracyclines, showing pathways for doxorubicin metabolism in human cardiac cytosol. ................................................................... 19 Figure 1.5 Schematic showing the one-electron reduction of doxorubicin leading to the generation of reactive oxygen species and iron reduction....................................... 22 Figure 1.6 Schematic of androgen receptor activation by intraprostatic androgens and its modulation of transcription of androgen responsive genes. .................................... 28 Figure 1.7 Metabolic scheme for the synthesis and deactivation of androgen steroids that can occur within the prostate. ................................................................................ 33 Figure 2.1 A translated protein product of the pET28a-AKR1A1 construct with (A) double 6#His tags and (B) a single 6#His tag.................................................................... 52 Figure 2.2 Purification of human recombinant (A) double 6#His-tagged AKR1A1 wildtype, (B) single 6#His-tagged AKR1A1 wild-type, (C) E55D variant and (D) N52S variant enzymes..................................................................................................... 58 Figure 2.3 In vitro enzymatic activities for the purified single and double 6#His-tagged AKR1A1 wild-types along with the GST-tagged AKR1A1 wild-type in the presence of 1 mM (A) p-nitrobenzaldhyde and (B) DL-glyceraldehyde test substrates as measured by following the initial rate of NADPH oxidation. ................................. 60 Figure 2.4 In vitro enzymatic activities for the purified single 6#His-tagged AKR1A1 wild-type and ns-SNP variants with 1 mM (A) p-nitrobenzaldehyde and (B) DLglyceraldehyde test substrates as measured by following the initial rate of NADPH oxidation. .............................................................................................................. 61 Figure 2.5 Generation of DAUNol in vitro of purified single 6#His-tagged AKR1A1 wild-type and ns-SNP variants. Measurement of DAUNol was performed using HPLC-fluorescence detection. ............................................................................... 63 Figure 2.6 Chemical structures of DOX and DAUN and their conversion to their corresponding carbon-13 alcohol metabolites, DOXol and DAUNol...................... 65 Figure 3.1 Purification of recombinant histidine-tagged human (A) AKR1A1 and (B) AKR1C2. .............................................................................................................. 78  viii  Figure 3.2 Generation of DAUNol and DOXol in vitro by purified recombinant AKR1A1 incubated with (A) DAUN and (B) DOX............................................................... 80 Figure 4.1 Scheme for reactions of DHT, 3!/"-diol, and 16!-bromoandrosterone (top to bottom) with 2-fluoro-1-methyl pyridinium p-toluenesulfonate to form stably charged fluoro methyl pyridinium ether derivatives for mass spectrometric detection. .............................................................................................................................. 90 Figure 4.2 Representative LC/MS/MS chromatogram for fluoro methyl pyridinium derivatives of DHT, 3!/"-diol, and 16!-bromoandrosterone. ................................. 91 Figure 5.1 Compounds tested for AKR1C2 activation. ................................................ 101 Figure 5.2 Rate of DHT reduction to 3!-diol by AKR1C2 as measured by LC/MS/MS. ............................................................................................................................ 106 Figure 5.3 Effect of modulators on AKR1C2 catalyzed reduction of DHT to 3!-diol. . 108 Figure 5.4 Dose-response relationships for modulator compounds fluoxetine, paroxetine, sertraline, and imipramine for their activation of AKR1C2 catalyzed reduction of DHT to 3!-diol.................................................................................................... 109 Figure 5.5 Effect of bovine serum albumin on AKR1C2 catalyzed reduction of DHT to 3!-diol in presence and absence of activating compounds. Inset graph is data in absence of BSA to show activation by modulators............................................... 110 Figure 5.6 AKR1C2-catalyzed reduction of DHT and DHP......................................... 113 Figure 6.1 Representative Western blots showing detection of AKR1C2 expression in transfected Sf9 cells and mean expression levels.................................................. 134 Figure 6.2 Representative Western blot showing detection of AKR1C2 expression in transfected T.ni cells and mean expression levels................................................. 135 Figure 6.3 Michaelis-Menten curves for fluorogenic probe reduction by wild-type and variant allele AKR1C2. ....................................................................................... 136 Figure 6.4 Association analysis of maximal rates, Vmax, measured for reduction of the fluorogenic probe and DHT by AKR1C2 expressed by transfected insect cells.... 140  ix  LIST OF ABBREVIATIONS 3!-diol 3"-diol °C #g µg µl µM ACN AKR ANOVA ARE BSA cDNA CEPH CLint CBR C.V. DAUN DAUNol DCM DHP DHT DNA DOX DOXol E. coli FMPTS g GABA h HPLC HSD i.d. K’ kDa Km LOQ  5!-Androstane-3!,17"-diol 5!-Androstane-3",17"-diol Degree Celsius Times gravity (centrifugal force) Microgram Microlitre Micromolar Acetonitrile Aldo-keto reductase Analysis of variance Antioxidant response element Bovine serum albumin Complementary DNA Centre d'Etude du Polymorphisme Humain Intrinsic clearance Carbonyl reductase Coefficient of variation Daunorubicin Daunorubicinol Dichloromethane 5!-Dihydroprogesterone 5!-Dihydrotestosterone Deoxyribonucleic acid Doxorubicin Doxorubicinol Escherichia coli 2-Fluoro-1-methylpyridinium toluene sulphonate Gram Gamma-aminobutyric acid Hill slope, describing allosteric sigmoidal kinetics High performance liquid chromatography Hydroxysteroid dehydrogenase Internal diameter Substrate concentration parameter described by the Hill equation, equal to Km when Hill coefficient is 1 Kilodalton Michaelis constant - the substrate concentration at which the reaction rate reaches half of its maximum value Limit of quantitation  x  m M mg ml mM min MRM mRNA MS/MS m/z n NADH NADPH NCBI n.d. nM nmol ns-SNP p PCR pH pKa pmol r2 RNA ROS rpm S.D. SDS-PAGE S.E.M. Sf9 siRNA SNP SSRI TEA T. ni UGT UPLC v/v Vmax  Metre Molar (mole/litre) Milligram Millilitre Millimolar Minute Multiple reaction monitoring Messenger ribonucleic acid Tandem mass spectrometry Mass to charge ratio Number of subjects or samples Nicotinamide adenine dinucleotide, reduced form Nicotinamide adenine dinucleotide phosphate, reduced form National Centre for Biotechnology Information Not determined Nanomolar Nanomole Non-synonymous single nucleotide polymorphism Statistical probability (of obtaining a result at least as extreme as the one that was observed) Polymerase chain reaction Negative logarithm of hydrogen ion concentration Negative logarithm of the acid dissociation constant Picomole Coefficient of determination Ribonucleic acid Reactive oxygen species Revolutions per minute Standard deviation Sodium dodecyl sulphate polyacrylamide gel electrophoresis Standard error of mean Spodoptera frugiperda Small interfering RNA Single nucleotide polymorphism Selective serotonin reuptake inhibitor Triethylamine Trichoplusia ni UDP-glucuonysyl transferase Ultra-high performance liquid chromatography Volume to volume ratio Maximal rate of velocity xi  HapMap population groups CEU Utah residents with ancestry from northern and western Europe, collected by Centre d'Etude du Polymorphisme Humain (CEPH) CHB Han Chinese in Beijing, China JPT Japanese in Tokyo, Japan YRI Yoruba in Ibadan, Nigeria Nucleotide one letter abbreviations A Adenine C Cytosine G Guanine T Thymine Amino acid one letter abbreviations A Alanine (Ala) C Cysteine (Cys) D Aspartic acid (Asp) E Glutamic acid (Glu) F Phenylalanine (Phe) G Glycine (Gly) H Histidine (His) I Isoleucine (Ile) K Lysine (Lys) L Leucine (Leu) M Methionine (Met) N Asparagines (Asn) P Proline (Pro) Q Glutamine (Gln) R Arginine (Arg) S Serine (Ser) T Threonine (Thr) V Valine (Val) W Tryptophan (Trp) Y Tyrosine (Tyr)  xii  ACKNOWLEDGEMENTS I express sincere thanks to my supervisor, Dr. Wayne Riggs, for his reliable support and encouragement throughout my graduate training. His thoughtfulness and care, both professionally and personally, are truly appreciated. Thank you to my cosupervisor, Dr. Ronald Reid, and Dr. Thomas Grigliatti for their research insight, constructive comments, and encouragement of lively scientific discussion. I am grateful to my supervisory committee members, Drs. Stelvio Bandiera, Tom Chang, and Kathleen MacLeod for their time, effort, and suggestions. Thank you to all members of the Riggs-Reid-Grigliatti research team, Onkar Bains, Dr. Joanna Lubieniecka, Dr. Randy Mottus, Dr. Tom Pfeifer, Madalene Earp, Omid Toub, Dr. Pamela Kalas, and Dr. Greg Doheny, for their technical assistance, suggestions, and support. Thank you to Andras Szeitz for his technical assistance with LC/MS/MS analysis and for consistently providing interesting conversation. I express special thanks to my colleagues and friends in the Faculty of Pharmaceutical Sciences for making these years memorable. I have utmost respect for their dedication and tenacity towards their work and value their support and friendship. I am deeply grateful for having family and friends – some who were lost and are sorely missed – that supported me. Thank you to my parents and sister, Trisha, for encouraging my education at universities for so many years. Thank you to friends for listening and providing perspective, opinion, and distraction when it was needed. Financial support was gratefully accepted from the Canadian Institutes of Health Research (CIHR) as project funding, and from CIHR, Canada’s Research-Based Pharmaceutical Companies, the Faculty of Pharmaceutical Sciences, Revelstoke District Health Foundation, Pfizer and Merck as training and personal funding. xiii  CO-AUTHORSHIP STATEMENT This thesis includes five manuscripts for which I am the co-author on one and the first author on four manuscripts. In the case of the co-authored paper (Chapter 2), equal work was performed by Mr. Onkar Bains and myself towards the research and preparation of the manuscript.  Specifically, I was responsible for: i) designing the  experiments to characterize enzyme metabolic function, including the indirect methods for measuring cofactor usage and direct method for measuring metabolite generation; ii) conducting experiments and data analysis to characterize the enzyme in anthracycline metabolism; and iii) preparing the sections of the manuscript that describe these experimental procedures and data. For the second manuscript (Chapter 3), the expression of the recombinant enzymes was performed by the second author, Mr. Onkar Bains. I was responsible for all other practical aspects of the work, data analysis, and manuscript preparation. In the case of the remaining manuscripts (Chapters 4 to 6), I was the primary individual responsible for the conceptualization and design of the research, practical aspects of the research, data collection and analyses, and manuscript preparation. Other than the exceptions specified above, the contribution of co-authors was through their involvements in intellectual discussion regarding the design and conduct of the research and preparation of manuscripts.  xiv  CHAPTER 1 BACKGROUND  1.1.  Aldo-Keto Reductases  1.1.1. Overview The aldo-keto reductases (AKRs) are a superfamily of proteins that have important roles in catalyzing the NAD(P)H-dependent reductive metabolism of carbonylcontaining compounds to their respective alcohol metabolites and the corresponding reverse oxidative reactions (Jin and Penning, 2007). The AKRs are generally described as monomeric proteins with an approximate length of 320 amino acids and are found in mammals, amphibians, plants, yeast, protozoa, and bacteria.  The crystal structures  determined for a number of AKRs reveal a general protein structure described as an (!/")8 barrel, which has eight repeated units of alpha helix and beta sheet with the beta sheets forming the sides of a barrel that lead to a conserved active site for catalysis (Figure 1.1) (Penning, 2003). Currently, 115 members of the AKR superfamily have been identified that are grouped into 15 families (Penning, 2003). A nomenclature system for these enzymes was developed based on amino acid sequence similarity that distinguishes the unique protein isoforms (Jez et al., 1997b). The designation “AKR” identifies the protein as a member of the superfamily; a numeric figure designates family, defined by 40% shared sequence identity; a letter designates subfamily, defined by 60% shared sequence identity; and numeric figure representing the unique protein sequence. Proteins that share 95% or  1  greater sequence identity must be demonstrated to have distinct functions to be considered unique proteins, rather than variant alleles of a single protein isoform. A full listing of AKRs is maintained and publicly available at the AKR website (http://www.med.upenn.edu/akr/).  Figure 1.1 AKR protein structure showing conserved (!/")8 barrel structure. Shown is AKR1A1 (Protein Data Bank id 2ALR) with !-helices coloured in red and "-sheets coloured in blue.  2  1.1.2. Human aldo-keto reductases Thirteen human AKR proteins, including members of the AKR1, 6, and 7 subfamilies, have been identified (Table 1.1) (Penning, 2003). The AKR1 proteins are the human homologs of reductases that are expressed across many species and include aldose reductases, aldehyde reductase, hydroxysteroid reductases, and steroid 5"reductase. The AKR6 and AKR7 proteins are human homologs of potassium channel subunits and aflatoxin aldehyde reductase, respectively. This latter group demonstrates the diversity of the AKRs, functioning in atypical non-enzymatic functions and utilizing non-carbonyl-containing substrates (Penning and Drury, 2007). The substrates used by AKRs that function as oxidoreductases in humans are diverse and include aliphatic and aromatic carbonyl containing compounds such as sugar aldehydes, lipid aldehydes, steroids, prostaglandins, and metabolites of polycyclic aromatic hydrocarbons (Jin and Penning, 2007). The activation and detoxification of many of these substrates are critical for complex processes such as drug metabolism, carcinogenesis, or toxicant elimination. Hence, the expression and function of AKRs are necessary for normal physiological balance and for conferring disease risk (Jin and Penning, 2007). For example, it is proposed that AKRs have a role in determining the risk of carcinogenesis due to smoking tobacco by being involved in the secondary metabolism of polyaromatic hydrocarbons.  The human AKR1A1 and AKR1C1-4  enzymes can oxidize polyaromatic hydrocarbon trans-dihydrodiols to form quinone species that can increase oxidative stress and DNA damage, potentially resulting in cancer development (Palackal et al., 2001; Palackal et al., 2002).  3  Table 1.1 Human AKR isoforms identified to-date. Gene AKR1A1 AKR1B1 AKR1B10 AKR1C1 AKR1C2 AKR1C3 AKR1C4 AKR1D1 AKR6A3 AKR6A5 AKR6A9 AKR7A2 AKR7A3  Protein  UniProt reference Aldehyde reductase; alcohol dehydrogenase P14550 Aldose reductase P15121 Aldose reductase; aldose reductase-related protein; small intestine reductase O60218 20!-Hydroxysteroid dehydrogenase; dihydrodiol dehydrogenase 1 Q04828 3!-Hydroxysteroid dehydrogenase, type 3; dihydrodiol dehydrogenase 2 P52895 3!-Hydroxysteroid dehydrogenase, type 2; 17"-hydroxysteroid P42330 dehydrogenase, type 5 3!-Hydroxysteroid dehydrogenase, type 1; dihydrodiol dehydrogenase 4; P17516 chlordecone reductase Steroid 5"-reductase; 3-oxo-5"-steroid 4-dehydrogenase P51857 Potassium voltage-gated channel, shaker-related subfamily, beta member 1 Q14722 Potassium voltage-gated channel, shaker-related subfamily, beta member 2 Q13303 Potassium voltage-gated channel, shaker-related subfamily, beta member 3 O43448 Aflatoxin aldehyde reductase O43488 Aflatoxin aldehyde reductase O95154  Chromosomal location 1p33-p32 7q35 7q33 10p15-p14 10p15-p14 10p15-p14 10p15-p14 7q32-q33 3q26.1 1p36.3 17p13.1 1p36.13 1p36.13  4  1.1.3. Mechanism of reduction Studies on AKR1C9, a rat protein that shares a high level (69%) of amino acid sequence identity with human AKR1C isoforms, have investigated the structure-function features of AKR enzymes that act as 3!-hydroxysteroid reductases (Penning et al., 2003). The high level of shared sequence identity and the overlap in substrate specificities suggest that the mechanism of action proposed by these studies is likely shared throughout the family. A tetrad of residues, namely tyrosine-55, lysine-84, histidine-117, and aspartic acid-50 (numbering based on rat AKR1C9), are highly conserved and form the active site (Schlegel et al., 1998). Site-directed mutagenesis has demonstrated critical roles for Tyr 55 and Lys 84 in the enzyme’s reductive actions, with the tyrosine acting as a general acid in 3-ketosteroid reduction and the lysine acting to facilitate proton transfer to substrates by lowering the pKa of the tyrosine residue through proton donation (Figure 1.2) (Schlegel et al., 1998). The histidine and aspartic acid also participate in proton donation or removal by the catalytic tyrosine in the reductive and oxidative actions of the enzyme, respectively, but were not essential to protein function.  5  Figure 1.2 Proposed catalytic mechanism for AKR-mediated reduction and oxidation reactions. Tyrosine 55 and lysine 84 residues are shown as acting as a general acid residue and facilitating proton transfer, respectively, in reduction reaction. AKR-catalyzed reduction occurs following an ordered sequence described as a bibi mechanism (Askonas et al., 1991). The co-ordinated steps begin with the sequential binding of cofactor, then substrate, forming a ternary complex within the active site of the enzyme. Chemical transformation occurs with the transfer of a proton and hydride to the substrate (Figure 1.2). Following substrate reduction, the ternary complex breaks apart in an ordered fashion; the enzyme releases the reduced product, then the oxidized cofactor. The kinetics of the AKR catalysis has been studied through complex stop-flow studies, which have shown that the chemical step (hydride transfer), release of the reduction product, and release of the oxidized cofactor are slow steps that are likely rate-limiting and determine the turnover rate of the catalytic cycle (Jin and Penning, 2006).  6  1.1.4. Polymorphism in AKRs Allelic variation in the form of single nucleotide polymorphisms (SNPs) exists in the human AKR genes. Though a majority of these genetic variations occur in noncoding regions, some will lead to amino acid changes. These non-synonymous SNPs in AKR genes are listed on the AKR website, compiled from submissions to the National Centre  for  Biotechnology  Information  (NCBI)  SNP  database  (http://www.ncbi.nlm.nih.gov/SNP/). To date, there have been few studies conducted for determining the effects of allelic variations on AKR function.  A novel variant of AKR1C4, coded by two  nucleotide substitutions and that results in two amino acid substitutions compared to the wild-type protein, has been characterized for its prevalence and effect on catalytic activity (Kume et al., 1999). These studies demonstrated that the in vitro activities of the variant AKR1C4 for metabolizing a panel of xenobiotics and steroid substrates were significantly reduced. Another novel polymorphism, this one in AKR1C3, was identified in a Swedish population, but was absent in a Korean population. In this case, the variant AKR1C3 affected protein function resulting in reduced biosynthesis of testosterone from androstenedione (Jakobsson et al., 2007). These results demonstrate that allelic variation may modulate AKR-mediated metabolism; however, the functional consequences of genetic polymorphism in other human AKRs are yet to be determined.  1.1.5. Aldo-keto reductase 1A1 Aldehyde reductase (AKR1A1) is a cytosolic reductase that catalyzes the NADP(H)-dependent reduction or reverse oxidation reactions of aliphatic aldehydes, aromatic aldehydes, biogenic amines, and monosaccharides (Jin and Penning, 2007). Though immunoblotting experiments using a variety of human tissues suggest ubiquitous 7  expression, a distinct pattern of tissue-specificity for AKR1A1 levels was observed (O'Connor et al., 1999).  The kidney and liver show high expression.  Though  immunoblotting experiments demonstrated that a high level of variability in liver expression of AKR1A1 exists, quantitative analysis of individual liver samples was not reported. Moderate expression was found in the brain, small intestine, and testis, while low expression was observed in lung, prostate, and spleen. Though in these studies expression of AKR1A1 was not observed in heart or skeletal muscle, other investigators report evidence of its function in human heart fractions (Mordente et al., 2001; Mordente et al., 2003). Based on its broad substrate specificity and elevated expression in organs of detoxification, it is expected that the primary role for AKR1A1 is deactivation to facilitate the elimination of endogenous and xenobiotic chemicals from the body (O'Connor et al., 1999). Two non-synonymous SNPs occur in the AKR1A1 gene that lead to the amino acid substitutions N52S and E55D and are listed at the AKR website (Table 1.2). The frequencies for these SNPs have been estimated using sampling from several human populations (Table 1.3); however, the impact of these genetic variations on protein function is not yet known. Though the E55D variant allele was not found in the human studies (ss9877871), the allelic frequency of the N52S variant allele has been measured in several investigations (ss3176571, ss48418114, ss68765032).  While there are some  discrepancies in the findings of these studies, overall, they report a low presence of this allele in the CEU (Utah residents with Northern or Western Europe ancestry), CHB (Han Chinese in Beijing, China), JPT (Japanese in Tokyo, Japan), YRI (Yoruba in Ibadan, Nigeria), and selected multi-national and North American populations (NCBI database online http://www.ncbi.nlm.nih.gov/sites/entrez?db=snp).  8  Table 1.2 Non-synonymous single nucleotide polymorphisms naturally occurring in the AKR1A1 gene compiled from the National Centre for Biotechnology Information (NCBI) SNP database. AKR website identifier AKR1A1*2 AKR1A1*3  NCBI SNP database identifier rs2229540 rs6690497  Symbol  Exon  SNP  Residue  Wild-type  Variant  N52S E55D  5 5  A155G G165C  52 55  Asparagine Glutamic acid  Serine Aspartic acid  9  Table 1.3 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1A1 compiled from the National Centre for Biotechnology Information (NCBI) SNP database. AKR website identifier AKR1A1*2  NCBI SNP database identifier rs222950  Amino acid substitution N52S  AKR1A1*3  rs6690497  E55D  SNP submission i.d. ss3176571 ss48418114 ss68765032 ss9877871  Variant allele frequencies (n) CEU  CHB  n.d. n.d. 4.4% (114) 0% (88) 2.5% (120) 5.6% (90) 0% (118) 0% (90)  JPT n.d. 0% (86) 3.3% (90) 0% (90)  YRI  Other  n.d. 1.9% (52)a 0% (116) 3.8% (78)b 1.7% (120) n/a 0% (120) n/a  CEU: Utah residents with ancestry from northern and western Europe, collected by Centre d'Etude du Polymorphisme Humain (CEPH) CHB: Han Chinese in Beijing, China JPT: Japanese in Tokyo, Japan YRI: Yoruba in Ibadan, Nigeria n.d.: Not yet determined n/a: No applicable a b  multi-national population (population ID MITOGPOP6) mixed North American population, Caucasian and African American, all female (population ID AGI_ASP population)  10  1.1.6. Aldo-keto reductase 1C2 Dihydrodiol dehydrogenase 2 (3!-hydroxysteroid dehydrogenase, type 3; AKR1C2) catalyzes the NADPH-dependent metabolism of several endogenous substrates, including sugar and lipid aldehydes, retinals, steroids, and prostaglandins, as well as xenobiotic compounds such as drugs and chemical carcinogens (Shiraishi et al., 1998; Jin and Penning, 2007). Though it shares at least 84% amino acid sequence with three related human AKR1C proteins (AKR1C1, 1C3, and 1C4), AKR1C2 demonstrated distinct substrate specificities and reductase function when the 3!-hydroxysteroid dehydrogenase actions of these proteins were compared (Penning et al., 2000). In this investigation, AKR1C2 catalyzed the reduction and oxidation reactions of androgens, estrogens, and progesterones, acting as a stereospecific 3!-, 17"-, or 20!-hydroxysteroid oxidoreductase. Of great interest was that these actions were uniquely distinct from the highly similar human protein AKR1C1 (20!-hydroxysteroid reductase, dihydrodiol dehydrogenase, type 1), which differs from AKR1C2 by only seven amino acid residues (Penning et al., 2000). Site-directed mutagenesis was used to demonstrate the critical nature of residue 54 in determining these isoform-specific substrate preferences and activities. Substitution of residue 54 in AKR1C1 from leucine to valine, the residue found in AKR1C2, provided the protein with substrate preferences that resembled AKR1C2 (Matsuura et al., 1997). Substitution of the remaining six individual residues in AKR1C1 to correspond to the amino acids found in AKR1C2 did not change the enzyme function. AKR1C2 is expressed less ubiquitously than other AKR genes in the human body with high numbers of gene transcripts being measured only in liver, heart, lung, testis, brain and prostate tissue (Shiraishi et al., 1998; Penning et al., 2000). Based on its abundant expression in the prostate, uterus, and brain, it was suggested that the primary 11  function of AKR1C2 is regulating hormone signaling by catalyzing the bioactivation or metabolism of steroids (Penning et al., 2000). The transcription of AKR1C2 has been shown to be specifically inducible through an antioxidant response element (ARE) located upstream from the start site for transcription. Treatment with the known phase II inducers "-naphthoflavone and ethacrynic acid resulted in increased gene transcription and mRNA stability, resulting in levels of induction of approximately 2-3-fold (Lou et al., 2006). It was suggested that this mechanism for induction is indicative of AKR1C2 being a part of a cellular detoxification system, which can be triggered in defense against toxicant insults. Earlier studies have demonstrated that the activities of AKR1C2 are altered through binding of modulator compounds.  Up to 10 to 30-fold greater enzyme  efficiencies reducing 5!-dihydroprogesterone to allopregnanolone were observed in vitro when recombinant AKR1C2 was in the presence of selective serotonin reuptake inhibitors, fluoxetine, sertraline, and paroxetine, due to proposed binding of these modulators at an allosteric site (Griffin and Mellon, 1999). It was suggested that this activation accounts for the increased levels of allopregnanolone in the brain and cerebrospinal fluid following fluoxetine treatment (Pinna et al., 2006). Imipramine, paroxetine, and sertraline, which are pharmacologically related to fluoxetine, provided similar activation. However, this modulation of AKR1C2 has remained unsubstantiated. A separate research group tested for AKR1C2 activation by fluoxetine using a similar in vitro system, but failed to observed activity modulation resulting from it or pharmacologically related chemicals (Trauger et al., 2002) Eleven naturally-occurring single nucleotide polymorphisms that result in single amino acid substitutions exist in AKR1C2 (Table 1.4); however, the functional consequences of these non-synonymous SNPs are unknown. The allelic frequencies for 12  some AKR1C2 variants have been reported with some ethnic-differences being observed (Table 1.5). For example, the allele leading to the F46Y substitution was found in CEU and YRI samples with frequencies of 5.9 and 15%, respectively; but was not found to occur in sampling from Han Chinese or Japanese populations (ss4041394, ss69068270). The L172Q variant has been tested only in CEPH samples, and found to occur with a 32.5% allele frequency (ss13684).  Genotyping conducted for the HapMap project did  not find the alleles leading to the H47R, V38I, or V38A variants (ss4438229, ss74818067, ss4438222, ss4438225). Also, the R258C variant was not found in a large sampling of CEU samples (ss38343033). Frequency data for S87C, H170R, L179E, V111A, and K185E variants have not yet been collected or reported (NCBI database online http://www.ncbi.nlm.nih.gov/sites/entrez?db=snp).  13  Table 1.4 Non-synonymous single nucleotide polymorphisms naturally occurring in the AKR1C2 gene compiled from the National Centre for Biotechnology Information (NCBI) SNP database. AKR website identifier AKR1C2*2 AKR1C2*3 AKR1C2*4 AKR1C2*5 AKR1C2*6 AKR1C2*7 AKR1C2*8 AKR1C2*9 AKR1C2*10 AKR1C2*11 Not listed  NCBI SNP database identifier rs13933 rs11474 rs10618 rs2518042 rs2854482 rs2854486 rs3207905 rs3207898 rs3207901 rs28943580 rs2518043  Symbol  Exon  SNP  Residue  Wild-type  Variant  S87C L172Q H170R K179E F46Y V111A H47R V38I V38A R258C K185E  6 8 8 8 5 6 5 5 5 10 8  A259T T515A A509G A535G T137A T332C A140G G112A T113C C772T A553G  87 172 170 179 46 111 47 38 38 258 185  Serine Lysine Histidine Lysine Phenylalanine Valine Histidine Valine Valine Arginine Lysine  Cysteine Glutamine Arginine Glutamic Acid Tyrosine Alanine Arginine Isoleucine Alanine Cysteine Glutamic Acid  14  Table 1.5 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1C2 compiled from the National Centre for Biotechnology Information (NCBI) SNP database. AKR website identifier AKR1C2*2 AKR1C2*3 AKR1C2*4 AKR1C2*5 AKR1C2*6  NCBI SNP database identifier rs13933 rs11474 rs10618 rs2518042 rs2854482  Amino acid substitution S87C L172Q H170R K179E F46Y  AKR1C2*7 AKR1C2*8  rs2854486 rs3207905  V111A H47R  AKR1C2*9 AKR1C2*10 AKR1C2*11 Not listed  rs3207898 rs3207901 rs28943580 rs2518043  V38I V38A R258C K185E  SNP submission i.d. ss13684 ss4041394 ss69068270 ss4438229 ss74818067 ss4438222 ss4438225 ss38343033 -  Variant allele frequencies (n) CEU  CHB  JPT  YRI  n.d. 32.5% (188) n.d. n.d. 5.9% (118) 5.9% (120) n.d. 0% (298)a 0% (114) 0% (116) 0% (116) 0% (606) n.d.  n.d. n.d. n.d. n.d. 0% (90) 0% (90) n.d. 0% (88) 0% (88) 0% (88) 0% (88) n.d. n.d.  n.d. n.d. n.d. n.d. 0% (88) 0% (90) n.d. 0% (84) 0% (84) 0% (86) 0% (86) n.d. n.d.  n.d. n.d. n.d. n.d. 15.0% (120) 14.2% (120) n.d. 0% (112) 0% (112) 0% (116) 0% (116) n.d. n.d.  CEU: Utah residents with ancestry from northern and western Europe, collected by Centre d'Etude du Polymorphisme Humain (CEPH) CHB: Han Chinese in Beijing, China JPT: Japanese in Tokyo, Japan YRI: Yoruba in Ibadan, Nigeria n.d.: Not yet determined a  pooled data for population ID CEPH (n=184) and population ID HapMap-CEU (n=114), both samples from CEPH  15  1.2.  Anthracycline Chemotherapeutics and Their Cardiac Toxicity  1.2.1. Anthracycline drugs The anthracycline antibiotics were first isolated from Streptomyces peucetius in the 1960s and since their discovery have become some of the most important anti-tumour drugs for the treatment of a range of malignancies (Minotti et al., 2004). Their chemical structure consists of a tetracyclic ring system that contains quinone and hydroquinone groups in neighbouring rings B and C, a methoxy group on ring D, and a short carbonylcontaining side chain and glycosidic bond to an attached sugar on ring A (Figure 1.3).  Figure 1.3 Chemical structures of doxorubicin, daunorubicin, and epirubicin. A high level of structural similarity exists between these anthracyclines. Compared to doxorubicin, daunorubicin terminates in a methyl group in the carbonyl-containing side chain; epirubicin has an equatorial to axial switch of the hydroxyl group on the daunosamine sugar. 16  The principal anthracyclines used clinically are daunorubicin (DAUN), doxorubicin (DOX), and epirubicin. These anthracyclines are structurally very similar (Figure 1.3).  DOX and DAUN differ only by small differences in the carbonyl-  containing side chain, with DOX terminating in a primary alcohol and DAUN terminating in a methyl group. Epirubicin differs from DOX by an equatorial to axial switch of the hydroxyl group on the daunosamine sugar. Though structurally very similar, these anthracyclines differ in their clinical utility (Minotti et al., 2004).  DOX has the widest spectrum of use in cancer therapy,  particularly in the treatment of breast and esophageal carcinomas, osteosarcoma, Kaposi’s sarcoma, soft-tissue sarcomas, and Hodgkin’s and non-Hodgkin’s lymphoma. In addition, its combination with other chemotherapeutics, such as taxanes and trastuzumab, has further extended its use by enhancing its chemotherapeutic effects (Singal and Iliskovic, 1998). In contrast, DAUN is primarily used for the treatment of acute lymphoblastic or myeloblastic leukemias. Epirubicin was developed in an effort to engineer a less toxic version of DOX; however, it has a limited spectrum of actions and its main clinical uses are in the treatment of breast and gastric cancers. Many other anthracycline analogues have been developed in attempts to improve the toxicity profile of DOX, but, to date, none have found extensive clinical application (Weiss, 1992). Despite their extended record of widespread use in patient care and experimental study, the exact mechanisms by which anthracyclines exert their cytostatic and cytotoxic actions are still not clear (Gewirtz, 1999; Minotti et al., 2004). The most commonly described action is interaction with topoisomerase II, the nuclear protein responsible for introducing DNA breaks to allow strand separation and replication during cell division (Minotti et al., 2004). The binding of the anthracycline stabilizes a DNA-topoisomerase II complex, which prevents the resealing of the DNA strand. This damage to DNA 17  results in cell growth arrest and programmed cell death. Several additional mechanisms of action may contribute to the cytotoxic actions of anthracyclines (Gewirtz, 1999; Danesi et al., 2002). These include inhibition of DNA synthesis, generation of free radicals leading to DNA damage or lipid peroxidation, DNA adduct formation and DNA cross linking, interference with DNA strand synthesis and helicase activity, direct membrane effects, induction of apoptosis, and signaling leading to growth arrest. It remains unclear to what extent these actions may contribute to the cytotoxic function because though they are observed in vitro, the drug concentrations or experimental conditions for these studies often did not reflect the in vivo situation (Gewirtz, 1999). Elimination of DOX occurs primarily by excretion in the bile, accounting for about 50% of the dosed amount over a 5-day period, and to a smaller extent in the urine (~10%) (Takanashi and Bachur, 1976). A large proportion of the excreted drug (60%) is as metabolites, indicating substantial metabolism of the drug occurs before elimination (Takanashi and Bachur, 1976). The major metabolite observed in tissues, plasma, bile, and urine is generated through the reduction of the C-13 carbonyl group to its corresponding secondary alcohol by cytosolic reductase enzymes, resulting in doxorubicinol (DOXol) (Bachur, 1971; Lovless et al., 1978). To a much lesser extent, deglycosylation plays a role in metabolism, resulting in poorly water-soluble aglycone metabolites such as doxorubicinone and 7-deoxydoxorubicinone (Figure 1.4) (Takanashi and Bachur, 1976; Lovless et al., 1978). Marked interpatient variation is observed in DOX pharmacokinetics, which is thought to be related in part to the individual differences in metabolism (Cummings et al., 1986).  18  Figure 1.4 Metabolic scheme for anthracyclines, showing pathways for doxorubicin metabolism in human cardiac cytosol. The major metabolite observed in humans is doxorubicinol, which is formed by the actions of aldo-keto and carbonyl reductase enzymes (left). 1.2.2. Cardiotoxicity Immediate and late, long-term side effects are associated with the clinical use of anthracyclines (Singal and Iliskovic, 1998; Wojtacki et al., 2000). The immediate side effects, including myelosuppression, nausea, vomiting, and arrhythmia, are largely reversible and clinically manageable.  Of much greater concern is cardiomyopathy  followed by congestive heart failure, a long-term effect that is largely unresponsive to drug treatment and often results in patient death (Singal and Iliskovic, 1998; Wojtacki et al., 2000).  This cardiac dysfunction develops through a progressive decrease in  ventricular contractile function, typically within the first year after completion of 19  anthracycline treatment (Singal and Iliskovic, 1998); however, it also occurs as a delayed form, presenting itself at 4 to 20 years after completion of therapy (Steinherz et al., 1991). Drug therapies have been unsuccessful in reversing cardiac damage, therefore the best strategies to minimize the impact of anthracycline-associated cardiotoxicity are active screening of cardiac function during treatment and the suspension of treatment if there are indications of cardiotoxicity (Wojitacki et al., 2000). The development of cardiomyopathy and congestive heart failure is dosedependent, escalating in its occurrence with increasing total lifetime dosing of anthracyclines. Congestive heart failure in patients receiving 551 to 600 mg/m2 DOX was observed in 18 percent of patients, compared to 4 percent in patients receiving 500 to 550 mg/m2 (Singal and Iliskovic, 1998). Based on these observations, the maximum cumulative DOX lifetime dose a patient receives is typically restricted to 500 mg/m2 (Wallace, 2003). It is clear that these recommendations do not fit all patients. Large patient-to-patient variation is observed with some patients developing cardiotoxicity well below their maximal cumulative dosing (300 mg/m2), whereas others have shown no cardiac damage with cumulative dosages of >1000 mg/m2 (Deng and Wojnowski, 2007). Presently, there are no markers to assess a patient’s risk for developing cardiac damage before initiating an anthracycline regimen. Hence, patients that are highly susceptible to cardiac damage are put at high risk for a lethal toxicity, whereas, patients that can tolerate higher doses may receive sub-optimal therapy. There is clear evidence that DOX accumulation in the heart is dangerous and results in damage to cardiac tissue (Menna et a., 2007). Slow infusion rather than bolus dosing and encapsulation, which prolong drug circulation, have been employed with some success in reducing cardiac toxicities (Speyer and Wasserheit, 1998; Minotti et al., 2004; Cortes-Funes and Coronado, 2007). However, the exact mechanism by which 20  anthracyclines damage cardiac tissue remains unclear. Two prevailing hypotheses exist that describe reductive metabolism of anthracyclines as generating chemical species that exert damage to the heart (Minotti et al., 1999; Mordente et al., 2001).  The first  hypothesis describes one-electron reduction generating free radicals that interact with cellular iron. The second hypothesis describes two-electron reduction generating alcohol metabolites, which exert toxic cellular effects. The free radical and iron hypothesis describes an initiating event of the oneelectron reduction of the quinone moiety of DOX to a semiquinone free radical, which triggers a cascade of redox reactions that generate highly reactive chemical species (e.g., hydroxyl radicals (OH.), hydrogen peroxide (H2O2), ferry ions (Fe(IV)=O)) (Figure 1.5) (Gewirtz, 1999; Xu et al., 2001). When the semiquinone regenerates the parent quinone it reacts with molecular oxygen forming the superoxide radical anion species (O2.-). The superoxide species is acted upon by superoxide dismutase to form hydrogen peroxide (H2O2), which, in turn, forms water and oxygen under the influence of catalase. If not deactivated by catalase, the decomposition of the superoxide radical can generate hydrogen peroxide or hydroxyl radicals (OH.), which are both reactive oxygen species (ROS) that can directly cause cellular damage. Of greater concern, these ROS may interact with iron in the intracellular environment of cardiomyocytes. Under normal cellular conditions, intracellular stores of iron are carefully regulated, in part by ferritin. This specialized protein sequesters iron, protecting it from reacting with ROS generated in normal cellular processes.  With anthracycline semiquinone cycling, abundant  superoxide and hydrogen peroxide are generated intracellularly and can overwhelm the protective capacity of ferritin. ROS can penetrate into ferritin and react with sequestered iron leading to its release as Fe(II), which can subsequently react with O2.- and H2O2 to form hydroxyl radicals or ferryl ions (Fe(IV)=O). These products are highly oxidizing 21  species that are prone to react with lipids, proteins, and DNA, and result in cellular dysfunction. The use of iron chelation therapy, such as the dosing of dexrazoxane, has shown some success in preventing cardiotoxicity due to anthracycline treatment, thus providing good clinical evidence for the iron and free-radical hypothesis (Hasinoff and Herman, 2007).  Figure 1.5 Schematic showing the one-electron reduction of doxorubicin leading to the generation of reactive oxygen species and iron reduction. The resulting hydroxyl radicals (OH.), hydrogen peroxide (H2O2), and ferryl ions (Fe(IV)=O) are highly oxidizing species that are may react with lipids, proteins, and DNA, leading to cellular dysfunction.  Alternatively,  two-electron  reductive metabolism  produces  anthracycline  metabolites that have potent, damaging effects. Several lines of evidence implicate DOXol, the major measured metabolite. Following dosing, DOX levels peak in the heart 22  and plasma within hours and diminish over days, clearly not corresponding with the delayed development of cardiotoxicity (Peters et al., 1981). In contrast, DOXol follows a concentration-time profile unique from DOX, increasing and reaching its peak tissue levels after DOX levels have fallen significantly. There is delayed accumulation of DOXol specifically in the heart, where concentrations have been observed to reach levels two to three-fold higher than in the liver (Peters et al., 1981). In addition, some studies indicate that DOXol may be more toxic than DOX in cardiomyocytes. In isolated cardiac preparations, DOXol acts as a potent inhibitor of ion pumps of the sacroplasmic reticulum, mitochondria, and sarcolemma, resulting in cardiac dysfunction (Boucek et al., 1987; Platel et al., 2001).  Other metabolites such as 7-deoxydoxorubicinol and  doxorubicinol aglycone, formed through reductive metabolism and glycosidic bond cleavage, are also speculated to have actions that result in cellular dysfunction (Sokolove, 1994; Licata et al., 2000). The free radical and the toxic metabolite hypotheses are useful for describing some of the clinical and biochemical observations of cardiac damage; however, neither has provided a clear strategy for preventing or treating cardiotoxicity in patients (Olson and Mushlin, 1990; Mordente et al., 2001). It is likely that both mechanisms, which are unified by involving anthracycline metabolism, contribute to the manifestation of the cardiotoxicity (Menna et al., 2007).  Simply considered, the rate and extent of  metabolism will modify the levels of parent anthracycline drug and its reactive metabolites within the body. Therefore, factors that alter anthracycline metabolism will likely contribute to the variability in the development of cardiotoxicity.  23  1.2.3. Role of reductases in anthracycline metabolism Biochemical studies provide evidence that identifies the enzymes responsible for the reductive metabolism of anthracyclines to their respective alcohol metabolites. When tested in vitro as isolated hepatic enzymes, AKR1A1, AKR1C2, and the carbonyl reductases were identified as utilizing DOX and DAUN as substrates, with the carbonyl reductases showing the greatest activities (Ohara et al., 1995). The involvement of the AKRs is further supported by the observed decrease in DOXol generation by liver and heart cytosolic fractions isolated from rats pre-treated with phenobarbital, a general AKR chemical inhibitor (Behnia and Boroujerdi, 1999). The formation of DOXol by human liver cytosol was decreased significantly by a potent and specific AKR1A1 chemical inhibitor, AL1576, leading investigators to conclude that DOX metabolism was mediated by AKR1A1 (Mordente et al., 2003). There is also evidence from these studies that differences exist between metabolism of DOX and DAUN.  Though AKRs were  demonstrated to be the primary enzymes involved in DOX metabolism, carbonyl reductases were identified as the primary enzymes involved in DAUN metabolism (Mordente et al., 2003). Finally, there is evidence that overexpression of metabolic enzymes may confer anthracycline resistance in cancer cells. For example, higher levels of the carbonyl reductase enzymes, AKR1B1, and AKR1C2, are shown to be associated with DAUN resistance in a human stomach cancer cell line, suggesting that they are important metabolic enzymes for DAUN (Ax et al., 2000).  1.2.4. Implications of altered anthracycline metabolism The association of reductase-mediated metabolism of anthracyclines with the development of cardiotoxicity is most strongly supported by evidence collected using animal models. In rats that were pre-dosed with phenobarbital, an inhibitor of AKR 24  enzymes, lower levels of creatine kinase, a marker of regular cardiac muscle function, were observed to coincide with decreases in DOXol generation (Behnia and Boroujerdi, 1999).  Furthermore, transgenic and knockout mouse models have been used to  significantly alter the development of heart tissue damage resulting from DOX treatment. Overexpression of human carbonyl reductase in mouse heart resulted in increased DOXol levels, earlier cardiotoxicity, and shorter survival (Forrest et al., 2000). Alternatively, protection from the anthracycline-induced cardiotoxicity was observed in mice deficient in carbonyl reductase 1 due to a null allele for the gene (Olson et al., 2003). Chemical alteration of DOX has also been shown to result in metabolic changes and altered cardiotoxicity potential.  Epirubicin was synthetically derived through  chemical modification of DOX and has unique pharmacokinetic properties. This has allowed cumulative dosing for epirubicin to reach double that for DOX without increasing cardiotoxicity (Robert, 1994). Though these changes may occur due to a number of mechanisms, the switch from predominantly reductive metabolism to a greater contribution by glucuronidation is thought to be an important factor.  25  1.3.  Androgens and Prostate Cancer  1.3.1. Prostate cancer Prostate cancer is the most diagnosed cancer in Canadian men. It was estimated that in 2008 there would be 24,700 cases diagnosed in Canada (Canadian Cancer Society, 2008). The known risk factors for men to develop this disease are increased age (65 years of age and older), family history of prostate cancer, and African ancestry (Bostwick et al., 2004). Early detection methods, such as the serum prostate-specific antigen (PSA) test and digital rectal examination, and intervention therapies have contributed to a decline in mortality due to prostate cancer over the past 10 years (LaSpina and Haas, 2008). Despite these improvements, prostate cancer remains the third most common cause of death due to cancer and it was estimated that 4,300 Canadian men would die from the disease in 2008 (Canadian Cancer Society, 2008).  1.3.2. Androgen receptor activation and prostate cancer The androgen receptor is a member of the steroid receptor family of transcription factors (Evans, 1988). It exists as part of a multiprotein complex associated with heat shock proteins and immunophilins in the cytoplasmic space. Upon binding an androgen ligand, the receptor is released from this complex and undergoes conformational change and homodimerization (Figure 1.6). The dimerized ligand-bound receptor translocates into the nucleus where it binds at androgen response elements in the promoter and enhancer regions of genes with the concomitant recruitment of coactivator or corepressor modulating proteins to form a multiprotein complex. Through its interactions with the transcriptional machinery, this complex modulates transcription of the target gene and  26  thereby mediates gene expression to regulate cell growth, survival, and function (Dehm and Tindall, 2007). The normal development and function of the prostate requires androgen receptor signaling (Chatterjee, 2003). Males with genetic dysfunction in receptor function are observed to have an absent or underdeveloped gland (Griffin, 1992).  This  underdevelopment of the prostate gland also results in individuals that are deficient in the enzyme responsible for converting testosterone to dihydrotestosterone (DHT), the most potent ligand for the androgen receptor in the prostate (Tindall and Rittmaster, 2008). The work of Huggins and Hodges in 1941 first associated the development of prostate cancer with the actions of androgen steroids. Through castration, they observed markedly reduced levels of testosterone and were able to effectively stop the development of prostate cancer (Huggins and Hodges, 1941).  Hence, it was  hypothesized that development of prostate cancer was an exaggeration of the normal growth and development of the prostate gland, resulting from excess androgen receptor signaling.  Supporting this hypothesis, the injection of testosterone induced prostate  cancer in rat models (Pollard et al., 1982). In addition, repression of androgen receptor using siRNA lead to the initiation of apoptosis in the human prostate cancer LNCaP cells (Yang et al., 2005). Consequently, a considerable amount of research has focused on elucidating the role of androgen receptor activation in the development and progression of prostate cancer (Chu et al., 2008).  27  Figure 1.6 Schematic of androgen receptor activation by intraprostatic androgens and its modulation of transcription of androgen responsive genes. Representative metabolic pathways leading to the synthesis of dihydrotestosterone (DHT) are shown. Following binding of DHT the androgen receptor homodimerizes and translocates to the nucleus, forming a complex with coactivators and corepressors at androgen-response elements to modulate gene transcription. (Adapted from Hsing et al., 2008.)  28  1.3.3. Pharmacological role of DHT Dihydrotestosterone (DHT, 5!-androstan-17"-ol-3-one) is the most potent natural ligand for the androgen steroid receptor (Kd=10-11 M).  In the systemic circulation,  testosterone is the major androgen; whereas in the prostate, DHT is predominant with concentrations several times higher than testosterone (Marks et al., 2008). Hence, DHT is expected to have an integral role in the androgen receptor signaling in the prostate and has been extensively studied for its role in development of the gland (Siiteri and Wilson, 1970; Hsing et al., 2002; Andriole et al., 2004). Methods for reducing androgen levels in the prostate have been extensively investigated as strategies for limiting the progression of prostate cancer (Hsing et al., 2002). These strategies aim to eliminate androgens from the prostate in order to reduce androgen receptor signaling, which is thought to promote cell proliferation. Castration decreases intraprostatic DHT levels by 50 to 80%, however, circulating androgens are also significantly lowered leading to undesirable side effects in bone density, libido, and muscle mass (Tindall and Rittmaster, 2008).  Alternatively, specific reduction of  intraprostatic DHT is accomplished through treatment with steroidal 5!-reductase inhibitors, such as finasteride and dutasteride, which reduce DHT synthesis in the prostate (Thompson et al., 2003; Rittmaster et al., 2008). Though in the initial stages of these intervention therapies, uncontrolled cell proliferation is successfully stopped. This commonly will only last temporarily and then cell proliferation resumes in what is referred to as androgen-independent cell growth. It is now known that this continued cancer progression does not occur fully independent of androgens but, rather, it occurs in an environment that contains residual amounts of DHT despite elimination of circulating  29  androgens and inhibition of the enzymes necessary for DHT synthesis. The residual DHT is sufficient to support androgen receptor activation and consequently, is thought to facilitate tumour cell survival and growth (Mohler et al., 2004; Page et al., 2006). The source of these residual tissue androgens is unknown, but it is hypothesized that the prostate cancer cells may adapt an alternative source for DHT synthesis (Mostaghel and Nelson, 2008). There has also been speculation that an environment of reduced, but persistent low-level androgens, induces an adaptation that sensitizes the cells to allow their proliferation with reduced presence of androgen receptor ligand. This may explain how the cells become unresponsive to further androgen suppression treatment, and suggests that the prostate cells that continue to proliferate may grow more aggressively and be more difficult to eliminate (Imamoto et al., 2008).  1.3.4. Androgen synthesis and degradation The classic model that describes androgen biosynthetic pathways in the prostate focuses on testosterone and DHT (Figure 1.7) (Luu-The et al., 2008). Testosterone is produced by the testes and delivered throughout the body in the circulation. DHT is formed in peripheral tissues from testosterone through reductive metabolism catalyzed by 5!-reductase enzymes for which two major isoforms exist: type 1 in skin and liver and type 2 in prostate tissue (Zhu and Imperato-McGinley, 2009). The simple scheme of DHT arising from the reductive metabolism of testosterone fails to describe how the prostate is supplied with DHT despite effective androgen ablation therapies. Hence, a more elaborate schematic of biosynthetic pathways for DHT describes the actions of various steroidogenic enzymes in the prostate (Figure 1.7) (Luu-  30  The et al., 2008). In summary, this scheme proposes that alternate adrenally produced androgens, such as 4-dione, act as precursors for DHT through metabolic biotransformation in the prostate (Mohler et al., 2004; Stanbrough et al., 2006). Alternatively, retinol dehydrogenase-like 3!-HSD (RL-HSD) can function to catalyze the oxidative formation of DHT from 3!-diol (Bauman et al., 2006).  A recent report  suggests that prostate cancer cells may act autonomously in DHT generation by expressing all of the steroidogenic enzymes for synthesizing DHT in situ from simple chemical building blocks such as acetic acid, eliminating the need for steroid precursors (Locke et al., 2008). Compared to the biosynthetic steps producing DHT, elimination pathways in peripheral tissues are poorly described (Negri-Cesi and Motta, 1994). In the prostate, reductive metabolism (Phase I) and glucuronidation (Phase II) are the major metabolic pathways for DHT elimination (Figure 1.7) (Hsing et al., 2002). Reduction by 3!hydroxysteroid (3!-HSD) or 3"-hydroxysteroid dehydrogenases (3"-HSD) produces 5!androstane-3!,17"-diol (3!-diol) or 5!-androstane-3",17"-diol (3"-diol), respectively (Penning et al., 2000). Alternatively, DHT can be converted to 5!-androstanedione through the oxidative actions of 17"-hydroxysteroid reductase, though it appears to a lesser extent than generation of the diols (Negri-Cesi and Motta, 1994). Deactivation and elimination through formation of glucuronide conjugates by the UDP-glucuronysyl transferase (UGT) enzymes also plays an important role in androgen metabolism in liver and peripheral tissues (Rittmaster et al., 1993; Pirog and Collins, 1997; Belanger et al., 2003). UGT2B7, B15, and B17 have been identified to be involved in the conjugative  31  metabolism of DHT and its metabolites 3!-diol and 5!-androstanedione.  Only the  UGT2B15 and B17 enzymes are expressed in the prostate (Chouinard et al., 2007).  32  Figure 1.7 Metabolic scheme for the synthesis and deactivation of androgen steroids that can occur within the prostate. Enzyme abbreviations: AKR, aldo-keto reductase; HSD3B, 3!-hydroxysteroid dehydrogenase; HSD17B, 17!-hydroxysteroid dehydrogenase; RLHSD, retinol dehydrogenase-like 3"-HSD; SRD5A, steroid 5"-reductase; UGT, UDP-glucuronysyl transferase.  33  1.3.5. Role of aldo-keto reductase 1C2 in androgen metabolism AKR1C2 acts as a 3!-hydroxysteroid dehydrogenase, catalyzing the NADPH dependent reduction of DHT in a stereospecific manner, to produce 3!-diol, a weak ligand for the androgen receptor (Kd=10-6 M) (Penning et al., 2008). This reductive metabolism effectively deactivates DHT, reducing its affinity for the androgen receptor by several levels of magnitude. Further, 3!-diol is a suitable substrate for conjugative metabolism by UGT2B15 leading to its elimination from the prostate (Penning et al., 2000; Ji et al., 2003). A variety of steroid metabolizing enzymes are expressed in the prostate and are capable of utilizing a large number of potential substrate and product androgens (LuuThe et al., 2008). For example, several AKR enzymes are expressed in the prostate that function as 3-, 17-, and 20-ketosteroid reductases and 3!-, 17"-, and 20!-hydroxysteroid oxidases to varying degrees with isoform-specific substrate preferences (Penning et al., 2000). In the case of AKR1C2, the recombinant human enzyme acted in vitro as both a reductase generating 3!-diol from DHT, and an oxidase generating DHT from 3!-diol (Penning et al., 2000). Several lines of evidence demonstrate that AKR1C2 preferentially acts as a reductase for ketosteroids in vivo.  In AKR1C2-transfection studies using  monkey COS-1 kidney and human PC-3 prostate cancer (hormone independent) cells, the enzyme predominantly acted as a reductase in the cellular environment (Rizner et al., 2003b). Further, the 3!-hydroxysteroid dehydrogenases show high selectivity in cofactor binding, preferring reduced cofactor (NADPH or NADH) to oxidized forms (NADP+ or NAD+) (Ma et al., 2000). This sensitivity to cofactor balance was also demonstrated by the potent inhibition of the enzyme’s oxidative actions by NADPH (Rizner et al., 2003b). Based on these substrate and cofactor specificities and activity, and its high expression in prostate, AKR1C2 was identified as the main deactivation enzyme for DHT, catalyzing 34  its reductive metabolism to generate 3!-diol. Hence, AKR1C2 has a critical role in modulating DHT levels and, consequently, regulating receptor signaling in androgendependent tissues such as the prostate (Ji et al., 2007; Luu-The et al., 2008; Penning et al., 2008).  1.3.6. Pharmacological role of 3!-diol Recent evidence suggests that 3!-diol plays a role in modulating neuronal excitability in peripheral tissues (Melcangi et al., 2008). Though the mechanism has yet to be elucidated, 3!-diol is produced in glial cells in the brain and exerts powerful anticonvulsant activities in animal models independent of interaction with the androgen receptor (Reddy, 2004). Studies suggest that the central action of 3!-diol may occur through allosteric activation of GABAA receptors and that it mediates the physiological effects observed for testosterone on the brain (Frye et al., 2001; Reddy, 2004). This proposed mechanism is similar to the modulation of GABAA receptor by allopregnanolone, which is structurally similar to 3!-diol, to relieve patient anxiety. Further, the LNCaP prostate cancer cell line has been shown to proliferate through stimulation by 3!-diol via a pathway that is independent of androgen receptor activation (Nunlist et al., 2004).  1.3.7. Implications of altered androgen metabolism The leading hypotheses describe prostate cancer arising from excess or abnormal androgen receptor signaling, most often attributed to high DHT levels in prostate tissue (Hsing et al., 2008). Since tissue DHT levels are a probable risk factor for prostate cancer, imbalances in DHT deactivation and elimination due to alterations in enzymes that catalyze these metabolic pathways is of particular interest in prostate cancer therapy. 35  For example, impaired DHT clearance would be speculated to increase the risk of a man for prostate disease progression. This hypothesis is supported by the measurement of significantly higher levels of DHT and lower expression levels of AKR1C2 observed in prostate cancer samples compared to paired benign tissues (Ji et al., 2003; Ji et al., 2007). Studies have investigated the significance of polymorphisms in UGT2B15 and UGT2B17 enzymes responsible for the conjugation of DHT leading to its deactivation and elimination and their association with prostate cancer (Belanger et al., 2003; Chouinard et al., 2007; Swanson et al., 2007; Barbier and Belanger, 2008). Patients homozygous for variant UGT2B15 (D85Y), which has reduced activity compared to the wild-type, have been identified as having increased risk of developing prostate cancer (MacLeod et al., 2000). Similarly, patients with a deletion polymorphism of UGT2B17 have been identified to be at increased risk (Park et al., 2007; Karypidis et al., 2008).  1.4.  Rationale It is clear that cardiotoxicity associated with anthracycline dosing and prostate  cancer are complex phenomena that result in significant patient morbidity and mortality. The risk of developing irreversible and lethal cardiac damage limits the highly effective use of anthracycline cancer therapies. The further success in detecting and treating prostate cancer depends on identifying more specific risk factors for disease development. Though these two clinical situations are diverse, they are unified by the hypotheses that metabolism is likely an integral component to the mechanisms by which they develop. The metabolism of the anthracyclines generates potentially reactive and toxic species that are expected to be damaging to heart tissue. The metabolism of androgen steroids regulates activation of the androgen receptor and, therefore, is critical for controlling prostate cell proliferation. 36  It is known that the human AKRs have important roles in the bioactivation and detoxification of a diverse range of chemical species and are involved in complex processes such as drug metabolism, carcinogenesis, and detoxification. Included among their substrates are the anthracycline drugs DOX and DAUN and the potent androgen steroid DHT. function.  There is a lack of understanding of the factors that modulate AKR  Hence, it is important to characterize these factors to understand their  contributions to the inter-patient variability seen in their metabolism of anthracyclines and androgen steroids and potentially, as determinants for developing anthracyclineinduced cardiotoxicity or prostate cancer.  1.5.  Research Overview and Hypotheses The original concept for these studies was to investigate the function of AKRs  and carbonyl reductases in metabolizing the anthracycline drugs DOX and DAUN and to characterize the effect of naturally-occurring allelic variation in altering these metabolic pathways. During these investigations, it was discovered that AKR1C2 is not involved in this reductive metabolism (Chapter 3). In order to study the function and allelic variants of AKR1C2, a pharmacologically-relevant substrate for this enzyme was selected. Therefore, the major component and focus of this thesis describes the development and application of in vitro techniques to characterize the deactivation of the androgen steroid DHT to its metabolite, 3!-diol, by AKR1C2.  The reported allosteric activation of  AKR1C2 by anti-depressant drugs and the effect of naturally occurring allelic variation in AKR1C2 on the metabolism of DHT are investigated. Overall, the studies described in this thesis are in vitro investigations designed to collect biochemical evidence that test three research hypotheses concerning cardiotoxicity associated with anthracycline therapy and the development of prostate cancer due to elevated prostate DHT levels. 37  •  Allelic variants of AKR1A1 have altered enzymatic activities compared to wildtype in metabolizing the anthracycline drugs DOX and DAUN to their corresponding alcohol metabolites and, therefore, contribute to interpatient variability in the generation of toxic metabolites or reactive species and the risk of anthracycline-induced cardiotoxicity.  •  Allosteric activation of AKR1C2, previously reported as increasing the efficiency of AKR1C2 to metabolize dihydroprogesterone, alters the enzyme’s activity in metabolizing DHT to 3!-diol.  •  Allelic variants of AKR1C2 compared to wild-type have altered enzymatic activities in metabolizing DHT to 3!-diol and, therefore, contribute to the interpatient variability in prostate DHT levels and risk of prostate cancer development.  1.6.  Specific Objectives The studies described in this thesis addressed the following specific objectives:  •  To measure the generation of the alcohol metabolites of DOX and DAUN by recombinant AKR1A1 and AKR1C2 enzymes in vitro.  •  To compare the kinetic properties of recombinant AKR1A1 wild-type and variant allele proteins in reducing DOX and DAUN to their alcohol metabolites in vitro.  •  To measure the generation of 3!-diol from DHT by recombinant AKR1C2 in vitro.  •  To compare the kinetic properties of recombinant AKR1C2 in the presence and absence of the anti-depressant drugs fluoxetine, imipramine, paroxetine, and sertraline in the reduction of DHT to 3!-diol in vitro. 38  •  To compare the kinetic properties of recombinant AKR1C2 wild-type and variant allele proteins in metabolizing DHT to 3!-diol using cell-based assays.  39  1.7.  References  Andriole G, Bruchovsky N, Chung LW, Matsumoto AM, Rittmaster R, Roehrborn C, Russell D and Tindall D (2004) Dihydrotestosterone and the prostate: the scientific rationale for 5alpha-reductase inhibitors in the treatment of benign prostatic hyperplasia. J Urol 172:1399-1403. Askonas LJ, Ricigliano JW and Penning TM (1991) The kinetic mechanism catalysed by homogeneous rat liver 3 alpha-hydroxysteroid dehydrogenase. Evidence for binary and ternary dead-end complexes containing non-steroidal antiinflammatory drugs. Biochem J 278 ( Pt 3):835-841. Ax W, Soldan M, Koch L and Maser E (2000) Development of daunorubicin resistance in tumour cells by induction of carbonyl reduction. Biochem Pharmacol 59:293300. Bachur NR (1971) Daunorubicinol, a major metabolite of daunorubicin: isolation from human urine and enzymatic reactions. J Pharmacol Exp Ther 177:573-578. Barbier O and Belanger A (2008) Inactivation of androgens by UDPglucuronosyltransferases in the human prostate. Best Pract Res Clin Endocrinol Metab 22:259-270. Bauman DR, Steckelbroeck S, Williams MV, Peehl DM and Penning TM (2006) Identification of the major oxidative 3alpha-hydroxysteroid dehydrogenase in human prostate that converts 5alpha-androstane-3alpha,17beta-diol to 5alphadihydrotestosterone: a potential therapeutic target for androgen-dependent disease. Mol Endocrinol 20:444-458. Behnia K and Boroujerdi M (1999) Inhibition of aldo-keto reductases by phenobarbital alters metabolism, pharmacokinetics and toxicity of doxorubicin in rats. J Pharm Pharmacol 51:1275-1282. Belanger A, Pelletier G, Labrie F, Barbier O and Chouinard S (2003) Inactivation of androgens by UDP-glucuronosyltransferase enzymes in humans. Trends Endocrinol Metab 14:473-479. Bostwick DG, Burke HB, Djakiew D, Euling S, Ho SM, Landolph J, Morrison H, Sonawane B, Shifflett T, Waters DJ and Timms B (2004) Human prostate cancer risk factors. Cancer 101:2371-2490. Boucek RJ, Jr., Olson RD, Brenner DE, Ogunbunmi EM, Inui M and Fleischer S (1987) The major metabolite of doxorubicin is a potent inhibitor of membrane-associated ion pumps. A correlative study of cardiac muscle with isolated membrane fractions. J Biol Chem 262:15851-15856. Canadian Cancer Society/National Cancer Institute of Canada: Canadian Cancer Statistics 2008, Toronto, Canada, 2008. Chatterjee B (2003) The role of the androgen receptor in the development of prostatic hyperplasia and prostate cancer. Mol Cell Biochem 253:89-101. 40  Chouinard S, Barbier O and Belanger A (2007) UDP-glucuronosyltransferase 2B15 (UGT2B15) and UGT2B17 enzymes are major determinants of the androgen response in prostate cancer LNCaP cells. J Biol Chem 282:33466-33474. Chu LW, Reichardt JK and Hsing AW (2008) Androgens and the molecular epidemiology of prostate cancer. Curr Opin Endocrinol Diabetes Obes 15:261270. Cortes-Funes H and Coronado C (2007) Role of anthracyclines in the era of targeted therapy. Cardiovasc Toxicol 7:56-60. Cummings J, Milstead R, Cunningham D and Kaye S (1986) Marked inter-patient variation in adriamycin biotransformation to 7-deoxyaglycones: evidence from metabolites identified in serum. Eur J Cancer Clin Oncol 22:991-1001. Danesi R, Fogli S, Gennari A, Conte P and Del Tacca M (2002) Pharmacokineticpharmacodynamic relationships of the anthracycline anticancer drugs. Clin Pharmacokinet 41:431-444. Dehm SM and Tindall DJ (2007) Androgen receptor structural and functional elements: role and regulation in prostate cancer. Mol Endocrinol 21:2855-2863. Deng S and Wojnowski L (2007) Genotyping the risk of anthracycline-induced cardiotoxicity. Cardiovasc Toxicol 7:129-134. Evans RM (1988) The steroid and thyroid hormone receptor superfamily. Science 240:889-895. Forrest GL, Gonzalez B, Tseng W, Li X and Mann J (2000) Human carbonyl reductase overexpression in the heart advances the development of doxorubicin-induced cardiotoxicity in transgenic mice. Cancer Res 60:5158-5164. Frye CA, Park D, Tanaka M, Rosellini R and Svare B (2001) The testosterone metabolite and neurosteroid 3alpha-androstanediol may mediate the effects of testosterone on conditioned place preference. Psychoneuroendocrinology 26:731-750. Gewirtz DA (1999) A critical evaluation of the mechanisms of action proposed for the antitumor effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol 57:727-741. Griffin JE (1992) Androgen resistance--the clinical and molecular spectrum. N Engl J Med 326:611-618. Griffin LD and Mellon SH (1999) Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc Natl Acad Sci U S A 96:1351213517. Hasinoff BB and Herman EH (2007) Dexrazoxane: how it works in cardiac and tumor cells. Is it a prodrug or is it a drug? Cardiovasc Toxicol 7:140-144.  41  Hsing AW, Chu LW and Stanczyk FZ (2008) Androgen and prostate cancer: is the hypothesis dead? Cancer Epidemiol Biomarkers Prev 17:2525-2530. Hsing AW, Reichardt JK and Stanczyk FZ (2002) Hormones and prostate cancer: current perspectives and future directions. Prostate 52:213-235. Huggins C and Hodges CV (1941) Studies on prostatic cancer. I. The effect of castration, of estrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1:293-297. Imamoto T, Suzuki H, Yano M, Kawamura K, Kamiya N, Araki K, Komiya A, Nihei N, Naya Y and Ichikawa T (2008) The role of testosterone in the pathogenesis of prostate cancer. Int J Urol 15:472-480. Jakobsson J, Palonek E, Lorentzon M, Ohlsson C, Rane A and Ekstrom L (2007) A novel polymorphism in the 17beta-hydroxysteroid dehydrogenase type 5 (aldo-keto reductase 1C3) gene is associated with lower serum testosterone levels in caucasian men. Pharmacogenomics J 7:282-289. Jez JM, Flynn TG and Penning TM (1997) A new nomenclature for the aldo-keto reductase superfamily. Biochem Pharmacol 54:639-647. Ji Q, Chang L, Stanczyk FZ, Ookhtens M, Sherrod A and Stolz A (2007) Impaired dihydrotestosterone catabolism in human prostate cancer: critical role of AKR1C2 as a pre-receptor regulator of androgen receptor signaling. Cancer Res 67:13611369. Ji Q, Chang L, VanDenBerg D, Stanczyk FZ and Stolz A (2003) Selective reduction of AKR1C2 in prostate cancer and its role in DHT metabolism. Prostate 54:275289. Jin Y and Penning TM (2006) Multiple steps determine the overall rate of the reduction of 5alpha-dihydrotestosterone catalyzed by human type 3 3alpha-hydroxysteroid dehydrogenase: implications for the elimination of androgens. Biochemistry 45:13054-13063. Jin Y and Penning TM (2007) Aldo-keto reductases and bioactivation/detoxication. Annu Rev Pharmacol Toxicol 47:263-292. Karypidis AH, Olsson M, Andersson SO, Rane A and Ekstrom L (2008) Deletion polymorphism of the UGT2B17 gene is associated with increased risk for prostate cancer and correlated to gene expression in the prostate. Pharmacogenomics J 8:147-151. Kume T, Iwasa H, Shiraishi H, Yokoi T, Nagashima K, Otsuka M, Terada T, Takagi T, Hara A and Kamataki T (1999) Characterization of a novel variant (S145C/L311V) of 3alpha-hydroxysteroid/dihydrodiol dehydrogenase in human liver. Pharmacogenetics 9:763-771. LaSpina M and Haas GP (2008) Update on the diagnosis and management of prostate cancer. Can J Urol 15 Suppl 1:3-13; discussion 13. 42  Licata S, Saponiero A, Mordente A and Minotti G (2000) Doxorubicin metabolism and toxicity in human myocardium: role of cytoplasmic deglycosidation and carbonyl reduction. Chem Res Toxicol 13:414-420. Locke JA, Guns ES, Lubik AA, Adomat HH, Hendy SC, Wood CA, Ettinger SL, Gleave ME and Nelson CC (2008) Androgen levels increase by intratumoral de novo steroidogenesis during progression of castration-resistant prostate cancer. Cancer Res 68:6407-6415. Lou H, Du S, Ji Q and Stolz A (2006) Induction of AKR1C2 by phase II inducers: identification of a distal consensus antioxidant response element regulated by NRF2. Mol Pharmacol 69:1662-1672. Lovless H, Arena E, Felsted RL and Bachur NR (1978) Comparative mammalian metabolism of adriamycin and daunorubicin. Cancer Res 38:593-598. Luu-The V, Belanger A and Labrie F (2008) Androgen biosynthetic pathways in the human prostate. Best Pract Res Clin Endocrinol Metab 22:207-221. Ma H, Ratnam K and Penning TM (2000) Mutation of nicotinamide pocket residues in rat liver 3 alpha-hydroxysteroid dehydrogenase reveals different modes of cofactor binding. Biochemistry 39:102-109. MacLeod SL, Nowell S, Plaxco J and Lang NP (2000) An allele-specific polymerase chain reaction method for the determination of the D85Y polymorphism in the human UDP-glucuronosyltransferase 2B15 gene in a case-control study of prostate cancer. Ann Surg Oncol 7:777-782. Marks LS, Mostaghel EA and Nelson PS (2008) Prostate Tissue Androgens: History and Current Clinical Relevance. Urology. Matsuura K, Deyashiki Y, Sato K, Ishida N, Miwa G and Hara A (1997) Identification of amino acid residues responsible for differences in substrate specificity and inhibitor sensitivity between two human liver dihydrodiol dehydrogenase isoenzymes by site-directed mutagenesis. Biochem J 323 ( Pt 1):61-64. Melcangi RC, Garcia-Segura LM and Mensah-Nyagan AG (2008) Neuroactive steroids: state of the art and new perspectives. Cell Mol Life Sci 65:777-797. Menna P, Recalcati S, Cairo G and Minotti G (2007) An introduction to the metabolic determinants of anthracycline cardiotoxicity. Cardiovasc Toxicol 7:80-85. Minotti G, Cairo G and Monti E (1999) Role of iron in anthracycline cardiotoxicity: new tunes for an old song? Faseb J 13:199-212. Minotti G, Menna P, Salvatorelli E, Cairo G and Gianni L (2004) Anthracyclines: molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol Rev 56:185-229.  43  Mohler JL, Gregory CW, Ford OH, 3rd, Kim D, Weaver CM, Petrusz P, Wilson EM and French FS (2004) The androgen axis in recurrent prostate cancer. Clin Cancer Res 10:440-448. Mordente A, Meucci E, Martorana GE, Giardina B and Minotti G (2001) Human heart cytosolic reductases and anthracycline cardiotoxicity. IUBMB Life 52:83-88. Mordente A, Minotti G, Martorana GE, Silvestrini A, Giardina B and Meucci E (2003) Anthracycline secondary alcohol metabolite formation in human or rabbit heart: biochemical aspects and pharmacologic implications. Biochem Pharmacol 66:989-998. Mostaghel EA and Nelson PS (2008) Intracrine androgen metabolism in prostate cancer progression: mechanisms of castration resistance and therapeutic implications. Best Pract Res Clin Endocrinol Metab 22:243-258. Negri-Cesi P and Motta M (1994) Androgen metabolism in the human prostatic cancer cell line LNCaP. J Steroid Biochem Mol Biol 51:89-96. Nunlist EH, Dozmorov I, Tang Y, Cowan R, Centola M and Lin HK (2004) Partitioning of 5alpha-dihydrotestosterone and 5alpha-androstane-3alpha, 17beta-diol activated pathways for stimulating human prostate cancer LNCaP cell proliferation. J Steroid Biochem Mol Biol 91:157-170. O'Connor T, Ireland LS, Harrison DJ and Hayes JD (1999) Major differences exist in the function and tissue-specific expression of human aflatoxin B1 aldehyde reductase and the principal human aldo-keto reductase AKR1 family members. Biochem J 343 Pt 2:487-504. Ohara H, Miyabe Y, Deyashiki Y, Matsuura K and Hara A (1995) Reduction of drug ketones by dihydrodiol dehydrogenases, carbonyl reductase and aldehyde reductase of human liver. Biochem Pharmacol 50:221-227. Olson LE, Bedja D, Alvey SJ, Cardounel AJ, Gabrielson KL and Reeves RH (2003) Protection from doxorubicin-induced cardiac toxicity in mice with a null allele of carbonyl reductase 1. Cancer Res 63:6602-6606. Olson RD and Mushlin PS (1990) Doxorubicin cardiotoxicity: analysis of prevailing hypotheses. Faseb J 4:3076-3086. Page ST, Lin DW, Mostaghel EA, Hess DL, True LD, Amory JK, Nelson PS, Matsumoto AM and Bremner WJ (2006) Persistent intraprostatic androgen concentrations after medical castration in healthy men. J Clin Endocrinol Metab 91:3850-3856. Palackal NT, Burczynski ME, Harvey RG and Penning TM (2001) The ubiquitous aldehyde reductase (AKR1A1) oxidizes proximate carcinogen trans-dihydrodiols to o-quinones: potential role in polycyclic aromatic hydrocarbon activation. Biochemistry 40:10901-10910. Palackal NT, Lee SH, Harvey RG, Blair IA and Penning TM (2002) Activation of polycyclic aromatic hydrocarbon trans-dihydrodiol proximate carcinogens by 44  human aldo-keto reductase (AKR1C) enzymes and their functional overexpression in human lung carcinoma (A549) cells. J Biol Chem 277:2479924808. Park JY, Tanner JP, Sellers TA, Huang Y, Stevens CK, Dossett N, Shankar RA, Zachariah B, Heysek R and Pow-Sang J (2007) Association between polymorphisms in HSD3B1 and UGT2B17 and prostate cancer risk. Urology 70:374-379. Penning TM (2003) Introduction and Overview of the Aldo-Keto Reductase Superfamily, in Aldo-Keto Reductases and Toxicant Metabolism (Penning TM, Petrash, J. M. ed) pp 3-22, American Chemical Society, Washington, DC. Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, Moore M, Palackal N and Ratnam K (2000) Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Biochem J 351:67-77. Penning TM and Drury JE (2007) Human aldo-keto reductases: Function, gene regulation, and single nucleotide polymorphisms. Arch Biochem Biophys 464:241-250. Penning TM, Jin Y, Heredia VV and Lewis M (2003) Structure-function relationships in 3alpha-hydroxysteroid dehydrogenases: a comparison of the rat and human isoforms. J Steroid Biochem Mol Biol 85:247-255. Penning TM, Jin Y, Rizner TL and Bauman DR (2008) Pre-receptor regulation of the androgen receptor. Mol Cell Endocrinol 281:1-8. Peters JH, Gordon GR, Kashiwase D and Acton EM (1981) Tissue distribution of doxorubicin and doxorubicinol in rats receiving multiple doses of doxorubicin. Cancer Chemother Pharmacol 7:65-69. Pinna G, Costa E and Guidotti A (2006) Fluoxetine and norfluoxetine stereospecifically and selectively increase brain neurosteroid content at doses that are inactive on 5HT reuptake. Psychopharmacology (Berl) 186:362-372. Pirog EC and Collins DC (1997) Androgen UDP-glucuronyl transferase activity is found primarily in the liver in the human. J Androl 18:116-122. Platel D, Bonoron-Adele S and Robert J (2001) Role of daunorubicinol in daunorubicininduced cardiotoxicity as evaluated with the model of isolated perfused rat heart. Pharmacol Toxicol 88:250-254. Pollard M, Luckert PH and Schmidt MA (1982) Induction of prostate adenocarcinomas in Lobund Wistar rats by testosterone. Prostate 3:563-568. Reddy DS (2004) Anticonvulsant activity of the testosterone-derived neurosteroid 3alpha-androstanediol. Neuroreport 15:515-518. 45  Rittmaster R, Hahn RG, Ray P, Shannon JB and Wurzel R (2008) Effect of dutasteride on intraprostatic androgen levels in men with benign prostatic hyperplasia or prostate cancer. Urology 72:808-812. Rittmaster RS, Zwicker H, Thompson DL, Konok G and Norman RW (1993) Androstanediol glucuronide production in human liver, prostate, and skin. Evidence for the importance of the liver in 5 alpha-reduced androgen metabolism. J Clin Endocrinol Metab 76:977-982. Rizner TL, Lin HK and Penning TM (2003) Role of human type 3 3alpha-hydroxysteroid dehydrogenase (AKR1C2) in androgen metabolism of prostate cancer cells. Chem Biol Interact 143-144:401-409. Robert J (1994) Clinical pharmacokinetics of epirubicin. Clin Pharmacokinet 26:428438. Schlegel BP, Jez JM and Penning TM (1998) Mutagenesis of 3 alpha-hydroxysteroid dehydrogenase reveals a "push-pull" mechanism for proton transfer in aldo-keto reductases. Biochemistry 37:3538-3548. Shiraishi H, Ishikura S, Matsuura K, Deyashiki Y, Ninomiya M, Sakai S and Hara A (1998) Sequence of the cDNA of a human dihydrodiol dehydrogenase isoform (AKR1C2) and tissue distribution of its mRNA. Biochem J 334 ( Pt 2):399-405. Siiteri PK and Wilson JD (1970) Dihydrotestosterone in prostatic hypertrophy. I. The formation and content of dihydrotestosterone in the hypertrophic prostate of man. J Clin Invest 49:1737-1745. Singal PK and Iliskovic N (1998) Doxorubicin-induced cardiomyopathy. N Engl J Med 339:900-905. Sokolove PM (1994) Interactions of adriamycin aglycones with mitochondria may mediate adriamycin cardiotoxicity. Int J Biochem 26:1341-1350. Speyer J and Wasserheit C (1998) Strategies for reduction of anthracycline cardiac toxicity. Semin Oncol 25:525-537. Stanbrough M, Bubley GJ, Ross K, Golub TR, Rubin MA, Penning TM, Febbo PG and Balk SP (2006) Increased expression of genes converting adrenal androgens to testosterone in androgen-independent prostate cancer. Cancer Res 66:2815-2825. Steinherz LJ, Steinherz PG, Tan CT, Heller G and Murphy ML (1991) Cardiac toxicity 4 to 20 years after completing anthracycline therapy. Jama 266:1672-1677. Swanson C, Mellstrom D, Lorentzon M, Vandenput L, Jakobsson J, Rane A, Karlsson M, Ljunggren O, Smith U, Eriksson AL, Belanger A, Labrie F and Ohlsson C (2007) The uridine diphosphate glucuronosyltransferase 2B15 D85Y and 2B17 deletion polymorphisms predict the glucuronidation pattern of androgens and fat mass in men. J Clin Endocrinol Metab 92:4878-4882.  46  Takanashi S and Bachur NR (1976) Adriamycin metabolism in man. Evidence from urinary metabolites. Drug Metab Dispos 4:79-87. Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, Lieber MM, Cespedes RD, Atkins JN, Lippman SM, Carlin SM, Ryan A, Szczepanek CM, Crowley JJ and Coltman CA, Jr. (2003) The influence of finasteride on the development of prostate cancer. N Engl J Med 349:215-224. Tindall DJ and Rittmaster RS (2008) The rationale for inhibiting 5alpha-reductase isoenzymes in the prevention and treatment of prostate cancer. J Urol 179:12351242. Trauger JW, Jiang A, Stearns BA and LoGrasso PV (2002) Kinetics of allopregnanolone formation catalyzed by human 3 alpha-hydroxysteroid dehydrogenase type III (AKR1C2). Biochemistry 41:13451-13459. Wallace KB (2003) Doxorubicin-induced cardiac mitochondrionopathy. Pharmacol Toxicol 93:105-115. Weiss RB (1992) The anthracyclines: will we ever find a better doxorubicin? Semin Oncol 19:670-686. Wojtacki J, Lewicka-Nowak E and Lesniewski-Kmak K (2000) Anthracycline-induced cardiotoxicity: clinical course, risk factors, pathogenesis, detection and prevention--review of the literature. Med Sci Monit 6:411-420. Xu MF, Tang PL, Qian ZM and Ashraf M (2001) Effects by doxorubicin on the myocardium are mediated by oxygen free radicals. Life Sci 68:889-901. Yang Q, Fung KM, Day WV, Kropp BP and Lin HK (2005) Androgen receptor signaling is required for androgen-sensitive human prostate cancer cell proliferation and survival. Cancer Cell Int 5:8. Zhu YS and Imperato-McGinley JL (2009) 5alpha-reductase isozymes and androgen actions in the prostate. Ann N Y Acad Sci 1155:43-56.  47  CHAPTER 2 TWO ALLELIC VARIANTS OF ALDO-KETO REDUCTASE 1A1 EXHIBIT REDUCED IN VITRO METABOLISM OF DAUNORUBICIN1  2.1.  Preface The anthracyclines are invaluable chemotherapy medicines; however, their use is  limited by the development of cardiotoxicity leading to lethal congestive heart failure in some patients receiving these drugs (Cortes-Funes and Coronado, 2007). The formation of cardiac damage is dose-dependent and therefore, its occurrence is minimized by limiting the total lifetime dose that a patient will receive (Singal and Iliskovic, 1998). These restrictions protect many patients from suffering cardiotoxicity; however, highly susceptible patients continue to develop and die of this disorder. Conversely, other patients may be expected to tolerate higher doses, which may provide more successful cancer treatment without cardiomyopathy. Unfortunately, there is currently no way to identify these patients prior to treatment (Deng and Wojnowski, 2007). These important concerns show that in order to achieve the optimal use of the anthracyclines, it is essential to understand the mechanism(s) whereby cardiotoxicity develops.  This information  should also contribute to a better understanding of the source(s) of the interpatient variability in toxicity seen with these drugs. The mechanism by which this side effect manifests itself has remained elusive, though there is good evidence that their metabolism is involved (Menna et al., 2007). As described in Section 1.2.2., the leading hypotheses describe reductive metabolism of the 1  A version of this chapter has been published. Bains OS, Takahashi RH, Pfeifer TA, Grigliatti TA, Reid RE and Riggs KW (2008) Two allelic variants of aldo-keto reductase 1A1 exhibit reduced in vitro metabolism of daunorubicin. Drug Metab Dispos 36:904910. 48  anthracyclines as generating either reactive chemical species or toxic metabolites, both of which have high liabilities for damaging cardiac tissue. Further, there is good evidence that when the metabolism of anthracyclines is reduced, the occurrence of cardiac damage is also reduced. DOX and DAUN represent the most clinically used anthracycline anticancer drugs and both are associated with anthracycline-induced cardiotoxicity as a side effect. Their metabolism is similar, with the alcohol metabolites DOXol and DAUNol being the primary species formed in both human and animal subjects (Takanashi and Bachur, 1976). The enzymes that are responsible for these transformations are AKR1A1, AKR1C2, and the carbonyl reductases 1 and 3. Genotyping of these genes reveal that non-synonymous SNPs occur in these genes; however, the functional consequences of these genetic variations are not known. Based on evidence demonstrating that SNPs are associated with impaired enzyme function in several AKR proteins, it is likely that SNPs in these reductase genes may result in enzymes that have an altered activity with respect to the metabolism of DOX or DAUN.  Consequently, the levels of toxic species  generated in the metabolism of DOX may be altered and dictate the development of cardiac damage. Such an association would provide valuable evidence for the genetic role in the marked interpatient variability observed in the manifestation of anthracyclineinduced cardiac damage. The objectives of this chapter are to characterize the function of AKR1A1 in metabolizing DOX and DAUN in vitro and to investigate the effects of naturally occurring non-synonymous SNPs in AKR1A1 in altering the enzyme function in these metabolic pathways.  The functional characterization of the wild-type and variant  AKR1A1 proteins were conducted by measuring the kinetic activities of the isolated recombinant proteins in the in vitro reductions of standard AKR substrates and the 49  anthracyclines DOX and DAUN. The variant AKR1A1 alleles that are associated with altered activities in metabolizing DOX and DAUN may be one factor that contributes to the interpatient variability in response to treatment with these drugs, and hence, may be good potential markers for patients at increased risk of cardiotoxicity.  2.2.  Materials and Methods  2.2.1. Chemicals and enzymes Agarose, chloramphenicol, daunorubicin (DAUN), doxorubicin (DOX), DLglyceraldehyde,  kanamycin  sulphate,  lysozyme,  methanol, p-nitrobenzaldehyde,  potassium phosphate (KH2PO4) RnaseI, sodium phosphate (NaH2PO4), N,N,N’,N’tetramethylethylenediamine  (TEMED)  and  "-nicotinamide  adenine  dinucleotide  phosphate reduced tetrasodium salt (NADPH) were supplied by Sigma-Aldrich (St. Louis, MO). HPLC-grade acetonitrile, agar, ammonium persulfate, formic acid, glycine, glycerol, glacial acetic acid, imidazole, and Tris (hydroxymethyl)aminomethane (Tris) were purchased from Fisher Scientific (Fair Lawn, NJ). Sodium chloride (NaCl) and yeast extract were acquired from EMD Chemicals Inc. (Darmstadt, Germany). BactoTMtryptone and isopropyl "-D-1-thiogalactopyranoside (IPTG) were obtained from BectonDickinson and Co. (Franklin Lakes, NJ) and Fermentas (Hanover, MD), respectively. Tween-20 was purchased from EMD Biosciences, Inc. (La Jolla, CA) while DNaseI was provided by Boehringer Manheim GmbH (Manheim, Germany). Klenow fragment and T4 DNA ligase were obtained from Fermentas (Burlington, ON). Restriction enzymes used for this study were purchased from New England Biolabs, Inc. (Ipswich, MA). Doxorubicinol (DOXol) was obtained from Qventas Inc. (Branford, CT, USA).  50  2.2.2. Molecular cloning of the human AKR1A1 gene The insect cell expression vector, p2ZOp2N, containing the wild-type AKR1A1 and the two variant genes were engineered to produce a 6#His tag and Factor Xa (FXa) cleavage site on the amino terminus of the AKR gene product. The 6#His tag-FXa-AKR coding regions were excised from the p2ZOp2N vector using DraI and NotI and subcloned into the NheI (blunt end with Klenow fragment)-NotI sites of the pET28a (EMD Chemicals Inc., San Diego, CA) prokaryotic expression vector with T4 DNA ligase. These constructs encoded an AKR with two 6#His tags separated by a 17 amino acid residue linker on the amino terminus (Figure 2.1A). The inner 6#His tag for the pET28a-AKR1A1 wild type and variant constructs was deleted (Figure 2.1B) using the QuikChange® Site-Directed Mutaganesis Kit (Stratagene, La Jolla, CA) with the 5’—CCATCCTACCCTCGATCATGTTAAGCTTTCTAG—3’  (forward)  and  5’—CTAGAAAGCTTAACATGATCGAGGGTAAGATGG—3’ (reverse) primers. All constructs were verified by dideoxy sequencing at the University of British Columbia Nucleic Acid Protein Service (NAPS) unit.  51  Figure 2.1 A translated protein product of the pET28a-AKR1A1 construct with (A) double 6#His tags and (B) a single 6#His tag. The inner 6#His tag for (A) was deleted through site directed mutagenesis.  2.2.3. Expression of recombinant human AKR1A1 The pET28a constructs (single and double 6#His tags) were heat-shock transformed into E. coli BL21 (DE3) pLysS and expressed under the control of an IPTGinducible T7 polymerase.  Bacterial cultures (500 ml) were grown at 37 °C in LB  medium (1% bacto-tryptone, 0.5% yeast extract, 0.5% NaCl) supplemented with 50 $g/ml kanamycin sulfate and 25 $g/ml chloramphenicol until an OD600 of 0.4 was reached. IPTG was added to a final concentration of 1 mM and cells were allowed to grow for an additional 3 h. Aliquots of cells (1 ml) were collected at 0 (uninduced), 1, 2 and 3 h post-IPTG administration for the assessment of AKR1A1 expression levels. The 52  aliquots and cultures remaining after the 3 h of induction were harvested by centrifugation (4000xg, 20 min, 4oC), and the resulting bacterial cell pellets were stored at -70 °C.  2.2.4. Purification of recombinant human AKR1A1 The frozen bacterial pellets were thawed on ice and resuspended at 5 ml per gram wet weight with Buffer A (300 mM NaCl, 50 mM NaH2PO4, pH 8.0) for the 1 ml aliquots while the suspension from the remaining cultures was further supplemented with 10 mM imidazole. A final concentration of 1 mg/ml lysozyme was added to all of the cell suspensions, followed by incubation on ice for 30 min. Cells were disrupted using six 10 sec bursts (with a 10 sec cooling period between each burst) from a sonic dismembranator with a microtip set at 200-300W. This was followed by incubation with DNaseI and RNAseI (5 and 10 $g/ml, respectively) for 15 min on ice and then centrifugation (10,000xg, 20 min, 4oC). The cell lysate supernatants of the aliquots collected from 0 to 3 hr were saved for Western blot analysis while the lysate from the remaining culture was subjected to Ni-NTA affinity chromatography, with the recombinant protein being isolated according to the manufacturer’s instructions (QIAGEN, Mississauga, Ontario). Briefly, the supernatant was incubated with Ni-NTA agarose beads for 1.5 h at 4oC and the mixture was transferred into a column. The 6#Histagged AKR1A1 protein bound to the agarose beads was washed with multiple fractions (2.5 bed volumes per fraction) of Buffer A containing 20 mM imidazole to remove nonspecific endogenous bacterial proteins bound to the beads, which was monitored by measuring the total protein concentration of each fraction using the Bio-Rad Protein Assay (Hercules, CA). AKRs were further eluted with multiple fractions of Buffer A with 30, 50, 100 and 250 mM imidazole. At each of these imidazole concentrations, 53  fractions (0.75 bed volume) were collected until the total protein dropped to baseline levels. Glycerol was added to the elution fractions to a final concentration of 20% and the samples stored at -20oC.  2.2.5. Total protein staining Purity of the AKR1A1 elution fractions was assessed visually after electrophoresis using an 18% SDS-polyacrylamide gel and staining for total protein. The gels were fixed in 50% methanol, 7% glacial acetic acid for two periods of 30 min each and then stained with SYPRO® Ruby (Invitrogen Canada, Inc., Burlington, Ontario) overnight. Following staining, the gels were washed in 10% methanol, 7% glacial acetic acid for 30 min and the protein detected using a Storm 860 Molecular Imager (GMI Inc., Ramsey, MN).  2.2.6. Western blotting Western blot analyses of the cell lysates from the aliquots and the purified fractions were performed according to the protocol described by OdysseyTM (LI-COR® Biosciences, Lincoln, NE).  After an 18% SDS-PAGE, proteins were transferred in  Towbin's buffer (25 mM Tris, 192 mM glycine, 20% v/v methanol) overnight at 20-30V to a HybondTM-C Extra nitrocellulose membrane (Amersham Biosciences, Piscataway, NJ). The membranes were blocked in OdysseyTM blocking buffer (BB) and the enzyme was detected using a monoclonal mouse anti-human AKR1A1 (Abnova® Corporation, Taipei City, Taiwan) antibody (diluted 1:5000) as the primary and IRDye 800CW goat anti-mouse IgG was used as the secondary antibody (diluted 1:5000) (LI-COR® Biosciences, Lincoln, NE).  Both primary and secondary antibodies were in BB  54  containing 0.1% Tween-20. The bound secondary antibodies were detected using the OdysseyTM Infrared Imaging system (LI-COR® Biosciences).  2.2.7. AKR1A1 enzymatic activity assays The activity of the purified single and double 6#His-tagged AKR1A1 was measured at 25oC using a Fluoroskan Ascent® FL (Thermo Fisher Scientific, Waltham, MA) by following the initial rate of NADPH oxidation at excitation and emission wavelengths of 355nm and 460nm, respectively.  The assays were conducted as  previously described for the characterization of AKR proteins (Palackal et al., 2001). In short, purified protein was incubated with 180 $M NADPH and 1 mM test substrate, either p-nitrobenzaldehyde or DL-glyceraldehyde, in a reaction mixture of 150 $l 100 mM potassium phosphate, pH 7.0, at 25ºC. Protein concentration and incubation times were selected for each enzyme and substrate concentration to ensure that measured rates were in the linear range of the enzyme kinetic curve. In these assays, the concentration of organic solvent, which was required to dissolve the substrate, was kept below 4% (v/v) in the final reaction mixture. Readings were collected at 1 min intervals for 3 h with shaking between each reading. Maximal rates were calculated from the Ascent® program (version 2.6) using a 5 min interval with the steepest slope. Enzymatic activity ($mol NADPH consumed/min/mg purified protein) was calculated from these rates using a standard curve constructed from the fluorescence measurements of solutions of known NADPH concentrations. Activity measurements for DAUN reduction were performed by incubating either DOX or DAUN with purified AKR1A1 protein in a total volume of 150 µl containing 25 mM potassium phosphate pH 7.4 and 1 mM NADPH at 37ºC. Protein concentrations 55  were based on the Bradford protein assay using bovine serum albumin as a standard. The reaction was stopped by adding 300 µl of ice-cold acetonitrile, which contained idarubicin as an internal standard, followed by vortex mixing and centrifuging at 10,000xg for 10 minutes at 4ºC to remove protein. The supernatant was removed for HPLC analysis. HPLC separation was carried out using a Waters Alliance 2695 system (Waters Corporation, Milford, MA) with an analytical column (Waters Symmetry C18, 75 # 4.6 mm i.d., 3 µm) and guard column (Phenomenex SecurityGuard C18, 40 # 4.6 mm i.d.; Phenomenex, Torrance, CA). HPLC conditions were as follows: mobile phase A, 0.1% formic acid and B, acetonitrile; gradient elution, 0 to 1 min, 15% B; 1 to 8 min, 15% B to 35% B; 8 to 10 min, 35% B; 10 to 10.1 min, 35% B to 15% B; column reequilibration for 2 minutes at a constant flow rate of 1 ml/min. The column was heated to 30ºC and the autosampler temperature was set to 10ºC with 5 µl of sample injected onto the column. Fluorescence detection of DOX and DAUN, and their carbon-13 hydroxy metabolites was performed with excitation and emission wavelengths of 460 and 550 nm, respectively (Waters 2475 Multi ! Detector). Quantitation of DOXol and DAUNol was performed based on a weighted (1/x2) linear regression determined from solutions of known concentrations of an authentic chemical standard for DOXol. Because a chemical standard for DAUNol could not be obtained, its concentrations were calculated as DOXol equivalents using a response ratio of 1.0. All HPLC data processes, including chromatogram integration, calibration and quantitative calculations, were performed with Waters Empower software (version 2.0). The kinetic constants of maximal rate of reaction, Vmax, and Km were determined by fitting rate measurement data using nonlinear least-squares fitting of a MichaelisMenten hyperbola (GraphPad Prism version 4.0; GraphPad Software Inc., San Diego, 56  CA). kcat values were calculated from Vmax values using the apparent molecular weight (Mr) for the 6#His-tagged AKR protein of 41 kDa.  2.2.8. Statistical analysis Statistical analyses were performed using GraphPad Instat® (version 3.6; GraphPad Software Inc., San Diego, CA). Results are expressed as means ± S.D. Enzyme activities were compared using a one-way ANOVA followed by Tukey–Kramer multiple comparisons tests. Differences were considered significant at p<0.05.  2.3.  Results  2.3.1. Expression and purification Successful expression of the single and double 6#His-tagged human AKR1A1 was obtained and visualized by performing a Western blot on the non-induced and IPTG induced (1, 2 and 3 h) cell lysates from the aliquots that were collected during the growth of the bacterial cultures. In order to purify the wild-type and ns-SNPs to homogeneity, the lysates from the 3 h induced cultures were subjected to affinity chromatography yielding fractions containing the 6#His tagged AKR1A1 without the presence of detectable contaminating proteins (Figure 2.2).  The majority of pure enzyme was  recovered in the elution fractions with 50 and 250 mM imidazole for the single and double 6#His-tagged enzymes, respectively. The bands for each of the purified samples have a mobility corresponding to the relative molecular weight of the 6#His tagged AKR1A1, which is ~41-42 kDa. The identities of these fractions were verified by Western blot analysis (Figure 2.2).  57  Figure 2.2 Purification of human recombinant (A) double 6#His-tagged AKR1A1 wildtype, (B) single 6#His-tagged AKR1A1 wild-type, (C) E55D variant and (D) N52S variant enzymes. (Left) Gels stained with SYPRO® Ruby following SDS-PAGE showing purified protein samples, 1A1 wild-type and variants (700 ng), free of contaminating proteins from bacterial lysates (L) (20 µg total protein). Removal of contaminating proteins is observed in fractions from Qiagen purification procedures (NiNTA column flow through (F), 15 µg total protein). (Right) Western blot detection of purified protein samples, 1A1 wild-type and variants (700 ng), confirms expression of the desired AKR protein with mobility at expected molecular weight (~41-42 kDa). Positive controls for antibody immunoreactivity are human liver cytosol (H; 20 $g total protein) and GST-tagged purified human recombinant AKR enzymes (G; 700 ng). No antibody immunoreactivity is observed for untransformed bacterial lysate (U; 20 $g total protein).  2.3.2. AKR1A1 enzymatic activities with test substrates The single 6#His-tagged AKR1A1 wild-type had greater metabolic activities for p-nitrobenzaldehyde and DL-glyceraldehyde compared to the enzyme tagged with double 6#His or the glutathione S-transferase (GST)-tagged wild-type construct (Abnova® Corporation) (Figure 2.3). The reaction rates for the single 6#-His-tagged wild-type 58  AKR1A1 were 5.09±0.16 and 1.24±0.17 µmol NADPH consumed/min/mg purified protein for p-nitrobenzaldehyde and DL-glyceraldehyde, respectively. These values were in accordance with the reaction rates of bacterially-expressed recombinant human untagged AKR1A1 reported in literature using identical or comparable assays: 6.0 µmol/min/mg for p-nitrobenzaldehyde and 1.26 µmol/min/mg for DL-glyceraldehyde (O’Connor et al., 1999; Palackal et al., 2001). This suggests that the amino acid linker and 6#His tag engineered on the N-terminus of the AKR gene product have no significant effect on the enzyme activity. Therefore, the recovery tag does not need to be cleaved off with Factor Xa (FXa) for subsequent activity assays involving the anthracyclines. Enzyme activity measurements of the single 6#His-tagged E55D allelic variant demonstrated significantly reduced metabolic activity with p-nitrobenzaldehyde while both ns-SNPs had significantly reduced activity with DL-glyceraldehyde (Figure 2.4). The enzymatic rates for reductions of p-nitrobenzaldehyde and DL-glyceraldehyde for the E55D variant were 1.04±0.26 and 0.52±0.17 $mol/min/mg, respectively, while the N52S variant had rates of 5.00±0.41 and 0.85±0.17 $mol/min/mg.  59  Figure 2.3 In vitro enzymatic activities for the purified single and double 6#His-tagged AKR1A1 wild-types along with the GST-tagged AKR1A1 wild-type in the presence of 1 mM (A) p-nitrobenzaldhyde and (B) DL-glyceraldehyde test substrates as measured by following the initial rate of NADPH oxidation. A single batch of each enzyme was purified with assays being performed in triplicate for each batch. Enzymatic activities are reported as mean ± S.D. (n=3) with the background levels subtracted. The background levels represented the reaction buffer, enzyme and NADPH cofactor only, without the addition of the test substrate. *, p<0.05; ***, p <0.001; significantly different from the single 6#His-tagged treatment group by one-way ANOVA followed by Tukey–Kramer multiple comparisons tests.  60  Figure 2.4 In vitro enzymatic activities for the purified single 6#His-tagged AKR1A1 wild-type and ns-SNP variants with 1 mM (A) p-nitrobenzaldehyde and (B) DLglyceraldehyde test substrates as measured by following the initial rate of NADPH oxidation. Three independent batches of each enzyme were purified. Assays were performed in quadruplicate with each batch. Enzymatic activities are reported as mean ± S.D. (n=12) with the background levels subtracted. The background levels represented the reaction buffer, enzyme and NADPH cofactor only, without the addition of the test substrate. **, p<0.01; ***, p <0.001; significantly different from wild-type treatment group by one-way ANOVA followed by Tukey–Kramer multiple comparisons tests.  2.3.3. AKR1A1 enzymatic activities with anthracyclines To evaluate the impact of the single amino acid substitutions on the reduction of the anthracycline drugs by AKR1A1, we measured the formation of the alcohol metabolites in vitro.  The in vitro conditions were selected to reflect physiological  61  conditions (i.e., pH 7.4, 37ºC) and both DAUN and DOX were studied.  Full  chromatographic resolution of DAUNol and DOXol from DAUN, DOX, and idarubicin (internal standard) was achieved for all chemical standards and in vitro samples. DOXol, DOX, DAUNol, DAUN, and idarubicin were observed to elute at 4.5, 5.5, 6.0, 6.8, and 7.3 minutes, respectively. Incubation of the single 6#His-tagged AKR1A1 protein with DOX generated a single new chromatographic peak that was identified as DOXol. Similarly, incubation with DAUN generated a single new chromatographic peak that was identified as DAUNol. The identification of the metabolite peaks was confirmed by incubation of DOX and DAUN with human liver cytosol and the generation of compounds that had identical chromatographic behaviors, as well as correspondence in retention time of the metabolite peak from DOX incubations with that for the chemical standard of DOXol.  No detectable amounts of DAUNol or DOXol were found in  incubations conducted in the absence of protein. Specific activity measurements were performed in order to compare anthracycline reductase function of the single 6#His-tagged AKR1A1 wild-type and ns-SNP variant proteins. Specific activities for the wild-type, N52S and E55D proteins were determined as 2.34±0.71, 0.72±0.47, and 1.10±0.42 nmol/min/mg protein, respectively, using DAUN as the substrate at an initial concentration of 1 µM (Figure 2.5). For DOXol, metabolite was only observed to form after 120 or 240 minutes by one 6#His tagged AKR1A1 wildtype and E55D samples, but the amount of metabolite produced over that time was less than the lower limit of quantification (25 nM) for the HPLC assay.  No DOXol  metabolite was observed using the tagged N52S allelic variant.  62  Figure 2.5 Generation of DAUNol in vitro of purified single 6#His-tagged AKR1A1 wild-type and ns-SNP variants. Measurement of DAUNol was performed using HPLCfluorescence detection. Three independent batches of each enzyme were purified. Assays were performed in triplicate with each batch. Enzymatic activities are reported as mean ± S.D. (n=9). ***, significantly different from wild-type treatment group by oneway ANOVA followed by Tukey–Kramer multiple comparisons tests, p <0.001.  To further understand the functional consequences of the single amino acid substitutions on anthracycline metabolism, we performed kinetic studies using DAUN as a prototype for anthracycline drugs, at concentrations from 12.5 to 1250 µM. It was noted during the experiments that some extrapolation of the enzyme kinetics curves was necessary to estimate Vmax and Km parameters; however, complete saturation with higher substrate concentrations was not experimentally feasible due to analytical and limited substrate constraints. We observed similar maximal rates of reduction, but the Km values for the N52S and E55D were both increased by approximately 2-fold (Table 2.1). The reduced substrate affinities of these enzymes from the wildtype isoform are in good agreement with the differences we observed in specific activity for DAUN in vitro at 1 µM concentration.  63  Table 2.1 Kinetic constants for DAUN reduction by recombinant AKR1A1 wildtype and variant allele proteins  wildtype  AKR1A1 N52S  E55D  mM  1.2 ± 0.3  2.3 ± 0.9*  2.3 ± 1.1*  Vmax  µmol/min! mg protein  4.4 ± 0.5  4.3 ± 2.0  3.1 ± 1.0  kcatb  s-1  3.0  2.1  3.0  2.5#103  9.3#102  1.3#103  Kinetic parameter a Km  kcat/Km  s-1 M-1  * significantly different from wildtype when tested using one-way ANOVA followed by Tukey-Kramer multiple comparison test (p<0.05) a  Values correspond to mean ± S.D. obtained from three experiments performed with three independent protein preparations (n=9) for each isoform. b  kcat calculated from Mr 41 kDa.  2.4.  Discussion AKRs play a fundamental role in the metabolism of a wide array of carbonyl-  containing compounds that are either endogenous or exogenous in nature. However, few studies have examined the effect of genetic polymorphisms of specific AKRs on physiological and biomolecular processes that can lead to increased or decreased susceptibility to human-related diseases. To our knowledge, this chapter describes the first experimental finding of the altered metabolism of DOX and DAUN by the single 6#His-tagged human AKR1A1 (wild-type and ns-SNPs). These anthracyclines have contributed to improved life expectancy in countless cancer patients through different mechanisms such as topoisomerase II inhibition, DNA intercalation, induction of cell apoptosis and RNA synthesis inhibition (Minotti et al., 2004; Rabbani et al., 2005). In  64  this study, in vitro metabolic assays using purified AKR1A1, which was expressed and induced in E. coli, were conducted with DOX and DAUN.  Incubations with test  substrates, p-nitrobenzaldehyde and DL-glyceraldehyde, were performed in order to ensure that purified enzyme was fully functional. The water soluble carbon-13 alcohol metabolites, DOXol and DAUNol, were selected for analysis since previous studies have acknowledged them to be the major metabolites in patients (Plebuch et al., 2007). It was shown that DOX is a poor substrate for the single 6#His tagged AKR1A1 unlike DAUN, which is readily metabolized to its respective carbon-13 metabolite. This is an interesting finding since the chemical structures of DOX and DAUN are very similar (Figure 2.6). This higher metabolic activity for DAUN compared to DOX is in agreement with what has been observed in vitro for formation of the alcohol metabolites by cytosolic fractions prepared from human cardiac tissue (Mordente et al., 2001).  It would appear that  AKR1A1 has high specificity for DAUN over DOX even though these anthracyclines are structurally similar.  Figure 2.6 Chemical structures of DOX and DAUN and their conversion to their corresponding carbon-13 alcohol metabolites, DOXol and DAUNol.  65  The amino acid linker modification of AKR1A1 was essential in exposing the 6#His tag to the Ni-NTA resin for successful purification using Ni-NTA affinity chromatography.  Using the single and double 6#His-tagged wild-type and variant  enzymes, our intent was to remove the tag(s) and amino acid linker using FXa, which cleaves after the arginine residue in the IEGR recognition sequence, in order to produce the native AKR1A1. This was unsuccessful with substantial cleavage occurring at a secondary site (data not shown). It is likely that the FXa cleavage site was hidden within the protein structure, thereby making it inaccessible for recognition by the FXa enzyme. The following attempts to remove the 6#His tag by FXa cleavage optimization were undertaken: (i) altering concentrations of the reaction buffer components, Tris-HCl, CaCl2 and NaCl; (ii) using different ranges of pH (6 to 9) and temperature (4 to 37oC) to increase FXa specificity; (iii) increasing FXa enzyme concentration; and (iv) incubating the 6#His tagged protein with thrombin to cleave the tag and a portion of the linker at the thrombin recognition site, which may alter the folding of AKR1A1 and expose the primary cleavage site for FXa.  Secondary site cleavage was also seen when FXa  incubations were conducted while the enzyme was bound to the Ni-NTA resin. However, as stated earlier, the activities of the single 6#His tagged wild-type enzyme were in accordance to activity values of non-tagged human AKR1A1 as reported in the literature using comparable assays. This was not the case for the double 6#His-tagged enzymes, which were demonstrated to have lower activity when using the test substrates. Based on these findings, we deemed it unnecessary to cleave the tag and linker from single 6#His-tagged AKR1A1. The single 6#His-tagged ns-SNP variants were demonstrated to have significantly reduced catalytic activity compared to that of the single 6#His-tagged wild-type. Reduction of the carbonyl group by AKR enzymes is thought to involve the co-operation 66  of four amino acids (tyrosine, lysine, aspartic acid and histidine), which form a catalytic tetrad. These four amino acids and their positions are conserved in the vast majority of the individual reductases within this superfamily (Jez et al., 1997a; Penning, 2001). In the case of AKR1A1, these amino acids would be found in the following positions: aspartic acid-45, tyrosine-50, lysine-85 and histidine-113. The phenolic hydroxyl group of tyrosine is considered to be the most likely candidate to function as the general acid/base catalyst of the reaction with AKR1A1 substrates while the other amino acid residues appear to have auxiliary roles in catalysis (Schlegel et al., 1998; Di Luccio et al., 2006; Jin and Penning, 2007). Since the single amino acid changes seen with the two nsSNPs, N52S and E55D, occur close to the catalytic tyrosine-50 residue, it is possible that these changes hinder the catalytic role of tyrosine-50 in the metabolism of the anthracyclines and the test substrates.  Though the chemical properties are mostly  retained in these amino acids substitutions (N52S: both residues are hydrophilic; E55D: both residues are negatively charged and hydrophilic), the size and shape differences in the amino acid side chains may affect the structure of the active site of the enzyme, thereby altering the efficiency of the metabolic process. Based on the observed metabolic activity differences between the single 6#His tagged AKR1A1 wild-type and allelic variants, we hypothesize that they contribute to the inter-individual variability seen in DOX or DAUN metabolism. We observed that the N52S and E55D substitutions result in a reduction of approximately 50% in catalytic efficiency (kcat/Km) of the AKR1A1 enzyme due to changes in substrate affinity. This suggests that the efficiency of anthracycline alcohol metabolite formation via these enzymes per unit time will be dramatically reduced in patients that carry this ns-SNP. There are several proposed mechanisms of anthracycline-induced cardiotoxicity, which include cellular toxicity from metabolites, selective inhibition of gene expression for 67  proteins associated with contraction of the myocardium and inhibition of topoisomerase II activity (Boucek et al., 1987; Ito et al., 1990; Cummings et al., 1991; Minotti et al., 1995; Mordente et al., 2001; Adamcova et al., 2003).  Although the cause of  cardiotoxicity is probably multifactorial, most studies support the view that an increase in reactive oxygen species (ROS) plays a key role in the pathogenesis of anthracyclineinduced cardiomyocyte damage (Doroshow, 1983; Rajagopalan et al., 1988; Yen et al., 1996; Arai et al., 2000; Singal et al., 2000; Kim et al., 2005; Wold et al., 2005). ROS are a major concern since they are capable of deleterious effects such as oxidation of cell membrane lipids, defects of the mitochondrial respiratory chain, DNA disintegration and dysfunction of enzymes containing sulfhydryl groups, all of which contribute to chronic cardiotoxicity (Wojtacki et al., 2000). ROS generation in the presence of DOX and DAUN is speculated to occur through metabolic processes involving oxidoreductases such as the AKRs (Tokarska-Schlattner et al., 2006). Therefore, it is conceivable that certain AKR alleles can contribute to a substantially higher, or lower, risk of producing ROS. In conclusion, this study illustrates that a one 6#His tagged AKR1A1 enzyme with similar activity to the untagged enzyme is an efficient metabolizer of DAUN and has little to no activity on DOX. We have also demonstrated here that the two naturally occurring allelic variations in AKR1A1 lead to reduced reductase activity towards DAUN as a substrate. Hence, the N52S and/or E55D SNPs in AKR1A1 may prove to be useful genetic biomarkers for assessing the risk of developing cardiotoxicity in cancer patients prior to treatment. Individuals treated with DAUN and who carry one or both of these alleles would be expected to have a reduced ability to eliminate DAUN via this pathway. Consequently, this may lead to its metabolism by a less favorable pathway and thus the  68  generation of other harmful metabolites or reactive intermediates. Studies to support these hypotheses are currently underway in our laboratory.  69  2.5.  References  Adamcova M, Pelouch V, Gersl V, Kaplanova J, Mazurova Y, Simunek T, Klimtova I and Hrdina R (2003) Protein profiling in daunorubicin-induced cardiomyopathy. Gen Physiol Biophys 22:411-419. Arai M, Yoguchi A, Takizawa T, Yokoyama T, Kanda T, Kurabayashi M and Nagai R (2000) Mechanism of doxorubicin-induced inhibition of sarcoplasmic reticulum Ca(2+)-ATPase gene transcription. Circ Res 86:8-14. Boucek RJ, Jr., Olson RD, Brenner DE, Ogunbunmi EM, Inui M and Fleischer S (1987) The major metabolite of doxorubicin is a potent inhibitor of membrane-associated ion pumps. A correlative study of cardiac muscle with isolated membrane fractions. J Biol Chem 262:15851-15856. Cortes-Funes H and Coronado C (2007) Role of anthracyclines in the era of targeted therapy. Cardiovasc Toxicol 7:56-60. Cummings J, Anderson L, Willmott N and Smyth JF (1991) The molecular pharmacology of doxorubicin in vivo. Eur J Cancer 27:532-535. Deng S and Wojnowski L (2007) Genotyping the risk of anthracycline-induced cardiotoxicity. Cardiovasc Toxicol 7:129-134. Di Luccio E, Elling RA and Wilson DK (2006) Identification of a novel NADH-specific aldo-keto reductase using sequence and structural homologies. Biochem J 400:105-114. Doroshow JH (1983) Effect of anthracycline antibiotics on oxygen radical formation in rat heart. Cancer Res 43:460-472. Ito H, Miller SC, Billingham ME, Akimoto H, Torti SV, Wade R, Gahlmann R, Lyons G, Kedes L and Torti FM (1990) Doxorubicin selectively inhibits muscle gene expression in cardiac muscle cells in vivo and in vitro. Proc Natl Acad Sci U S A 87:4275-4279. Jin Y and Penning TM (2007) Aldo-keto reductases and bioactivation/detoxication. Annu Rev Pharmacol Toxicol 47:263-292. Kim DS, Kim HR, Woo ER, Hong ST, Chae HJ and Chae SW (2005) Inhibitory effects of rosmarinic acid on adriamycin-induced apoptosis in H9c2 cardiac muscle cells by inhibiting reactive oxygen species and the activations of c-Jun N-terminal kinase and extracellular signal-regulated kinase. Biochem Pharmacol 70:10661078. Menna P, Recalcati S, Cairo G and Minotti G (2007) An introduction to the metabolic determinants of anthracycline cardiotoxicity. Cardiovasc Toxicol 7:80-85.  70  Minotti G, Cavaliere AF, Mordente A, Rossi M, Schiavello R, Zamparelli R and Possati G (1995) Secondary alcohol metabolites mediate iron delocalization in cytosolic fractions of myocardial biopsies exposed to anticancer anthracyclines. Novel linkage between anthracycline metabolism and iron-induced cardiotoxicity. J Clin Invest 95:1595-1605. Minotti G, Menna P, Salvatorelli E, Cairo G and Gianni L (2004) Anthracyclines: molecular advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol Rev 56:185-229. Mordente A, Meucci E, Martorana GE, Giardina B and Minotti G (2001) Human heart cytosolic reductases and anthracycline cardiotoxicity. IUBMB Life 52:83-88. Palackal NT, Burczynski ME, Harvey RG and Penning TM (2001) The ubiquitous aldehyde reductase (AKR1A1) oxidizes proximate carcinogen trans-dihydrodiols to o-quinones: potential role in polycyclic aromatic hydrocarbon activation. Biochemistry 40:10901-10910. Plebuch M, Soldan M, Hungerer C, Koch L and Maser E (2007) Increased resistance of tumor cells to daunorubicin after transfection of cDNAs coding for anthracycline inactivating enzymes. Cancer Lett 255:49-56. Rabbani A, Finn RM and Ausio J (2005) The anthracycline antibiotics: antitumor drugs that alter chromatin structure. Bioessays 27:50-56. Rajagopalan S, Politi PM, Sinha BK and Myers CE (1988) Adriamycin-induced free radical formation in the perfused rat heart: implications for cardiotoxicity. Cancer Res 48:4766-4769. Schlegel BP, Jez JM and Penning TM (1998) Mutagenesis of 3 alpha-hydroxysteroid dehydrogenase reveals a "push-pull" mechanism for proton transfer in aldo-keto reductases. Biochemistry 37:3538-3548. Singal PK and Iliskovic N (1998) Doxorubicin-induced cardiomyopathy. N Engl J Med 339:900-905. Singal PK, Li T, Kumar D, Danelisen I and Iliskovic N (2000) Adriamycin-induced heart failure: mechanism and modulation. Mol Cell Biochem 207:77-86. Takanashi S and Bachur NR (1976) Adriamycin metabolism in man. Evidence from urinary metabolites. Drug Metab Dispos 4:79-87. Tokarska-Schlattner M, Zaugg M, Zuppinger C, Wallimann T and Schlattner U (2006) New insights into doxorubicin-induced cardiotoxicity: the critical role of cellular energetics. J Mol Cell Cardiol 41:389-405. Wojtacki J, Lewicka-Nowak E and Lesniewski-Kmak K (2000) Anthracycline-induced cardiotoxicity: clinical course, risk factors, pathogenesis, detection and prevention--review of the literature. Med Sci Monit 6:411-420. 71  Wold LE, Aberle NS, 2nd and Ren J (2005) Doxorubicin induces cardiomyocyte dysfunction via a p38 MAP kinase-dependent oxidative stress mechanism. Cancer Detect Prev 29:294-299. Yen HC, Oberley TD, Vichitbandha S, Ho YS and St Clair DK (1996) The protective role of manganese superoxide dismutase against adriamycin-induced acute cardiac toxicity in transgenic mice. J Clin Invest 98:1253-1260.  72  CHAPTER 3 ALDO-KETO REDUCTASE 1C2 FAILS TO METABOLIZE DOXORUBICIN AND DAUNORUBICIN IN VITRO1  3.1.  Preface Though the anthracycline drugs have been proven to be effective cancer  chemotherapeutic drugs, their use is limited by the development of cardiotoxicity in some patients (Weiss, 1992). The mechanism by which this toxicity manifests is not well understood, however, there is good evidence that metabolism of the anthracycline drugs is involved. The major metabolites of DOX and DAUN that are formed in humans are DOXol and DAUNol, the alcohol metabolites formed by the reduction of their respective C-13 carbonyl functions (Takanashi and Bachur, 1976). The human enzymes that are identified as catalyzing this reductive metabolism are the AKR1A1, AKR1C2, and carbonyl reductase 1 and 3 proteins (Licata et al., 2000; Martin et al., 2006; Jin and Penning, 2007; Oppermann, 2007). The identification of the enzymes involved and their respective contributions to the total metabolism of the anthracyclines will be important for understanding how biotransformation may generate species that are damaging to the heart. This knowledge may also identify the best protein targets for pharmacological therapies that can slow or prevent the evolution of this side effect. In vitro analysis of individual enzymes as isolated recombinant proteins is one technique that provides estimates for kinetic parameters that describe a specific enzyme’s intrinsic abilities for metabolizing a  1  A version of this chapter has been published. Takahashi RH, Bains OS, Pfeifer TA, Grigliatti TA, Reid RE and Riggs KW (2008) Aldo-keto reductase 1C2 fails to metabolize doxorubicin and daunorubicin in vitro. Drug Metab Dispos 36:991-994. 73  substrate of interest (Crespi and Miller, 1999). Thus, the findings from these studies can be useful for determining the contribution of individual enzymes to the total metabolism of a chemical compound. The primary objective of this chapter is to characterize the in vitro function of AKR1C2 in the metabolism of DOX and DAUN. The data collected in these studies will provide a basis to estimate the contribution that this enzyme has to the total metabolism of these drugs. These data are important for allowing us and other researchers to focus research efforts on the relevant enzyme systems that are likely associated with the cardiotoxicity of anthracycline treatment.  3.2.  Materials and Methods  3.2.1. Chemicals and enzymes. 1-acenaphthenol,  p-nitrobenzaldehyde,  DL-glyceraldehyde,  DAUN,  DOX,  idarubicin, postassium phosphate, and NADPH were purchased from Sigma-Aldrich (St. Louis, MO). HPLC-grade acetonitrile and formic acid were purchased from Fisher Scientific (Fair Lawn, NJ). DNA modifying and restriction enzymes were obtained from Fermentas (Burlington, ON) and New England Biolabs, Inc. (Ipswich, MA).  An  authentic chemical standard for DOXol was obtained from Qventas Inc. (Branford, CT, USA).  3.2.2. Cloning of human AKR1A1 and 1C2. The AKR1A1 and AKR1C2 genes were sub-cloned from lab vector constructs (p2ZOp2N-AKR1A1 and p2ZOp2N-AKR1C2) using standard molecular cloning techniques to the prokaryotic expression vector pET28a (Novagen, Madison, WI) as 74  previously described (Bains et al., 2008). The expression constructs encoded an AKR with a six-histidine affinity tag separated by a 23-amino acid residue linker attached to the amino terminus of the expressed protein.  3.2.3. Expression of recombinant enzymes. The AKR proteins were expressed and purified from Escherichia coli cultures. Briefly, pET28a-AKR constructs were transformed by heat-shock into BL21 (DE3) pLysS and grown in 500 ml low-salt LB media supplemented with 50 $g/ml kanamycin sulfate and 25 $g/ml chloramphenicol at 37ºC. Expression was induced by addition of isopropyl "-D-1-thiogalactopyranoside administration to a final concentration of 1 mM and the cultures grown for a further 3 hours.  Bacterial pellets were collected by  centrifugation (4000 rpm for 20 mins at 4ºC), then resuspended at 5 ml per gram wet weight with 300 mM NaCl, 50 mM NaH2PO4, (pH 8.0). Lysis of the cell suspensions and purification of the AKR proteins using Ni-NTA agarose were carried out according to the manufacturer’s recommendations (QIAGEN, Mississauga, Ontario). Glycerol was added to the tagged purified protein to a final concentration of 20% and the samples stored at -20oC. The recombinant proteins were detected by Western blot analysis using the Odyssey Infrared Imaging System (LI-COR® Biosciences, Lincoln, NE) with the primary monoclonal AKR1A1 and AKR1C2 antibodies (Abnova Corporation, Taipei City, Taiwan) diluted 1:5000 and 1:3000, respectively, and secondary IRDye 800CW goat anti-mouse IgG antibody (LI-COR® Biosciences), diluted 1:5000. Protein concentrations were determined by the Bradford method using bovine serum albumin as a standard (BioRad, Hercules, CA). Protein purity was assessed by SDS-PAGE stained with Sypro Ruby (Invitrogen, Burlington, ON). 75  3.2.4. Measurement of AKR enzyme activities. AKR activity was measured by monitoring the initial rate of NADP(H) oxidation/reduction reactions using standard conditions (Penning, 2004). Fluorescence measurements of reduced cofactor were made with excitation and emission wavelengths of 355 and 460 nm, respectively, using a Fluoroskan Ascent FL (Thermo Fisher Scientific, Waltham, MA). The assays were carried out with 200-275 ng purified protein, cofactor (180 $M NADPH for AKR1A1; 2.3 mM NADP+ for AKR1C2) and 1 mM test substrate (p-nitrobenzaldehyde or DL-glyceraldehyde for AKR1A1; 1-acenaphthenol for AKR1C2) in a reaction mixture of 150 $l 100 mM potassium phosphate, pH 7, at 25ºC. Maximal rates (min-1) were calculated from the Ascent® program (version 2.6) using a 5 min interval. Enzymatic activities (µmol cofactor consumed/min/mg purified protein) were calculated based on a standard curve constructed from fluorescence measurements for known NADPH concentrations. Organic solvents at a concentration below 4% (v/v) in the final mixture were used to solubilize some substrates and were not observed to affect enzyme function.  All incubations were conducted in quadruplicate for each  purified protein preparation.  3.2.5. Measurement of anthracycline reduction. Identification and quantification of DOXol and DAUNol were performed using high-performance liquid chromatography (HPLC; Waters Alliance 2695, Milford, MA) and fluorescence detection with excitation and emission wavelengths of 460 and 550 nm, respectively (Waters 2475 Multi ! Detector). Quantification was based on peak area ratio of DOXol to the internal standard. Linear calibration was determined over the concentration range 25-500 nM using a 1/x2 weighting scheme. Because an authentic 76  chemical standard for DAUNol could not be obtained, DAUNol was quantified as DOXol equivalents using a response ratio of 1.0. The limit of quantification for DOXol was estimated as 25 nM. All data processes were performed with Waters Empower software (ver 2.0). Purified tagged AKR1A1 and AKR1C2 proteins were incubated in 150 µl reaction mixtures containing 25 mM KH2PO4, pH 7.4; and 1 µM of DOX or DAUN at 37ºC. Reaction conditions were selected to reflect a human physiological environment. Reactions were initiated with the addition of 1 mM NADPH.  Incubations were  conducted with up to 1 µg of purified protein for 120 and 240 minutes for DAUN and DOX, respectively, to maximize metabolite generation given the apparent low activity of these enzymes. At the end of the incubations, reductase activity was stopped with the addition of 150 µl of ice-cold acetonitrile that contained idarubicin as an internal standard. Protein was removed from the sample by vortex mixing and centrifugation at 10,000#g for 10 minutes at 4ºC, and the supernatant was removed for HPLC analysis. All incubations were conducted in triplicate for each purified protein preparation.  3.3.  Results and Discussion The expression of the recombinant tagged AKR1A1 and AKR1C2 was confirmed  by Western blot analysis showing bands with mobility corresponding to the molecular weight of the tagged AKRs (~41-42 kDa) (Figure 3.1). Total protein staining of a SDSPAGE gel demonstrated that the only detectable bands for the purified preparations corresponded to the AKR1A1 and AKR1C2 proteins identified by Western blotting (Figure 3.1).  77  Figure 3.1 Purification of recombinant histidine-tagged human (A) AKR1A1 and (B) AKR1C2. (Left) SDS-PAGE stained with Sypro Ruby shows purified protein samples, 1A1 or 1C2 (700 ng), free of contaminating proteins from bacterial lysates (L) (20 µg total protein). Removal of contaminating proteins is observed in fractions from Qiagen purification procedures (Ni-NTA column flow through (F), 15 µg total protein). (Right) Western blotting of purified protein samples, 1A1 or 1C2 (700 ng), confirms expression of the desired AKR protein with mobility at expected molecular weight (~41-42 kDa). Positive controls for antibody immunoreactivity are human liver cytosol (H) and GSTtagged purified human recombinant AKR enzymes (G; 700 ng). No antibody immunoreactivity is observed for untransformed bacterial lysate (U; 20 $g total protein).  78  Activities of the recombinant AKR1A1 using the test substrates pnitrobenzaldehyde and DL-glyeraldehyde were 5.39±0.35 and 1.56±0.16 µmol/min/mg purified protein, respectively, while the activity of the recombinant AKR1C2 using 1acenaphthenol was 2.23±0.08 µmol/min/mg. These activities are in good agreement with data reported for purified recombinant and native proteins by other laboratories: AKR1A1, p-nitrobenzaldehyde, 6.0 µmol/min/mg (Palackal et al., 2001); DLglyceraldehyde, 1.26 µmol/min/mg (O'Connor et al., 1999); AKR1C2, 1-acenaphthenol, 2.5 µmol/min/mg (Burczynski et al., 1998), providing confidence that the recombinant proteins have retained full reductase function. Formation of DAUNol and DOXol by AKR1A1 and absence of metabolite generation by AKR1C2 were measured by HPLC with fluorescence detection (Figure 3.2).  The C-13 metabolites were positively identified by identical chromatographic  retention times to an authentic chemical standard of DOXol and to the metabolites generated in incubations with human liver cytosolic fractions. Mordente et al. (2001) measured the formation of DAUNol and DOXol by human cardiac cytosol reductases with average rates of 27.5 and 2.3 pmol/min/mg protein, respectively. The high substrate specificity of AKR1A1 was also demonstrated in our results from recombinant enzymes where DAUN was metabolized to a much greater extent than DOX. The levels of DOXol generated were too low for quantification using the HPLC assay, whereas specific reductase activities of 1.71±0.09 pmol/min/mg were determined for DAUN reduction.  79  Figure 3.2 Generation of DAUNol and DOXol in vitro by purified recombinant AKR1A1 incubated with (A) DAUN and (B) DOX. Measurement of DAUNol and DOXol was performed using HPLC-fluorescence. Representative chromatograms show clear resolution of DAUNol and DOXol from DAUN, DOX, and idarubicin (internal standard). Retention times observed for DOXol, DOX, DAUNol, DAUN, and idarubicin (internal standard) are 4.5, 5.5, 6.0, 6.8, and 7.3 minutes, respectively.  Our findings of a lack of AKR1C2 involvement in DOX and DAUN metabolism differ from previously published results (Ohara et al., 1995). This discrepancy may be due to contaminating proteins in previous enzyme preparations or non-specific measurement of reductase activities.  SDS-PAGE gel staining for our enzyme  preparations shows the absence of any detectable contaminating proteins following 80  affinity chromatography. Sypro Ruby total protein staining provides detection of proteins at 1-2 ng (Berggren et al., 2000) and therefore we estimate protein contamination in our preparations at less than 0.01%. Monitoring NADPH absorbance is the standard assay to characterize reductase enzyme function; however, this method cannot distinguish contaminating protein-catalyzed or non-enzyme catalyzed cofactor usage, potentially leading to erroneous identification of reductase function or substrate use. In our attempts to use the spectroscopic method to measure cofactor usage in AKR1C2-catalyzed DOX and DAUN metabolism, we could not distinguish enzyme and non-enzyme catalyzed signal decrease with substrate concentrations of 10, 50, and 250 µM (data not shown). To specifically measure AKR activities for the anthracycline drugs DOX and DAUN, we have used a direct and sensitive HPLC-fluorescence assay and do not identify any alcohol metabolites formed by AKR1C2 that exceed the limit of quantification of the assay (25 nM).  In our in vitro system, the initial anthracycline concentrations are 1 µM, a  physiologically-relevant concentration for patient cardiac tissue (Stewart et al., 1993). The failure of AKR1C2 to generate alcohol metabolites in vitro suggests that it is unimportant for the generation of DOX or DAUN metabolites in vivo. We recognize that in vivo studies are needed to clearly define the role of AKR1C2 in anthracycline metabolism; however, the conduct of definitive studies will be difficult.  Several  reductase enzymes are involved in the metabolism of these drugs and distinguishing their individual contributions to the total metabolism will be a challenge using the traditional approaches of correlation studies and chemical or antibody inhibition. It is with these considerations that we present our in vitro findings to focus future work on anthracycline metabolism on the other AKR isoforms and carbonyl reductases.  81  3.4.  References  Bains OS, Takahashi RH, Pfeifer TA, Grigliatti TA, Reid RE and Riggs KW (2008) Two allelic variants of aldo-keto reductase 1A1 exhibit reduced in vitro metabolism of daunorubicin. Drug Metab Dispos 36:904-910. Berggren K, Chernokalskaya E, Steinberg TH, Kemper C, Lopez MF, Diwu Z, Haugland RP and Patton WF (2000) Background-free, high sensitivity staining of proteins in one- and two-dimensional sodium dodecyl sulfate-polyacrylamide gels using a luminescent ruthenium complex. Electrophoresis 21:2509-2521. Burczynski ME, Harvey RG and Penning TM (1998) Expression and characterization of four recombinant human dihydrodiol dehydrogenase isoforms: oxidation of trans7, 8-dihydroxy-7,8-dihydrobenzo[a]pyrene to the activated o-quinone metabolite benzo[a]pyrene-7,8-dione. Biochemistry 37:6781-6790. Crespi CL and Miller VP (1999) The use of heterologously expressed drug metabolizing enzymes--state of the art and prospects for the future. Pharmacol Ther 84:121131. Jin Y and Penning TM (2007) Aldo-keto reductases and bioactivation/detoxication. Annu Rev Pharmacol Toxicol 47:263-292. Licata S, Saponiero A, Mordente A and Minotti G (2000) Doxorubicin metabolism and toxicity in human myocardium: role of cytoplasmic deglycosidation and carbonyl reduction. Chem Res Toxicol 13:414-420. Martin HJ, Breyer-Pfaff U, Wsol V, Venz S, Block S and Maser E (2006) Purification and characterization of akr1b10 from human liver: role in carbonyl reduction of xenobiotics. Drug Metab Dispos 34:464-470. O'Connor T, Ireland LS, Harrison DJ and Hayes JD (1999) Major differences exist in the function and tissue-specific expression of human aflatoxin B1 aldehyde reductase and the principal human aldo-keto reductase AKR1 family members. Biochem J 343 Pt 2:487-504. Ohara H, Miyabe Y, Deyashiki Y, Matsuura K and Hara A (1995) Reduction of drug ketones by dihydrodiol dehydrogenases, carbonyl reductase and aldehyde reductase of human liver. Biochem Pharmacol 50:221-227. Oppermann U (2007) Carbonyl reductases: the complex relationships of mammalian carbonyl- and quinone-reducing enzymes and their role in physiology. Annu Rev Pharmacol Toxicol 47:293-322. Palackal NT, Burczynski ME, Harvey RG and Penning TM (2001) The ubiquitous aldehyde reductase (AKR1A1) oxidizes proximate carcinogen trans-dihydrodiols to o-quinones: potential role in polycyclic aromatic hydrocarbon activation. Biochemistry 40:10901-10910.  82  Penning TM (2004) Aldo-keto reductases and formation of polycyclic aromatic hydrocarbon o-quinones. Methods Enzymol 378:31-67. Stewart DJ, Grewaal D, Green RM, Mikhael N, Goel R, Montpetit VA and Redmond MD (1993) Concentrations of doxorubicin and its metabolites in human autopsy heart and other tissues. Anticancer Res 13:1945-1952. Takanashi S and Bachur NR (1976) Adriamycin metabolism in man. Evidence from urinary metabolites. Drug Metab Dispos 4:79-87. Weiss RB (1992) The anthracyclines: will we ever find a better doxorubicin? Semin Oncol 19:670-686.  83  CHAPTER 4 QUANTITATION OF DHT AND ITS REDUCTION METABOLITES BY LC/MS/MS1  4.1.  Preface AKR1C2 is an important human metabolic enzyme responsible for the reduction  of DHT to 3!-diol in the prostate (Rizner et al., 2003b). This metabolic pathway is critical for deactivating DHT, the most potent natural ligand for the androgen receptor and a necessary compound for normal growth of the prostate gland. Hence, AKR1C2 is predicted to be an important pre-receptor regulator of androgen signaling by dictating tissue levels of DHT with a potential role in the development and/or progression of prostate cancer (Ji et al., 2007; Penning et al., 2008). The study of AKR1C2 function in the deactivation of DHT requires a methodology for the accurate and reliable quantitation of 3!-diol formation. Conventional  methods  chromatography-mass  of  immunoassay,  spectrometry  require  thin-layer intensive  chromatography, sample  or  gas  preparation  and  chromatography and suffer from the limitations of poor specificity and accuracy due to interference from other steroids and lipids (Salerno et al., 1988; Taieb et al., 2003; Wang et al., 2008). Liquid chromatography-tandem mass spectrometry (LC/MS/MS) is an attractive alternative analytical technology that provides high levels of sensitivity and selectivity. In addition, it is readily available in today’s research laboratories and is  1  A version of this chapter will be submitted for publication. Takahashi RH, Grigliatti TA, Reid RE and Riggs KW. Analysis of Dihydrotesterone and Its Reduction Metabolites by LC/MS/MS for the Study of Aldo-Keto Reductase 1C2 Activities. 84  commonly the standard instrumentation for drug, food, environmental, and clinical sample testing. However, the use of LC/MS/MS for androgen analysis is not without challenge.  The steroid molecule is known to be poorly ionized and therefore, less  amenable to achieving low levels of detection needed for biological samples (Quirke et al., 1994; Higashi and Shimada, 2004). This difficulty has been overcome by a number of strategies, such as the use of various derivatization reagents, a variety of extraction, concentration, and chromatographic techniques, and utilizing positive and negative ion monitoring for precursor, multiple reaction, or cluster ions. These method development efforts have resulted in a number of published LC/MS/MS methods for testosterone and DHT analysis (Zhao et al., 2004; Higashi et al., 2005; Kalhorn et al., 2007; Licea-Perez et al., 2008; Wang et al., 2008). To date, there are only a few reports providing protocols that measure 3!-diol, an important DHT metabolite (Zhao et al., 2004; Reddy, 2008). This chapter describes a procedure for the measurement of DHT and its two stereoisomeric reduction products 3!-diol and 3"-diol in in vitro samples using LC/MS/MS. Since a suitable method was not found in the literature, a novel method was developed and tested for its range, specificity, accuracy, lower limits of quantitation, and precision (intra- and inter-day variability). The results reported herein demonstrate the suitability of this method for the in vitro study of AKR1C2 reduction of DHT. This method has been subsequently used for the characterization of AKR1C2 function and the effect of allelic variation on the enzyme activities (described in Chapters 5 and 6).  85  4.2.  Materials and Methods  4.2.1. Chemicals Chemical standards for DHT, 3!-diol, 3"-diol, and 16!-bromoandrosterone were purchased from Steraloids, Inc. (Newport, RI).  Formic acid, 2-fluoro-1-methyl  pyridinium p-toluenesulfonate (FMPTS), and triethylamine were purchased from Sigma Chemical Co (St Louis, MO). Acetonitrile, dichloromethane, methanol, and methyl tertbutyl ether, all HPLC grade or equivalent, were purchased from Fisher Scientific (Toronto, ON). Deionized, distilled water was obtained from a Milli-Q filtration system (Millipore, Bedford, MA).  4.2.2. Extraction and derivatization of DHT and metabolites The samples that were analyzed were in vitro incubation reactions prepared for studying the kinetic activities of AKR1C2 in the reduction of DHT. These samples consisted of either insect cell culture media or 100 mM sodium phosphate buffer containing transfected insect cells or purified recombinant protein as the source of AKR1C2, respectively. The total volume of the samples was 150 µl and they contained DHT, 3!-diol and 3"-diol.  An equivalent volume of ice-cold acetonitrile or 1 N  hydrochloric acid was added to terminate the reaction, resulting in a total sample volume of 300 µl.  The solutions also contained 0.5 µg/ml of the internal standard 16!-  bromoandrosterone. The androgens were extracted from samples by adding 1 ml of methyl tert-butyl ether containing 0.1% (by volume) triethylamine and mixing by rotation for 10 minutes, followed by centrifuging for 5 minutes (>13,000 rpm). The upper  86  organic phase was transferred to tubes and evaporated to dryness under a gentle stream of nitrogen at 35ºC. The residue was redissolved in 50 µl dichloromethane and 100 µl of 0.01 M triethylamine in dichloromethane and 100 µl of 0.01 M FMPTS in dichloromethane were added. The tubes were capped and the derivatization reaction was carried out for 30 minutes at 28ºC with gentle shaking (125 rpm).  Solvent was  evaporated under a gentle stream of nitrogen at 35ºC and the residue redissolved in 150 µl of methanol-water (1:1). A 15 µl aliquot was injected onto the LC column.  4.2.3. Chromatographic separation and mass spectrometric analysis of derivatized DHT and metabolites Chromatography of the methyl pyridinium derivatives was performed using a Waters Acquity system (Milford, MA). Separation was achieved using a Waters Acquity BEH C18 column (100 mm # 2.1 mm i.d.) with 1.7 µm particle size packing, maintained at a constant temperature of 30ºC. Gradient elution using mobile phases of (A) 0.1 % formic acid in water and (B) 0.1% formic acid in methanol was performed according to the following program: 0 min, 35% B; 3.5 min, 70% B; 3.6 min, 98% B; 5.5 min, 98% B; 5.6 min 35% B; at a constant flow rate of 0.2 ml/min. Flow from the column was diverted to waste for the first 1.5 minutes and at 4.0 minutes following injection for each run. The total run-time for each injection was 7.5 minutes. The column effluent was monitored using a Waters Premiere tandem-quadrupole mass spectrometric detector (Milford, MA) equipped with an electrospray interface and computer control using MassLynx version 4.1 software (Waters). Positive electrospray ionization was performed with a nebulizer temperature of 350ºC using a cone voltage set at 20 V. Multiple reaction  87  monitoring of three ion transitions was used, with the mass transitions m/z 382.2!254.9 for DHT (collision energy (CE)=20 V); 384.1!257.1 for 3!-diol and 3"-diol (CE=20 V); and 460.5!109.5 for 16!-bromoandrosterone (CE=20 V). All mass spectrometric data processes including peak smoothing and integration, calibration, and calculation of unknown concentrations were performed using Waters MassLynx software version 4.1 (Waters).  4.2.4. Assay validation The characteristics of the analytical method were assessed by testing the range, specificity, accuracy, precision, and the lower limit of quantitation (LOQ) using fortified samples that contained the in vitro incubation buffers and a known concentration of the analytes. The range of quantitation was assessed by preparing eight calibration standards, including a zero standard, for each batch. The calibration standards contained DHT, 3!diol, and 3"-diol at the following eight concentrations (listed as DHT/3!-diol/3"-diol): 0/0/0, 15/3/3, 30/6/6, 75/15/15, 250/50/50, 500/100/100, 1000/200/200, 1500/300/300 ng/ml. Specificity was assessed by performing the sample preparation procedures on three separate aliquots of buffered media without the addition of the internal standard. Accuracy was assessed as the % bias (from predicted concentration) of six separately fortified quality control (QC) samples for each batch. Low, medium, and high QC levels represented 3-times the LOQ, mid-range, and 80% of the upper limit of the calibration range, respectively. Precision was assessed as the % coefficient of variation (CV) for six aliquots of a pooled QC sample that were individually prepared and analyzed for each batch. The LOQ was determined in the same manner as accuracy and precision at a  88  single concentration level equivalent to the lowest calibration standard. Range, accuracy, precision, and LOQ were assessed using three separate batches of samples prepared and analyzed on three consecutive days. The specificity was judged in a single experiment conducted on a single day.  4.3.  Results Methyl pyridinium derivatives were successfully formed for DHT, 3!-diol, 3"-  androstanediol, and 16!-bromoandrosterone (Figure 4.1). These derivatives represent the substrate of interest for AKR1C2, the two isomeric alcohol products that are formed through reductive metabolism, and a synthetic androgen selected as an internal standard for quantitation purposes. The observed retention times for these analytes were 2.79, 2.52, 3.51, and 3.00 minutes, respectively. Multiple reaction monitoring provided clear mass resolution of the compounds. For 3!-diol and 3"-diol, where no mass difference exists, the two analytes were clearly distinguished by their chromatographic separation (Figure 4.2).  89  Figure 4.1 Scheme for reactions of DHT, 3!/"-diol, and 16!-bromoandrosterone (top to bottom) with 2-fluoro-1-methyl pyridinium p-toluenesulfonate to form stably charged fluoro methyl pyridinium ether derivatives for mass spectrometric detection. Derivatives of 3!/"-diol can occur through reactions of 3- or 17-position hydroxyl functional groups – both are shown.  90  Figure 4.2 Representative LC/MS/MS chromatogram for fluoro methyl pyridinium derivatives of DHT, 3!/"-diol, and 16!-bromoandrosterone. Peaks monitored for quantitative purposes are labeled with retention times.  To validate the suitability of the method for assessing the in vitro reduction of DHT by AKR1C2, the range, accuracy, intra-day and inter-day precision, and specificity of the method were examined with fortified samples. The linear range of the method was determined to be 15-1500 ng/ml for DHT, and 3-300 ng/ml for 3!-diol and 3"-diol with all eight calibration standards for each batch meeting an acceptance criteria of calculated concentrations of less than 10% bias from predicted. The best-fit calibration line was determined using weighted (1/y2) linear regression without inclusion of the zero standard, which showed no signal at the expected retention time. Accuracy and precision met the  91  suitability criteria of less than 15% CV for calculated concentrations and less than 15% bias from expected concentration with six replicate samples prepared at three QC levels (low, medium, high) within the range of the method (Table 4.1). The lowest limit of quantitation was judged at the lowest point of the calibration range that met the suitability criteria for QC levels, and was estimated to be 15 ng/ml for DHT, and 3 ng/ml for 3!diol and 3"-diol (Table 4.1). No interfering signals were determined in blank in vitro samples. The criteria used to assess the suitability of the method were based on the best practices developed for bioanalytical method validation that fulfill the current regulatory guidance (Shah et al., 2000; Nowatzke and Woolf, 2007).  92  Table 4.1 Accuracy and precision, intra-day and inter-day, for the LC/MS/MS determinations of DHT, 3!-diol, and 3"-diol in in vitro samples. Acceptance criteria for the method was set at less than 15% CV (precision) and less than 15% bias at all QC levels. Each determination was based on six replicate samples as described in the Methods section. Batch 1  Batch 2  Batch 3  Analyte  QC level  Conc. (ng/ml)  Intra-day precision (% CV)  Accuracy (% bias)  Intra-day precision (% CV)  Accuracy (% bias)  Intra-day precision (% CV)  Accuracy (% bias)  Inter-day precision (% CV)  DHT  LOQ  15  8.19  1.04  6.91  10.44  7.83  6.66  4.84  low  45  3.18  -6.45  5.92  3.64  3.30  -13.3  10.2  med  400  2.88  3.07  8.47  2.19  1.58  -3.23  8.57  high  1200  2.60  4.83  5.48  6.33  8.85  -0.44  7.46  LOQ  3  6.48  1.04  6.99  3.22  7.78  14.4  6.23  low  9  6.52  -9.63  5.80  -0.47  14.0  -14.6  2.53  med  80  5.74  2.65  3.83  6.64  14.9  -0.18  3.04  high  240  2.44  7.50  1.81  10.7  14.8  9.85  9.34  LOQ  3  9.43  7.45  14.7  4.06  11.2  7.41  6.62  low  9  8.10  -6.01  7.87  1.44  9.04  -6.68  7.86  med  80  3.22  3.01  8.15  4.76  6.61  1.12  7.00  high  240  2.63  5.00  1.49  10.1  11.0  7.38  5.45  3!-diol  3"-diol  93  4.4.  Discussion We report the use of a simple derivatization procedure to form stably charged  methyl pyridinium derivatives for mass spectrometric detection. The extraction and derivatization steps are convenient, one-step procedures that use commercially available reagents. In addition, they are easily scaled up and this allows for processing of a number of samples in parallel on a daily basis, which may be expected from in vitro studies. The method was found to be specific, accurate, precise, and of appropriate range to assess the kinetics of the AKR1C2-catalyzed reduction of DHT to 3!-diol. There are two potential stereoisomeric products that can be generated through reduction of DHT, 3!- and 3"-diol. Although AKR1C2 is stereospecific in its reduction of DHT, producing 3!-diol, we analyzed and separately quantified the two metabolites to ensure the method was sufficiently specific to accurately quantitate each isomer independently. The two diol analytes can react with the derivatization agent at the 3- and 17-hydroxyl positions, forming four unique singly-derivatized chemical species, which share a common molecular mass, and two doubly-derivatized chemical species, which share a common molecular mass. Mass spectral scan analyses were used to detect the formation of the doubly-derivatized species for either 3!- or 3"-diol; however, it was not found. This result was interpreted as demonstrating that there is low occurrence of second derivatization reaction, hypothetically due to charge-charge repulsion or loss of steroid reactivity after the first ether formation. Mass spectral signals were observed for the four singly-derivatized diol species with both a major and a minor peak for 3!- and 3"-diol after the derivatization reactions (Figure 4.2). These signals were adequately separated chromatographically and acceptable quantitative accuracy and precision were  94  achieved when only the major peak was monitored and used for quantifying 3!- or 3"diol. A major inherent difficulty in the mass spectrometric analysis of steroids and their metabolites is inefficient ion generation. For this reason, a variety of derivatization reagents have been used to enhance the detection of steroids in liquid chromatographymass spectrometry applications (Higashi and Shimada, 2004). In particular, the use of oxime,  2,3-pyridinedicarboxylic  anhydride,  and  2-hydrazino-1-methylpyridine  derivatives have allowed the quantitative analysis of DHT (Higashi et al., 2005; Kalhorn et al., 2007; Licea-Perez et al., 2008).  We have successfully reacted the hydroxyl  functionalities of DHT, 3!-diol, 3"-diol, and 16!-bromoandrosterone to form methyl pyridinium ether derivatives to enhance their ionization and facilitate their mass spectrometric measurement (Figure 4.1). The formation of these ether derivatives has previously been reported for the analysis of testosterone and DHT (Kashiwagi et al., 2005). The method we report here extends the utility of this derivatization reaction to the reduction products of DHT and provides a simpler sample preparation protocol. Based on the report by Kashiwagi et al., the simultaneous analysis of testosterone or other androgens, and the analysis of androgen levels of biological fluids should be possible with minimal modifications of the current method. As the pharmacological importance of 3!-diol is further elucidated, there will be a growing need for simple and specific methods for measuring this metabolite. Currently, few validated methods are reported that include 3!-diol as an analyte. A report by Zhao et al demonstrates excellent performance in simultaneously measuring DHT and 3!-diol, and other steroids, in small volume biological samples (Zhao et al., 2004). Their method measures surprisingly low levels of DHT and 3!-diol, 0.02 and 0.2 ng/ml, respectively, 95  without any specified techniques for overcoming the well-noted difficulties in ionizing steroids (Higashi and Shimada, 2004). We were unable to attain the reported detection limits in our laboratory without performing a derivatization step. A method by Reddy was developed specifically to measure 3!-diol and shows good sensitivity and specificity (Reddy, 2008). This method was successfully used to demonstrate increases in plasma 3!-diol levels following testosterone treatment in rats.  Our method provides the  advantages of the simultaneous measurement of DHT and the stereoisomer 3"-diol, which are required for the application of the method to studying the reductive metabolism of DHT to 3!-diol.  However, the sensitivity and range for 3!-diol of the method  described here or the method by Reddy is not suitable for measurement of physiological levels, especially for central nervous tissues where neurosteroids are present at very low concentration (e.g., low nanomolar) (Schumacher et al., 2003). Modifications that could increase the sensitivity of the method are extracting a larger sample, redissolving the sample in a smaller volume to increase its concentration, monitoring and summing additional ion fragmentation reactions, or modifying the mobile phases to stabilize charged species at the ion source. In summary, we have developed a novel method using LC/MS/MS for the simultaneous quantitation of DHT and its two reduction products, 3!- and 3"-diol. The method shows acceptable accuracy, precision, specificity, LOQ, and range without difficult or intensive sample preparation, and was deemed suitable for studying the ability of AKR1C2 to reduce DHT to 3!-diol in vitro.  96  4.5.  References  Higashi T and Shimada K (2004) Derivatization of neutral steroids to enhance their detection characteristics in liquid chromatography-mass spectrometry. Anal Bioanal Chem 378:875-882. Higashi T, Yamauchi A, Shimada K, Koh E, Mizokami A and Namiki M (2005) Determination of prostatic androgens in 10 mg of tissue using liquid chromatography-tandem mass spectrometry with charged derivatization. Anal Bioanal Chem 382:1035-1043. Ji Q, Chang L, Stanczyk FZ, Ookhtens M, Sherrod A and Stolz A (2007) Impaired dihydrotestosterone catabolism in human prostate cancer: critical role of AKR1C2 as a pre-receptor regulator of androgen receptor signaling. Cancer Res 67:13611369. Kalhorn TF, Page ST, Howald WN, Mostaghel EA and Nelson PS (2007) Analysis of testosterone and dihydrotestosterone from biological fluids as the oxime derivatives using high-performance liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 21:3200-3206. Kashiwagi B, Shibata Y, Ono Y, Suzuki R, Honma S and Suzuki K (2005) Changes in testosterone and dihydrotestosterone levels in male rat accessory sex organs, serum, and seminal fluid after castration: establishment of a new highly sensitive simultaneous androgen measurement method. J Androl 26:586-591. Licea-Perez H, Wang S, Szapacs ME and Yang E (2008) Development of a highly sensitive and selective UPLC/MS/MS method for the simultaneous determination of testosterone and 5alpha-dihydrotestosterone in human serum to support testosterone replacement therapy for hypogonadism. Steroids 73:601-610. Nowatzke W and Woolf E (2007) Best practices during bioanalytical method validation for the characterization of assay reagents and the evaluation of analyte stability in assay standards, quality controls, and study samples. Aaps J 9:E117-122. Penning TM, Jin Y, Rizner TL and Bauman DR (2008) Pre-receptor regulation of the androgen receptor. Mol Cell Endocrinol 281:1-8. Quirke JME, Adams CL and Van Berkel GJ (1994) Chemical derivatization for electrospray ionization mass spectrometry. 1. alkyl halides, alcohols, phenols, thiols, and amines. Analytical Chemistry 66:1302-1315. Reddy DS (2008) Mass spectrometric assay and physiological-pharmacological activity of androgenic neurosteroids. Neurochem Int 52:541-553. Rizner TL, Lin HK and Penning TM (2003) Role of human type 3 3alpha-hydroxysteroid dehydrogenase (AKR1C2) in androgen metabolism of prostate cancer cells. Chem Biol Interact 143-144:401-409.  97  Salerno R, Moneti G, Forti G, Magini A, Natali A, Saltutti C, Di Cello V, Costantini A and Serio M (1988) Simultaneous determination of testosterone, dihydrotestosterone and 5 alpha-androstan-3 alpha,-17 beta-diol by isotopic dilution mass spectrometry in plasma and prostatic tissue of patients affected by benign prostatic hyperplasia. Effects of 3-month treatment with a GnRH analog. J Androl 9:234-240. Schumacher M, Weill-Engerer S, Liere P, Robert F, Franklin RJ, Garcia-Segura LM, Lambert JJ, Mayo W, Melcangi RC, Parducz A, Suter U, Carelli C, Baulieu EE and Akwa Y (2003) Steroid hormones and neurosteroids in normal and pathological aging of the nervous system. Prog Neurobiol 71:3-29. Shah VP, Midha KK, Findlay JW, Hill HM, Hulse JD, McGilveray IJ, McKay G, Miller KJ, Patnaik RN, Powell ML, Tonelli A, Viswanathan CT and Yacobi A (2000) Bioanalytical method validation--a revisit with a decade of progress. Pharm Res 17:1551-1557. Taieb J, Mathian B, Millot F, Patricot MC, Mathieu E, Queyrel N, Lacroix I, SommaDelpero C and Boudou P (2003) Testosterone measured by 10 immunoassays and by isotope-dilution gas chromatography-mass spectrometry in sera from 116 men, women, and children. Clin Chem 49:1381-1395. Wang C, Shiraishi S, Leung A, Baravarian S, Hull L, Goh V, Lee PW and Swerdloff RS (2008) Validation of a testosterone and dihydrotestosterone liquid chromatography tandem mass spectrometry assay: Interference and comparison with established methods. Steroids 73:1345-1352. Zhao M, Baker SD, Yan X, Zhao Y, Wright WW, Zirkin BR and Jarow JP (2004) Simultaneous determination of steroid composition of human testicular fluid using liquid chromatography tandem mass spectrometry. Steroids 69:721-726.  98  CHAPTER 5 ALLOSTERIC ACTIVATION OF AKR1C2 REDUCTION OF DIHYDROTESTOSTERONE BY ANTIDEPRESSANT DRUGS AND ITS IN VIVO SIGNIFICANCE1  5.1.  Preface The function of systems that regulate the extent, timing, and duration of androgen  receptor activation is important for normal prostate growth and, potentially, may determine the risk of developing prostate cancer (Chu et al., 2008). One such system is comprised of the enzymes that catalyze the metabolic pathways of androgen synthesis and degradation, which will dictate the levels of potent androgens in the prostate (Mostaghel and Nelson, 2008). Some of today’s most effective clinical treatments for prostate cancer impact the metabolic pathways that synthesize androgens, particularly targeting the biogenesis of DHT from testosterone within the prostate (Suzuki et al., 2007; Mostaghel and Nelson, 2008). The successes of these therapies demonstrate the importance of the DHT biosynthetic pathways for cancer development. In a similar manner, the degradative pathways for DHT are likely of great importance, and there is evidence that impairment of these pathways result in elevated DHT concentrations and abnormal prostate tissue proliferation (Morimoto et al., 1980). The elucidation of these metabolic systems is complicated by the variety of steroid metabolizing enzymes expressed in the prostate and the number of potential substrate and product androgens 1  A version of this chapter will be submitted for publication. Takahashi RH, Grigliatti TA, Reid RE, and Riggs KW. Allosteric Activation of Aldo-Keto Reductase 1C2 Reduction of Dihydrotestosterone by Antidepressant Drugs.  99  (Luu-The et al., 2008). For example, there are several AKR enzymes that function to varying degrees as 3!-, 17"-, and 20!-hydroxysteroid oxidoreductases with overlapping substrate preferences (Penning et al., 2000). A variety of metabolites have been identified that result from the deactivating metabolism of DHT including 3!-diol, 3"-diol, and 5!-androstanedione. In addition, DHT and these metabolites are substrates for the UDP-glucuronysyl transferase enzymes, which can conjugate these steroids for their deactivation and elimination from prostate tissue (Chouinard et al., 2008). Of these deactivation/elimination pathways, it appears that the generation of 3!-diol occurs to the greatest extent and represents a highly important route to terminating androgen receptor signaling in the prostate (Negri-Cesi and Motta, 1994; Jin and Penning, 2001). Based on its high expression in the prostate and its activity as a 3-ketosteroid reductase, AKR1C2 has been identified as the major reductive enzyme responsible for catalyzing the biotransformation of DHT to 3!-diol (Rizner et al., 2003b). Steroid metabolism is also becoming recognized for synthesizing neurosteroids that act in the central nervous system where they play key roles for determining complex behaviours such as anxiety disorders (Melcangi et al., 2008).  For example,  allopregnanolone, a progesterone metabolite, has anxiolytic and anticonvulsant actions through positive allosteric modulation of the GABAA receptor. It was shown with in vitro studies that levels of allopregananolone generated by the reduction of 5!dihydroprosterone (DHP) were increased by fluoxetine through activation of the enzyme catalyzing this metabolism, AKR1C2 (Griffin and Mellon, 1999). A second report does not agree, as an effect by fluoxetine to increase allopregnanolone generation was not  100  observed (Trauger et al., 2002). It remains unclear after these two conflicting studies whether or not AKR1C2 is activated by fluoxetine or other antidepressant drugs. The major objectives of the studies described in this chapter are to determine the kinetic parameters that describe the reduction of DHT to 3!-diol by AKR1C2 in vitro and to characterize the effect of the antidepressant drugs, specifically, fluoxetine, imipramine, paroxetine, and sertraline (Figure 5.1), on these kinetic parameters.  Based on the  previous findings of the anti-depressant activation of AKR1C2 in the reductive metabolism of DHP, we hypothesized that similar activation would be observed for DHT reduction. This modulation represents a factor that could alter the function of AKR1C2 and thereby, have important effects on the metabolism of steroids in vivo.  Figure 5.1 Compounds tested for AKR1C2 activation.  101  5.2.  Materials and Methods  5.2.1. Chemicals DHT, 3!-diol, 3"-diol, and 16!-bromoandrosterone were purchased from Steraloids, Inc. (Newport, RI).  A chemical reference standard for fluoxetine was  obtained from Eli Lilly & Co. (Indianapolis, IN). Fluoxetine, imipramine, paroxetine, and sertraline were obtained in tablet forms in the following formulations and suppliers: generic (PMS Pharmascience), generic (PMS Pharmascience), Paxil (SmithKline Beecham), and Zoloft (Pfizer), respectively. Formic acid, 2-fluoro-1-methyl pyridinium p-toluenesulfonate (FMPTS), NADPH, sodium phosphate, and triethylamine were purchased from Sigma Chemical Co (St Louis, MO). Acetonitrile, dichloromethane, methanol, and methyl tert-butyl ether, all HPLC grade or equivalent, were purchased from Fisher Scientific (Toronto, ON). Deionized, distilled water was obtained from a Milli-Q filtration system (Millipore, Bedford, MA).  5.2.2. AKR1C2 protein AKR1C2 protein was expressed and isolated from Escherichia coli cultures as previously reported (Takahashi et al., 2008).  Purification of the amino terminus  histidine-tagged protein was performed using Ni-NTA agarose and carried out according to the manufacturer’s recommendations (Qiagen, Mississauga, ON). Glycerol was added to the purified protein to a final concentration of 20% and aliquots were stored at -20ºC until used. Protein concentrations were determined by the Bradford method using bovine serum albumin as a standard (BioRad, Hercules, CA).  102  5.2.3. In vitro reduction of DHT by AKR1C2 Reaction mixtures contained purified AKR1C2 protein (300 ng), DHT, and NADPH (2 mM) in a final volume of 150 µl of 100 mM sodium phosphate buffer, pH 7.4. DHT was dissolved in methanol such that the final concentration of methanol in the incubation mixture was 1% by volume.  Reactions were started by the addition of  NADPH, 2 mM, to the mixture after 25 minutes of pre-incubation at 37ºC with gentle shaking (100 rpm). After 30 minutes at the same incubation conditions, reactions were stopped by the addition of 150 µl of ice-cold acetonitrile containing 16!bromoandrosterone (0.5 µg/ml) as an internal standard.  The in vitro samples were  prepared and analyzed for DHT and 3!-diol using LC/MS/MS according to the procedures described in Chapter 4. For each batch of samples, fresh calibration standards and QCs at low, medium, and high concentrations were prepared and analyzed. All mass spectrometric data processes including peak smoothing and integration, calibration, and calculation of unknown concentrations were performed using Waters MassLynx software version 4.1 (Waters).  5.2.4. Kinetics of DHT reduction Eight concentrations of DHT, ranging from 0.2 to 10 µM, were used to analyze the kinetics of the AKR1C2 protein. Three separate aliquots of protein were incubated at each DHT concentration level. An allosteric sigmoidal model described by the following equation was fit to the data:  103  Vmax ! [S]h v= K 'h + [S]h  Where Vmax is the maximal velocity, [S] is the substrate concentration, K' is the substrate concentration at half maximal velocity, and h is the parameter that describes the sigmoidicity from a typical Michaelis-Menten expression. Curve-fitting was performed using non-linear least squares regression (Graphpad Prism version 5, San Diego, CA).  5.2.5. Effect of modulators on AKR1C2 reduction of DHT Fluoxetine, imipramine, paroxetine, and sertraline tablets were crushed and dissolved in methanol and the insoluble materials were removed by centrifugation. Kinetic analysis of 3!-diol formation from DHT by AKR1C2 was performed in the presence of 50 µM fluoxetine, imipramine, paroxetine, and sertraline. Each modulator compound was added separately to the incubation for the 25-minute pre-incubation period. Stock solutions of the modulators used methanol as a solvent, but were diluted to keep the final incubation mixture at less than 1% organic solvent, which was observed to not affect enzyme activities. For each modulator compound, three separate incubations were performed at each DHT concentration. Modeling to estimate kinetic parameters, Vmax, K', and H, was performed in the same manner as for data obtained in absence of any modulating chemicals. For the compounds that showed modulating effects at 50 µM, enzyme activities were determined using a range of modulator concentrations of 0.4 to 65 µM with the DHT concentration constant at 5 µM. Activities were determined as percentages of the  104  activities measured in the absence of modulator compound and dose-response curves were constructed for each modulator compound.  5.2.6. Effect of bovine serum albumin on AKR1C2 reduction of DHT The effect of bovine serum albumin (BSA) on DHT reduction was performed with DHT concentration constant at 5 µM and in presence and absence of modulator at 50 µM. BSA at final concentrations of 0.001 to 1 mg/ml was added to the incubation mixtures for the 25-minute pre-incubation period. Incubations and sample preparation were performed as described above. The activities were calculated as percentages of the activities measured in absence of BSA and in absence of modulator. Differences between activities in presence of modulator and no modulator at each BSA concentration were tested by one-way ANOVA with Tukey post-hoc analysis with a threshold of p<0.05 for significance.  5.3.  Results Recombinant AKR1C2 was successfully expressed in E coli and isolated as a  histidine-tagged (amino terminus) protein by affinity chromatography as previously shown (Takahashi et al., 2008).  In incubations with DHT, 3!-diol was the major  metabolite generated by AKR1C2 with only low levels of 3"-diol observed in the in vitro samples. The substrate versus reduction velocity plot and corresponding Eadie-Hofstee plot for formation of 3!-diol by AKR1C2 are shown as Figure 5.2. The Eadie-Hofstee plots had a clear convex curvature. The kinetics of the reaction were sigmoidal and consistent with substrate activation, therefore modeling was performed with the Hill  105  equation, which provided a good fit to the data. This allowed the parameters of K', h, and Vmax to be derived. The K' and Vmax data in absence of modulator compound were 2.26 µM and 0.948 nmol/min/mg protein, respectively. These values are consistent with the Km and Vmax that have been previously reported by other investigators (Griffin and Mellon, 1999; Dufort et al., 2001; Jin and Penning, 2006).  Figure 5.2 Rate of DHT reduction to 3!-diol by AKR1C2 as measured by LC/MS/MS. The Eadie-Hofstee plot (inset) demonstrates clear convex curvature, consistent with activation kinetics. Kinetic data was fitted with the Hill equation. Data points represent mean ± standard error (n=3).  The Vmax and K’ describing the kinetics of DHT reduction by AKR1C2 in the presence of 50 µM fluoxetine, imipramine, paroxetne, and sertraline were determined (Table 5.1). For each of these reactions, sigmoidal kinetics was observed and data was analyzed using the Hill equation (Figure 5.3). The sigmoidicity of the kinetics appeared unchanged in the presence of the modulator compounds, as described by no significant differences in the determined Hill coefficient values.  Significant increases in Vmax, 106  reaching approximately 2, 2.5, and 5-fold greater activities than measured in absence of modulator, were observed in the presence of sertraline, imipramine, and paroxetine, respectively. Concurrently, reductions in K’ (60% lower) were observed in presence of paroxetine. These alterations resulted in a significant increase in the enzyme efficiency in the presence of imipramine and paroxetine, as judged by the Vmax/K’ ratio. Similar alterations in Vmax, K’, and Vmax/K’ were observed in the presence of sertraline, though these difference were not statistically significant. The addition of fluoxetine, tested as either a pure chemical standard or extracted from a formulated product, did not alter the enzyme activities.  Table 5.1 Summary of AKR1C2 catalyzed reduction of DHT in absence and presence of modulating compounds. Data were fit with the Hill equation describing activation kinetics. Reported values are mean kinetic parameters and standard deviations for three separate kinetic curves. * indicates significant differences from DHT alone when tested using Student’s two-tailed t-test, p<0.05. Vmax, nmol/mg/min  K', µM  h  Vmax/K’  DHT alone  0.948 ± 0.100  2.26 ± 0.50  2.1 ± 0.3  0.44 ± 0.11  + fluoxetine  0.907 ± 0.145  2.82 ± 1.25  2.4 ± 0.3  0.38 ± 0.21  + imipramine  2.340 ± 0.297 *  1.38 ± 0.32  2.4 ± 0.2  1.76 ± 0.48 *  + paroxetine  4.791 ± 0.967 *  1.01 ± 0.32 *  2.0 ± 0.1  5.21 ± 2.20 *  + sertraline  1.907 ± 0.278 *  1.66 ± 0.55  2.3 ± 0.2  1.25 ± 0.53  107  Figure 5.3 Effect of modulators on AKR1C2 catalyzed reduction of DHT to 3!-diol. Rates of DHT reduction in absence of modulators (closed squares, overlapping) and in the presence of fluoxetine (closed triangles, overlapping), imipramine (open triangles), paroxetine (open squares), and sertraline (open circles) indicate activation of metabolism. All modulators were tested at 50 µM final concentration. Data points represent mean ± standard error (n=3).  Dose-response curves constructed for fluoxetine, sertraline, imipramine, and paroxetine are shown as Figure 5.4. Similarly shaped curves were observed in the presence of paroxetine and imipramine with increasing enzyme activities with greater modulator concentrations up to a maximal activation at approximately 50 µM. Beyond this concentration, there was a sharp decline in activation. Similar levels of activation were observed for setraline, though the maximal activation occurred at concentrations ranging from 10 to 50 µM. Lower levels of activation were observed with the addition of fluoxetine, where 10 µM showed an increase in enzyme activity, but declined at higher modulator concentrations. 108  Figure 5.4 Dose-response relationships for modulator compounds fluoxetine, paroxetine, sertraline, and imipramine for their activation of AKR1C2 catalyzed reduction of DHT to 3!-diol. Activities were normalized to the activity measured in absence of modulator (control) and shown in percent activity. Data points represent mean ± standard error (n=3).  The addition of BSA to the in vitro reaction mixture resulted in large increases in activities whether in the presence or absence of imipramine, paroxetine, or setraline (Figure 5.5). BSA concentrations of 0.05 or 0.1 mg/ml in the incubation mixtures resulted in the highest levels of enzyme activity, which reached approximately 10-fold higher activities than in absence of this protein. Beyond these BSA concentrations, activity levels decreased slightly such that activities were approximately 5 or 6-fold higher for BSA at 1 mg/ml.  When the incubation mixtures contained no BSA,  significantly higher activities were observed in incubations to which paroxetine and  109  sertraline were added. Upon addition of BSA, these differences were diminished and only at a low level of BSA (0.001 mg/ml) were activities for incubations containing paroxetine significantly higher than in absence of modulator. However, at 1 mg/ml BSA, the highest protein concentration tested, activation by paroxetine, imipramine, and sertraline were distinguished by activities that were significantly greater than control (no modulator) reactions (Figure 5.5).  Figure 5.5 Effect of bovine serum albumin on AKR1C2 catalyzed reduction of DHT to 3!-diol in presence and absence of activating compounds. No modulator is shown as crosses with solid line, imipramine is closed circles with dotted line, sertraline is closed squares with dot-dash line, and paroxetine is open squares with dashed line. DHT was 5 µM for all incubations; modulators were 50 µM. Rates of DHT reduction are normalized to activity measured for no modulator in absence of BSA. Data points represent mean ± standard error (n=3). Differences from no modulator were tested separately at each BSA level for each compound using the Student’s two-tailed t-test, p<0.05; *, paroxetine; †, sertraline; ‡, imipramine.  110  5.4.  Discussion Our study has demonstrated that AKR1C2 activity, measured by the reduction of  DHT to 3!-diol, was significantly increased in the presence of 50 #M sertraline, imipramine, or paroxetine (Table 5.1). The modulators were tested at 50 #M to be directly comparable to previous studies, which reported that this concentration provided maximal enzyme activation. We report these data as confirmatory evidence to the report by Griffin and Mellon that paroxetine, sertraline, and imipramine directly alter the activities of AKR1C2 acting as allosteric activators (Griffin and Mellon, 1999). We tested fluoxetine as a pure chemical standard and extract from a formulated tablet and neither form showed activation. This is a surprising result as it contradicts the findings of the earlier report, and because fluoxetine shares many chemical and pharmacological features with the compounds that activated AKR1C2. The dose-response curves support that the concentrations used were appropriate to observe maximal activation for imipramine, paroxetine, and sertraline. We have planned future studies that will confirm these dose-response relationships. The pure chemical standards will be obtained and tested for all of the modulators and fluoxetine will be tested at the concentration that provided the maximal activation, which is estimated to be ~10 µM. Preliminary experiments that were conducted using the modulating compounds at a lower concentration (6 #M) failed to demonstrate activation for any of the compounds (data not shown). Based on these results and the marginal level of activation that was observed for fluoxetine at 10 µM (less than two-times activity) in the completed dose-response studies, we do not anticipate that the data provided by these confirmatory experiments will alter our conclusions.  111  One potential explanation for the discrepancies between our findings and earlier reports is that the previous investigators measured the reduction of DHP to allopregnanolone, rather than the reduction of DHT to 3!-diol. We studied the reduction of DHT based on its importance in the normal development and disease progression of androgen-dependent tissues, such as the prostate.  Furthermore, 3!-diol has been  identified as an important positive allosteric ligand for the GABAA receptor (Frye et al., 2001; Reddy, 2004), and it was of interest whether AKR1C2 activation increased its rate of production. At the initiation of our studies we predicted that since DHT and DHP share the same steroid structure at the site of reduction (Figure 5.6), similar enzyme kinetics would be observed. This hypothesis was supported by the report by Griffin and Mellon where they observed similar reductase activation using DHP or DHT as substrates (Griffin and Mellon, 1999). Based on our data, we cannot rule out that differences in the steroid structure (DHP is a C-21 steroid, whereas DHT is a C-19 steroid) may influence the binding of a ligand at an allosteric binding site. Clearly, additional data using various substrates are needed before any conclusions can be made on substrate-dependency of allosteric activation.  112  Figure 5.6 AKR1C2-catalyzed reduction of DHT and DHP. DHT is a C-19 steroid, whereas DHP is a C-21 steroid; however, the site for metabolism by AKR1C2 is the same with reduction of the carbonyl functionality at carbon-3 to the corresponding secondary alcohol for both steroids. Our in vitro conditions were selected to replicate those used in the previous reports. Some experimental details, however, deserve some special note for interpreting the current findings. First, we attribute the activation to the active chemical components, but as with the previous studies, our experiments do not distinguish modulation from soluble components of the formulations used.  Though we feel it is an unlikely  explanation for our results, polymers or other excipients extracted from the tablets used to obtain imipramine, paroxetine, and sertraline may act to promote substrate-enzyme interaction or may directly alter the enzyme function. Notably, fluoxetine extracted from a formulated product did not activate the enzyme. Our future studies using the pure chemical standards will provide solid evidence that it is the drug component of these formulations that acts to activate the enzyme. Second, as with the previous studies, the  113  recombinant proteins were expressed and purified from bacteria. While the other groups purified proteins by preparation of bacterial inclusion bodies or glutathione-S-transferase fused protein, we used a histidine-tagged protein that we have demonstrated to be highly purified and fully functional when tested with standard reductase substrates (Takahashi et al., 2008).  Finally, for the quantitation of 3!-diol, we have used a liquid  chromatography-mass spectrometry method that was confirmed to be specific and accurate. It was ruled out that ion suppression or enhancement contributed to the larger mass spectral signals for 3!-diol that we observed in the presence of the modulator compounds. Additional chromatographic and mass spectral scan analyses clearly showed that the modulator compounds did not co-elute with 3!-diol or the internal standard, and therefore did not lead to the enhanced 3!-diol signals that were concluded to result from AKR1C2 activation. The kinetics that we observed for AKR1C2 reduction of DHT followed nonMichaelis-Menten kinetics and the data were best fit by the Hill equation. To the best of our knowledge, this is the first time that sigmoidicity has been described for this reaction. Several examples of non-typical kinetics due to autoactivation have been observed using in vitro systems with purified recombinant enzymes studying cytochrome P450 metabolism (Houston and Kenworthy, 2000). These authors speculate that autoactivation kinetics exist for many reactions, but are infrequently reported due to the difficulty distinguishing atypical kinetics when using mixed enzyme sources (e.g., microsomes) or when insufficient data points have been collected. It is difficult to judge whether a mixed protein source contributed to the observation of Michaelis-Menten kinetics for DHT reduction in previous investigations. The recombinant enzyme used in our studies was  114  bacterially-expressed as a histidine-tagged (amino terminus) protein, and was shown, using protein electrophoresis and staining, to be highly purified by affinity chromatography (Takahashi et al., 2008). This report also described evidence that the activities of contaminating proteins may have been attributed to AKR1C2 in previous investigations, leading to the incorrect identification of substrates for this enzyme. It is interesting that in our experiments, the Hill coefficient (h) did not vary with the binding of an allosteric activator and, further, was not influenced by the chemical nature of the activating compound.  These results support the concept that the  mechanisms for autoactivation and allosteric activation can occur independently of each other. Neither our data, nor those of the previous report for AKR1C2 activation, provide the protein structure-function information to propose how activation may occur. Future studies are needed to identify the location of an allosteric binding site in the AKR1C2 structure, provide a structural description for allosteric modifiers, and determine whether autoactivation is observed for other steroid and non-steroid AKR1C2 substrates. The data from these studies should help to assess whether enzyme activation is pharmacologically important.  For example, potent AKR1C2 activators could  significantly increase enzyme activities, potentially leading to complex and unpredictable metabolism of endogenous chemicals, toxicants, or drugs. In the case of drugs that are predominantly metabolized by AKR1C2, this could be important for predicting drugdrug/toxicant interactions or non-linear dose scaling. We have clearly demonstrated that the addition of BSA to in vitro samples results in increases in activity that are substantially larger than those that result from enzyme activation by paroxetine or any of the other tested modulators. The effect of albumin  115  may explain why some investigators did not observe AKR1C2 activation in their experiments. Specifically, the incubations conducted by Trauger et al. included BSA (0.5 mg/ml) and their reported kinetic constants differ widely from those determined by Griffin and Mellon. For example, apparent Km values in the low micromolar and low nanomolar concentrations are reported for DHP reduction measured without addition of any modulating compounds (Griffin and Mellon, 1999; Trauger et al., 2002). Though these discrepancies have many potential sources, it is probable that the inclusion of BSA is an important factor for the disparate conclusions of these two studies. As our results show, enzyme activation was not observed for in vitro incubations containing 0.01-0.5 mg/ml BSA. We expect that the inclusion of BSA in the experiments by Trauger et al. may be the probable explanation for why they did not observe AKR1C2 activation. The mechanism for the enhancement in AKR1C2 activity in the presence of BSA is unknown at this point. Alterations of in vitro kinetics have been observed due to the addition of albumin using microsomal or liver slice metabolic systems to study cytochrome P450-mediated metabolism (Ludden et al., 1997; Tang et al., 2002; Wang et al., 2002) and in microsomal systems to study UDP-glucuronysyl transferase metabolism (Rowland et al., 2008; Kilford et al., 2009).  These previous reports provide good  precedence for a significant albumin effect when using in vitro metabolic systems. In the report by Rowland et al., the investigators demonstrated that albumin acts to sequester fatty acids released into the in vitro systems that are inhibitory to metabolic enzymes (Rowland et al., 2008). It is not clear whether albumin is acting by the same mechanism to impact AKR-mediated metabolism. Alternative explanations for the enhancement in  116  enzyme activities are: i) soluble BSA interacts with the reductase protein, affecting its tertiary structure, or ii) soluble BSA facilitates the interaction of substrate and reductase. Importantly, the recent investigations of an albumin effect on metabolic enzymes suggest that in vitro kinetic results obtained in presence of albumin provide better predictions of in vivo results (Rowland et al., 2008; Kilford et al., 2009). These authors used hepatic clearance values collected in vivo to test the goodness of their in vitro results. Unfortunately, due to the overlapping activities of several enzymes, kinetic estimates for AKR1C2 reduction of DHT in vivo have not been collected and therefore, there is no appropriate data to compare with the in vitro estimates. We speculate that the in vivo situation is better simulated by in vitro experiments that include albumin (or other soluble proteins); however, there are not yet sufficient data to quantitatively support this practice. Activation of AKR1C2 by fluoxetine was suggested to explain the elevation of allopregnanolone levels in the central nervous systems of patients receiving this drug (Griffin and Mellon, 1999).  Our data, particularly the effect of soluble protein on  enzyme activities, indicate that in vitro activation was overinterpreted and may not be relevant to the clinical situation. In the in vivo situation, soluble proteins would be expected to enhance AKR1C2 activities, as was observed in vitro using BSA as a standard laboratory protein. Therefore, the activity increases due to the antidepressant drugs are relatively small. This prediction is supported by results of cell-based assays that we conducted to model the in vivo situation (data not shown). Using a transfected insect cell model expressing AKR1C2, we did not observe activation by any of the modulator compounds, presumably due to the soluble proteins in the intracellular  117  environment.  Furthermore, based upon the micromolar concentrations of modulator  needed to observe activation in vitro, the steady-state plasma concentrations in patients receiving antidepressants drugs, which range from 10-600 ng/ml (DeVane, 1999), may not be adequate to significantly increase the activities of AKR1C2. Clearly, additional investigations are necessary to understand the importance of the in vitro conditions for AKR1C2 activation and how to scale these findings for accurate in vivo predictions. In summary, we have demonstrated that the reductive metabolism of DHT by AKR1C2 was activated by compounds that are commonly used for treating depression and anxiety, namely paroxetine, sertraline, and imipramine. In contrast, fluoxetine was not observed to activate the enzyme. Furthermore, the presence of BSA enhanced the rates of DHT metabolism in vitro, such that activation by the antidepressants was not observed. Though these findings confirm previous reports of the proposed allosteric activation of AKR1C2 in vitro, the predicted in vivo significance of this activation may have been overstated because earlier extrapolations did not account for the effects of solubilized protein.  118  5.5.  References  Chouinard S, Yueh MF, Tukey RH, Giton F, Fiet J, Pelletier G, Barbier O and Belanger A (2008) Inactivation by UDP-glucuronosyltransferase enzymes: The end of androgen signaling. J Steroid Biochem Mol Biol 109:247-253. Chu LW, Reichardt JK and Hsing AW (2008) Androgens and the molecular epidemiology of prostate cancer. Curr Opin Endocrinol Diabetes Obes 15:261270. DeVane CL (1999) Metabolism and pharmacokinetics of selective serotonin reuptake inhibitors. Cell Mol Neurobiol 19:443-466. Dufort I, Labrie F and Luu-The V (2001) Human types 1 and 3 3 alpha-hydroxysteroid dehydrogenases: differential lability and tissue distribution. J Clin Endocrinol Metab 86:841-846. Frye CA, Park D, Tanaka M, Rosellini R and Svare B (2001) The testosterone metabolite and neurosteroid 3alpha-androstanediol may mediate the effects of testosterone on conditioned place preference. Psychoneuroendocrinology 26:731-750. Griffin LD and Mellon SH (1999) Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc Natl Acad Sci U S A 96:1351213517. Houston JB and Kenworthy KE (2000) In vitro-in vivo scaling of CYP kinetic data not consistent with the classical Michaelis-Menten model. Drug Metab Dispos 28:246-254. Jin Y and Penning TM (2001) Steroid 5alpha-reductases and 3alpha-hydroxysteroid dehydrogenases: key enzymes in androgen metabolism. Best Pract Res Clin Endocrinol Metab 15:79-94. Jin Y and Penning TM (2006) Multiple steps determine the overall rate of the reduction of 5alpha-dihydrotestosterone catalyzed by human type 3 3alpha-hydroxysteroid dehydrogenase: implications for the elimination of androgens. Biochemistry 45:13054-13063. Kilford PJ, Stringer R, Sohal B, Houston JB and Galetin A (2009) Prediction of drug clearance by glucuronidation from in vitro data: use of combined cytochrome P450 and UDP-glucuronosyltransferase cofactors in alamethicin-activated human liver microsomes. Drug Metab Dispos 37:82-89. Ludden LK, Ludden TM, Collins JM, Pentikis HS and Strong JM (1997) Effect of albumin on the estimation, in vitro, of phenytoin Vmax and Km values: implications for clinical correlation. J Pharmacol Exp Ther 282:391-396.  119  Luu-The V, Belanger A and Labrie F (2008) Androgen biosynthetic pathways in the human prostate. Best Pract Res Clin Endocrinol Metab 22:207-221. Melcangi RC, Garcia-Segura LM and Mensah-Nyagan AG (2008) Neuroactive steroids: state of the art and new perspectives. Cell Mol Life Sci 65:777-797. Morimoto I, Edmiston A and Horton R (1980) Alteration in the metabolism of dihydrotestosterone in elderly men with prostate hyperplasia. J Clin Invest 66:612-615. Mostaghel EA and Nelson PS (2008) Intracrine androgen metabolism in prostate cancer progression: mechanisms of castration resistance and therapeutic implications. Best Pract Res Clin Endocrinol Metab 22:243-258. Negri-Cesi P and Motta M (1994) Androgen metabolism in the human prostatic cancer cell line LNCaP. J Steroid Biochem Mol Biol 51:89-96. Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, Moore M, Palackal N and Ratnam K (2000) Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Biochem J 351:67-77. Reddy DS (2004) Anticonvulsant activity of the testosterone-derived neurosteroid 3alpha-androstanediol. Neuroreport 15:515-518. Rizner TL, Lin HK and Penning TM (2003) Role of human type 3 3alpha-hydroxysteroid dehydrogenase (AKR1C2) in androgen metabolism of prostate cancer cells. Chem Biol Interact 143-144:401-409. Rowland A, Knights KM, Mackenzie PI and Miners JO (2008) The "albumin effect" and drug glucuronidation: bovine serum albumin and fatty acid-free human serum albumin enhance the glucuronidation of UDP-glucuronosyltransferase (UGT) 1A9 substrates but not UGT1A1 and UGT1A6 activities. Drug Metab Dispos 36:1056-1062. Suzuki K, Nishiyama T, Hara N, Yamana K, Takahashi K and Labrie F (2007) Importance of the intracrine metabolism of adrenal androgens in androgendependent prostate cancer. Prostate Cancer Prostatic Dis 10:301-306. Takahashi RH, Bains OS, Pfeifer TA, Grigliatti TA, Reid RE and Riggs KW (2008) Aldo-keto reductase 1C2 fails to metabolize doxorubicin and daunorubicin in vitro. Drug Metab Dispos 36:991-994. Tang C, Lin Y, Rodrigues AD and Lin JH (2002) Effect of albumin on phenytoin and tolbutamide metabolism in human liver microsomes: an impact more than protein binding. Drug Metab Dispos 30:648-654.  120  Trauger JW, Jiang A, Stearns BA and LoGrasso PV (2002) Kinetics of allopregnanolone formation catalyzed by human 3 alpha-hydroxysteroid dehydrogenase type III (AKR1C2). Biochemistry 41:13451-13459. Wang JS, Wen X, Backman JT and Neuvonen PJ (2002) Effect of albumin and cytosol on enzyme kinetics of tolbutamide hydroxylation and on inhibition of CYP2C9 by gemfibrozil in human liver microsomes. J Pharmacol Exp Ther 302:43-49.  121  CHAPTER 6 THE EFFECT OF ALLELIC VARIATION IN ALDO-KETO REDUCTASE 1C2 ON THE IN VITRO METABOLISM OF DIHYDROTESTOSTERONE1  6.1.  Preface The development of prostate cancer is unpredictable, with only general risk  factors (age, family history of disease, and ethnic background) being described to-date (Bostwick et al., 2004).  Thus, there is a heavy reliance on wide-spread screening  techniques to increase the chances that abnormal prostate growth is detected in its earliest stages and then, pharmacological or surgical intervention is inititated to minimize the mortality from this disease (LaSpina and Haas, 2008). A decline in the death rates following the implementation of these strategies has been cited as a clear demonstration of the successes of earlier detection and better treatment. A description of accurate and more specific risk factors for prostate cancer will identify patients that should be more carefully followed with diagnostic procedures or should initiate preventative therapies. It is reasonable to expect that this would further reduce death rates. Many targets have been suggested as potential markers of increased prostate cancer risk (D'Amico et al., 2007; Imamoto et al., 2008). Several of these are based on the proposed mechanisms by which the disease manifests, where abnormal or excess androgen signaling in the prostate is expected to result in uncontrolled cell proliferation 1  A version of this chapter has been published. Takahashi RH, Grigliatti TA, Reid RE, and Riggs KW (2009) The Effect of Allelic Variation in Aldo-Keto Reductase 1C2 on the In Vitro Metabolism of Dihydrotestosterone. J Pharmacol Exp Ther 329:1032-1039.  122  and carcinogenesis.  Thus, the androgen receptor and androgen levels have been  extensively investigated for their involvement and association with disease risk (Eaton et al., 1999; Hsing et al., 2008). Though these studies have brought a better understanding of the roles and function of androgens and their activation of the androgen receptor, to date they have not provided a clear marker for prostate cancer. As described in Section 1.3.7., the association of excess levels of androgens in the prostate with cancer development is supported by in vitro and patient data. Metabolic enzymes have critical roles in regulating these levels by catalyzing the metabolic pathways that synthesize and deactivate androgens that can act as ligands for the androgen receptor. Therefore, metabolic enzymes are important modulators of androgen receptor signaling and potential determinants for prostate cancer development. Consequently, factors that alter the function of these enzymes may significantly impact an individual’s risk of cancer developing.  One potential factor, which to date has  received little attention, is naturally-occurring genetic variation in the reductase genes. Based on current knowledge, AKR1C2 is suggested to be the major metabolic enzyme involved in the deactivation of DHT, which is the most potent natural ligand for the androgen receptor and, in concentration, the predominant androgen within the prostate tissue (Penning and Drury, 2007). The pivotal role for AKR1C2 in intracrine metabolism of DHT and its potential association with prostate cancer is underscored by the findings that its expression levels and activities are reduced in prostate cancer tissue when compared with paired benign tissue (Ji et al., 2003; Ji et al., 2004). Genotyping experiments demonstrate that naturally-occurring non-synonymous SNPs occur in AKR1C2 (Section 1.1.6.); however, the functional consequence of these genetic  123  variations are not known. Based on the evidence demonstrating that SNPs are associated with impaired enzyme function in several AKR proteins (Section 1.1.4.), it is highly likely that some of the eleven SNPs in AKR1C2 that lead to amino acid substitution will alter its function. Therefore, these SNPs may result in reduced DHT deactivation in vivo, and potentially be associated with abnormal androgen levels in the prostate.  It is  associations such as these that will provide a better understanding of the etiology of prostate cancer development or rate of progression and may serve as markers to identify men who are at increased risk of developing this disease. The objective of this study is to characterize the effects of the eleven naturally occurring non-synonymous SNPs in AKR1C2 on the activity of the expressed protein by comparing their in vitro activities to the activity of the wild-type protein. The functional characterization of the wild-type and variant AKR1C2 proteins are conducted using cellbased assays that monitor the reduction of two substrates specifically metabolized by AKR1C2, a synthetic fluorogenic probe substrate and DHT. The variant AKR1C2 alleles that are associated with altered activities in reducing DHT may represent one factor involved in the variable degradation of DHT in vivo and, hence, may be potential markers for patients with increased risk of prostate cancer.  6.2.  Materials and Methods  6.2.1. Materials DHT, 3!-diol, 3"-diol, and 16!-bromoandrosterone were purchased from Steraloids (Newport, RI). The fluorogenic AKR1C2 probe substrate (8-phenylketone2,3,5,6-tetrahydro-1H, 4H-11-oxa-3a-aza-benzo[de]anthracen-10-one) (Yee et al., 2006) was 124  kindly provided by Dr. Dalibor Sames, Columbia University, New York, NY. DNA primers were synthesized by Integrated DNA Technologies (San Diego, CA). Zeocin, Cellfectin, and NuPAGE gels and reagents were supplied by Invitrogen (Carlsbad, CA). ESF AF and Grace’s growth media were obtained from Expression Systems LLC (Woodland, CA) and Gibco (Grand Island, NY), respectively. Human liver cytosol was from Cellzdirect (Durham, NC). Bradford assay reagent and bovine serum albumin were supplied by BioRad (Hercules, CA). Primary antibodies for detection of AKR1C2 and GAPDH were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA) and Abnova (Taipei, Taiwan), respectively. Secondary antibodies were from LI-COR Biosciences (Lincoln, NE).  All solvents were HPLC grade or better and obtained from Fisher  Scientific (Toronto, ON).  Deionized, distilled water was obtained from a Milli-Q  filtration system (Millipore, Bedford, MA).  All other chemicals were from Sigma  Chemical Co (St Louis, MO).  6.2.2. Construction of expression plasmids The insect expression vector, p2ZOpie2N (Hegedus et al., 1998), carrying the human AKR1C2 cloned cDNA was used as a template to generate all of the AKR1C2 variant alleles according to the protocol provided with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). Substituting the essential catalytic tyrosine residue at position 55 with a phenylalanine (Y55F) was reported to result in a nonfunctional AKR1C2 protein (Matsuura et al., 1997). This substitution was constructed for use as a negative control for in vitro experiments. The forward and reverse primers used for site-directed mutagenesis are listed as Table 6.1. The wild-type and variant  125  AKR1C2 cDNAs were subsequently subcloned into the p2ZOpie2F insect expression vector for cell transfection and translation of an unmodified protein. All constructs were verified by dideoxynucleotide sequencing at the University of British Columbia Nucleic Acid Protein Service unit.  DNA plasmid stocks were produced by heat-shock  transforming Escherichia coli DH5! cells with the p2ZOpie2F constructs and selecting a single transformant grown under Zeocin antibiotic selection. Transformed bacteria were amplified by culturing overnight in low salt LB media containing Zeocin and plasmids isolated using an E.N.Z.A. high purity plasmid miniprep kit (Omega Bio-tek, Doraville, GA).  DNA concentrations were determined from spectrometric measurements of  absorbance at 260 nm, and purity was checked by the ratio of ultraviolet absorbance measured at 260 and 280 nm.  126  Table 6.1 PCR primers used for site-directed mutagenesis to create AKR1C2 variant alleles. Bold highlighting in primer sequences shows single nucleotide polymorphisms introduced by site-directed mutagenesis. Amino acid variation V38A  NCBI Identifier (rs number) rs3207901  V38I  rs3207898  F46Y  rs2854482  H47R  rs3207905  S87C  rs13933  V111A  rs2854486  H170R  rs10618  L172Q  rs11474  K179E  rs2518042  K185E  rs2518043  R258C  rs28943580  Y55F *  Primer forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse forward reverse  Primer sequence GTA CTT GTA CTT  AAG CTC TAG AGG CCG CCA AAT TGG CAA TAG AAG CTA TTG CCA ATT TGG CGG CCT CTA GAG CTT TAC AAG CTC TAG AGG CCA TCA AAT TGG CAA TAG AAG CTA TTG CCA ATT TGA TGG CCT CTA GAG CTT TAC CAA TAG AAG CCG GGT ACC ACC ATA TTG ATT C GAA TCA ATA TGG TGG TAC CCG GCT TCT ATT G CAA TAG AAG CCG GGT TCC GCC ATA TTG ATT CTG CAC GTG CAG AAT CAA TAT GGC GGA ACC CGG CTT CTA TTG CAC TTC AAA GCT TTG GTG CAA TTC CCA TCG AC GTC GAT GGG AAT TGC ACC AAA GCT TTG AAG TG CTT CAA TTG GAC TAT GCT GAC CTC TAT CTT ATT C GAA TAA GAT AGA GGT CAG CAT AGT CCA ATT GAA G GTG TCC AAC TTC AAC CGC AGG CTG CTG GAG ATG CAT CTC CAG CAG CCT GCG GTT GAA GTT GGA CAC CAA CTT CAA CCA CAG GCA GCT GGA GAT GAT CCT C GAG GAT CAT CTC CAG CTG CCT GTG GTT GAA GTT G GAT GAT CCT CAA CGA GCC AGG GCT CAA G CTT GAG CCC TGG CTC GTT GAG GAT CAT C CAG GGC TCA AGT ACG AGC CTG TCT GCA AC GTT GCA GAC AGG CTC GTA CTT GAG CCC TG CTG ATT GCC CTG TGC TAC CAG CTG C GCA GCT GGT AGC ACA GGG CAA TCA G GAT TCT GCA CAT GTT TTC AAT AAT GAG GAG CAG G CCT GCT CCT CAT TAT TGA AAA CAT GTG CAG AAT C  * Y55F is a non-naturally occurring substitution that removes the catalytic tyrosine necessary for reductase activity (Matsuura et al., 1997). It was constructed for use as a negative control for in vitro experiments.  127  6.2.3. Expression of AKR1C2 in insect cells AKR1C2 was expressed in Trichoplusia ni (T. ni) and Spodoptera frugiperda (Sf9) cell lines.  A general procedure was used for all cell transfections with the  conditions of cell densities, amounts of DNA and transfection reagent, media volumes, and time until testing protein expression optimized for the specific insect cell lines used. Briefly, cells cultured in ESF AF media were plated and allowed to adhere.  The  expression plasmids were incubated with Cellfectin reagent in Grace’s media for 30 minutes at room temperature to form DNA-liposome complexes. Cells were briefly washed with Grace’s media and then incubated with DNA-liposomes for 4 hours at 27°C to allow cell transfection with the vector constructs. ESF AF media was added to the cells and the plates were incubated undisturbed at 27°C until protein expression and enzyme activity were tested. For T. ni cell transfections, 5!105 cells were plated into each well of a 12-well plate. The use of 250 ng of plasmid DNA and 1.5 µl of Cellfectin resulted in optimal AKR1C2 expression. A volume of 1 ml of ESF AF media was added post-transfection and expression was tested 48 hours later. For Sf9 cell transfections, 1.5!106 cells were plated into each well of a 6-well plate. The use of 1000 ng of plasmid DNA and 5 µl of Cellfectin resulted in optimal AKR1C2 expression. A volume of 2 ml of ESF AF media was added post-transfection and expression was tested 72 hours later. Cells were removed from the plates as suspensions by gently washing the plate surface with the growth media until they were visibly detached from the plate surface.  128  6.2.4. Western blotting of AKR1C2 Cells were collected by centrifugation (1000!g for 10 minutes) and the growth media removed. The cell pellets were resuspended in 0.1! volume of a lysis buffer (100 mM Tris-HCl, adjusted to pH 8.0, containing 0.1% Triton X-100) and lysed by freezing for at least one hour at -80°C and subsequently warming at 42°C for 2-3 minutes or until samples were visibly judged to be completely thawed. Lysates were collected as the supernatant fraction following centrifugation (13,000!g for 10 minutes).  The total  protein content of the cell lysates was determined by the Bradford method with bovine serum albumin as a standard. SDS-PAGE of the cell lysates was performed with NuPAGE 4-12% Bis-Tris gels using denaturing, reducing conditions with MES running buffer according to the manufacturer’s standard procedures (Invitrogen). For cell lysate samples, 25 µg of total protein per lane was used. Human liver cytosol, 2 µg, was co-analyzed as a source of human AKR1C2 protein to confirm antibody binding. Proteins were electrophoretically transferred to nitrocellulose membranes (Amersham Biosciences, Piscataway, NJ), which were then prepared according to the protocol described by Odyssey (LI-COR Biosciences, Lincoln, NE). AKR1C2 protein was detected using goat polyclonal antihuman dihydrodiol dehydrogenase and donkey anti-goat IgG IR800 antibodies. GAPDH protein was measured as a loading control for all samples. Detection of GAPDH was performed with mouse anti-GAPDH and donkey anti-mouse IgG IR700 antibodies. Fluorescence detection and data processing were performed using the Odyssey Infrared Imaging system and software.  129  6.2.5. Kinetic analysis of AKR1C2 Fluorogenic probe reduction. The reduction of the AKR1C2 specific probe substrate was monitored by detection of the fluorescent signal arising from the alcohol reduction product of the probe. Cell suspensions were incubated with 1 to 25 µM of the fluorogenic probe in a total volume of 150 µl in a 96-well plate format. Stock solutions of the substrate were made with DMSO as the solvent and were diluted so that the DMSO content of the incubation samples was 1%. The fluorescent signal from the samples was measured every 5 minutes for two hours, which included both linear and non-linear phases for the reduction, using a Fluoroskan Ascent FL (Thermo Fisher Scientific, Waltham, MA) with excitation and emission wavelengths of 390 and 518 nm, respectively. Plates were incubated at 37°C and shaken gently before each fluorescence measurement.  Kinetic rates, in fluorescence units per time, were determined using  Ascent® software, version 2.6 (Thermo Fisher Scientific) from data observed to be within the linear phase of the two hour reaction.  Dihydrotestosterone reduction. The reduction of DHT to its 3"-diol metabolite by AKR1C2 was performed as follows. Cell suspensions were first lysed by sonication with a probe dismembranator using three 3-second bursts. Incubation mixtures contained the treated cell suspension and 2.5 to 60 µM of DHT in a total volume of 140 µl. These samples were pre-incubated at 37°C for 25 minutes, then 10 µl of a NADPH solution was added, 1 mM final concentration, to initiate the reaction. Samples (150 µl total volume) were incubated at 37°C for 30 minutes with continuous gentle plate shaking (100 rpm). The reaction was terminated by the addition of 125 µl of ice-cold 1 N hydrochloric acid  130  and 25 µl of 2.5 µg/ml 16"-bromoandrosterone was added as an internal standard. The androgens were extracted with 1 ml of methyl tert-butyl ether containing 0.1% triethylamine and subsequently evaporated to dryness under a gentle stream of nitrogen. Following the evaporation step, methyl pyridinium ether derivatives of the extracted androgens were formed in a total volume of 250 µl dichloromethane containing 0.01 M triethylamine and 0.01 M 2-fluoro-1-methyl pyridinium sulphonate. The reaction was maintained at a temperature of 28°C in a heated oven for 30 minutes. Samples were then evaporated to dryness under nitrogen and resolubilized as an 150 µl aqueous methanol solution (1/1) for LC/MS/MS analysis. A 15 µl aliquot was injected onto the column for analysis. The analytical system consisted of an Acquity HPLC system (Waters, Milford, MA) coupled with a Premiere tandem quadrupole mass spectrometer with an electrospray ion source (Waters). The derivatives of DHT, 3"-diol, 3#-diol, and the internal standard were chromatographically separated using an Acquity BEH C18 column, 100 mm ! 2.1 mm i.d. with 1.7 µm particle size packing by gradient elution with mobile phases of 0.1 % formic acid in water (A) and 0.1% formic acid in methanol (B) according to the following program: 0 min, 35% B; 3.5 min, 70% B; 3.6 min, 98% B; 5.5 min, 98% B; 5.6 min 35% B. The flow rate was constant at 0.2 ml/min and the column was heated to 30ºC. Positive electrospray ionization was performed with a nebulizer temperature of 350ºC and a cone voltage of 20 V. The analytes were detected using multiple reaction monitoring of three ion transitions with collision energies set at 20 V.  The mass  transitions used were m/z 382.2!254.9 for DHT, 384.1!257.1 for 3"-diol and 3#-diol, and 460.5!109.5 for 16"-bromoandrosterone.  All acquisition and data processes  131  including peak smoothing and integration, calibration, and calculation of unknown concentrations were performed using MassLynx software version 4.1 (Waters). The lower limit of quantitation (LOQ) was 15 ng/ml for DHT, and 3 ng/ml for 3"-diol and 3#diol. Accuracy and variability (intra- and inter-day) were determined to be less than 20% bias and less than 20% CV at the LOQ, and less than 15% bias and less than 15% CV at low, medium, and high QC levels.  6.2.6. Data analysis The kinetic parameters Vmax and Km were determined by fitting the data by nonlinear least squares regression with the Michaelis-Menten model (Graphpad Prism version 4, San Diego, CA). The intrinsic clearance (CLint) was determined as the ratio of Vmax to Km.  All kinetic values are reported as mean ± S.D. determined for three  independent cell transfections. Statistical comparison of data collected for variant alleles to the wild-type was performed using the Student’s t test (Graphpad Prism version 4). A two-tailed p-value of < 0.05 was considered statistically significant. Correlation between kinetic data obtained with the fluorogenic probe and DHT was calculated and the strength of the association was assessed by the Pearson correlation coefficient (r) and two-tailed p value (Graphpad Prism version 4).  132  6.3.  Results  6.3.1. Expression of AKR1C2 in insect cells Wild-type AKR1C2 and variants were transiently expressed in T. ni and Sf9 cells as demonstrated by Western blotting prepared lysates of the transfected cells (Figure 6.1 and Figure 6.2). AKR1C2 expressed in the insect cells was observed at an estimated protein size of 37 kDa as judged by electrophoretic migration observed in Western blots. The same migration was observed for the detected protein found in human liver cytosol. Control transfections with the p2ZOpie2F expression vector containing no insert or the cDNA coding for a related reductase protein, AKR1A1, which does not metabolize DHT, did not contain proteins that were recognized by the antibody against AKR1C2. All samples were positive for the GAPDH signal, which was used to normalize the Western blot signals for AKR1C2. The intensity of the normalized AKR1C2 signal did not differ significantly between wild-type and the variants in the T. ni and Sf9 cells as shown in Figure 6.1 and Figure 6.2, respectively.  133  Figure 6.1 Representative Western blots showing detection of AKR1C2 expression in transfected Sf9 cells and mean expression levels. GAPDH was detected as a loading control and used to normalize AKR1C2 signal. No differences were observed in expression for AKR1C2 variants compared to wild-type (WT). Results represent the mean ± S.D. of three independently performed cell transfections.  134  Figure 6.2 Representative Western blot showing detection of AKR1C2 expression in transfected T.ni cells and mean expression levels. GAPDH was detected as a loading control and used to normalize AKR1C2 signal. No differences were observed in expression for AKR1C2 variants compared to wild-type (WT). Results represent the mean ± S.D. of three independently performed cell transfections.  6.3.2. Probe substrate and DHT reduction by wild-type and variant alleles of AKR1C2 Michaelis-Menten kinetic curves were constructed for wild-type and variant AKR1C2 in vitro (Figure 6.3). The kinetic parameters describing the reduction of the probe substrate and DHT are provided in Table 6.2 and Table 6.3, respectively. DHT reduction rates were determined as rates of 3"-diol generation because there were no changes in 3#-diol formation upon AKR1C2 expression. The rates of reduction were normalized to the total amount of soluble protein used in the incubations. The wild-type 135  and all variants were tested within one sample group for the fluorogenic probe studies. Due to the number of variants to be tested, the samples were split between two groups for DHT reduction. Both groups contained the wild-type AKR1C2 and the non-functional Y55F variant, which was included as a negative control.  Figure 6.3 Michaelis-Menten curves for fluorogenic probe reduction by wild-type and variant allele AKR1C2. Representative curves have been selected to show background activity of controls (AKR1A1, p2ZOp2F-empty, and Y55F are overlapping at the bottom of the graph) and significant alterations in Vmax (F46Y), Km (H47R), and intrinsic clearance (S87C) compared to wild-type. Error bars and curves for other variant alleles have been removed to provide clarity. Results represent the mean of three independently performed cell transfections that were tested in separate experiments.  The kinetic parameters for a number of enzyme variants, as determined by reduction of the fluorogenic probe, differed significantly from the wild-type enzyme (Table 6.2). Significant reductions in Vmax, ranging from 39.3 to 78.9% of wild-type, were observed for the F46Y, H47R, S87C, H170R, L172Q, K179E, K185E, and R258C variants. The H47R, L172Q, K185E, and R258C variants also had significantly lower Km 136  values, ranging from 2.9 to 3.4 µM compared to 4.5 µM determined for wild-type AKR1C2. The intrinsic clearance (CLint; Vmax/Km) values observed for the F46Y and S87C variants were significantly lower (approximately 40-50%) than the values for the wild-type enzyme. The formation of 3"-diol was measured for all of the DHT incubations with the transfected Sf9 cell preparations. Some variability was observed in the DHT kinetic values determined for wild-type AKR1C2 between the two groups (Table 6.3). To account for this, each variant allele was compared with the values determined for wildtype AKR1C2 prepared within the same group. The Vmax determined for DHT reduction was significantly lower for the F46Y and L172Q variants, with both reaching approximately 60% of the maximal reduction rate measured for the wild-type protein. The Km and intrinsic clearance values varied widely with V38A, L172Q, K185E, and R258C variants exhibiting significantly reduced Km values, and the L172Q, K185E, and R258C variants showing higher CLint values (2.1, 1.7, and 3.8 times higher than wildtype, respectively). A moderate positive association, though not statistically significant, was observed between the maximal rate (Vmax) for probe substrate reduction and DHT reduction (Figure 6.4), however, no correlation was observed for Km or intrinsic clearance.  137  Table 6.2 Kinetic parameters determined for the reduction of the fluorogenic probe by wild-type and variant AKR1C2 expressed in T. ni insect cells. Probe reduction was measured in fluorescence units (FL) per unit time per mg protein. Results represent the mean ± S.D. determined for three independently performed cell transfections. Significant differences between the wild-type and variant AKR1C2 were assessed using the Student’s two-tailed t test. *, p<0.05.  Allele  Vmax (FL unit/min/mg protein)  Km (µM)  Vmax /Km (volume/min)  AKR1C2 WT  7.05 ± 0.66  4.5 ± 0.5  1.58 ± 0.29  p2ZOp2F-empty  1.19 ± 0.65 *  11 ± 10  0.13 ± 0.04 *  1.08 ± 0.22 *  9.6 ± 2.8 *  0.12 ± 0.03 *  Y55F  1.09 ± 0.09 *  8.6 ± 1.5 *  0.13 ± 0.03 *  V38A  6.01 ± 1.46  5.1 ± 1.0  1.24 ± 0.50  V38I  8.37 ± 1.24  5.3 ± 1.7  1.65 ± 0.32  F46Y  2.77 ± 0.66 *  4.1 ± 1.9  0.77± 0.35 *  H47R  5.01 ± 0.70 *  3.2 ± 0.6 *  1.65 ± 0.57  S87C  4.68 ± 0.50 *  7.2 ± 1.9  0.66 ± 0.11 *  V111A  6.21 ± 0.32  3.8 ± 0.2  1.65 ± 0.11  H170R  5.56 ± 0.56 *  3.6 ± 0.6  1.56 ± 0.29  L172Q  5.02 ± 0.42 *  3.4 ± 0.2 *  1.49 ± 0.19  K179E  4.45 ± 0.11 *  3.4 ± 0.7  1.33 ± 0.27  K185E  5.02 ± 0.74 *  3.2 ± 0.5 *  1.60 ± 0.40  R258C  4.71 ± 0.27 *  2.9 ± 0.5 *  1.63 ± 0.32  a  AKR1A1 b  a  The related reductase protein AKR1A1, which is not involved in DHT metabolism, was included in experiments as a negative control. b The functionally inactive Y55F variant was included in experiments as a negative control.  138  Table 6.3 Kinetic parameters determined for the reduction of DHT by wild-type and variant AKR1C2 expressed in Sf9 insect cells. Data are grouped due to two separate batches of transfections and DHT reduction measurements. Each group contained wildtype AKR1C2 and Y55F variant. Results represent mean ± S.D. determined for three independently performed cell transfections. Significant differences between the wildtype and variant AKR1C2 were assessed using the Student’s two-tailed t test. *, p<0.05.  Vmax (nmol/min/mg protein)  Km (µM)  Vmax /Km (ml/min)  0.168 ± 0.036  8.3 ± 1.1  0.020 ± 0.003  0.075 ± 0.016 *  12 ± 8.2 *  0.009 ± 0.006 *  Y55F  0.072 ± 0.020 *  5.2 ± 1.7 *  0.014 ± 0.004  V38A  0.185 ± 0.059  5.8 ± 0.5 *  0.032 ± 0.012  V38I  0.192 ± 0.097  9.4 ± 0.6  0.020 ± 0.010  F46Y  0.106 ± 0.014 *  8.6 ± 4.7  0.018 ± 0.016  H47R  0.190 ± 0.075  23 ± 21  0.015 ± 0.016  S87C  0.147 ± 0.064  7.2 ± 1.4  0.021 ± 0.010  0.141 ± 0.026  15 ± 3.8  0.009 ± 0.001  0.040 ± 0.003 *  4.0 ± 1.7 *  0.011 ± 0.004  Y55F  0.047 ± 0.007 *  3.9 ± 1.0 *  0.012 ± 0.002  V111A  0.096 ± 0.017  6.1 ± 3.3  0.018 ± 0.006  H170R  0.125 ± 0.031  9.7 ± 8.3  0.017 ± 0.008  L172Q  0.083 ± 0.009 *  4.4 ± 1.0 *  0.020 ± 0.006 *  K179E  0.100 ± 0.006  7.1 ± 3.4  0.017 ± 0.008  K185E  0.106 ± 0.007  6.7 ± 0.9 *  0.016 ± 0.003 *  R258C  0.082 ± 0.022  2.7 ± 1.6 *  0.035 ± 0.014 *  Allele Group 1 AKR1C2 WT a  AKR1A1 b  Group 2 AKR1C2 WT p2ZOp2F-empty b  c  a  The related reductase protein AKR1A1, which is not involved in DHT metabolism, was included in experiments as a negative control. b  The functionally inactive Y55F variant was included in experiments as a negative control. c  The p2ZOp2F-empty vector was included in experiments as a negative control. 139  Figure 6.4 Association analysis of maximal rates, Vmax, measured for reduction of the fluorogenic probe and DHT by AKR1C2 expressed by transfected insect cells. Probe reduction was measured in fluorescence units (FL) per unit time per amount protein. Data are grouped due to two separate batches of transfections and DHT reduction measurements being performed for Sf9 cells. The Pearson correlation coefficient (r) and two-tailed p value were determined separately for each group.  6.4.  Discussion AKR1C2 is an important human metabolic enzyme that regulates the extent and  duration of androgen receptor activation in androgen-responsive tissues (Rizner et al., 2003b; Penning et al., 2007). We have investigated the effect of naturally-occurring allelic variations on the activity of AKR1C2 in reducing DHT to 3"-diol. The wild-type and each of the eleven variant alleles were expressed separately in an insect cell line and their expression levels and kinetic properties were measured. Michaelis-Menten kinetic parameters were determined for the reduction of two different substrates, a fluorogenic probe and DHT. The fluorogenic probe was previously demonstrated to be selectively reduced by AKR1C2 to a fluorescent alcohol product that is easily measured (Yee et al.,  140  2006). DHT reduction to 3"-diol was measured directly by LC/MS/MS. Combined, these studies provide the most comprehensive data to-date on the effect of naturally occurring allelic variation on the function of AKR1C2. There is consensus data from the two separate substrate systems showing the F46Y and L172Q variants have decreased maximal rates of reduction (significantly lower Vmax), and the L172Q, K185E, and R258C variants have increased substrate affinities (significantly lower Km) compared to wild-type. Further, the F46Y and S87C variants show significant reductions in intrinsic clearance values when characterized with the fluorogenic probe (Table 6.2). We conclude that these differences in kinetic parameters result from functional alterations in the enzyme since no differences exist in the expression of AKR1C2 (Figure 6.1 and Figure 6.2). The association between the Vmax determined using the two separate substrates suggests that factors that affect rates of DHT metabolism also impact the rates of utilization of the fluorogenic probe. Given the simple fluorescence detection of the probe and its utility in cell-based assays, it may provide a simple and reliable method to assess whether other factors alter the AKR1C2-mediated metabolism of DHT. We suggest the lack of association between the Km values determined using the two systems is due to the bias introduced by the high level of endogenous reductase activity of the insect cells used. This endogenous activity did not differ between control transfections with empty expression vector, vector coding for AKR1A1, or vector coding for functionally-inactive AKR1C2 (Y55F variant) (Table 6.2 and Table 6.3). Western blotting demonstrated that in these cells, AKR1C2 or a cross-reactive protein was not expressed, indicating that the observed background activity results from a distinct, insect  141  genome-encoded reductase protein(s). In developing our cell assays, we selected cell lines that showed the lowest background activities (i.e., for fluorogenic probe substrate reduction, T. ni cells were used, whereas for DHT reduction, Sf9 cells were used); however, the background levels in the Sf9 cells for DHT reduction remained substantial, contributing ~30-40% of the total activity of cells expressing wild-type AKR1C2. This compares to a low level of endogenous activity for the reduction of the fluorogenic probe in the T. ni cells (~15%). We expect that these Km and intrinsic clearance data collected using the fluorogenic probe are less biased, and may therefore, more accurately describe the effect of allelic variation on AKR1C2. It would be ideal to have activities for the two substrates measured in both cell lines to distinguish substrate specificity differences of the wild-type or variant AKR1C2.  Unfortunately, when we tested reduction of the  fluorogenic probe in the Sf9 cells or reduction of DHT in the T. ni cells, the background metabolism was too large to distinguish the contribution of the expressed AKR1C2 to the overall measured activities. The variances in our data reflect the biological variability of cell-based assays, which are typically larger than in vitro assays using fractionated cell systems or purified recombinant enzymes. While it is generally accepted that subtracting background will reduce bias, we have chosen to report activity data without background subtraction because of the increases in variances that this data treatment introduces. Further, we do not know the identities of the enzymes responsible for this background activity. Therefore, we cannot ensure that comparable levels of these enzymes exist between the control and wild-type AKR1C2 transfected cells.  142  Isolation of expressed AKR1C2 was considered to assess the enzyme against a null background. However, reports suggest that upon isolation, the AKR1C2 enzyme is quite labile and this labile nature of reductase enzymes is responsible for significantly different activities that were measured with the isolated enzyme (Dufort et al., 1999; Dufort et al., 2001). In addition, the reductase enzymes are highly sensitive to the availability of cofactors (Penning et al., 2000; Rizner et al., 2003a). Because of these complications we did not pursue measuring the enzyme activity of the wild-type and variant forms of the purified enzyme isoforms. By performing activity measurements with the fluorogenic probe in intact AKR1C2-expressing cells, we expect that the measured activity is more comparable, and thus relevant, to what would be observed in vivo. In comparing the insect cell system to other heterologous expression systems (e.g., bacteria, yeast, mammalian cells), we expect that the insect cell expressed AKR1C2 proteins would undergo similar post-translational modifications found in mammalian cell systems (Kidd and Emery, 1993; Hood et al., 1998; Hegedus et al., 1999) while providing an AKR1C2-null background. The impact of reduced clearance of DHT is hard to predict because, despite being extensively studied, it is not yet clear how androgen steroid levels, and specifically, the prostate tissue levels of DHT, affects the development or progression of prostate cancer. In addition, the synthetic and degradation pathways for DHT in the prostate are complex and are catalyzed by enzymes that can utilize alternate substrates when challenged, which suggests a high potential for compensatory metabolic pathways (Penning et al., 2000). Hence, it is difficult to extrapolate our in vitro results to the in vivo situation. With these caveats, we propose that the kinetic properties we have measured will be useful for  143  comparative studies that will help elucidate the role of AKR1C2 in vivo. Previous studies have performed comparisons of differences in AKR1C2 expression in prostate cancer and benign prostate tissues (Ji et al., 2003; Ji et al., 2007). These investigators demonstrated that the cancerous tissues had higher levels of DHT accompanied by lower expression of AKR1C2. They suggest that this loss of AKR1C2 results in impaired DHT clearance and an increased risk of prostate disease progression. Similarly, we hypothesize that lower AKR1C2 activities resulting from genetic variation may be associated with elevated DHT levels and potentially, the progression of prostate cancer. Clearly, patient data are needed to test this hypothesis. Genetic variation in non-coding regions also likely has key roles for regulating the tissue, temporal and amount of expression of AKR1C2, and will be important data for understanding the clinical significance of AKR1C2 polymorphism and DHT metabolism. If AKR1C2 is, indeed, a critical enzyme for determining DHT levels in the prostate, the F46Y variant is of special interest given its altered function compared to wild-type and population-specific occurrence in men. The frequency of this allele varies based on ethnicity, with individuals of African descent having the highest occurrence of the variant allele (15%), followed by Europeans (5.9%) and then Asian populations (undetected) (HapMap data; http://www.hapmap.org/) (Table 6.4).  These allele  frequencies parallel the risk for prostate cancer observed in these populations (Bostwick et al., 2004). The clinical measurement of prostate DHT levels in men carrying or not carrying the F46Y allele would be of interest to test for the association of AKR1C2 impairment with elevated prostate DHT levels.  144  It is difficult to assess the relevance of the other variant alleles because there is limited data on their prevalence in men (Table 6.4). The variant alleles leading to V38A, V38I, and H47R substitutions were not found in genotyping conducted for the HapMap project. The L172Q allele, which demonstrated lower activities, has a 32.5% allelic frequency measured in a pooled genomic DNA from 94 unrelated Centre d’Etudes Polymorphism Humaine (CEPH) individuals (NCBI SNP database submission ss13684). Using the NCBI SNP database and HapMap project genome browser as our sources for population allelic frequency data, studies to measure the frequencies for the remaining variant alleles have not been conducted.  145  Table 6.4 Allelic frequencies determined for non-synonymous single nucleotide polymorphisms in AKR1C2 compiled from the National Centre for Biotechnology Information (NCBI) SNP database. AKR website identifier AKR1C2*2 AKR1C2*3 AKR1C2*4 AKR1C2*5 AKR1C2*6  NCBI SNP database identifier rs13933 rs11474 rs10618 rs2518042 rs2854482  Amino acid substitution S87C L172Q H170R K179E F46Y  AKR1C2*7 AKR1C2*8  rs2854486 rs3207905  V111A H47R  AKR1C2*9 AKR1C2*10 AKR1C2*11 Not listed  rs3207898 rs3207901 rs28943580 rs2518043  V38I V38A R258C K185E  SNP submission i.d. ss13684 ss4041394 ss69068270 ss4438229 ss74818067 ss4438222 ss4438225 ss38343033 -  Variant allele frequencies (n) CEU  CHB  JPT  YRI  n.d. 32.5% (188) n.d. n.d. 5.9% (118) 5.9% (120) n.d. 0% (298)a 0% (114) 0% (116) 0% (116) 0% (606) n.d.  n.d. n.d. n.d. n.d. 0% (90) 0% (90) n.d. 0% (88) 0% (88) 0% (88) 0% (88) n.d. n.d.  n.d. n.d. n.d. n.d. 0% (88) 0% (90) n.d. 0% (84) 0% (84) 0% (86) 0% (86) n.d. n.d.  n.d. n.d. n.d. n.d. 15.0% (120) 14.2% (120) n.d. 0% (112) 0% (112) 0% (116) 0% (116) n.d. n.d.  CEU: Utah residents with ancestry from northern and western Europe, collected by Centre d'Etude du Polymorphisme Humain (CEPH) CHB: Han Chinese in Beijing, China JPT: Japanese in Tokyo, Japan YRI: Yoruba in Ibadan, Nigeria n.d.: Not yet determined a  pooled data for population ID CEPH (n=184) and population ID HapMap-CEU (n=114), both samples from CEPH  146  Like its role in biotransformation pathways for androgens, AKR1C2 has roles in the inactivation and formation of female sex hormones (Penning et al., 2000). As a result of 17-keto and 20-keto steroid reduction, estrone and progesterone are transformed to 17!-estradiol and 20"-hydroxyprogesterone, respectively, therefore allelic variation in AKR1C2 has potential to affect other receptor-mediated signaling pathways.  Other  known substrates for AKR1C2 include lipid aldehydes that are produced under conditions of oxidative stress, polycyclic aromatic hydrocarbons, and nicotine-derived nitrosaminoketones (Palackal et al., 2002; Penning and Drury, 2007). By altering the detoxification or activation of these endogenous and xenobiotic compounds, allelic variation in AKR1C2 may also have an integral role in carcinogenesis or the development of other diseases. In summary, we have shown that a number of naturally occurring AKR1C2 allelic variants exhibit altered reductase activities in vitro, most notably in Vmax, but also in Km and CLint. Future clinical studies are needed to test whether polymorphic AKR1C2 variants are associated with altered androgen levels in vivo, and to provide genotype frequency data for the other variant alleles to identify their relevance for further study.  147  6.5.  References  Bostwick DG, Burke HB, Djakiew D, Euling S, Ho SM, Landolph J, Morrison H, Sonawane B, Shifflett T, Waters DJ and Timms B (2004) Human prostate cancer risk factors. Cancer 101:2371-2490. D'Amico F, Biancolella M, Margiotti K, Reichardt JK and Novelli G (2007) Genomic biomarkers, androgen pathway and prostate cancer. Pharmacogenomics 8:645661. Dufort I, Labrie F and Luu-The V (2001) Human types 1 and 3 3 alpha-hydroxysteroid dehydrogenases: differential lability and tissue distribution. J Clin Endocrinol Metab 86:841-846. Dufort I, Rheault P, Huang XF, Soucy P and Luu-The V (1999) Characteristics of a highly labile human type 5 17beta-hydroxysteroid dehydrogenase. Endocrinology 140:568-574. Eaton NE, Reeves GK, Appleby PN and Key TJ (1999) Endogenous sex hormones and prostate cancer: a quantitative review of prospective studies. Br J Cancer 80:930934. Hegedus DD, Pfeifer TA, Hendry J, Theilmann DA and Grigliatti TA (1998) A series of broad host range shuttle vectors for constitutive and inducible expression of heterologous proteins in insect cell lines. Gene 207:241-249. Hegedus DD, Pfeifer TA, Theilmann DA, Kennard ML, Gabathuler R, Jefferies WA and Grigliatti TA (1999) Differences in the expression and localization of human melanotransferrin in lepidopteran and dipteran insect cell lines. Protein Expr Purif 15:296-307. Hood SR, Shah G and Jones P (1998) Expression of cytochromes P450 in a baculovirus system. Methods Mol Biol 107:203-218. Hsing AW, Chu LW and Stanczyk FZ (2008) Androgen and prostate cancer: is the hypothesis dead? Cancer Epidemiol Biomarkers Prev 17:2525-2530. Imamoto T, Suzuki H, Yano M, Kawamura K, Kamiya N, Araki K, Komiya A, Nihei N, Naya Y and Ichikawa T (2008) The role of testosterone in the pathogenesis of prostate cancer. Int J Urol 15:472-480. Ji Q, Aoyama C, Nien YD, Liu PI, Chen PK, Chang L, Stanczyk FZ and Stolz A (2004) Selective loss of AKR1C1 and AKR1C2 in breast cancer and their potential effect on progesterone signaling. Cancer Res 64:7610-7617. Ji Q, Chang L, Stanczyk FZ, Ookhtens M, Sherrod A and Stolz A (2007) Impaired dihydrotestosterone catabolism in human prostate cancer: critical role of AKR1C2  148  as a pre-receptor regulator of androgen receptor signaling. Cancer Res 67:13611369. Ji Q, Chang L, VanDenBerg D, Stanczyk FZ and Stolz A (2003) Selective reduction of AKR1C2 in prostate cancer and its role in DHT metabolism. Prostate 54:275289. Kidd IM and Emery VC (1993) The use of baculoviruses as expression vectors. Appl Biochem Biotechnol 42:137-159. LaSpina M and Haas GP (2008) Update on the diagnosis and management of prostate cancer. Can J Urol 15 Suppl 1:3-13; discussion 13. Matsuura K, Deyashiki Y, Sato K, Ishida N, Miwa G and Hara A (1997) Identification of amino acid residues responsible for differences in substrate specificity and inhibitor sensitivity between two human liver dihydrodiol dehydrogenase isoenzymes by site-directed mutagenesis. Biochem J 323 ( Pt 1):61-64. Palackal NT, Lee SH, Harvey RG, Blair IA and Penning TM (2002) Activation of polycyclic aromatic hydrocarbon trans-dihydrodiol proximate carcinogens by human aldo-keto reductase (AKR1C) enzymes and their functional overexpression in human lung carcinoma (A549) cells. J Biol Chem 277:2479924808. Penning TM, Bauman DR, Jin Y and Rizner TL (2007) Identification of the molecular switch that regulates access of 5alpha-DHT to the androgen receptor. Mol Cell Endocrinol 265-266:77-82. Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, Moore M, Palackal N and Ratnam K (2000) Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution reveals roles in the inactivation and formation of male and female sex hormones. Biochem J 351:67-77. Penning TM and Drury JE (2007) Human aldo-keto reductases: Function, gene regulation, and single nucleotide polymorphisms. Arch Biochem Biophys 464:241-250. Penning TM, Jin Y, Rizner TL and Bauman DR (2008) Pre-receptor regulation of the androgen receptor. Mol Cell Endocrinol 281:1-8. Rizner TL, Lin HK, Peehl DM, Steckelbroeck S, Bauman DR and Penning TM (2003a) Human type 3 3alpha-hydroxysteroid dehydrogenase (aldo-keto reductase 1C2) and androgen metabolism in prostate cells. Endocrinology 144:2922-2932. Rizner TL, Lin HK and Penning TM (2003b) Role of human type 3 3alphahydroxysteroid dehydrogenase (AKR1C2) in androgen metabolism of prostate cancer cells. Chem Biol Interact 143-144:401-409. 149  Yee DJ, Balsanek V, Bauman DR, Penning TM and Sames D (2006) Fluorogenic metabolic probes for direct activity readout of redox enzymes: Selective measurement of human AKR1C2 in living cells. Proc Natl Acad Sci U S A 103:13304-13309.  150  CHAPTER 7 SUMMARY  7.1.  Overall Conclusions The AKRs are a superfamily of metabolic enzymes that are being revealed to  have important roles in humans, catalyzing the bioactivation and deactivation of a wide range of substrates (Jin and Penning, 2007). Focus on these enzymes has been growing since many of these substrates or active metabolites have important pharmacological roles. Hence, it is becoming increasingly realized that AKRs have critical roles in drug metabolism, toxicity, disease development and progression. To date, the major investigations of these proteins have identified isoform-specific functions, characterized their expression and substrate specificities, and provided a description of their mechanisms of action. However, research of AKR-mediated metabolism is still in its infancy and many aspects of their function remain poorly understood. Two examples of factors affecting function that are not well described are the effects of modulating compounds and genetic variation.  The gap in knowledge regarding genetic  polymorphisms is described in a recent review of the state of research for the AKR superfamily,  stating:  “Knowledge  of  the  phenotypes  associated  with  these  polymorphisms [polymorphisms resulting in amino acid changes] is limited, and their impact on the metabolism of xenobiotics remains to be elucidated” (Jin and Penning, 2007). The development of lethal cardiotoxicity is the major limitation to the use of anthracyclines in cancer treatment.  This side effect is dose-dependent; however, it  151  displays high levels of patient-to-patient variability leading to unpredictable and irreversible cardiac damage. Though there is no consensus on the mechanisms that lead to cardiac damage, the leading hypotheses describe metabolism of these drugs resulting in reactive species or anthracycline metabolites that are toxic to heart tissue (Menna et al., 2007). Hence, unpredictable development of cardiotoxicity due to anthracycline drugs, DOX and DAUN being the most commonly used clinically, may, in part, result from variability in the function of enzymes responsible for their metabolism. The enzymes AKR1A1 and AKR1C2 were characterized for their function in metabolizing DOX and DAUN with interest to determine the impact that naturally-occurring allelic variants have on the kinetic parameters describing their reductive metabolism. Prostate cancer is a prevalent cancer in men that results in a large number of deaths each year (estimated 4,300 Canadian deaths per year (Canadian Cancer Society, 2008)). The risk factors that describe increased susceptibility are very general. For screening and early stage treatment to become more effective in minimizing the impact of prostate cancer, better predictors of the disease are needed. The androgens levels in the prostate are an important factor in its normal growth, and are expected to be a major determinant for cancer development (Mostaghel and Nelson, 2008).  The metabolic  enzymes that catalyze the interconversion between the highly potent androgen, DHT, and less potent precursors and metabolites form an important system that regulate androgen levels in the prostate. Hence, the study of these enzymes has considerable potential to contribute to a description of the mechanism by which prostate cancer develops and, ideally, will provide specific and accurate markers for identifying those individuals who have an increased risk of developing the disease. Therefore, AKR1C2 was characterized  152  for its role in deactivating DHT with a focus on the effects of modulator compounds on enzyme function and a comparison of the kinetics of the naturally-occurring AKR1C2 allelic variants to the wild-type enzyme. In the first set of experiments, the wild-type and the naturally-occurring variant AKR1A1 proteins were isolated as recombinant proteins and were functionally characterized for their reductive activities (Chapter 2).  Two variant alleles were  associated with reduced in vitro activities compared to wild-type in the reduction of DOX and DAUN and these data were in concordance with the results of parallel activity assays that used standard AKR substrates. Similarly designed studies characterized the function of AKR1C2 (Chapter 3). Using standard AKR substrates, recombinant AKR1C2 protein was confirmed to have comparable activities to the reductase isolated from human tissues. In vitro studies with DOX and DAUN were used to demonstrate that AKR1C2 does not metabolize the anthracycline drugs, a result that is in stark contrast to what had previously been published (Ohara et al., 1995). Based on these findings, and judging the likelihood that an amino acid substitution would result in new protein function is low (based on assumed evolutionary selection for the wild-type allele to code for a fully functional protein), the allelic variants of AKR1C2 were not characterized for their function in metabolizing DOX and DAUN. Further investigations of the AKR1C2 enzyme focused on its role in metabolizing the androgen steroid DHT. First, a novel methodology utilizing ultra high performance liquid chromatography-tandem mass spectrometry was developed for the analysis of DHT and its reduction products 3"-diol and 3!-diol (Chapter 4). The suitability of this  153  method for quantitative analysis of these analytes was demonstrated through the validation of the precision, accuracy, lower limit of quantitation, specificity, and range of the method. The kinetic parameters describing the function of AKR1C2 in reducing DHT to 3"-diol in vitro were determined and the effects of the anti-depressant drugs paroxetine, sertraline, imipramine, and fluoxetine, as activators of this metabolism were measured (Chapter 5). We confirmed that several of these compounds increased the activities of AKR1C2.  Our investigation of the source of the discrepancies between various  researchers in observing this activation identified that it is only observed in low-protein conditions (Griffin and Mellon, 1999; Trauger et al., 2002).  In the presence of  physiologically relevant protein concentrations, enhanced AKR1C2 action was observed and any additional activation resulting from the presence of the anti-depressant drugs was not observed. The allelic variants of AKR1C2 were characterized using a transfected insect cell model in two independent cell-based metabolic assays (Chapter 6).  The kinetic  parameters describing reduction of a specific fluorescent probe substrate and DHT were determined for the wild-type and variant alleles of AKR1C2. The combined data sets from these assays identified several variant alleles of AKR1C2 that showed significantly reduced rates for metabolizing DHT to 3"-diol. Overall, the findings in this thesis provide new evidence describing the roles of two human AKRs, AKR1A1 and AKR1C2, in the metabolism of anthracyclines and androgens. Several of the allelic variants of these AKR genes were associated with significantly reduced activities, supporting the hypothesis that genetic variation may  154  contribute to the interpatient variability observed in the metabolism of these chemicals. It also became clear from these investigations that some of the data that exist in the AKR literature are misleading. Most surprising was the finding that AKR1C2 does not use DOX or DAUN as a substrate (Chapter 3). Finally, our data reveal a significant effect of soluble proteins on in vitro derived kinetic data by enhancing reductase activities (Chapter 5). The in vitro kinetic data is ultimately most useful when it provides insight to the in vivo situation. Though it is tempting to extrapolate these in vitro findings to the in vivo situation, it appears that the in vitro activities of AKR proteins are highly complex with sensitivities to several influences. Hence, given the state of knowledge, this extrapolation may lead to false predictions. This emphasizes the importance of continuing studies of AKRs in vitro for developing reliable methods for in vitro-in vivo correlation, and the initiation of additional animal and clinical studies to collect much needed data regarding the in vivo actions of these proteins.  7.2.  Implications of Findings The characterization of the naturally occurring allelic variations in AKR1A1 and  AKR1C2 in this thesis provides the most comprehensive data sets to date on the impact of ns-SNPs on functional activities of these enzymes. The findings of the in vitro and cell-based assays answer whether or not allelic variants may contribute to variability in metabolism of the anthracyclines DOX and DAUN and the androgen steroid DHT. The biochemical evidence collected demonstrates that they, indeed, may. Several variant alleles are associated with decreased AKR activities in the reductive metabolism of these  155  compounds. In addition, these alleles are observed to occur with varying frequencies in human populations. Together, these data form a solid foundation of biochemical data for further research that will determine the importance of allelic variation and associated variability in metabolism for determining patient phenotype. In the case of anthracycline metabolism, significant reductions in the enzymatic efficiencies of AKR1A1 were measured for the alleles coding for the N52S and E55D substitutions. Consequently, patients carrying these alleles and expressing the enzyme variants will be expected to have reduced rates for metabolizing DOX and DAUN to their alcohol metabolites.  Clearly, the simple scenario in which the anthracyclines are  metabolized solely by AKR1A1 and there are no compensatory pathways for metabolizing or eliminating the drugs is highly unlikely. Therefore, the presence or absence of an AKR1A1 variant allele will not provide a clear distinction between patients that are extensive or poor metabolizers of anthracyclines or patients that are at heightened risk for developing cardiotoxicity and those that are not. Rather, atypical AKR function may contribute to the well-documented interpatient variability in response to DOX or DAUN treatment. In testing this prediction we will advance our knowledge for the contribution of AKR1A1 to the metabolism of anthracyclines. To date, this information has been extrapolated from studies using genetically-engineered animal models, in vitro or cell-based models, or through non-specific induction or inhibition of specific enzymes. Based on the data collected for the AKR alleles, if AKR1A1 contributes substantially to the total metabolism of DOX or DAUN and few compensatory deactivation and elimination pathways exist, large differences in phenotypes are expected between patients that carry and those that do not carry the variant alleles. Alternatively, if AKR1A1  156  metabolism has a minimal contribution, phenotypic differences are not expected to be strongly associated with occurrence of the variant alleles. Further, since metabolism is hypothesized to be an integral component to anthracycline-associated cardiotoxicity, comparisons of cardiac dysfunction between patients carrying and not carrying the variant allele should reveal if there is strong dependence on AKR1A1-mediated metabolism for the development of this side effect. In the deactivation of DHT, reduced AKR1C2 activities are expected for patients expressing F46Y and L172Q variants for this enzyme. The prevailing evidence to date suggests that AKR1C2 is the major enzyme responsible for DHT deactivation, therefore, it is expected that individuals with alleles associated with impaired AKR1C2 function will have elevated or persistent levels of DHT in the prostate. This is predicted to result in excess activation of the androgen receptor and therefore, these individuals should have an increased risk for developing prostate disease. As described above for the contribution of AKR1A1 to anthracycline metabolism, it is unlikely that a single enzyme catalyzing a single metabolic pathway will be the sole determinant for DHT levels. Many enzymes expressed in the prostate catalyze the synthetic and degradative metabolism of DHT, and it is the total overall flux of this system of enzymes that determines DHT levels. The contribution of AKR1C2 or other specific isoforms is not yet known, and similar to the case of anthracycline metabolism, future studies should aim to elucidate this information. The large enhancements of AKR1C2 activity observed in presence of bovine serum albumin have not been previously reported.  This finding has important  implications for how in vitro data should be used for making in vivo predictions. A comparison of the in vitro and cell-based kinetic parameters determined for AKR1C2  157  reduction of DHT to 3"-diol demonstrates that biases can be introduced by the experimental model used. The in vitro determined Vmax and K’ estimates were 0.948 nmol/min/mg protein and 2.26 µM, respectively (Chapter 5). The comparable cell-based determined Vmax and Km estimates were 0.168 nmol/min/mg protein and 8.3 µM, respectively (Chapter 6); however, these activity values were normalized to the total cellular protein and notably, these activities include a substantial contribution by endogenous insect reductases. Assuming that AKR1C2 makes up 1-2% of the total protein for these cells, which was confirmed by a Western blot using purified recombinant enzyme as a standard (data not shown), this provides us with a cell-based determined Vmax value of 10-20 nmol/min/mg protein, which is markedly different than the in vitro estimate. It is tempting to speculate that an enhanced rate of enzyme activity is being observed in the cell-based assay due to solubilized intracellular proteins, which is therefore, more comparable to data collected with the addition of BSA. Alternatively, this discrepancy may reflect differences between the bacteria and insect-expressed proteins.  However, there is not yet sufficient data to judge the merit of these  speculations. Some of the data collected in these studies contradict what has previously been reported in the AKR literature. Without doubt, inter-laboratory variation is a likely contributor to these data discrepancies; however, the primitive state of knowledge (relative to other metabolic systems such as cytochrome P450 enzymes) also stems from the underdevelopment of tools and methodologies for the study of AKRs. Clearly, this reflects the fact that research interest for these enzymes has only developed relatively recently and demonstrates that much more research attention is necessary to provide a  158  reasonable understanding of the importance and impact of AKR-mediated metabolism to health. It will continue to be important to critically evaluate the sources for discrepancies as data from future studies contest or support our and other researcher’s results. With proper attention, these reviews should reveal the best practices for collecting and interpreting AKR information and benefit the translation of this information to patient care.  7.3.  Scope and Limitations of Research The research presented in this thesis was conducted using a combination of in  vitro and cell-based assays. These model systems were selected because they provided environments that lacked the enzymes being investigated (AKR1A1 or AKR1C2), thus, they allowed the function of the recombinant proteins to be studied in isolation. This is particularly important since the differences in function expected to result from single amino acid substitutions are small and may be difficult to distinguish against the background signal of more complex systems (e.g., in vivo).  The results that were  collected using cell-based assays (Chapter 6) point to this challenge. The transfected insect cells used, though shown to not express AKR1C2, contained endogenous reductases that contributed to the observed metabolism.  Hence, subtle enzymatic  differences from wild-type became more difficult to distinguish. An in vitro system using purified recombinant proteins could have been used for all of the experiments and provides a null background, but there are good indicators that it may generate data that is disparate from the in vivo situation. Reports of protein lability (Dufort et al., 1999; Dufort et al., 2001) and our observation of enhanced  159  activities in presence of other dissolved protein (Chapter 5) show that the reductases are highly sensitive to in vitro conditions. Unfortunately, AKR research has not yet provided sufficient in vivo data to properly judge the value of collected in vitro data. This is in stark contrast to examples of cytochrome P450-catalyzed metabolism where relevant in vivo data have been collected and provide good direction on the best practices for conducting and applying the findings of in vitro studies (Obach, 1999; Ito and Houston, 2004). The cell-based assays used to study the effect of allelic variation in AKR1C2 (Chapter 6) were selected because they were expected to better simulate the in vivo environment for the reductase. Human tissue-derived cell cultures may provide a more relevant model to study the human reductases; however, the ubiquitous expression of the reductases suggests that they will be expressed in most, if not all, of these cell lines and will contribute to a higher magnitude background signal. Thus, the use of these more complex cell systems will require surveying each of these cell lines for AKR expression and, with high likelihood, employing techniques such as gene silencing to lower endogenous expression or validating methods to mathematically subtract the background signal. However, there is growing evidence that gene-silencing systems, such as RNAi, are not specific and often, the expression of a number of off-target products are affected. Consequently, extensive research effort will need to be invested in order to use human cell lines to collect relevant experimental data for select AKR proteins. Bacteria and insect cell expression systems were used for in vitro studies and cellbased assays, respectively. These two expression systems provide various benefits for these applications (Crespi and Miller, 1999). Bacteria-expressed proteins were isolated for in vitro studies because of the ease and speed of growing and scaling up cultures for  160  high yield production. The cell wall of bacterial cells is a barrier to chemicals reaching the intracellular space that does not exist for human cells. Therefore, insect cells were used for the cell-based assays. Benefits of insect expression are that it provided proteins that have undergone the post-translational modifications expected for human proteins (Kidd and Emery, 1993; Hood et al., 1998; Hegedus et al., 1999) while providing a null background for the human reductase. Notably, insect cells could have been used to provide proteins that could be purified for in vitro assay use. However, the functional characterization of the bacterially-expressed AKR1A1 and AKR1C2 provided comparable data to what has been reported for the proteins isolated from human tissue (Chapters 2 and 3) and were expected to be fully functional and appropriate for their applications. For some of the variant alleles associated with altered activities there is insufficient data to accurately describe allele frequencies in human populations. Therefore, studies on the clinical relevance of these alleles would be premature. In the case of AKR1C2, the V38A, V38I, H47R, and R258C variant alleles were not found in the genotyping studies performed for the HapMap project and there is no frequency data reported yet for the S87C, V111A, H170R, K179E, or K185E variant alleles. If these alleles, in fact, do not occur with any significant presence in humans, they will have little or no physiological importance for determining patient variability in metabolism. However, the data for these variant alleles do provide new protein structure-function information that identifies critical residues for the function of the enzyme, which may be valuable for basing predictions of the likelihood for newly identified SNPs to alter enzymatic activities.  161  The polymorphisms in AKR genes studied in these experiments are the single nucleotide polymorphisms that result in amino acid changes. A larger proportion of the identified genetic variation in these genes occurs in non-coding regions, and will be important for determining regulation of AKR expression or variation in splicing in specific tissues or organs. These changes in gene expression may occur in a constitutive manner or in response to stimuli. Hence, these variants will also contribute to the differences between individuals in drug metabolism and patient outcomes. The study of these genetic variants will provide necessary data to understand the full impact of genetics on these enzymes. Though there is strong belief that gene sequences and gene expression are intricately linked to disease states, the correlation of genetics and disease will not be simple. The studies conducted here have focused on a single aspect (AKR-mediated metabolism) of what are highly complex patient responses (development of anthracycline-associated cardiotoxicity and prostate cancer). The likelihood that a single variant allele in a single enzyme will be the sole determinant of an individual’s susceptibility to one of these disorders is low. Further, the data show that the AKR functions are compromised, not eliminated, by allelic variation; therefore, dramatic patient phenotype differences are unlikely to result from these single variations.  It  follows that if genetic information can be related to these disorders, it is more likely going to be a combination of alleles occurring in genes of several target proteins. For example, though androgen levels are important for regulating androgen receptor signaling, polymorphism in the androgen receptor itself, or co-activator and co-repressor proteins will also alter androgen signaling and need to be considered. Likewise, other  162  drug metabolizing enzymes, transporters, receptors, or modulator proteins, are all components of an individual’s response to anthracyclines or androgen steroids. Complicating this tremendously complex realm of a variety of genetic manipulations that can occur in a vast number of genes is the fact that genetic factors will only in part determine individual response. Non-genetic environmental factors, including nutrition, lifestyle, or concomitant disease, will need to be considered together with genetic data in order to ascertain if genetics indeed contributes significantly enough to be a determinant for specific diseases. Clearly, in order to establish these multi-factorial associations, an exceedingly large amount of data must be collected and analyzed.  Therefore, the  information collected towards understanding the function of AKRs should be considered one step towards elucidating a source for these disorders and identifying the potential genetic factors that contribute to an individual patient’s response. Towards this goal, these data sets describing AKR1A1 and AKR1C2 will need to be pooled with large amounts of genomic data from various research focuses and collectively used to identify genetic markers that correlate strongly with disease. AKR1A1 and AKR1C2 are involved in the metabolism of a variety of other chemicals with pharmacological or toxicological effects. Therefore, the role of allelic variation in altering these metabolic pathways has implications for individual responses or disease risk conferred by these substrates. For example, AKR1C2 is involved in the detoxification of the nicotine-derived nitrosamine-ketone, a highly carcinogenic compound that is derived from tobacco smoke (Atalla et al., 2000). Hence, AKR1C2 allelic variants may be associated with heightened risk of cancer development due to cigarette smoking. The functional characterizations of AKR1A1 and AKR1C2 clearly  163  demonstrate that the polymorphic enzymes have variable reductase activities; however, these findings cannot be immediately extrapolated to all substrates for these enzymes. The impact of amino acid substitution is potentially substrate-dependent and the variant proteins may function normally or atypically when metabolizing other substrates. Thus, the polymorphic AKRs will need to be tested experimentally with these alternative substrates to ascertain if these differences do, indeed, exist.  7.4.  Suggested Future Research Directions From the findings of this thesis, a number of experiments can be proposed. The  research directions suggested here will focus on: (i) using the methods and technologies that have been developed for future applications, (ii) future biochemical studies, and (iii) future patient genotype-phenotype correlation studies.  7.4.1. Application of the developed technologies Few of the AKR proteins have been studied for the impact of polymorphism on the function (Section 1.1.4.). Using the in vitro systems and cell-based models that were used to study AKR1A1 and AKR1C2, the allelic variants for each of the human AKR genes can be systemically tested and compared to wild-type to characterize their functional consequences. The methods that are used are fairly standard including cloning the human gene, introducing the known polymorphisms using site-directed mutagenesis, expressing the human protein, and characterizing its in vitro activities. Therefore, these systems should provide a developed model that can be applied to other AKRs to express, purify, and characterize the function and impact of allelic variation for these other AKRs.  164  In addition, it is likely that the use of the insect cell expression system (Chapter 6) will provide a null background for all of the human AKR proteins, therefore simple modification should allow these cell-based systems to be used. A LC/MS/MS methodology was developed and validated for the measurement of DHT and its reduction products, 3"-diol and 3!-diol, for in vitro metabolic studies (Chapter 4). With simple modifications the utility of this method can be extended. The procedures, as described, will allow the simultaneous measurement of testosterone, dihydroepiandrosterone, or androsterone, which are precursors or metabolites of DHT. These androgens all contain a free hydroxyl group that can be reacted to form fluoro methylpyridinium ether derivatives. In order to analyze other precursors of DHT that do not contain a hydroxyl group, specifically 5"-dione and 4-dione, an alternative derivatization procedure will need to be considered.  Oxime and 2-hydrazine-1-  methylpyridine derivatives formed by reacting ketone groups are good candidates because they have already been shown to be useful for mass spectrometric analysis of testosterone and DHT (Higashi et al., 2005; Kalhorn et al., 2007). The analysis of biological samples will be critical to understanding androgen metabolism in vivo, and the extraction, derivatization, and mass spectrometric protocol of these methods can be applied to these samples. Hence, with simple modifications this LC/MS/MS method will provide measurement of the DHT, and a number of its precursor and metabolites, making it a valuable tool for providing a complete profile of the androgens present in the prostate and allowing the study of the enzymes that catalyze their synthesis and metabolism.  165  7.4.2. Biochemical investigations A major limitation to evaluating the significance of the functional differences resulting from allelic variation stems from not knowing the contribution of these enzymes to the overall metabolism of the substrates of interest (i.e., DOX and DAUN with respect to anthracycline-associated cardiotoxicity, and DHT with respect to prostate cancer). Several biochemical experiments can be suggested to address this knowledge gap. If a specific enzyme contributes largely to the full metabolism, altering its function within a complete metabolic system will result in a significant change in levels of metabolites and parent compound.  This strategy has been used to demonstrate the critical role of  carbonyl reductase in anthracycline metabolism using mouse models. Levels of the alcohol metabolites after anthracycline dosing and markers of heart toxicity were both reduced in a carbonyl reductase knock-out mouse (Olson et al., 2003). Overexpression of this reductase in the heart provided the opposite effects, increasing the development of cardiotoxicity (Forrest et al., 2000).  Experiments of this nature should be used to  measure the contributions of AKR1A1 to anthracycline metabolism, and AKR1C2 to DHT metabolism in relevant cell lines or tissues.  For example, if DHT levels  significantly increase upon specific inhibition or suppression of AKR1C2 in prostate cells, this is clear evidence that AKR1C2 is critical to DHT levels.  Then, the  polymorphic nature of AKR1C2 may be a crucial factor affecting cellular DHT levels. The use of in vitro and cell-based assays offers the possibility of combining AKRmediated metabolism with toxicological measurements within a single system.  For  example, the toxicities of anthracyclines can be investigated in cell-based assays to distinguish whether metabolites or reactive species are the toxic species. If the alcohol  166  metabolites are directly responsible, DOX or DAUN may still appear to have similar toxicities to DOXol or DAUNol using cells that contain functioning metabolic reductases that can generate the alcohol metabolites. However, the differences in toxicity should become more apparent if reductase function is impaired, such as the N52S or E55D variants for AKR1A1, which are less capable at generating the metabolites. It will also be valuable to measure reactive species generation or iron release to support or refute the reactive species generation hypothesis for cardiotoxicity. Models can also be set up to further our understanding of androgen metabolism and its regulation of androgen receptor activation. A cell system can be reconstituted with a functional androgen receptor, and individual or combinations of reductases that metabolize androgens. Treatments with various androgens would be expected to result in variable levels of androgen receptor activation depending on whether these androgens are, or can be, converted to potent receptor ligands, such as DHT. Combined with measurement of the androgen levels, this provides a very representative system for studying the synthetic and degradative androgen pathways and the resulting receptor activation in the prostate. A valuable use of in vitro-derived metabolic kinetic data is to predict the in vivo clearance of drugs (Houston, 1994).  This has been explored extensively for the  cytochrome P450 enzymes. The current understanding of AKR-mediated metabolism is considerably less established and provides many potential avenues for in vitro research. The problems that still exist in scaling in vitro kinetic data for cytochrome P450mediated pathways demonstrate the difficulty in establishing these in vitro-in vivo relationships. For example, there are clear differences in the kinetic data collected from the use of recombinant enzymes, microsomes, hepatocytes, or liver slices, such that there  167  is not yet consensus on which system best predicts hepatic clearance (Brandon et al., 2003). In vitro-in vivo extrapolation for cytochrome P450s uses scaling factors such as P450 protein content per liver cell and number of liver cells in the full organ to estimate hepatic clearance based on in vitro data (Houston and Galetin, 2003).  It is not  unimaginable that similar scaling factors for extrapolating in vitro AKR data can be developed. Towards this goal, it will be necessary to quantitate expression levels of specific AKR isoforms in many tissues to derive population-averaged estimates. Potentially, these experiments will identify differential expression patterns due to factors such as disease states, age, diet, sex, or pregnancy, which will be relevant for understanding temporal or environmental influences for altered AKR metabolism. It will also be valuable to systematically test the effects of in vitro experimental conditions on the determined kinetic parameters. The combination of these data will provide a pool of in vitro kinetic data and various extrapolation strategies that can be judged for their accuracy by their comparison with in vivo data collected using probes that are solely or predominantly metabolized by a single AKR isoform, another area for future research. The outcome of these investigations will develop methods to scale in vitro data to reliably predict in vivo observations, and recommend best practices for conducting AKR research.  7.4.3. Patient genotype-phenotype correlation studies An important component to predicting the significance of pharmacogenetic investigations is having accurate estimate of the frequency of each variant allele in relevant patient populations. Currently, the data sets for several AKR1A1 and AKR1C2 variant alleles are not complete and the population frequencies for a number of SNPs are  168  unknown. It is especially important to establish the frequency of those alleles that have dramatically altered function. These alleles have the most potential to alter a patient’s metabolism of anthracyclines, androgens, and other relevant xenobiotic and endogenous compounds. Should thorough and accurate genotyping experiments show that the variant alleles are absent or occur only in an exceedingly low frequency, it is unlikely that they will be useful in describing an individual’s risk of a specific disorder, especially for conditions such as drug toxicities or cancers that occur in relatively large proportions of their relevant populations. Alternatively, a variant allele associated with dramatically altered function and occurring with a measurable prevalence in a population of individuals with the specific condition or disease has potential to be related with the phenotypic differences. Clearly, it is this latter group of alleles that are of most interest for elucidating the roles for specific genes and protein products in determining a patient’s response to treatment or disease risk, and their potential use as biomarkers for patient susceptibility. The frequencies of occurrence for the variant alleles of AKR1A1 or AKR1C2 will become more important if they are shown to exist in patients that have anthracyclineassociated cardiotoxicity or prostate cancer development, respectively. In those cases, future studies should compare the frequency of an allele’s occurrence between patients showing anthracycline-induced cardiotoxicity or prostate cancer development and the normal population. A significant difference between the affected and normal populations would support that the role of metabolism by these specific enzymes is associated with the development of these disorders and that the presence of these alleles is predictive of cardiotoxicity or prostate cancer.  169  If an association between allelic variants and increased risk for cardiotoxicity or prostate cancer can be established, we can apply this understanding to minimize or eliminate the risk of these disorders.  Genotyping patients before the initiation of  anthracycline therapy will identify individuals that are at increased risk for developing cardiotoxicity, thereby allowing patients and physicians to choose alternate therapies that will minimize their risk. Dosage regimens of anthracyclines, with increased likelihood to be effective in eliminating cancer, could be increased for patients that are at low risk of developing cardiotoxicity. Similarly, men who are at high risk for developing prostate cancer can be identified and undergo more frequent screening to increase the likelihood that aggressive cancers are detected early in their development. Ideally, the functional characterization of the AKRs will provide this level of personalized medicine and will have important benefits for minimizing or eliminating the impacts of anthracyclineassociated cardiotoxicity and prostate cancer.  170  7.5.  References  Atalla A, Breyer-Pfaff U and Maser E (2000) Purification and characterization of oxidoreductases-catalyzing carbonyl reduction of the tobacco-specific nitrosamine 4-methylnitrosamino-1-(3-pyridyl)-1-butanone (NNK) in human liver cytosol. Xenobiotica 30:755-769. Brandon EF, Raap CD, Meijerman I, Beijnen JH and Schellens JH (2003) An update on in vitro test methods in human hepatic drug biotransformation research: pros and cons. Toxicol Appl Pharmacol 189:233-246. Canadian Cancer Society/National Cancer Institute of Canada: Canadian Cancer Statistics 2008, Toronto, Canada, 2008. Crespi CL and Miller VP (1999) The use of heterologously expressed drug metabolizing enzymes--state of the art and prospects for the future. Pharmacol Ther 84:121131. Dufort I, Labrie F and Luu-The V (2001) Human types 1 and 3 3 alpha-hydroxysteroid dehydrogenases: differential lability and tissue distribution. J Clin Endocrinol Metab 86:841-846. Dufort I, Rheault P, Huang XF, Soucy P and Luu-The V (1999) Characteristics of a highly labile human type 5 17beta-hydroxysteroid dehydrogenase. Endocrinology 140:568-574. Forrest GL, Gonzalez B, Tseng W, Li X and Mann J (2000) Human carbonyl reductase overexpression in the heart advances the development of doxorubicin-induced cardiotoxicity in transgenic mice. Cancer Res 60:5158-5164. Griffin LD and Mellon SH (1999) Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc Natl Acad Sci U S A 96:1351213517. Hegedus DD, Pfeifer TA, Theilmann DA, Kennard ML, Gabathuler R, Jefferies WA and Grigliatti TA (1999) Differences in the expression and localization of human melanotransferrin in lepidopteran and dipteran insect cell lines. Protein Expr Purif 15:296-307. Higashi T, Yamauchi A, Shimada K, Koh E, Mizokami A and Namiki M (2005) Determination of prostatic androgens in 10 mg of tissue using liquid chromatography-tandem mass spectrometry with charged derivatization. Anal Bioanal Chem 382:1035-1043. Hood SR, Shah G and Jones P (1998) Expression of cytochromes P450 in a baculovirus system. Methods Mol Biol 107:203-218.  171  Houston JB (1994) Utility of in vitro drug metabolism data in predicting in vivo metabolic clearance. Biochem Pharmacol 47:1469-1479. Houston JB and Galetin A (2003) Progress towards prediction of human pharmacokinetic parameters from in vitro technologies. Drug Metab Rev 35:393-415. Ito K and Houston JB (2004) Comparison of the use of liver models for predicting drug clearance using in vitro kinetic data from hepatic microsomes and isolated hepatocytes. Pharm Res 21:785-792. Jin Y and Penning TM (2007) Aldo-keto reductases and bioactivation/detoxication. Annu Rev Pharmacol Toxicol 47:263-292. Kalhorn TF, Page ST, Howald WN, Mostaghel EA and Nelson PS (2007) Analysis of testosterone and dihydrotestosterone from biological fluids as the oxime derivatives using high-performance liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 21:3200-3206. Kidd IM and Emery VC (1993) The use of baculoviruses as expression vectors. Appl Biochem Biotechnol 42:137-159. Menna P, Recalcati S, Cairo G and Minotti G (2007) An introduction to the metabolic determinants of anthracycline cardiotoxicity. Cardiovasc Toxicol 7:80-85. Mostaghel EA and Nelson PS (2008) Intracrine androgen metabolism in prostate cancer progression: mechanisms of castration resistance and therapeutic implications. Best Pract Res Clin Endocrinol Metab 22:243-258. Obach RS (1999) Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: An examination of in vitro half-life approach and nonspecific binding to microsomes. Drug Metab Dispos 27:1350-1359. Ohara H, Miyabe Y, Deyashiki Y, Matsuura K and Hara A (1995) Reduction of drug ketones by dihydrodiol dehydrogenases, carbonyl reductase and aldehyde reductase of human liver. Biochem Pharmacol 50:221-227. Olson LE, Bedja D, Alvey SJ, Cardounel AJ, Gabrielson KL and Reeves RH (2003) Protection from doxorubicin-induced cardiac toxicity in mice with a null allele of carbonyl reductase 1. Cancer Res 63:6602-6606. Trauger JW, Jiang A, Stearns BA and LoGrasso PV (2002) Kinetics of allopregnanolone formation catalyzed by human 3 alpha-hydroxysteroid dehydrogenase type III (AKR1C2). Biochemistry 41:13451-13459.  172  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0070830/manifest

Comment

Related Items