A STUDY OF DT-DIAPHORASE AS A PRODRUG-ACTIVATINGENZYME
Veet Misrri
A thesis submitted in confomity with the requirements for the degree of Doctor of Philosophy, Department of Medical Biophysics University of Toronto
8 Copyright by Veet Misra 2000 National Library Bibliothèque nationale 1*1 ofCanada du Canada Acquisitions and Acquisitioris et Bibliographie Services services bibliographiques 395 Wellington Street 395, rue WdIiigMm Onawa ON K1A ON4 OttrrwaON K1AON4 canada canada
The author has granted a non- L'auteur a accordé une licence non exclusive Licence aiiowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or sel1 reproduire, prêter, distribuer ou copies of this thesis in microfonn, vendre des copies de cette thèse sous paper or electronic formats. la forme de micro fi ch el^ de reproduction sur papier ou sur format électronique.
The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or othewise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. A Study of DT-Diaphnrase as a Prodrug-Activating Enzyme Doctor of Philosophy, 2000 Veet Misrzr Graduate Department of Meâical Biophysics University of Toronto
Abstract
DT-diaphorase (DTD) is an enzyme that acts on its substrates in a single-step two electron transfer reaction. It has a dual role both as a prodrug-activating enzyme and as a protector against carcinogenesis mediated by xenobiotics. A fiequently occurrïng base change, a C to T transition in exon six of the DTD gene, has been associated with low to undetectable DTD activity. Also, tumor cells Vary in their DTD levels, which poses a limitation for the use of drugs that rely on DTD for their chemotherapeutic effects.
To test directly the effect of the base change on enzyme activity, mammalian cells were transfected with expression vectors containing DTD cDNA with a C or T nucleotide at position 609. Results indicated that this base change is a mutation in the DTD gene which results in impaired activity and reduced protein stability.
The relationship between the status of the DTD nucleotide 609 allele and enzyrnatic activity was assessed in a panel of human fibroblast ce11 strains. DTD activity was, on average, Zfold higher in cells that were homozygous for the wild-type allele compared to heterozygotes. Cells that were homozygous for the variant allele displayed low to undetectable DTD activity and trace Ievels of DTD protein. A unique ce11 strain that is heterozygous yet has low to undetectable DTD activity and protein was shown to express DTD mRNA exclusiveIy fiom the variant allele. These results demonstrate that the statu of the nucleotide 609 ailele can, to a first approximation, predict for DTD
activity and function.
The possibility of using DTD in gene therapy was investigated by developing
adenoviral vectors that detiver DTD to sensitize cells that are deficient in this enzyme.
Infection of mitomycin C-resistant cells with an adenoviral vector that canies a minigene
for wild-type DTD rendered them sensitive to this dmg. The degree of sensitization
depended on multiplicity of infection..
These studies demonstrate the role of the C to T mutation in affecting DTD
activity and function and provide "proof of principle" data for the use of DTD as an enzyme for vims-directed enzyme-prodrug therapy. Dedicared ro the memory of my Mother, Shanti Misra. You are a part of me. Acknowledgements
As 1 look back on the six years that I carried out my project, ifeel hwly blessed that 1 have gotten to know so mony kind and generous people. Some have corne and gone from this Depurtrnent, but will always be rernembered. For the ones who rernained close by, 1 consider them piena5 for Ive.
To my parents: Mother, Ifeel the gifi of your love and strength in my very being. I believe rhat rhis is how one's spirit lives on. My memories of you are as real as what I see before my eyes. Everything in my ive has been made possible because of you. Father, when 1see you, I see someone who has dedicated his heart and soul to ensuring a good life for my Morher and myself: Ipledge to aïways love and take cure of you. Of all the people 1 have ever known. you are truly one of the most kindest. And 1 hope that I have done you proud.
To my colleagues and dearji-iends, Dr. Tricia Meh and Dr. Jonathan Tunggal, I cannot ask for more in a fiiend You have my complete trust. I don 't wanr to think about how ordinary my Ph.D days would have been without the laughs that we had. I also don? how hm!you two put up with al1 my complaining. But that's whar it S al1 abour, isn 't if? We know that we can count on each other for understanding. I consider you both remarknble people and will always look with great admiration to see where your lives are taking you.
I have never known anyone ro be so consistently generous and an upbeat fî-iend ro so many as Bob Kuba. it is hard to imagine my Ph. LX experience without the light- hearted sense of humor thut I have gotten used ro almosr each day. You can always count on Dr. David Cowan to teIl it like ir is. By doing so, he is not onIy easy to trust. bur lets us in on his insighrfùl perspectives on Ive. Dave is honourable and forthright. Guys. lunch will never be the same again.
To Dr. Andrerv Michael Raurh: I wgs extremely fortunate to have a reacher wirh fhe ~~tmostsense of what il is to be a decent person. You are a Stzident S reacher. ' You were amuzingly diligent in making sure your students were successfül. I enjoyed seeing your excellent analyfical skills at work during our discussions. When it came ru giving me rhe righr advice, you were 100%. The day that you leave teaching will be a major loss for rhe Department. I hope that you will look back on your career as a triumph.
I save my most special rhanks for last. This is for the love of my liye, Susan. il didn 'f fake me long to realize that you are everything I want in a life partner. I see the love rhaf ive share as a higher plane of existence. Norhing else even comes close to taking me rhere. You 're my bestji-iend. Time is precious and I am so grateful to the powers that be fhar the rest of my days wiII be with you. i swear to you that I will always rry to earn your love. You give me happiness to a level I never thought possible, kindness, and so much tvarmth. Also, you have given me one of the greafesf gifis a person can ever receive; a gifr rhut a lot people never get a chance to have during their days on this Earth: Inspiration. Table of Contents
Chapter 1: Introduction ...... 1 1.1 OverviewofIntroduction ...... 2 1.2 DT-Diaphorase @TD) ...... 4 1 2.1 Molecular Biology of DTD ...... 5 1 2.2 Biochemiçal Functions of DTD ...... 7 1.2.3 DTD as a Drug Activating Emme...... 11 1 .2.4 DTD as a Detoxification Enzyme ...... 14 1 .2.5 Substrate-Mediated Induction of DTD ...... 16 1.3 The Role of a Single Base Change at Nucleotide 609 of DTD cDNA ...... 19 1.3.1 Identification of a 609C to T Nucleotide Substitution ...... 20 1.3.2 Frequency Analyses of the Nucieotide 609 Base Change ...... 22 1.3 -3 Heterogeneity of DTD Enzymatic Activity and Chemotherapy ...... -23 1.4 Cancer Gene Therapy ...... 24 1.4.1 Non-Viral Gene Therapy Vectors ...... 25 1 .4.2 Retroviral Gene Therapy Vectors ...... 26 1 43Basic Adenovims Biology ...... 27 1.4.4 First Generation Ad5 Gene Therapy Vectors ...... 31 1.4.5 Rationale for Combination of Gene and Bioreductive Drug Therapy ...... 35 1.5 Thesis Outhe...... ,...... 40 1 -8 References ...... 43
Chapter 2: Transfection of COS- l Cells with DTD cDNA: Role of a Base Change at Position 609 ...... 53
2.1 Abstract ...... 54 3.2 Introduction ...... 54 2.3 Materials and Methods ...... 56 2.3.1 Chernicals and Reagents ...... 56 2.3.2 Preparation of Eukaryotic Expression Vectors ...... 57 2.3.3 Ce11 Culture ...... 61 2.3.4 Transient Transfection of DTD cDNA ...... 61 2.3.5 Assay for DTD Enzymatic Activity ...... 63 2 3.6 Western Blot Analysis ...... 63 2.3.7 Statistical Anaiysis ...... 65 2.4 Results ...... 65 2.4.1 Expression of DTD- and I3TDmT in COS- 1 Cells ...... 65 2.4.2 Western Blot Analysis ...... 66 2.5 Discussion ...... 69 2.6 References ...... 73 Chapter 3: Assessment of the Relationship Between Genotypic Status of a DT- Diaphorase Point Mutation and Enzymatic Activity ...... 76
3.1 Abstract ...... 77 3 -2 Introduction ...... 78 3 -3 Matenais and Methods ...... 80 3 .3.1 Chernicals and Reagents ...... 80 3.3.2 Cell Cultwe ...... 81 3 -3-3 Genomic PCR-Kinf 1 RFLP Assay ...... 81 -)? 3 .3 -4DTD Assay ...... 83 3.3.5 mRNA Profile Assessrnent of the Nucleotide 609 Base Change...... 83 3.3.6 Western Blot Analysis ...... 86 3 -4 Resdts...... 87 3.4.1 The 609T Nucleotide Substitution Correlates with Reduced DTD Activity in a Panel of 45 Normal Human Fibroblast Ce11 Strains ...... 87 3 .4.2 DTD mRNA Profile of 3437T Fibroblasts ...... 89 3.5 Discussion ...... 95 3 -6 References ...... 99
Appendix 3.1: Characteristics of the Human Skin Fibroblast Panel Used For DTD Nucleotide 609 Base Change Frequency Analysis ...... 103
Chapter 4: Exogenous Expression of the Prodrug-Activating Enzyme DT-Diaphorase Via AD5 Delivery ...... 106
4.1 Abstract ...... 107 4.2 Introduction ...... 108 4.3 Materials and Methods ...... 109 4.3.1 Chemicais and Reagents ...... 109 4.3.2 Ce11 Culture ...... 110 4.3.3 Preparation of the Ad5 CMV-Based Shuttle Vector ...... 111 4.3 -4Preparation of Ad5 CMV-Based Shuttle Vectors Containing DTD cDNA ...... 111 4.3.5 Large Scale Production and Purification of Ad5.DTD Vectors ...... 115 4 -3-6 Determination of Ad5 .DTD Viral Titres...... 118 4.3 -7Assay for Wild-Type Ad5 Contamination ...... 119 4.3.8 PCR Diagnostic for Ad5.DTD ...... 119 4.3.9 infection of BE and HT29 Cells with Ad5.B-Galactosidase ...... 121 4.3.10 Western Blot Analysis of HT29 and BE Cells or BE Cells lnfected with Ad5.DTD ...... 122 4.3.1 1 DTD Assay ...... 123 4.3.12 A~~.DTD~~Infection of HT29 Cells ...... 123 4.3.13 DTD Enzyme Activity Kinetics in BE Cells ...... 123 4.3.14 Clonogenic Survival of Cells Exposed to MMC ...... 124 4.3.15 Statistics ...... 126 4.4 Results ...... 126 4.4.1 Generation of Ad5.DTD to Transduce Cells Expressing Low Levels of DTD . 126 4.4.2 Infection of HT29 and BE Cells with Ad5.B-gal ...... 128 4.4.3 Protein Expression of DTD Isoforms in Human Cells Homozygous for the Variant DTD Allele by Ad5.DTD Transduction ...... 131 4.4.4 A~s.DTD- Infection of Human Cells Containing the Wild-Type DTD Allele ...... 131 4.4.5 Activity Kinetics of Exogenous Wild-Type DTD Mediated by A~S.DTD- in BE Cells ...... 134 4.4.6 Clonogenic Survival of Ad5.DTD Infected Cells Exposed to MMC ...... 137 4.4.7 Effect of Titration of MOI of A~S.DTD- on MMC Sensitivity of BE Cells . 140 4.5 Discussion ...... 143 4.6 References ...... 150
Chapter 5: Discussion and Future Directions...... 154
5.1 Role of the 609C to T Nucleotide Substitution in DTD cDNA ...... ,.,. 155 5.1.1 Relationship of C to T Base Change with DTD Enzymatic Activity and Protein Stability ...... 156 5.1.2 Nature of the C to T Mutation...... 157 5.1.3 Genotype vs . Phenotype I : Frequency Analysis of the Variant Allele and the Role for DTD in Xenobiotic Detoxification ...... 159 5.1.4 Genotype vs . Phenotype 2: The C to T Mutation as a Predictor of Drug Response ...... 165 5.1 .5 A~S.DTDm as a Gene Therapy Vector ...... 166 5.2 Future Directions ...... 171 5.2.1 Structure and Properties of the Serine 187 Isofonn ...... 171 5.2.2 Evaluation of the Bystander Effect of CB 1954 In Vitro and In Vivo ...... 172 5.3 Summary ...... 175 5.4 References ...... 176
Appendix 5.1: Transduction Efficiency of Ad5 in Hypoxic versus Aerobic Human Tumor Cells ...... 182
AS .1.1 Introduction ...... 183 AS .1.2 Materials and Methods ...... 183 A5 . I .3 Results ...... 184 A5.1.4 Conclusions and Discussion ...... ~...... 184 List of Figures and Tables
Chapter 1:
Figure 1.1: Schematic diagram of the human DTD gene ...... 6 Figure 1.2: The secondary and tertiary structure of DTD ...... 9 Figure 1 -3: Proposed scheme for DTD activation of MMC ...... 12 Figure 1.4: Scheme for quinone detoxification by DTD ...... 15 Figure 1.5: Gross morphology of adenovim capsid ...... 28 Figure 1.6: Examples of Ad5 packaging and shuttle vecton ...... 33 Figure 1.7: Mechanisrn for the activation of CB1954 by DTD ...... 37
Chapter 2:
Figure 2.1 : Cloning scheme for ~Rc/CMV.DTD- and p~c/~~~.DTD~~~~ ....-58 Figure 2.2: Schematic of pXGH5 hGH expression vector ...... 62 Figure 2.3: Mean DTD activities in COS- 1 cells transfected with plasmid constnicts...... 67 Figure 2.4: Western blot analysis COS- 1 cells transfected with plasmid constmcts...... 68
Chapter 3:
Figure 3.1 : Genomic PCR-RFLP schematic for DTD nucleotide 609 base change ...... 82 Figure 3.2: RT-PCR-RFLP schematic ...... 85 Figure 3.3: Relationship between allelic status of nucleotide 609 base change and DTD activity ...... 90 Figure 3.4: DTD allelic mRNA expression profiles in skin fibroblast strains..... 93 Figure 3.5: Westem Blot analysis of DTD protein levels in normal human skin fibroblast strains and 3437T cells...... 94
Table 3.1: Charactenstics of skin fibroblast strains used for allelic status and DT-diaphorase activity determination...... -88
Table 3.2: Allelic status and DTD activities of 7 skin fibroblast strains taken fiom panel and ce11 strain 343 7T used for DTD mRNA profile studies and Western blot analysis...... 92
vii Appendix 3.1 :
Table A3.1.1: Characteristics of the human skin fibroblast strains used for DTD nul1 allele fiequency and expression studies discussed in Chapter3...... 104
Chapter 4:
Figure 4.1: Scheme for construction of pAE 1sp 1BCMV. a CMV promoter-based Ad5 shuttle vector...... 1 12 Figure 4.2: Cloning scheme for pAE 1sp 1B.DTD- and pAE 1spl B.DTD~~.. 1 14 Figure 4.3: PCR diagnostic anal ysis of ~d5.DTD- and Ad5 .DTD-~ ...... 1 27 Figure 4-4: Western Blot analysis of DTD protein levels in HT29 and BE human colon carcinoma cells...... 129 Figure 4.5: Eficiency of HT29 and BE cells towards Ad5.p-gal infection...... 130 Figure 4.6: Western blot analysis of BE cells der48h infection with A~~.DTD-...... 132 Figure 4.7: Western blot analysis of BE cells &er 48h infection with A~S.DTD~'...... 133 Figure 4.8: DTD activities of uninfected HT29 cells or cells infected with AdSDTD609T ataMOIofI00 ...... 135 Figure 4.9: DTD enzyme activity kinetics in BE cells infected with A~S.DTD~"~' ...... 1 36 Figure 4.10: Clonogenic survival of cells infected with A~SDTD- or A~SDTD~~and exposed to MMC ...... ~...... 1 38 Figure 4.1 1: Clonogenic survival of BE cells infected with A~~DTD~'~'and exposed to MMC as a fiinction of viral titration ...... 141 Figure 4.12: Fold increases of BE cells to MMC sensitization via A~S-DTD- transduction ...... 142
Chapter 5:
Figure 5.1 : Structures of MeDZQ and its analogue, RH1 ...... 170
Table 5.1 : Sumrnary of reports descnbing the occurrence of the nucleotide 609 base change in human populations, its association with either toxin exposure, or cancer incidence...... 161
Appendix 5.1:
Table AS. 1.1 : Relative transduction efficiency of AdS.CMV.luciferase under aerobic vs. hypoxic conditions...... -184 Abbreviations
3 -MC 3-methylcholanthrene AAV adeno-associated virus Ad2 Mastadenovirus type 2 Ad5 Mastadenovirus type 5 Ah aromatic hydrocarbon-responsivenesslocus AHH aromatic hydrocarbon hydroxylase (enzyme system) AHR aromatic hydrocarbon receptor ampR ampicillin resistance gene AP activator protein ARE antioxidant response e lement BGH bovine growth hormone CAR coxsackievirus-adenovirus receptor CHO Chinese hamster ovary CMV cytomegalovinis COQ coenzyme Q Da Daltons DCPIP 2,6-dichlorophenol-indophenol DIC dicurnarol DPNH diphosphopyridine nucteotide reduced DTD DT-Diaphorase EC Enzyme Commission FAD flavin adenine dinucleotide FBS fetal bovine serum GST glutathione S-transferase hGH human growth hormone HSV-TK herpes simplex virus thymidine kinase Kz quinone-containhg vi tamin Ka, catalytic constant K* dissociation constant K, Michaelis constant LTR long terminal repeat mAb monoclonal antibody MCS multiple cloning site MeDZQ methy ldiazindinequinone MHC 1 major histocornpatibility complex class 1 MMC mitomycin C mMP-1 mouse metallothionein-1 promoter M.U. map unit NADH nicotinamide adenine dinucleotide reduced NADPH nicotinamide adenine dinucleotide phosphate reduced NCI National Cancer Institute NQOl NAD(P)H:(quinone acceptor) oxidoreductase 1 NQ02 NAD(P)H:(quinone accepter) oxidoreductase 2 NRU dihydronicotinamide riboside NTR E. coli nitroreductase ORF open reading frame on bacterial origin of replication PAH polycyclic aromatic hydrocarbon PBS phosphate-buf5ered saline PCR-RFLP polymerase chah reaction - restriction fiagrnent length pol ymorphism Ref- 1 redox factor- 1 RGD Arg-Gly-Asp, recognition motif of integnn CO-receptorfor adenovinis ROI reactive oxygen intemediate RT reverse transcriptase TBS Tris-bufTered saline tetR tetracycline tesistance gene TP terminal protein TPNH triphosphopyridine nucleotide reduced UTR untranslated region XRE xenobiotic response element CHAPTER 1
Introduction 1.1 Overview of Introduction
DT-diaphorase (DTD) rnay play an important role in cancer chemotherapy as well as in cancer prevention. As a background to the new data presented in Chapters 2-4 of this thesis, four areas are reviewed: 1) the dual nature of DTD; 2) DTD poIymorphisms;
3) a role for DTD in gene therapy; and 4) an overview of work presented in the data chapters.
The first part of this chapter focuses on the dual nature of DTD as 1) a prodrug activating enzyme; and 2) as a xenobiotic detoxifier; both roles arising fiom its biochemical function as a single-step two electron reductase (Ross et al., 1993). Although this enzyme acts on a wide variety of substrates, most attention has been given to naturally occurring and synthetic quinone-based compounds. Quinones such as mitomycin C (MMC) cm serve as prodnigs that are activated through DTD-rnediated reduction to DNA-alkylating anti-cancer agents. Exploiting the role of DTD as a chemotherapeutic drug-activating enzyme has become an important area of investigation in the field of bioreductive prodrug therapy. In addition, naturally occurring quinones in the environrnent, such as those found in industrial pollutants, can be rendered non-toxic when rnetabolized by DTD. In this regard, DTD can also be viewed as a xenobiotic detoxification enzyme that functions as a defense mechanism against carcinogenesis.
Cloning of the human DTD gene has led to the identification of upstream transcriptional regulatory elements involved in substrate (xenobiotic)-mediated upregulation of this enzyme.
The second part of this chapter describes a commonlysccurring base change at nucleotide 609 of DTD cDNA (located in exon 6 of the human DTD gene) that is predicted to resdt in a proline to serine change at amino acid 187 and has ken associated with reduced catalytic activity (Traver et al., 1992) and decreased protein stability. A number of reports have linked this mutation to a range of diseases, including some cancers, supporting the notion that DTD has a protective role against carcinogenesis, and providing some insight into the physiological Function of DTD. These studies have revealed that this mutation occurs in many diseased and normai reference populations.
The high fiequency of this mutation is important when considering the role of DTD as a xenobiotic detoxifier and as a cancer chemotherapeutic dmg activator.
The third part of this introductory chapter focuses on a possible role for reductively activated dmgs and their activation enzymes in gene therapy. The cornmon occurrence of the nul1 nucleotide 609 allele rnay contribute to the heterogeneity in DTD enzyme activities between individuals. Lack of suficient activating enzyme levels has been an obstacle to the ïmplementation of enzyme-prodnig therapy. The use of DTD in a gene therapy approach using a first generation Ad5 delivery vector is descnbed. Delivery and overexpression of an exogenous DTD minigene may provide a non-invasive means to ensure adequate levels of this enzyme prior to application of bioreductive prodrugs that are DTD substrates.
The final section of this chapter provides an introduction to the work to be presented in Chapters 2 - 4. It also indicates the curent publication status of the work and the roles of collaborators in carrying out the work. 1.2 DTD
DTD was fim characterized by Ernster and Navazio (1958) as a cytosolic-soluble diaphorase (catalyzing loss of color) using 2,6-dichlorophenol-indophenol (DCPIP) as a substrate and was discovered during snidies of the intracellular distribution of nicotinamide adenine dinucleotide reduced (NADFI)- and nicotinarnide adenine dinucleotide phosphate reduced (NADPH)-dependent oxidoreductases. It was found that
DTD utilizes both these cofactors, NADH and NADPH, with equal reactivity; the original nomenclature for these compounds was diphosphopyridine nucleotide reduced (DPNH) and triphosphopyridine nucleotide reduced (TPNH), respectively, hence the narne DT- diaphorase (Ernster et al ., 1960).
Demonstration of DTD activity in cellular extracts or in intact cells has, in general, relied on the specific inhibitory effect of the anticoagulant dicumaroi (DIC) towards DTD. DIC is cornpetitive with respect to the electron donors NADH and
NADPH (Ernster et al., 1960). A number of other names for the enzyme have arisen in the literature based on studies identieing genes coding for enzymatic activities using various substrates. The names include NAR(P)H:menadione reductase, quinone reductase and azo-reductase. A comrnon name for this enzyme, based on its Enzyme Commission
(EC) classification, is NAD(P)H:(quinone acceptor) oxidoreductase 1 (NQO,) and its EC designation is 1.6.99.2. However, in this and subsequent chapters, the term DTD will be used fiom this point forward. 1.2.1 Molecular Biology of DTD
The ability to puri@ DTD fiom various species by inhibitor-based afinity
chromatography techniques (Rase et al., 1976) led to the availability of anti-DTD
monoclonal antibodies (mAbs). Two groups (Williams et al., 1986; Robertson et al.,
1986), isolated rat DTD cDNA clones fiom protein expression Iibraries that were probed
with anti-rat DTD rnAbs. Using rat DTD cDNA as a probe, Jaiswal et al. (1988) cloned
the human DTD cDNA and localized the human DTD gene to chromosome 16 through
human-hamster somatic ce11 hybridization.
Several overlapping cDNA clones were isolated from a human liver cDNA library
giving a composite sequence of 2448 bp and containing a candidate open reading he
(ORF) encoding a protein of 273 amino acids with a rnolecular weight of 30,880 Daitons
(Da). The ORF within the composite human DTD cDNA sequence was followed by a
long 3' untrznslated region (UTR) 1679 bp in length. The 3' UTR contains four potential
polyadenylation signds (AATAAA) at positions 986. 1460. 1838, and 2419 bp. Northem
blot anal ysis (Alwine et al ., 1977) revealed mRNAs approxirnately 1 -2, 1.7, and 2.7 kb in
size corresponding to the la, 2", and 4thpolyadenylation sites, respectively, in the human
hepatocellular carcinoma ce11 line, HepG2 (Jaiswal et al., 1988). Sequence analysis
indicated that the hurnan DTD cDNA and protein are 83 and 85% similar to the rat DTD
cDNA and protein, respectively.
Jaiswal (1991) also identified and characterized a number of genomic clones
corresponding to the hurnan DTD gene by screening a Lphage based human liver
genomic library with the human DTD cDNA. The human DTD gene is approximately 20 kb in length and contains 6 exons interrupted by 5 introns (Figure 1.1). The first exon NF-kB AP-1 AP-2 TATA Exon 1 Exon 2 Exon 3 Exon 4 Exon 5 Exon 6
-740 -645 -447 XRE CCAAT ARE (~0~bp) (116 bp) (-3500 bp) (-2500 bp) (4000 bp)
Figure 1.1: Schematic diagram of the human MD gene containing 6 exons and 5 introns. The 5' flanking region contains a NF-& binding site, a consensus XRE site, a CCAAT box, an ARE (containing an AP-1 binding site), an AP-2 site, and a TATA box. The 3' region contains four potential polyadenylation site, three of which are used. Diagram based on work of Jaiswal et al. (1991) and Yao and O'Dwyer (1995). is 118 bp in length and codes for the initiating methionine and one G for the fmt codon
of the second exon. The largest exon, exon 6, is 1833 bp in length. The extent of exon 6
was based on subcloned gene fragments that were probed with short oligonucleotides
derived fiom the cDNA sequence and, as expected, contained the four potential
pol yadenylation sites.
1850 bp of the 5' flankuig region were sequenced and shown to contain a TATA
box at -34 bp, an activator protein (AP)-2 binding site at -157 bp, an AP-1 site at 462
bp (within an antioxidant response element [ARE] at 447bp), a CCAAT box at 645bp,
and a xenobiotic response element (XRE)starting at -740 bp. A NF-KB binding site was
later localized at -820 bp (Yao and O'Dwyer. 1995). NF-& binding was demonstrated to
play a role in MMC induction of DTD mRNA (peaking at 5-fold in 24 h) by
electrophoretic mobility shift assay. This effect could be abrogated by prior
immunodepletion of the p50 and p65 subunits of NF-icB (Yao and O'Dwyer, 1995).
A related form of human diaphorase, designated as NAD(P)H:(quinone acceptor)
oxidoreductase 2 (NQO?), was identified by Jaiswal and CO-workers (1990) using a
human DTD cDNA probe to screen a human liver cDNA library under low stringency
hybridization conditions. Hurnan liver NQ02 cDNA and protein are 54% and 49%
identical to DTD cDNA and protein, respectively. The NQ02 protein consists of 23 1 amino acids (with a molecular weight of 25,956 Da). A single NQO? mRNA species of
1.2 kb was identified and was not inducible by compounds that induce DTD transcription and is 50-100 times less active in reducing menadione than DTD. NQO, utilizes dihydronicotinamide riboside 0instead of NAD(P)H as a cofactor (Wu et al., 1997). 1.2.2 Biochemical Functions of DTD
At present, a definitive physiological role for DTD has not ken established. It is
primarily a cytosolic enzyme with 5-10% of total cellular activity membrane-bound in
mitochondrial, microsomal, and Golgi fiactions (Riley and Workman, 1992). The liver is
the richest source of the enzyme but other tissues, including heart, lung, and kidney also
exhibit varying DTD activities (Maritus, 1963).
The enzyme functions as a homodimer with each subunit containing a non-
covalently bound with a flavin adenine dinucleotide (FAD) prosthetic group. Li and co-
workers (1995) detennined the crystal structure of rat liver cytosolic DTD by x-ray
diffraction to 2.1 A resolution. Residues 1-220 in each monomer (Figure 1.2A) constitute
a major catalytic domain with alternating a / f3 secondary structure (seven a-helices and
seven f3-sheets separated by loop domains) while residues 221-273 represent a small C-
terminal domain. Like other flavoproteins, the P-sheets tend to twist together and form a
central core surrounded by a-helices but the overall topology does not resemble the
cataIytic domain of other FAD-containing proteins. Rather, the topology of each
monomer resembles Clostridiurn flavodoxin, a bacterial flavin mononucleotide-
containing protein involved in electron transport (Martin et al., 1989) but without
significant sequence identity with DTD. Recently, the crystal structure for human DTD
has been determined (Faig et al., 2000). One of the major differences between rat and human DTD is the manner in which they bind FAD, rendering the rat isoform more efficient in mediating hydride transfer fiom the flavin group to the substrate.
Identification of key residues involved in binding of DTD substrates, cofactors, and inhibitors based on its crystai structure provided insights into its reaction mechanism Figure 1.2: The secondary and tertiary structure of DTD. A) The structure of the monomer showing the a-helices (al - a8) and P-sheets (pl- B9) which are comected by loop domains (Ll-L7) that fold to form the monomer. B) Schematic showing two identical catalytic sites (black dots) fomed at the interface of the dimer by the Ll/L4 (marked with an asterisk) region of one subunit aligning with the L3L5 region (marked with an asterisk) of the other subunit. Taken fiorn Rauth et al. (1 997) which was adapted from Li et al. (1995). and allowed the assignment of specific roles in the reaction mechanism to residues found in the active sites of DTD. Formation of two identical substrate binding sites (Figure
1.2B). occurs upon dimerization of each D'Il3 monomer (Li et al., 1995). Dimerization ais0 positions the isoalloxazine ring of FAD in an appropriate orientation for reduction of the substrate with anchoring interactions fiom loop domains contributed by each monomer. The active sites overlap with the NAD(P)H cofactor binding sites, accounting for the 'ping-pong' reaction mechanisms of DTD (cofactor on-off, then substrate on-off).
Tyrosine 155 and histidine 16 1 stabilize the negative charge of the isoalloxazine ring of
FAD as a result of hydride transfer from NAD(P)H. A second hydride transfer fiom
FADH, to the quinone substrate results in a singly ionized hydroquinonolate product that again is stabilized by both active site residues.
Early studies (Emster, 1987) suggested that one metabolic fùnction of DTD was in the biosynthesis of the natural forms of a quinone-containing vitamin (K2) and coenzyme Q (COQ, and COQ,,). The high sensitivity of DTD to anticoagulant dmgs such as DIC suggested a possible involvement in vitarnin K-dependent biosynthesis of prothrombin and related coagulants, with DTD serving as a vitamin K reductase to generate its hydroquinone form. This species can accept carboxyl groups to act as a carboxyl donor to peptides involved in prothrombin synthesis (Stenflo et al. 1974).
In relation to its mitochondrial location, DTD was also implicated in the maintenance of membrane bound COQ in its reduced antioxidant state, thereby providing protection against fiee radical damage (Beyer et al., 1996). This hypothesis stemmed from studies involving homologues of COQ,including COQ, and COQ,,,incorporated into multilarnellar vesicles; the presence of DTD prevented the oxidation of reduced COQ and inhibited lipid peroxidation. This protective effect was inhibited by DIC. However,
current views (Rauth et al., 1997) of the possible physiological role(s) of DTD focus on
its dual nature as a bioreductive dnig activator and as a xenobiotic detoxifier.
1.2.3 DTD as a Drug Activating Enzyme
Contributions fiom the field of antimicrobials (Müller et al., 19761, dong with
studies of MMC (Iyer and Szybalski, 1963) and the design of hypoxic-ce11 specific
eiectron-afEnic radiosensitizers that react with radiation induced DNA radicals (Adams,
1992), have led to the concept that bioreductive agents can be used as anti-cancer
prodrugs. Interest in their mechanism of action as hypoxic ce11 cytotoxins resulted fiom their initial use as antimicrobiai agents (McCalla et al., 1970). The fact that nitroimidazoles and nitrofurans had a greater effect on anaerobic bacteria than on aerobes, coupled with the discovery of several nitroreductases in E. coli, two of which are oxygen- sensitive, ernphasized the importance of nitroreduction since the cytotoxic metabolite can revert back to the inactive prodrug in the presence of oxygen (Amis, 1957; Müller et al.,
1976).
MMC (Figure 1.3), initially used as an antibiotic isolated from Streptomyces caespitous (Hata et al., 1957), was demonstrated by Iyer and Szybalski (1 963) to alkylate and crosslink DNA upon chemicai or enzymatic reduction. Sartorelli and CO-workers(Lin et al.. 1974) observed enhanced cytotoxicty of MMC under hypoxic exposure conditions.
A mode1 of bioreductive activation of MMC, shown in Figure 1.3, was proposed to account for its differential cytotoxicity under hypoxic versus aerobic conditions Aziridine Ring Opning OH
Figure 1.3: Proposed scheme for DTD activation of MMC and subsequent DNA crosslink damage adapted fiom Ross et al. (1993). subsequent to elucidation of its reaction mechanism (Iyer and Szybalski, 1964). DTD can
reduce MMC in a concerted two electron and oxygen-independent process. However,
MMC can k converted to its semiquinone radical form by one electron reductases such
as NADPH:cytochrome P450 reductase, xanthine oxidase, cytochrome P450, aldehyde
dehydrogenase. and NADH:cytochrome b, reductase (Sartorelli et al., 1994). Metabolism
of MMC by one electron reductases is an oxygen-dependent process since the
semiquinone form of MMC is susceptible to back oxidation to the parent prodmg by
molecular oxygen.
Such one electron reductases are thought to play a major factor in the selective
hypoxic toxicity of MMC. Two electron reduction of the MMC prodrug, which bypasses
semiquinone formation, generates a relatively redox-stable hydroquinone. However, the
hydroquinone undergoes intemal rearrangement with the loss of a methoxy group
followed by azindine ring opening thereby forming an electrophilic center at the C-I
position (Figure 1.3). Alkylation of DNA results fiom attack of electrophilic C-1 by
nucleophilic bases, especially by guanine and cytosine (Iyer and Szybalksi, 1964).
Subsequent loss of the carbarnate group of MMC generates another electrophilic center at
the C-10 position and stimulates fiirther reaction with DNA leading to intra-strand
linkage or crosslinking of complementary strands; the latter of which is thought to be the
basis for the lethal effect of MMC. Tomasz and CO-workers(1987) have identified a
number of specific rnonofünctional and bifunctional DNA adducts upon chernical or
enzymatic activation of MMC.
MMC typically displays only 1.5 - 3.0 fold hypoxic versus aerobic ce11 toxicity
(Ross et al., 1993, in lems of concentration required to achieve 1 log ce11 kill). Gustafson et al. (1 996) determined the relationship between DTD levels and the toxicities of various bioreductive cytotoxins in Chinese hamster ovary (CHO) cells transfected with human
DTD. In this study, in agreement with previous reports (Riley and Workman, 1992),
MMC was considered a poor DTD substrate since a saturation effect, i.e. the amount of
MMC required to achieve one log ce11 kill, was constant at enzyme activities greater than
500 nrnol2,6-dichlorophenol-indophenol(DCPIP) reduced 1 min 1 mg protein.
1.2.4 DTD as a Detoxification Enzyme
DTD metabolizes a diverse range of substrates, including quinone-based xenobiotics that act as potential carcinogens via redox cycling and subsequent DNA darnage (Tikkanen et al., 1983; Chesis et al., 1984). As shown in Figure 1.4 for rnenadione, it has been demonstrated that a nurnber of endogenous and exogenous quinones are reduced by enzymes such as NADH: cytochrome b, reductase and NADPH: cytochrorne P450 reductase by one electron to semiquinones which can be readily reoxidized by molecular oxygen with concomitant formation of the superoxide radical
(O,--) (Bachur et al., 1979; Powis et al., 198 1). By interaction with superoxide dismutase, the superoxide radical can be converted to hydrogen peroxide (HzO,) which subsequently, in the presence of intracellular metals such as ferrous ion, can be cleaved to give rise to
DNA-damaging reactive oxygen intermediates (ROIs) such as the hydroxyl radical (OH.).
ROIs have been shown to cause chromosomal damage and to act as mutagens and thus promote tumor progression (Cerutti, 1985). Conversely, H,O, can be cleaved by catalase to water. Quinone snpcroxi& 1Dismutase
H20 + O2 DNA Damage
Figure 1.4: Scheme for quinone detoxification by DTD illustrated with menadione. One step reduction of the parent quinone by DTD bypasses the formation of the relatively redox-labile semiquinone radical. The semiquinone is susceptible to back-oxidation to the parent quinone by reaction with rnolecular oxygen and subsequent generation of DNA- darnaging ROIS whereas the hydroquinone is relatively redox-stable and can be Mer metabolized and excreted. Adapted fiom Ross et al. (1 993). See text for details. Work by Lind and colleagues, using liver microsomes from normal and 3-
rnethylcholanthrene (3-MCI a polycyclic aromatic hydrocarbon PAH] and a cornrnon
industrial pollutant) treated rats, has demonstrated a relationship between one electron reductases and DTD in ROI toxicity (Lind et al., 1982). Treatment with 3-MC caused a
several-fold increase in DTD activity without affecting the activity of
NADPH:cytochrorne P450 reductase. Competition between the two enzymes was demonstrated in a spectrophotometric assay measuring NADPH levels where addition of menadione resulted in limiting the extent of NADPH oxidation; this decline occurred more rapidly in 3-MC induced than in control microsomes. Cellular response to DTD substrate exposure often results in an increase in DTD enzymatic activity. Early ches that this represents a protective response to the toxic effects of these substrates date back to the mid 1960's. Huggins and CO-workersdiscovered that treatment of mice with small doses of PAHs and other xenobiotics that act as DTD substrates caused an increase in
DTD activity in the liver and other tissues. This rendered the mice resistant to the carcinogenic effects usually observed at higher doses of PAHs (Huggins et al., 1964 &
1965).
1.2.5 Su bstrate-Mediated Induction of DTD
In al1 cases investigated to date, substrate-mediated induction of DTD activity has been shown to involve a net increase in newly synthesized DTD as indicated by increased mRNA levels (Pickett et al., 1984) and the ability to block induction with cyclohexamide
(Lind and Ernster, 1977). In the late 1960's Nebert and CO-workerspioneered the concept of a molecular - genetic basis for the inductive response of various detoxiQuig enzymes
in response to xenobiotic exposure, thereby contributing to the establishment of the field
of pharmacogenetics (Nebert and Gelboin, 1968). This group concluded that the
induction of DTD is controlled by the Ah (aryl hydrocarbon) locus encoding the aromatic
hydrocarbon receptor (AHR) (Kumaki et al., 1977). This conclusion sternmed from their
previous studies of other detoxification enzyme batteries (mainly the cytochrome P450s)
which established a link between the Ah locus, the AHR and the aromatic hydrocarbon
hydroxylase (AHH) enzyme system, and is based on correlative evidence of the
upregulation of both AHH and DTD upon substrate exposure.
Prochaska and Talalay (1 988) categorized the range of inductive compounds that
can be metabolized by various enzyme batteries as bifunctional and monofunctional
inducers. Bifunctiond inducers induce transcription of both Phase I and Phase II
detoxifj4ng enzymes whereas monofùnctional inducers, which are highly electrophilic species and have a variety of chernical structures, induce transcription of Phase II enzymes only (reviewed by Begleiter et al., 1997). Phase 1 enzymes, which include the cytochrome P450s, create functional groups to generate more electrophilic and water- soluble species upon which Phase II enzymes can act (Prestera et al., 1993). In general,
Phase II enzymes, which include glutathione S-tramferase (GST), and UDP- glucuronosyltransferase, participate in conjugation reactions of the newly created functional group to generate species that can be further metabolized and excreted (Figure
1.4). DTD, although not serving as a conjugating enzyme, is considered to be a Phase II enzyme since it contributes to modification of xenobiotics to facilitate their elimination. Validation of the hypothesis that upregulation of enzyme batteries associated with
xenobiotic detoxification involves a transcriptional induction process via an AHR-inducer complex arose fkom the discovery of the cis-acting xenobiotic response element O(RE) in the cytochrome P450c gene (Fujisawa-Sehara et al., 1987). It was confinned that DTD
transcription induction can be mediated by the AHR upon identification of a cis-acting
XRE by Jaiswal(1991) in the 5' flanking region of the human DTD gene at position -740
(Figure 1.1).
However, sequence analysis of the 5' UTR of the human DTD gene also revealed the presence of a single copy of another transcriptional enhancer element, an antioxidant response element (ARE, Figure 1.1)' previously shown to mediate transcriptional activation of genes such as GST in response to exposure to aromatic planar compounds and phenolic antioxidants (Rushmore and Pickett, 1990). Such compounds promote oxidative stress and generate a yet to be identified 'redox signal' (Prochaska and Talaiay,
1988). Also, bifunctionai inducers, such as PAHs, which upregulate transcription of phase 1 enzymes, such as CYP IAI, via the XRE can be metabolized to generate radical species that alter the cellular redox state (Faweau and Pickett, 1993). This appears to lead to upregulation of phase II enzymes via the ARE.
A theory emerged that the aforementioned redox signal promotes dimerization of products of the early responsive genes, c-Fos and c-Jun, to form an AP-1 complex
(Mitchell and Tjian, 1989). This was based on the observation that AP-1 members can be activated by oxidative stress and the ARE core sequence is similar to the sequence recognized by AP-1 proteins. An AP-1 consensus site (TGACTCA) is present within the
ARE starting at position 462 of the hman DTD gene promotor (Jaiswal, 1991, Figure 1.1). However, recent evidence suggests that a transcription factor consisthg of a
NrfUsmall Maf heterodirner mediates the upregulation of phase 11 enzymes via the ARE
(Itoh et al., 1997). This was initidly show by failure of phenolic antioxidants to
upregulate GST and DTD transcription in nrf-2 knockout mice. The cytosolic protein
Keapl has been demonstrated to sequester Nrf-2 in the cytosol, thereby supressing Nrf-2
transcnptional activity until exposure to phase II enzyme inducers (Itoh et al., 1999). This
implicates Keapl as a sensor for oxidative stress leading to the shuttling of Nrf2 to the
nucleus to mediate induction of phase II enzymes.
Begleiter and CO-workers are developing the concept of inducing DTD using
analogues of 1,2-dithiole-3-thiones in order to sensitize turnor cells to bioreductive drugs
that require DTD activation. This group demonstrated that this approach provides larger
increases in sensitization to agents such as MMC in tumor ce11 panels representing a
number of different tissue types compared to normal cells of the corresponding tissue
type (Dohem et al., 1998, Wang et al., 1999). The extent of induction varied between
tissue types. The mechanism responsible for this variation is unknown but it may reflect
differences in transcription factor levels, alternative splicing, or the presence of a base
change at nucleotide 609 of the DTD coding region that is associated with low enzymatic
activity .
1.3 The RoIe of a Single Base Change at Nucleotide 609 of DTD cDNA
Marshall et al. (1991) investigated the relationship between DTD activity and
MMC sensitivity in a panel of five human skin fibrobIast strains donated by a cancer- prone fmily. Two of the fibroblast strains, 3437T and 3701T, were found to be approximately 6-fold more resistant to MMC than the other thestmins and cmtained
low to undetectable DTD activity. Consistent with a saturation effect between DTD activity and MMC toxicity, the three MMC-responsive strains displayed similar MMC
sensitivities despite possessing a wide range of individual DTD activities (400 - 1800
nrnol DCPIP reduced / min / mg protein). Southern blot analysis of the human DTD gene
in the MMC-resistant strains indicated no gross abnormalities. However, although
Northem blot analysis indicated the presence of equal levels of the 1.2 kb DTD mRNA
species between ce11 strains, Westem blot analysis demonstrated undetectable DTD
protein in 343 7T and 3 70 1T cells, suggesting that a pst-transcriptional or post-
translational defect results in low to undetectable DTD protein levels in these two ce11 strains.
1.3.1 Identification of a 609C to T Nucleotide Substitution
An expianation for the variation in DTD activities and MMC sensitivities observed in the ce11 strains discussed above was provided by work by Traver et al. (1 992) using a panel of hurnan colon carcinoma cells which exhibited a range of DTD enzymatic activities. HT29 cells contained moderate to high DTD activity whereas BE cells displayed low to undetectable activity despite exhibiting similar levels of DTD mRNA expression (assessed by semi-quantitative RT-PCR). cDNA sequencing of these two ce11 lines revealed that the only difference was in nucleotide position 609, a C nucleotide in
HT29 cells and a T nucleotide in BE cells, resulting in a predicted proline to se~e change in amino acid 187. BE cells were homozygous for this substitution and did not display Westem blot reactivity to anti-DTD mAbs (although this group subsequently reported low levels of protein using more stringent Western blot conditions with protease
inhibitors and prolonged cherniluminescence exposure time (Seigel es al., 199%
suggesting that the DTD serine 187 isoform is unstable.
cDNA sequencing of the aforementioned fibroblast ce11 strains taken from the
cancer-prone family uncovered a homozygous nucleotide 609C to T change in 3701T
cells, which were maximally resistant to MMC under aerobic conditions (Kuehl et al.,
1995). The remaining cells in the fibroblast panel were either homozygous for the 609C
nucleotide or were heterozygotes. However, the other resistant ce11 strain, 3437T, was a
heterozygote, which appeared inconsistent with the hypothesis that the proline 187 and
serine 187 isoforms represent a wild-type and functionally impaired DTD, respectively
(see Chapter 3).
Based on the x-ray structure of rat DTD (Li et al., 1999, the proline 187 to serine change is located in the j35 region (Figure 1.2A) and therefore does not obviously participate in reactions involving the DTD active site nor does it appear to contribute to the dimerization process. Two groups reported kinetic studies of the serine 187 isoform puified fiom an E. coli expression system. Traver et al. (1997) reported that the senne
187 isoform was only 2% as active as the wild-type enzyme, measured by DCPIP reduction. Wu et al. (1998) found it to possess a low but significant level of DCPIP- reductase activity - approximately 30-fold lower than the proline 187 isoform - thereby suggesting that the 609C to T nucleotide change constitutes a îùnctional mutation. Traver ef al. (1997) did not use menadione as a substrate to test the properties of their E. coli punfied serine 187 isoform, but Wu et al. (1998) reported it had 10-fold lower rnenadione-reductase activity than the proline 187 isoform. In addition, COS-1 monkey kidney cells, which contain undetectable background DTD activity, when msiently
transfected with a mamrnalian expression vector containing wild-type DTD cDNA
displayed on average IO-fold greater enzymatic activity compared to transfectants
receiving mutant cDNA (Misra et al., 1998, see Chapter 2). In addition, mutant COS-1
transfectants had at least a 3-fold reduction in protein levels compared to wild-type
tram fectants.
Both isoforms have similar Michaelis constant (&) values towards menadione
and DCPIP but the serine 187 isoform has a 20-fold greater dissociation constant (K,,)
value towards FAD (Wu et al., 1998). Therefore, the lower afkity of the serine 187 for
the FAD prosthetic group may lx the basis for its defective enzymatic activity. Another
possibility is that the mutation causes a change in enzyme conformation resulting in
decreased enzymatic activity (Wu et al., 1998) andor stability (Traver et al., 1997).
1.3.2 Frequency Analyses of the Nucleotide 609 Base Change
The postulated roles of DTD as a dmg activating enzyme, as an anti-cancer defense mechanism, and as an antioxidant enzyme, coupled with the establishment of the
609C to T nucleotide change as a fünctionai mutation, raises the issue that this mutation may be associated with disease arising fiom impaired DTD activity. Although the mutation wasn't identified at the time, Edwards el al. (1980) concluded that 4% of 628
British unrelated individuals lacked a diaphorase activity that was later identified as DTD as detected by starch-gel electrophoresis and staining with NAD or FAD.
Frequency studies of the nucleotide 609 base change (sumrnarized in Table 5.1) were aided by the development of a polyrnerase chah reaction - restriction fiagrnent length polymorphism (PCR-RFLP,Figure 3.1) screening assay that allows assignment of the genomic statu of the DTD 609C to T nucleotide change by virtue of the substitution resulting in a novel H& 1 recognition site (Eickelmann et al., 1994). An estimate of the fiequency of the DTD nucleotide 609 base change in the normal population was provided in a study of genomic DNA fiom 44 normal bone marrow donors where 9% of the subjects were homozygous for the mutation while 40% were carriers (Kuehl et al., 1995).
Therefore, this base change is fiequent and a number of studies to date (Rosvold et al, 1995) in addition to the one by Kuehl et al. suggested that it was in accordance with
Hardy- Weinberg equilibrium (Table 5.1 ).
The hypothesis that the variant allele is associated with decreased DTD protein stability has been more definitive than its postulated role in disease, as evidenced by 1) lower DTD levels in marnrnalian cells transfected with the serine 187 isoform than in wild-type transfectants (Misra et al., 1998, see Chapter 2) 2) the absence of detectable
DTD protein in biopsy material @th normal and tumor) homozygous for the nul1 ailele
(Siegel er al., 1999) and 3) the observation of undetectable, or, at kt, minor levels of
DTD protein in homozygous mutant ce11 lines (Phillips et al., 1995; Siegel et al., 1999).
1.3.3 Heterogeeieity of DTD Enzymatic Activity and Chemotherapy
The existence of a fiequent mutation that impairs DTD enzyrnatic huiction becomes important when using bioreductive prodrugs that depend on DTD activation.
Although the association of the variant allele with the cancer phenotype is unclear, DTD enzymatic activity is heterogeneous across and within spectra of both normal and turnor tissue types (Belcourt et al., 1998; Siegel et al., 1999). Whether this heterogeneity is due to the prevalence of the mutant ailele or to other factors that control DTD expression remains to be determined. However, treatment modalities that involve bioreductive procimg therapy that depend on DTD activity should ensure that high levels of Dm protein are present. One such approach is substrate-mediated induction as proposed by
Begleiter and CO-workers(discussed previously). This resulted in upregulation of both the proline and serine 187 DTD isoforms. One novel strategy is to deliver exogenous DTD to tumor cells through the use of vectors developed for gene therapy.
1.4 Cancer Gene Therapy
The concept of gene therapy originaily developed fiom the observation that some diseases are caused by the inheritance of one or two copies of a defective gene; it follows that such diseases are monogenic and therefore can be cured by insertion of a normal copy of the mutated gene (Roth and Cristiano, 1997). Since efficient delivery of therapeutic genes continues to be a major limitation to gene therapy, current gene replacement strategies are most effective for renewable ce11 populations such as bone marrow stem cells (Kiem et al., 1995). Since cancer arises through a multi-step process culminating in a variety of genetic abnormalities, and it is beyond the capability of current gene therapy vectors to restore normal gene function to every ce11 in a tumor population, a more effective strategy may be to use gene therapy to deliver genes encoding cytotoxic products that specifically eliminate deleterious cancer cells (Roth and
Cristiano, 1997). 1.4.1 Non-Vin1 Gene Therapy Vectors
'The simplest fom of gene delivery does not involve use of synthetic molecules or
a viral vector, but rather the delivery of naked DNA (Cheng et al., 1993). This is usually
done mechanically through direct injection or by high velocity bombardment with gold
particles with DNA attatched. Although this method has resulted in gene expression in
skeletal muscle tissue in vivo, it lacks targeting specificity at the cellular level. Expression
is localized to the site of injection with the added disadvantage of low uptake eficiency
in the targeted ce11 population. This method can also be invasive since surgery is often
required to allow access to the targeted tissue.
Non-viral vectors have the common advantage of not king limited in the size of
DNA that they can deliver (Roth and Cnstiano, 1997). Liposomes cm combine with
DNA to form lipid-DNA complexes, which can improve the eficiency of DNA delivery
to many ce11 types (Nabel et al., 1992). A wide vax-iety of synthetic liposomal vectors are
available. and most are non-toxic and non-imrnunogenic (Nabel et al., 1993). However,
the major drawback in the use of liposomes is the lack of targeting specificity towards specific ce11 types, with transduction usually limited to the site of administration (Roth and Crisitano, 1997). Also, the lipid component of liposomes gives it a high propensity
for non-specific uptake by cells of the reticuloendothelia1 system, resulting in short systemic half life (Wagner et al., 1990).
The problem of targeting specificity of non-viral vectors has been partially solved by the development of gene delivery vectors composed of protein-DNA conjugates and the use of various ligands (Cnstiano, 1993). A major limitation of protein-DNA complexes as a gene delivery vehicle is their inability to escape fiom endosornes
subsequent to receptor-mediated endocytosis.
1.4.2 Retroviral Gene Therapy Vectors
Unlike liposomes and naked DNA vehicles, viral vectors have the advantage of
tissue tropism, whether natural, or artificially through the use of packaging ce11 lines that
supply viral envelope ligands for specific ce11 types (Bitzer et al., 1997). The remahder
of this section describes some fiequently-used viral vectors as gene delivery vehicles.
Retroviruses have an RNA genome which is converted to DNA upon host ceil
infection and integration Field et al., 1996). The genome of simple retroviruses consists
of three genes, termed gag, pl, and env, flanked by elements called long terminal repeats
(LTRs) that act as viral transcriptional promoters/enhancers and also are needed for
replication. Simple retrovirai-based gene delivery vectors are constructed by substituting the transgene for the gag, pol, and env regions (the LTR continues to provide transcriptionai induction). The requirement of ce11 division for retroviral infection can be an advantage for cancer therapy. Also, there does not appear to be any pre-existing immunity to current simple retroviruses that are engineered for gene therapy.
However, there are limitations of the use of simple retroviral vecton including 1) their inability to infect non-dividing cells; 2) their limited capacity (8 kb) for foreign
DNA; 3) difficulties in large scale production - currently achievable titers are on the order of 10' particles I mL, which may be insuficient for the treatment of large tumon; 4) unstable levels of transgene expression, which has been known to shut off after five to seven days in some in vivo studies for unknown reasons (Verma and Somia, 1997); and 5) their Iimited host tissue range (Roth and Cristiano, 1997). Attempts to overcome the fifth
limitation involves the technique of 'pseudotyping', in which packaging lines supply
envelope proteins that have the desired host range (Yam et al., 1998).
1.4.3 Basic Adenovims Biology
Adenovirus infection has ken implicated to be the cause of a range of hurnan
upper respiratory tract, eye, and gastrointestinal diseases (Straus, 1984). They are
widespread in nature, infecting birds, many mammals and man. There are 2 genera of the
Adenovindae family: Aviadenovirus (avian) and Mastadenovirus (mammalian). Many
classification schemes have been used for members of the Adenoviridae family, with the
serotype classification king the most common (Nermut, 1984). Each serotype elicits a
di fferent set of serurn antibodies with some cross-reactivity between serotypes. The
serum reactivity of each serotype is assayed by degree of haemagglutination of
erythrocytes that carry various antibody sets. Mastadenovirus types 2 (Ad2) and 5 (Ad5) have been the most fiequently studied and their genomes have been completely sequenced; the remainder of this section will concentrate on these two viral types.
Adenoviruses are non-enveloped and have an icosahedral symmetry (Figure 1.5) of around 90 nm in diameter with molecular weight estimates of the whole virus ranging fiom 175,000 - 185,000 kDa (Nemut, 1984). The icosahedral capsid shell consists of at least 10 different proteins to form 252 capsomeres that are partitioned into 240 'hexons' and 12 'pentons.' Hexons are the major component of adenovinws, accounting for 46% of the mas of the virus. Each hexon is 10.5-1 1 nm in length and consists of three polypeptides forming a triangular shape with a centrai channel in the middle that is 10-15 Figure 1.5: Gross morphology of adenovins capsid. The capsomeres consist of at least 10 diflerent polypeptides. The virus has an overall icosahedral shape with 240 capsomeres foming tripartite hexons and 12 capsomeres forming pentons. Hexons (1 0.5 - f 1 nm in length) are the major molecular component believed to control the transport of solutes and larger molecules. Pentons consist of a 10 nm wide base supporting a fiber 28 nm in length that is terminated by 42 A sphencal knob. The fiber and knob structures contain domains that attach to host cellular receptors. White and black triangles show two different groups of hexons each forming a triangular icosahedral face. Adapted from Nermut, 1984. A wide. It is thought that these channels give hexons a functional role, in addition to a structural one, by controlling the transport of solutes or larger molecules (such as enzymes) across the capsid. The 12 capsorneres at the vertices of the icosahedron are called pentons because they are surrounded by 5 neighboring hexons. Each penton is formed by a 10 nm wide base, a thin antema-like projection called a 'fiber' that is 28 nrn in length, with each fiber terminated by a spherical 'knob' that is 42 A in diameter.
The adenovird genome consists of a linear DNA molecule around 36 kb in length which may aiso exist in a circular form mediated by the DNA terminal protein (TP), which is covalently attached to each 5' strand (Robinson et al., 1973). The entry of adenovirus into susceptible cells is sequential. The virion's fiber knob first attaches to a ce11 surface receptor. Fiber-mediated attachment of Ad2 and Ad5 has recently been show to occur through the major histocompatibility complex class 1 (MHC 1) and the coxsackievirus-adenovinis receptor (CAR) of the host (Bergelson et al., 1997).
Intemalization requires the interaction of five conserved Arg-Gly-Asp (RGD) motifs at the piIton base with specific membea of the heterodimeric vitronectin-binding integrin family, a$, and a#, @ai et al., 1993). The virus is endocytosed via coated pits and vesicles and enters the cytoplasmic cornpartment within 5 to 10 min following absorption
(Chardonnet and Dales, 1970).
Very rarely do Ad2 and 5 integrate into the host genome. Rather, they continue as independent minichromosomes that acquire cellular histones (Kasamatsu and Nakanishi,
1998). Before genome replication, adenovirai mRNAs are transcribed from the input
DNA. The genome is commonly divided into 100 map units (mm) where one m.u. is approximately 365 nucieotides and O m.u. corresponds to the first base (Sussenbach, 1984). Because of the temporal nature of adenoviral transcription, mRNAs are classified
as either immediate early (fiom Ela gene transcription), edy (fiom El b, E2& E2b, E3
and E4 gene traascription as well as nom genes for some viral proteins), or late (from LI
- L5 gene transcription, which are mostly for viral structural proteins).
nie gene products of the El region have a number of interesting activities. Three
splice variants of the E 1a transcription unit have been identified (Pettersson, 1984) and
are the first mRNAs to be expressed? usually within 1 hour after infection. Two of the
mRNAs encode a set of proteins ranging in molecular weight between 35 and 53 kDa. It
is not known at present how each mRNA can specifj. two or more proteins that differ in
molecular weight. The precise mode of action of the E1A proteins is unknown since they
do not bind DNA, despite their ability to activate transcription. However, one of the
pnmary functions of EIA proteins appears to be at the level of gene expression of early viral regions El b, E2, E3, and E4 as well as LI (McGrory et al., 1988). Two splice variants of the E 1b transcription unit have ken identified (Pettersson, 1984); one encodes a 55 kDa protein, and the other a 19 kDa protein.
EIA proteins are sunicient to induce transformation of primary cells in vitro by inducing oncogenic activity via the Ha-ras gene (McGrory et al., 1988). EIA may also mediate its oncogenic effect by binding to the himor suppressor p 105-RB (Boulanger and
Blair, 199 1). E 1B proteins do not transform cells on their own, but CO-operatewith El A proteins to stably transform cells. During adenoviral infection, EIA proteins cause an accumulation of the tumor suppressor gene pS3 (Lowe and Ruley, 1993) which can tngger apoptosis; however, the El B 19 kDa protein subsequently acts as a functional homologue of the anti-apoptotic protein Bcl-2, whereas the EIB 55 kDa protein binds to and inactivates p5 3. Together. these observations indicate that adenovinises, in the course of sequestering cellular machinery, alter the intracellular environment to favor virai replication by preventing apoptosis. Oncogenesis is a rare outcorne as a result of these events (Boulanger and Blair, 199 1).
At the onset of DNA replication, occurring as early as 6h post-infection, transcription switches pnmarily fiom the early to the late genes (Sussenbach, 1984). The five mRNA families transcnbed fiorn L1-L5 encode structural and or structure- supporting proteins for the virion and capsid, as welt as proteases involved in post- translational processing of late gene protein products.
1.4.4 First Generation Ad5 Gene Thenpy Vecton
The design of systems to create replication-defective recombinant Ad5 vectors that carry transgenes for gene therapy sternmed from extensive work examining the role of the El region in viral replication (see above) (McGrory et al., 1988). These studies usually invoived the construction of mutant El genes that were cloned into E. coli-based expression plasmids. Although the fùnctional consequences of these mutations could be studied by transfection of these plasmids into rnarnrnalian cells, it was more desirable to
'rescue' the mutations made in cloned DNA into infectious virus. One widely used rnethod was to digest the viral DNA with a restriction enzyme that has a single cut site near the left end of the genome (within the El region), and then to ligate the large fragment to, or cotransfect it with, the expression plasmids containing mutated El sequences (Stow, 1981). However, this method is highly inefficient in that a large number of plaque isolates have to be screened to identiQ the desired mutant recombinant clone. A major contribution by Graham and CO-worken to these studies was the development of a more efficient technique to rescue El mutations into infectious ~d5
(McGrory et al., 1988). Previous work by this group provided four impomt
observations and developments that led to this technique; these were: 1) adenoviral DNA
can circularize in idected cells; and this ability can be exploited to clone adenoviral
genomes as bacterial plasmids that can generate infectious virus following transfection
into mamrnaiian cells (Graham, 1984); 2) two cotransfected. non-infectious viral
plasmids can recombine, without the need to first linearize them, and generate infectious
virus (Ghosh-Choudhury et al., 1987); 3) the maximum amount of viral DNA that can be
packaged into capsids is limited to only 2 kb in excess of the size of the wild-type genome; and 4) the development of a 293 human embryonic kidney ce11 line that is stably transfected with the leil 1 1% of the Ad5 genome and thus is able to support replication of
E l defective recombinant adenovinis (Graham et al., 1977).
n+seobservations and developments led to the construction of pJM 17 (McGrory et al., 1988, Figure 1.6), a non-infectious plasmid, which consists of the complete Ad5 genome plus the 4.3 kb bacterial plasmid pBRX inserted at 3.7 m-u. (Figure 1.6A), which dismpts the El region and renders pM17 unable to mediate viral production in normal tells, and too large to be packaged in 293 cells. Since it is circular and contains the ampicillin resistance (ampR)and tetracycline resistance (tetR) genes, as well as a bacterial ongin of replication (on), pJM17 can be grown in bactena in selective media and isolated in large quantities. pJM17 is approximately 2 kb in excess of the maximum amount that cmbe packaged into adenovirus capsids, therefore it can only give rise to infectious O mou. 3.7 mou. pJM17 Packaging Vector (40.3 kb) 100 mou.
\ "- '.A, ' ,' \ O m.". 'AEI ~ectorl16.1mu.
1 (0.5 - 3.7 m.u.)
(78.3 - 85.8 m.u. , pBHG10) (77.5 - 86.2 mou., pBHG11)
Packaging Vector
pBHGlO = 34,783 bp pBHG11 = 34,304 bp Shuttle Vector
Figure 1.6: Two examples of Ad5 packaging and shuttle vectors and site of recombination. A) Cotransfection of the two vectors may result in homologous recombination by overlapping El sequences resulting in an infectious plasmid incapable of generating replication-competent virus except in packaging ce11 lines such as 293 cells that supply E 1 proteins in trans. pJM 17 contains a bacterial plasmid, pBRX, inserted into the 3.7 m.u. position that interrupts the El region and renders it non-fiinctional. pBRX also contains ampRand tep genes that enable pJM17 to be cloned in bacteria in selective media. The desired insert is cloned into a shuttle vector in the region that contains an El deletion. Although both plasmids in this figure are represented as linear, they are transfected as circular. Adapted fiorn McGrory et al., 1988. B) pBHGlO and pBHG11 contain deletions in both the El and E3 regions, allowing the incorporation of larger inserts than into pJM17. pBHGlO and pBHGl 1 also lack a Y packaging signal that can be restored only upon recombination with a shuttle vector containing Y, enswing that a replication-competent virus is not created. Also shown is the pAElsplB and its MCS which is located within the El deletion for cloning of inserts. Adapted fiom Ben et al., 1994. Vectors are not shown to scale. particles by rearrangements that excise the pBRX insert. This occurs at a very Iow
fiequency and only after long incubation times.
Recombination between pJM17 and a shuttle plasmid containing overlapping lefi
end Ad5 sequences which displace pBRX (Figure 1.6A) gives rise to an infectious
plasmid much more efficiently (McGrory et al., 1988). If this shuttle plasmici contains a
defective El region (which was the intention of the previously described studies
involving rescue of E 1 mutations), the recombinant infectious plasmid will yield
replication-defective virus in 293 cells, following homologous recombination. These
recombinant vimes can replicate in 293 cells since El proteins are supplied in tram by
the host ce11 genome. It is not known by what mechanism recombination occurs, but a
likely possibility is that it involves linearization of both plasmids (McGrory et al., 1988).
This strategy was the basis for the design of the fint generation replication-
defective recombinant Ad5 systems. As illustrated in Figure 1.6A, an Ad 5 shuttle vector
such as pAElspl B (Figure 1.6B), which contains Ad5 sequences from bp 22 to 5790 (O -
16.1 m-u.) with a deletion of E 1 sequences fiom bp 342 to 3523 (1 -0 - 9.8 m.u.) aiong
with a multiple cloning site (MCS) for minigene insertion, is cotransfected with an El -
defective Ad5 packaging plasmid, such as pJM17, into a packaging ce11 line such as 293 cells or Hela cervical carcinoma cells that supply El proteins in trans. The size of the
minigene is resincted to around 2.3 kb if pJM 17 is used as the packaging vector. Further modifications in packaging plasmid design subsequent to the development of pJM17 were created by Graham and CO-workersto accommodate larger rninigenes (Bett et al.,
1994). This involved deletions of the El and E3 region where E3 is nonessential for viral replication and both can be deleted and still produce a functional vector (Berkner and Sharp, 1983). The E3 gene is believed to fûnction by circurnventing the immune response towards adenovirai infection by binding to MHC 1 and interferhg with antigen presentation during immune surveilIance (Schowaiter et al., 1997).
As shown in Figure 1.6B, the pBHGlO and pBHGl 1 packaging vectoa (Ben er al., 1994) do not contain the complete Ad5 genome as with pJM17. These packaging vectors have deletions in the El region (fiom bp 188 to 1339 (0.5 - 3.7 m.u.)) and in the
E3 region (fiom bp 28,133 to 3O,8 18 [78.3 - 85.8 m.u.1 for pBHGlO and fiom bp 27,865 to 30,995 [77.5 - 86.2 m.u.1 for pBHGl1). The use of these modified packaging vectors results in a capacity for inserts of up to 8.3 kb. Minigenes can be incorporated into the El or E3 deletions or both.
The deletions in the E L region (AE 1) of pBHG l O and 1 1 removes the cis-acting Y packaging signal (AY) required for virion assembly (Ben et al., 1994) and cm only be restored upon recombination with a Y-containing adenoviral shuttle vector. This requirement adds an extra safety feature in ensuring that the rare event of recornbination between the packaging vector and the host El region does not yield a replication- competent virus, because of lack of the Y signal. In the present work the pAElsplB shuttle vector and the pJM17 packaging vector were utilized (see Chapter 4).
1.4.5 Rationale for Combination of Gene and Bioreductive Drug Therapy
One of the first applications of the idea of seiective transduction of tumor cells with a gene whose product converts a prodrug to a cytotoxic metabolite was the use of a retrovirus carrying the herpes simplex virus thymidine kinase (HSV-TK) gene which activates the nucleoside analogue gancyclovir (Culver et al., 1992). It was found in vivo that HSV-TIC therapy resulted in significant tumor mass regression despite observations that only a small fraction of the tumor was infected with the retrovirus (Freernan et al.,
1993). This cytotoxic effect mediated by transduced cells towards non-transduced cells has been termed a 'bystander effect.' In the case of HSV-TK and gancyclovir, the activated metabolite, gancyclovir-triphosphate, is transferred fiom cell to ce11 through gap j unctions (Bi et al., 1993). Ad5 .HSV-TK in combination with gancyclovir has recentl y been show to be effective in the treatment of human prostate (Martiniello-Wilks et al.,
1998) and retinoblastoma (Hurwitz et al., 1999) tumors in mice.
Bioreductive hgthetapies have traditionally been designed to: 1) take advantage of tumor hypoxia to provide selective drug activation; andor 2) exploit individual differences in reductase levels in order to utilize a particular bioreductive agent.
Workman and Stratford (1993) suggested that examination of enzyme levels in a particular individual prior to choosing a bioreductive agent could provide an optimal enzyme-drug match, a strategy they termed 'enzyme profiling.' However. to overcome the invasiveness (due to the need to obtain tissue biopsies) and time delay (due to growth of ce11 strains and assaying for enzyme activities) associated with enzyme profiling, approaches to de1iver and over-express a given bioreductive enzyme in tumor ce1ls including engineered viruses are in development.
Studies with CB1954 (Figure 1.7), synthesized at the Chester Beatty Institute in
England in the late 1960s demonstrated cure of rodent in vivo turnours of the Walker 256 carcinoma ce11 line (Cobb et al., 1969). Although CB1954 was shown to have high potency against Wdker turnor cells, it was ineffective against a range of human tumor DTD
Thioester I e.g. Acetyl CoA
DNA DNA C rosslinks
NO2 I-amino derivative
Figure 1.7: Mechanism for the activation of CB 1954 by DTD.Four electron reduction by DTD results in the formation of a Chydroxylamine compound that is diffusable in aqueous media and a non-cytotoxic Carnino derivative. Alternatively, the 4- hydroxylamino species can react with intracellular thioesters to generate a species (shown in parentheses) that can form adducts with DNA. Scheme is adapted fiom that of Knox et al. (1993). xenografts (Workman et al., 1986). Knox and CO-workers (1991) showed that the sensitivity of Walker 256 tumors to CB f 954 resdted fiom efficient activation of the dmg by high levels of rat DTD where the 4-hydroxylamine group of reduced CB1954 can further react with intracellular thiols to generate a reactive center for DNA alkylation
(Figure 1.7). DNA crosslinking car, result via the electrophilic 5-aziridinyl and the newly generated 4-amide groups. The 4-hydroxylamino form of CB 1954 (generated after four electron reduction by reductases such as DTD, Knox et al., 1993, Figure 1.7) has also been shown to induce a bystander effect in vitro (Bridgewater er al., 1997) making it an attractive candidate for gene therapy involving enzymes that eficiently metabolize
CB1954. Conditioned media taken fiom cells transduced with enzymes able to activate the prodnig and treated with CB1954 has been shown to be cytotoxic toward non- transduced cells. However, unlike gancyclovir-triphosphate, the 4-hydroxylamino derivative of CB1954 mediates its bystander effect via aqueous media with cellular uptake by passive diffusion and, therefore, does not involve gap junctions (Bridgewater et ai., 1997).
Knox and colleagues investigated the basis for the difference in CB1954 sensitivity between human and rat nimors in vitro using ce11 line panels that contained similar levels of DTD (Boland er al., 1991). Human ce11 lines were sirnilarly affected by the 4-hydroxylamino (activated) form of CB1954 as rat cells suggesting that their resistance was not due to further inactivation of the hydroxylamine group nor to intrinsic resistance to the formation of DNA adducts. Therefore, it was proposed that human DTD was ineficient at activating the parent CB1954 prodnig. Studies with rat and human
DTD purified to homogeneity revealed that the human isoform was indeed less able to carry out the reduction of the CB1954 prodrug to the 4-hydroxylamho derivative. The rat isoform displayed a catalytic constant &J value more than 6-fold higher than the hurnan isoform.
The rat isoform has been shown to be more effective than the human isoform in
reducing a number of bioreductive agents, including MMC (Siegel et al., 1990). Chen et
al. (1 997) created rat-human DTD chimenc enzymes to determine the region of DTD that
is responsible for the catalytic difference. Initially, this was done ushg large regions
spanning the rat amino and carboxyl terminal sequences, and it was found that the latter
region determined DTD catalytic activity. A senes of rat and hurnan DTD mutants containing minor substitutions in the carboxyl region were generated and analyzed. Of the 26 amino acid differences in the carboxyl region between the rat and human form, amino acid 104 (glutamine (Q) in human and tyrosine (Y) in rat) was the most important residue responsible for the catalytic differences between the rat and human enzymes; a human DTD mutant with a glutamine to tyrosine substitution in amino acid 104
(hQ104Y)behaved like the rat enzyme in its ability to metabolize CB 1954, making it an attractive prodrug-enzyme combination for gene therapy or delivery by conjugated anti body.
In addition to hQ104Y, two other enzymes have been identified recently that are more efficient in the metabolic activation of CB1954 than DTD. E. coli nitroreductase
(NTR) has 90-fold greater catalytic activity towards CB1954 than DTD (Bailey SM and
Hart IR, 1997). A recombinant retrovirus carrying NTR cDNA dong with CB 1954 has demonstrated a marked bystander effect where treatment of mixed populations consisting of 50% each of transduced and non-transduced tumor ce11 lines resulted in complete ce11 killing (Green et al., 1997). Hurnan NQO, has a very high activity towards CB1954, being 3000-fold more effective than DTD in CB1954 reduction (Wu et al., 1997). TO date, use of NQO in gene therapy has not been reported.
Therefore, enyme-prodrug therapy involving overexpression of an appropriate reductase in tumor cells may represent a solution to the problem of heterogeneity in reductase expression among cancer patients. This approach has the advantage of being re lativel y non-invasive.
1.5 Thesis Outline
The purpose of the work presented in this thesis was to examine the role of DTD as a prodrug-activating enzyme. Three specific aspects of DTD activity were studied: 1) the effect of a DTD 609C to T nucleotide change on DTD activity and protein stability; 2) the use of this base change as a prognostic indicator for cellular responses that are mediated by DTD; and 3) the feasibility of using Ad5 gene therapy to deliver DTD activity to tumor cells to render them drug sensitive.
C hapter 2 describes transient transfection studies of rnammalian COS- 1 cells with a plasmid vector resulting in high expression of a DTD minigene with or without the nucleotide 609C to T change. Since COS-1 cells contain undetectabie background DTD activity, these experiments provided for direct cornparisons of the cellular enzyme activities and protein stabilities of the proline and serine 187 DTD isofoms. This work has been published in the British Journal of Cancer (Misra et al., 1998).
Chapter 3 describes studies to determine if the allelic status of the 609C to T nucleotide change can predict DTD activity, and thus its use as an indicator for degree of drug activation and xenobiotic detoxification medîated by the enzyme. This issue was assessed by examining the relationship of allelic status and DTD enzyme activity in 45 human skin fibroblast ce11 strains that were provided by Nancy Cracknell and Dr. Peter
Ray (Hospital for Sick Children, Toronto). An explanation for the unique observation that 3437T cells are heterozygous for the 609C to T nucleotide change yet contain low to undetectable DTD activity and trace amounts of DTD protein is provided. This work was performed with the help of a summer student, Annic Grondin, and has been accepted for publication in the British Journal of Cancer. Appendix 3.1 provides a table of the characteristics of the human ce11 strains used for the studies described in Chapter 3.
Chapter 4 addresses the feasibility of the use of DTD in an Ad5-directed cancer gene therapy approach. Ad5 vectors were designed to express the proline or serine 187
DTD isoform under the control of the cytomegalovirus promoter. The relationship between multiplicity of infection and recombinant DTD protein expression as well as its expression kinetics in tems of enzyme activity were examined in BE cells. The possibility of a dominant negative effect mediated by the DTD 609T allele was tested by infection of HT29 cells with Ad5 carrying DTD cDNA encoding the serine 187 isoform.
FuI-thermore, the degree of MMC sensitization of BE cells infected with an Ad5 vector carrying a minigene for the proline 187 DTD isoform was deterxnined using an in vitro colony formation assay. The MMC sensitivity of these transduced BE cells was compared with MMC-sensitive HT29 cells. This work will be submitted to the International Journal of Cancer.
Chapter 5 provides a summary and discussion of the experimental results in context with progress in the field during the course of the work described in Chapter 2, 3, and 4. As well, possible fiiture experiments are described for fùrther investigation into the nature of the serine 187 DTD isofonn and the use of DTD in Ad5-directed enzyme- procimg gene therapy. Appendix 5.1 describes initial experiments exploring the relative transduction efficiencies of recombinant Ad5 vectors in human tumor cells in hypoxic versus aerobic conditions. 1.8 References
Adams, GE. Failla Memorial Lecture: Redox, radiation and reductive bioactivation. Radial Res 132: 129- 139, 1992.
Alwine JC, Kemp DJ and Stark GR. Method for detection of specific RNAs in agarose gels by transfer to diazobenzyloxymethyl-pa~rand hybndization with DNA probes. Proc Nat1 Acad Sci USA 74: 5350-5354,1977.
Asnis RE. The reduction of hcin by cell-free extracts of furacin-resistant and parent- susceptible strains of Escherichia coli. Arch Biuchem Biuphys 66: 208-2 16, 1957.
Bachur NR, Gordon SL, Gee MV and Kon H. NADPH cytochrome P-450 reductase activation of quinone anticancer agents to fiee radicals. Proc Natl Acad Sci USA 76: 954- 7, 1979.
Bai M. Harfe B, Freimuth P. Mutations that alter an Arg-Gly-Asp (RGD) sequence in the adenovirus type 2 penton base protein abolish its cell-rounding activity and delay virus reproduction in flat cells. J Virol67: 5 198-5205, 1993.
Baiiey SM and Hart IR. Nitroreductase activation of CB1954--an alternative 'suicide' gene system. Gene Ther 4: 80-8 1, 1997.
Beall HD, Mulcahy RT, Siegel D, Traver RD, Gibson NW and Ross D. Metabolism of bioreductive antitumor compounds by purified rat and human DTDs. Cancer Res 54: 3 196-320 1, 1994.
Begleiter A, Leith MK, Curphey TJ and Doherty GP. Induction of DTD in cancer chemoprevention and chemotherapy. Oncol Res 9: 37 1-382, 1997.
Belcourt MF, Hodnick WF, Rockwell S and Sartorelli AC. Exploring the mechanistic aspects of mitomycin antibiotic bioactivation in Chinese hamster ovary cells overexpressing NADPH:cytochrome C (P-450) reductase and DTD. Adv Enzyme Regul 38: 11-33, 1998,
Bergelson JM, Cunningham JA, Droguett G, Kurt-Jones EA, Krithivas A, Hong JS, Horwitz MS, Crowell RL and Finberg RW. Isolation of a comrnon receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science 275: 1320-1323, 1997.
Berkner KL and Sharp PA. Generation of adenovirus by transfection of plasmids. Nucleic Acids Res 11: 6003-6020, 1983.
Bett AJ, Haddara W, Prevec L and Graham FL. An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3. Proc Natl Acad Sci USA 91 :8802-8806, 1994. Beyer RE, Segura-Aguilar J, Di Bernardo S, Cavazzoni M, Fato R, Fiorentini D, Gdli MC, Setti M, Landi L and Lenaz G. The role of DTD in the maintenance of the reduced antioxidant form of coenzyme Q in membrane systems. Proc NatZ Acad Sci USA 93: 2528-2532,1996.
Bi WL, Parysek LM, Wamick R and Stambrook PJ. In vitro evidence that metabolic cooperation is responsible for the bystander effect observed with HSV tk retroviral gene therapy . Hum Gene Ther 4: 725-731, 1993.
Bitzer M, Lauer U, Baurnann C, Spiegel M, Gregor M and Neubert WJ. Sendai virus eficiently infects cells via the asialoglycoprotein receptor and requires the presence of cleaved F, precursor proteins for this alternative route of ce11 entry. J Virol 71: 5481- 5486.1997.
Boland MP, Knox RJ and Roberts JJ. The differences in kinetics of rat and hwnan DT diaphorase result in a differential sensitivity of denved ce11 Iines to CB 1954 (5-(aziridin- 1-y1)-2.4-dinitrobenzarnide). Biochem Pharmacol 41: 867-875, 1991.
Boulanger PA and Blair GE. Expression and interactions of human adenovirus oncoproteins. Biochern J 275: 28 1-299,199 1.
Bndgewater JA, Knox RI, Pitts JD, Collins MK and Springer CJ. The bystander effect of the nitroreductase/CB 1954 enzyme/prodrug system is due to a cell-permeable metabolite. Hum Gene Ther 8: 709-717,1997.
Cerutsi PA. Prooxidant States and tumor promotion. Science 227: 375-38 1, 198 5.
Chardonnet Y and Dales S. Eariy events in the interaction of adenoviruses witb HeLa cells. 1. Penetration of type 5 and intracellular release of the DNA genome. Viroloay 40: 462-477, 1970.
Chen S, Knox R, Wu K, Deng PSK, Zhou D, Bianchet MA and Amzel LM. Molecular bais of the catalytic differences among DTD of human, rat, and mouse. J Bi01 Chem 272: 1437- 1439, 1997.
Cheng L, Ziegelhoffer PR and Yang NS. In vivo promoter activity and transgene expression in mammalian somatic tissues evduated by using particle bombardment. Proc Nafl Acad Sci USA 90: 443-4459, 1993.
Chesis PL, Levin DE, Smith MT, Emster L and Ames BN. Mutagenicity of quinones: pathways of metabolic activation and detoxification. Proc Nat1 Acad Sci USA 81: 1696- 1700' 1984. Cobb LM, Connors TA, Elson LA, Khan AH, Mitchley BC, Ross WC and Whisson ME. 2,4-dinitro-5-ethyleneiminobellzafnide(CB 1954): a potent and selective inhibitor of the growth of the Walker carcinoma 256. Biochem Pharm 18: 15 19-1527, 1969.
Cxistiano RJ, Smith LC, Kay MA, Brinkley BR and Woo SL. Hepatic gene therapy: efficient gene delivery and expression in primary hepatocytes utilizing a conjugated adenovirus-DNA cornplex. Proc Natl Acad Sci USA 90: 1 1548-11552, 1993.
Culver KW, Ram 2, Wallbridge S, Ishii H, Oldfield EH and Blaese RM. In vivo gene transfer with retroviral vector-producer cells for treatment of experimental brain tumors. Science 256: 1550- 1552, 1992.
Doherty GP, Leith MK, Wang X, Curphey TJ and Begleiter A. Induction of DTD by 1,2- di thiole-3 -thiones in human turnour and normal cells and e ffect on anti-tumour activity of bioreductive agents. Br J Cancer 77: 124 1- 1252, 1998.
Edwards YH, Potter J and Hopkinson DA. Hurnan FAD-dependent NAD(P)H diaphorase. Biochem J 187: 429-436, 1980.
Eickelmann P, Schulz WA, Rohde D, Schmitz-Drager B and Sies H. Loss of heterozygosity at the NAD(P)H:quinone oxidoreductase locus associated with increased resistance against mitomycin C in a human bladder carcinoma cell line. Biol Chem Hoppe SeyIer 375: 439-445,1994.
Ernster L and Navazio F. Soluble diaphorase in animal tissues. Acta Chem Scand 12: 595-602,1958.
Ernster L, Ljunggren M and Danielson L. Purification and some properties of a highly dicoumarol-sensitive liver diaphorase. Biochem Biophys Res Commun 2: 88-92,1 960.
Emster L. DT-diaphorase: A histoncal review. Chernica Scripfa 27A: 1 - 13, 1987.
Faig M, Bianchet MA, Talalay P, Chen S, Winski S, Ross D and Arnzel LM. Structures of recombinant human and mouse NAD(P)H:quinone oxidoreductases: species coniparison and structural changes with substrate binding and release. Proc Natl Acad Sci USA 97: 3 177-3 182,2000.
Favreau LV and Pickett CB. Transcriptional regulation of the rat NAD(P)H:quinone reductase gene. Characterization of a DNA-protein interaction at the antioxidant responsive element and induction by 12-O-tetradecanoylphorbol 13-acetate. J Biol Chem 268: 19875- 1988 1,1993.
Field BN, Knipe DM and Howley PM (eds). Virology, Lippincott-Raven, Philadelphia PA, 1996. Freeman SM, Abboud CN, Whartenby KA, Packman CH, Koeplin DS, Moolten FL and Abraham GN. The "bystander effect": tumor regression when a fraction of the tumor mass is genetically modified. Cancer Res 53: 5274-5283, 1993.
Fujisawa-Sehara A, Sogawa K, Yarnane M and Fujii-Kuriyama Y. Characterization of xenobiotic responsive elements upstrearn fiom the dmg-metabolizing cytochrome P-450c gene: A similarity to glucocorticoid regulatory elements. Nucleic Acids Res 15: 4 179- 4191, 1987.
Ghosh-Choudhury G, Haj-Ahmad Y and Graham FL. Protein IX, a minor component of the human adenovirus capsid. is essential for the packaging of full length genomes. EMBO J6: 173301739, 1987.
Goldberg ZL, Cummings BI, Chapman WB, Klamut HJ and Rauth AM. Role of a DTD mutation in the response of anal canal carcinoma to radiation, 5-fluorouracil and mitomycin C. Int J Radiation Oncology 42: 33 t -334, 1998.
Graham FL, Srniley J, Russell WC and Nairn R. Characteristics of a hurnan ce11 line transformed by DNA from human adenovirus type 5. J Gen Virol36: 59-74, 1977.
Graham FL. Covalently closed circles of human adenovirus DNA are infectious. EMBO J 3: 391 7-2922, 1984.
Green NK, Youngs DJ, Neoptolemos P,Friedlos F, Knox RJ, Springer CJ, Anlezark GM, Michael NP, Melton RG, Ford MJ, Young LS, Kerr DJ, Searle PF. Sensitization of colorectal and pancreatic cancer ce11 lines to the procimg 5-(azindin- 1-yl)-2,4- dinitrobenzamide (CB1954) by retroviral transduction and expression of the E. coli nitroreductase gene. Cancer Gene Ther 4: 229-238, 1997.
Gustafson DL, Beall HD, Bolton EM, Ross D and Waldron CA. Expression of human NAD(P)H:quinone oxidoreductase (DTD) in chinese hamster ovary cells: effect on toxicity of antitumor quinones. Mol Pharm 50: 728-735, 1996.
Hata T, Sano Y. Sugawara R, Matsurnae A, Kanarnori K, Shima T and Hoshi T. J An~ibiot(Tokyo) Ser. A., 9: 14 1- 146, 1957.
Huggins CB, Ford E and Fukunishi R. J Erp Med 119: 923-942, 1964
Huggins CB, Ford E and Jensen EV. Carcinogenic aromatic hydrocarbons: Special vulnerability of rats. Science 147: 1 153-1 154, 1965.
Hurwitz MY, Marcus KT, Chevez-Barrios P, Louie K, Aguilar-Cordova E and Hurwitz RL. Suicide gene therapy for treatrnent of retinoblastoma in a murine model. Hum Gene Ther 10: 44 1-448, 1999. Itoh K, Chiba T, Takahashi S, Ishii T, Igarashi K, Katoh Y, Oyake T, Hayashi N, Satoh K. Hatayama I. Yamamoto M, Nabeshima Y. An Nrfî/small Maf heterodimer mediates the induction of phase 11 detoxifiing enzyme genes through anîioxidant response elements. Biochem Biophys Res Commun 236: 3 13-322, 1997.
Itoh K, Wakabayashi N, Katoh Y, Ishii T, Igafashi K, Engel JD and Yamamoto M. Keapl represses nuciear activation of antioxidant responsive elements by Nd2 rhrough binding to the amino-terminal Neh2 domain. Genes Dev 13: 76-86, 1999.
Iyer VN and Szybalski W. Molecular mechanism of mitomycin action. Link of complementary DNA strands. Proc Nat1 Acad Sci USA 50: 355-362, 1963.
Iyer VN and Sqbalski W. Mitomycins and Ponfomycui: chernical mechanism of activation and cross-linking of DNA. Science 145: 55-58, 1964.
Jaiswal AK, McBride OW, Adesnik M and Nebert DW. Hurnan dioxin-inducible cytosolic NAD(P)H:menadione oxidoreductase. cDNA sequence and localization of gene to chromosome 16. J Biol Chem 263: 13572- 13578,1988.
Jaiswal AK, Burnett P, Adesnik M and McBride OW. Nucleotide and deduced amino acid sequence of a human cDNA (NQ02) corresponding to a second member of the NAD(P)H:quinone oxidoreductase gene farnily. Extensive polymorphism at the NQ02 gene locus on chromosome 6. Biochemistry 29: 1899- 1906, 1990.
Jaiswal AK. Human NAD(P)H:quinone oxidoreductase (NQO1) gene structure and induction by dioxin. Biochernisfry30: 10647- 10653, 199 1.
Kasamatsu H and Nakanishi A. How do animal DNA viruses get to the nucleus? Annu Rev Microbiol52: 627-686, 1998.
Kiem HP, von Kalle C. Schuening F and Storb R-Gene therapy and bone marrow transplantation. Curr Opin Oncol 7: 10%1 14, 1995.
Knox RJ, FriedIos F, Marchbank T and Roberts JJ. Bioactivation of CB 1954: reaction of the active 4-hydroxylamino derivative with thioesters to form the ultimate DNA-DNA interstrand crosslinking species. Biochem Pharmacol 42: 169 1- 1697, 199 1.
Knox RJ, Friedlos F and Boland MP. The bioactivation of CB 1954 and its use as a prodrug in antibody-directed enzyme prodrug therapy (ADEPT). Cancer Mer Rev 12: 195- 212, 1993.
Kuehl BL, Paterson JW, Peacock JW, Paterson MC and Rauh AM. Presence of a heterozygous substitution and its reiationship to DTD activity. Br J Cancer 72: 555-561, 1995. Kumaki K, Jensen NM, Shire JG and Nebert DW. Genetic differences in induction of cytosol reduced-NAD(P):menadione oxidoreductase and microsomal aryl hydrocarbon hydroxylase in the mouse. J Bi01 Chem 252: 1 57- 165, 1977.
Li R. Bianchet MA, Talalay P and Amzel LM. The three-dimensional structure of NAD(P)H:quinone reductase, a flavoprotein involved in cancer chernoprotection and chemotherapy: Mechanism of the two-electron reduction. Proc Nat1 Acad Sei USA 92: 8846-8850, 1 995.
Lin AJ, Cosby LA and Sartorelli AC. Quinones as anticancer agents: Potential bioreductive alkylating agents. Cancer Chem Rep 4 @art 2): 23-25, 1974.
Lind C and Ernster L., in Microsornes and Dmg Oxidations. ed. V. Ullrich et al., pp.362- 369. Pergamon Press, Oxford. 1977.
Lind C, Hochstein P and Emster L. DTD as a quinone reductase: a cellular control device against semiquinone and superoxide radical formation. Arch Biochem Biophys 216: 178- 185, 1982.
Lowe SW and Ruley HE. Stabilization of the p53 tumor suppressor is induced by adenovirus 5 E 1A and accompanies apoptosis. Genes Dev 7: 535-545, 1993.
Marshall RS, Paterson MC and Rauth AM. DTD activity and mitomycin C sensitivity in non-transformed ce11 strains derived fiom members of a cancer-prone family. Curcinogenesis 12: 1 175- 1 180, 1991.
Maritus C. The Enzymes, ed. Boyer PD, Lardy H and Myrback K, vol VII, pp. 517-532. Academic Press, New York, 1963.
Martin AE, Burgess BK, Iismaa SE, Smartt CT, Jacobson MR and Dean DR. Construction and characterization of an Azotobacter vinelandii strain with mutations in the genes encoding flavodoxin and ferredoxin 1. J Bacteriol 171: 3 162-3 167, 1989.
Martiniello-Wilks R, Garcia-Aragon J, Daja MM, Russell P, Both GW, Molloy PL, Lockett LJ and Russell PI. In vivo gene therapy for prostate cancer: preclinical evaluation of two different enzyme-directed prodmg therapy systems delivered by identical adenovirus vectors. Hum Gene Ther 9: 1617- 1626, 1998.
McCalla DR, Reuvers A and Kaiser C. Mode of action of nitrofiirazone. J Bacteriol 104: 1 126- 1 134, 1970.
McGrory WJ, Bautista DS and Graham FL. A simple technique for the rescue of early region 1 mutations into infectious human adenovirus type 5. Virology 163: 614-6 17, 1988. Misra V, Klamut HJ and Rauth AM. Transfection of COS-1 cells with DTD cDNA: Role of a base change at position 609. Br J Cancer 77: 123 6- 1240, 1998.
Mitchell PJ and Tjian R. Transcriptional regdation in mamrnalian cells by sequence- specific DNA binding proteins. Science 245: 37 1-378, 1989. iMüller M, Lindmark DG and McLaughlin J. Mode of action of metronidazole on anaerobic microorganisms. In: Finegold SM, McFadzean JA, Roe FCC, editors. Meroinidazole, proceedings of the International Metroinidazole Conference. Montreal: Excerpta Medica 1976. p. 12- 19.
Nabel EG. Gordon D, Yang ZY, Xu L, San H, Plautz GE, Wu BY, Gao X, Huang L and Nabel GJ. Gene transfer in vivo witb DNA-liposome complexes: Lack of autoimrnunity and gonadal localization. Hum Gene Ther 3: 649-656, 1992.
Nabel GJ, Nabel EG, Yang ZY, Fox BA, Plautz GE, Gao X, Huang L, Shu S, Gordon D and Chang AE. Direct gene transfer with DNA-liposome complexes in melanoma: Expression, biologic activity. and lack of toxicity in humans. Proc Nat1 Acad Sci U S A 90: 11307-1 131 1, 1993.
Nebert DW and Gelboin HV. Substrate-inducible microsomal aryl hydroxylase in mammalian ce11 culture. 1. Assay and properties of induced enzyme. J Biol Chem 243: 6242-6249, 1968.
Nermut MV. The architecture of adenovinws. In The Adenoviruses. pp. 5-34, Ginsberg, HS ed. Plenum Press, New York, 1984.
Pettersson U. Structural and nonstructural adenovirus proteins. In The Adenoviruses. pp. 205-270, Ginsberg, HS ed. Plenum Press, New York, 1984.
Phillips RM, Gibson NW and Ross D. A point mutation in both human lung and colon carcinoma ce11 lines leading to a loss of DTD activity. Proc Amer Assoc Cancer Res 36: 525, 1995.
Pickett CB, Williams JB, Lu AY and Carneron RG. Regulation of glutathione transferase and DTD mRNAs in persistent hepatocyte nodules during chernical hepatocarcinogenesis. Proc Natl Acad Sci USA 8 1: 509 1-5095, 1984.
Powis G, Svingen BA and Appel P. Quinone-stimulated superoxide formation by subcellular fractions, isolated hepatocytes, and other cells. Mol Pharmacol 20: 387-394, 1981. Prestera T, Zhang Y, Spencer SR, Wilczak CA and Talalay P. The electrophile counterattack response: protection against neoplasia and toxicity. Adv Enryme Re& 33: 28 1-296, 1993.
Prochaska HJ and Talalay P. Regdatory mechanisms of monofunctional and bifunftional anticarcinogenic enzyme inducers in murine liver. Cancer Res 48: 477604778, 1988.
Rase B. Bartfai T and Emster L. Purification of DTD by afinity chromatography. Occurrence of two subunits and nonlinear Dixon and Scatchard plots of the inhibition by anticoagulants. Arch Biochem Biophys 172: 380-386, 1976.
Rauth AM, Goldberg Z and Misra V. DT-diaphorase: possible roles in cancer chernotherapy and carcinogenesis. Oncol Res 9: 339-349, 1997.
Riley RJ and Workman P. DTD and cancer chemotherapy. Biochem Pham 43: 1657- 1669,1992.
Robertson JA, Chen HC and Nebert DW. NAD(P)H:menadione oxidoreductase. Novel purification of enzyme cDNA and complete arnino acid sequence, and gene regulation. J Biol Chem 261: 15794-1 5799, 1986.
Robinson AJ- Younghusband HB and Bellett AJ. A circular DNA-protein complex fiom adenoviruses. Virology 56: 54-69, 1973.
Ross D, Siegel D, Beall H, Prakash AS, Mulcahy TR and Gibson N. DTD in activation and detoxification of quinones. Cancer Met Rev 12: 83- 1O 1, 1993.
Rosvold EA, McGlynn KA, Lustbader ED and Buetow KH. Identification of an NAD(P)H:quinone oxidoreductase polymorphism and its association with lung cancer and smoking. Pharrnacogenerics 5: 1 99-206, 1995.
Roth JA and Cnstiano W.Gene therapy for cancer: What have we done and where are we going? J Nat2 Cancer lnsr 89: 2 1-39, 1997.
Rushmore TH and Picken CB. Transcriptional regulation of the rat glutathione S- transferase Ya subunit gene. Characterization of a xenobiotic-responsive element controlling inducible expression by phenolic antioxidants. J Biol Chem 265:14648- 14653, 1990.
Sartorelli AC, Hodnick WF, Belcourt MF, Tomasz M, Haffly B, Fischer JJ and Rockwell S. Mitomycin C: a prototypical bioreductive agent. Oncol Res 6: 501-508, 1994.
Schowalter DB, Tubb JC, Liu M, Wilson CB and Kay MA. Heterologous expression of adenovirus E3-gp19K in an El a-deleted adenovirus vector inhibits MHC 1 expression in vitro, but does not prolong transgene expression in vivo. Gene Ther 4: 35 1-360, 1997. Siegel D, Gibson NW, Preusch PC and Ross D. Metabolisrn of mitomycin C by DTD: role in mitomycin C-induced DNA damage and cytotoxicity in human colon carcinoma cells. Cancer Res 50: 7483-7489, 1990.
Siegel D, McGuinness SM, Winski SL, and Ross D. Genotype-phenotype relationships in studies of a pol ymorphism in NAD(P)H:quinone oxidoreductase 1. Pharmacogenetics 9: 113-121. 1999.
Stenflo J, Ferlund P, Egan W and Roepstorff P. Vitamin K dependent modifications of glutarnic acid residues in prothrombin. Proc Nat! Acad Sci USA 7:2730-2733, 1974.
Straus SE. Adenovirus infections in humans. In The Adenoviruses. pp. 45 1-487 Ginsberg, HS ed. Plenum Press, New York, 1984.
Stow ND. Cloning of a DNA fragment from the left-hand terminus of the adenovirus type 2 genome and its use in site-directed mutagenesis. J Virol37: 1 7 1- 180, 198 1 .
Sussenbach JS. The structure of the genome. In The Adenovinws. pp. 35-124, Ginsberg, HS ed. Plenum Press, New York, f 984.
Tikkanen L, Matsushima T, Natori S and Yoshihira K. Mutagenicity of natural naphthoquinones and benzoquinones in the Salmonella/microsome test. Mutat Res 124: 25-34. 1983.
Tomasz M, Lipman R, Chowdary D, Pawlak J, Verdine GL and Nakanishi K. Isolation and structure of a covalent cross-link adduct between mitomycin C and DNA. Science 235: 1204- 1208, 1987.
Traver RD, Horikoshi T, Danenberg KD, Stadlbauer TH, Danenberg PV, Ross D and Gibson NW. NAD(P)H:quinone oxidoreductase gene expression in human colon carcinoma cells: characterization of a mutation which modulates DTD activity and mitomycin sensitivity. Cancer Res 52: 797-802, 1992.
Traver RD, Siegel D, Beall HD, Phillips RM, Gibson NW, Franklin WA and Ross D. Characterization of a polymorphism in NAD(P)H: quinone oxidoreductase (DT- diaphorase). Br J Cancer 75: 69-75, 1997.
Verma IM and Somia N. Gene therapy - promises, problems and prospects. Nature 389: 239-242, 1997.
Walton MI, Smith PJ and Workman P. The role of NAD(P)H: quinone reductase (EC 1.6.99.2, DTD) in the reductive bioactivation of the novel indoloquinone antitumor agent E09. Cancer Commun 3: 199-206, 1 991. Wagner E, Zenke M, Cotten M, Beug H and Bimstiel ML. Transfemn-polycation conjugates as carriers for DNA uptake into cells. Proc Nat1 Acad Sci U S A 87: 3410- 3414. 1990.
Wang X, Doherty GP, Leith MK, Curphey TJ and Begleiter A. Enhanced cytotoxicity of rnitomycin C in human tumour cells with inducers of DTD. Br J Cancer 80: 1223-1230, 1999.
Williams JB, Lu AY, Cameron RG and Pickett CB. Rat liver NAD(P)H:quinone reductase. Construction of a quinone reductase cDNA clone and regulation of quinone reductase mRNA by 3-methylcholanthrene and in persistent hepatocyte nodules induced by chemical carcinogens. J Bi01 Chem 261: 5524-5528, 1986.
Workman P, Morgan JE, Talbot K, Wright KA, Donaldson J and Twentyman PR. CB 1954 revisited. II. Toxicity and antitumour activity . Cancer Chernocher Pharmacol 16: 9- 14, 1986.
Workman P and Stratford IJ. The experimental development of bioreductive dmgs and their role in cancer therapy. Cancer Metastasis Rev 12: 73-82, 1993.
WU K, Knox R, Sun XZ, Joseph P, Jaiswal AK, Zhang D, Deng PS and Chen S. Catalytic properties of NAD(P)H:quinone oxidoreductase-2 (NQ02), a dihydronicotinamide nboside dependent oxidoreductase. Arch Biochem Biophys 347: 22 1-228, 1997.
Wu K. Deng PS, and Chen S. Catalytic properties of a naturally occurring mutant of human NAD(P)H:quinone acceptor oxidoreductase (DT-diaphorase), pro-187 to ser. Pafhophysiology of Lipid Peroxides and Related Free Radicals. pp. 135-148, Japan Sci Soc Press, Tokyo, Japan, 1998.
Yam PY, Yee JK, Ito JI, Sniecinski 1, Doroshow JH, Forman SJ and Zaia JA. Cornparison of amphotropic and pseudotyped VSV-G retroviral transduction in human CD34+ peripheral blood progenitor cells frorn adult donors with HIV-1 infection or cancer. Exp Hematol26: 962-968, 1998.
Yao KS and O'Dwyer P.J. Involvement of NF-kappa B in the induction of NAD(P)H:quinone oxidoreductase (DTD) by hypoxia, oltipraz and mitomycin C. Biochem Pharmacol 49: 275-282, 1995. Transfection of COS4 Cells with DTD cDNA: Role of a Base Change at Position 609 2.1 Abstract
DT-diaphorase (Dm) is a homodimeric flavoenzyme that can provide a defense mechanism against carcinogenesis mediated by dietary or environmental quinones as well as bioactivate quinone-containing chemotherapeutic dmgs. Huma.ce11 lines and strains have been identified with very low to undetectable enzymatic activity associated with a C to T transition at nucleotide 609 of the DTD cDNA. This single base change is predicted
to result in a proline to serine change in arnino acid 187. Human cells homo ygous for
this base transition fail to exhibit Western blot reactivity for DTD, suggesting that this
substitution results in protein instability. To test directly whether this base change affects
DTD enzymatic activity and/or protein stability in vivo, mammalian expression vectors c;ontaining DTD cDNA with or without the nucleotide 609 base transition were transiently transfected into COS-1 cells. Cotransfection with a human growth hormone
(hGH) expression vector allowed normalization for transfection efficiency. COS-1 transfectants expressing the C to T base change displayed at least 10-fold less DTD activity (p < 0.001) and 2- to 3- fold lower protein levels as compared to wild-type transfectants. These results provide, for the first time, evidence that DTD protein encoded by the C to T 609 base transition can be detected in marnmaiian cells and confirm that this protein has reduced enzymatic activity.
2.2 Introduction
DTD [NAD(P)H (reduced nicotinamide adenine dinucleotide, with or without phosphate):quinone oxidoreductase (NQO ,), Enzyme Commission No. 1.6.99.21 is a homodimeric flavoprotein that acts on its substrates by two-electron reduction (Emster, 1967). It utilizes a wide range of substrates, such as aromatic nitro and nitroso cornpounds, phenolic antioxidants. azo dyes, and quinone-containing compounds (Ross et al., 1993; Horie, 1990). Reduction of the quinone ring to its semiquinone radical form can be mediated by one-electron reductases such as NADPH:cytochrome P-450 reductase and NADH:cytochrome b, reductase (Powis, 1987). However, DTD converts the parent quinone to its hydroquinone form in a single step two-electron transfer reaction, thereby bypassing semiquinone radical formation (Iyenagi, 1987). Redox cycling between the parent quinone and the semiquinone species in aerobic cells has ken implicated in carcinogenesis (Koster, 199 1 ). Reduction of various dietary and environmental quinones by DTD rnay protect DNA and cellular organelles against insults fiom reactive oxygen intermediates (Powis, 1987; Lind et al., 1992; Chesis et al., 1984). Conversely, a nurnber of quinone-containing chernotherapeutic drugs such as MMC, the indoloquinone E09, and the aziridinylquinones can be activated by DTD (Begleiter et al., 1992; Walton et al.,
1992; Siegel et al., 1990) into DNA alkylating agents by conversion to their hydroquinone forms in a single step two-elecwn transfer reaction.
In the BE human coIon carcinoma ce11 line, a homozygous C to T base transition in nucleotide 609 of DTD cDNA has ken implicated in causing low to undetectable
DTD activity (Traver et al., 1992). A wide range of DTD activities was observed in human fibroblast strains taken fiom a cancer prone family and unrelated donors (Marshall ei al., 1991). DTD activity appeared to be related to the allelic status at nucleotide 609 (C or T) in these strains, furiher suggesting that this substitution may impair enzyme activity
(Kuehl et al., 1995). In addition, the BE ce11 Iine and human ce11 strains which have no or very low DTD activity expressed nomal mRNA levels but no DTD protein could be detected with polycloaal ador monoclonal antibodies directed against DTD (Traver et al.. 1997: Marshall et al., 1991). To test the hypothesis that this base change impairs
DTD enzymatic activity, mammalian expression vectors containing DTD CDNAS
(derived fiom cells that are homozygous for either the C or T nucleotide at position 609) were prepared and transiently transfected into COS- 1 monkey kidney cells which express very low ievels of endogenous DTD activity. Analysis of recombinant DTD protein levels and enzymatic activity in transfected COS-1 ce11 iysates ailowed for a direct examination of the effect of the C to T transition on DTD protein stability and fiinction.
2.3 Materials and Methods
2.3.1 Chemicals and Reagents
DCPIP, DIC, FAD, bovine serum albumin, B-NADPH, Tween-20, and Tris-HCl were obtained from Sigma Chernical Co. (St. Louis, MO). Lipofectamine reagent was obtained from Life Technologies Inc. (Gibco-BRL, Burlington, ON, Canada).
Recombinant human growth hormone (hGH) levels in culture medium were determined using a commerciaily available mdioimmunoassay kit (loldan Diagnostics, Aurora, ON,
Canada). Hybridoma supematants containing a mixture of two anti-DTD monoclonal antibodies (B771, rat/human DTD reactive; A180 human DTD specific) as well as purified human recombinant DTD were supplied by Dr. David Ross (University of
Colorado Health Sciences Center, Denver, CO). 2.3.2 Preparation of Euhyoîic Expression Vectors
Total RNA was isolated fiom the human fibroblast ce11 strains GM38 and 3701T which are homozygous for the C and T nucleotide at position 609, respectively, as previously described (Kuehl et al, 1995). Bnefly, reverse transcriptase @T)-PCR was used to ampli@ cDNAs using Superscript II reverse transcriptase (Gibco) and AmpliTaq polymerase (Perkin Elmer, Norwaik CT) corresponding to the DTD open reading frame using the following primers:
5' DTD sense: ATGCAAGCTAATCAGCGCCCCGGACTG(bases 23-40 of NQO,;
Hind III restriction site indicated by underline);
3'DTD antisense: CGACGTCGACAAGGAAATCCAGGCTAAGGA (bases 879-898 of
DTD; Sa1 1 site indicated by underline).
As show in Figure 2.1, the resdting 895 base pair fragments @TD 609C or T), containing the full-length DTD coding region were inserted into the Hind III and Sal I sites of pBAPr-1 -ne0 (Invitrogen Co., San Diego, CA)(Kuehl, 1995). However, in these constructs the DTD cDNA is in the wrong orientation with respect to the p-actin promoter of pf3APr-1 -neo. These constructs were subsequently digested with K~ndIII and
Sa1 1 to release DTD cDNAs, which were then gel-purified and subcloned into the promotor-less plasmid pBLCAT3 and designated as pBLCAT3.DTD (Invitrogen Co.).
This plasmid provided Hind III and Xba 1 sites flanking the DTD inserts to allow their cloning in the proper orientation with respect to the CMV promoter of pRcICMV. The
DTD cDNA inserts were isolated and gel-purified by digestion of pBLCAT3.DTD with
Hind III and Xba I and ligated into the pRc/CMV expression vector (Invitrogen Co.). Figure 2.1 : Cloning scheme for p~c/~~~.~~~M>9Cand ~RC/CMV.DTD~~, shown on the following pages. ppAPr-1-neo.DTD contained DTD cDNA (609C or T) in improper orientation with respect to the B-actin promoter. DTD cDNA inserts were released by digestion with Hind III and Sa1 1 and ligated into these sites in pBLCAT3 resulting in pBLCAT3.DTD. The flanking Hind III and Xba 1 sites allowed cloning of the DTD inserts in proper orientation with respect to the CMV promoter of the mammalian expression vector pRc/CMV. The Hind III - Xba 1 fiagrnent containing the DTD cDNA was inserted into these sites in pRc/CMV to yield ~Rc/CMV.DTD- and ~RC/CMV.DTD~'~~. I \ / \ Cut with I I '\ Hind 111 & Sa1 I I \ / \ I \ I \ I Sr ' cutwith Hind 111 SPI 1 Hind III & al I I I \ DTD609C or T I J 5' 3' / /
iI i
1
Hind III + \ DTD609C or T
P~YA Cut with ------_____Hind III & Xk1 'confinued on next page \ \ \ \ \ \ \ '4 continued from previous page t / Hind III Xbal I ./' Cut w ith \ DTD609C or T ,' Hind III & Xba I 5' 872bp 3' ---_-- i - 1 I l Ligate I I Hind III Xbal \ DTD609C orT / Constmcts containing DTD cDNA inserts with a C or T nucleotide at position 609
were designated as ~RC/CMV.DTD- and ~RC/CMV.DTD~~,respectively (Figure 2.1).
Constructs were purified for transfection experiments by two rounds of CsCl continuous
density gradient centrifugation. The sequence integrity of ~RC/CMV.DTD~and
~Rc/CMV.DTD~'~~were verified by Sanger sequencing of both strands using Sequenase
Version 2.0 T7 DNA polymerase (United States Biochemical, Cleveland, OH).
2.3.3 Cell Culture
COS-1 monkey kidney cells were obtained fiom the Amencan Type Culture
Collection (Manassas, VA) and grown in Alpha Minimum Essential Medium
supplemented with 10% fetal bovine senim (FBS, Sigma Chernical Co., St. Louis, MO,
growth medium) and maintained in a humidified atmosphere containing 5% CO2 at
37°C.
2.3.4 Transient Transfection of DTD cDNA
COS4 cells were seeded on 100 mm diarneter tissue culture dishes (NUNC,
Denmark) 24h pior to transfection at a density of 90-100 cells/mm2 in growth medium and maintained in a humidified atmosphere with 5% CO2 at 37°C. Lipofectamine transfection was performed according to the manufacturer's protocol (Gibco-BRL).
Briefly, transfections were performed using 20 pL (2 mg / mL) Lipofectamine and 5 pg of ~P,C/CMV.DTD- or ~RC/CMV.DTD~~coûansfected with 5 pg of pXGH5 (Selden el al., 1986, Figure 2.2) in 1.6 ml, of antibiotic-fiee Alpha Minimal Essential Medium. Figure 2.2: Schematic of pXGHS hGH expression vector, used to monitor transfection efficiency. This vector provides for hGH secretion under control of the mouse metallothionein-1 promoter (mMT-1). It also contains a MCS to allow cloning of gene expression regulatory elements where hGH secretion acts as a reporter. Adapted fiom Seldon et a[., 1996. Mock-trans fected cells (Lipo fectamine only) and vec tor-control transfectants @Rc/CMV vector alone) were similarly treated. Cells were incubated with this mixture for 5h, followed by an overnight incubation with the addition of 10% FBS at which the the medium was replaced with fkesh growth medium. Twenty-four hours later, an aliquot of growth medium was retained for analysis of recombinant hGH levels and cells were harvested for recombinant DTD enzymatic assays.
2.3.5 Assay for DTD Enzymatic Activity
Transfectants were harvested by scraping, centrifuged at 250 x g for 5 min at 4"C, resuspended in 1 mL phosphate-buffered saline (PBS), and lysed by exposure to five 10 s ultrasound pulses at 10 s intervals using a Vibra Ce11 sonicator (Sonics and Materials,
Danbury CT). Total ce11 extracts were centrifuged in an Eppendorf (Rexdale, Ontario,
Canada) 541 5 C microcentrifuge at 16000 x g for 10 minutes at 4°C. The supematants were retained. DTD activity. expressed as nrnol/min/mg total protein, was determined according to a modification (Kuehl et al., 1995) of an assay developed by Benson el al.
(1 980) and is expressed as dicumarol inhibitable activity measured by the loss of DCPIP at 600 nrn. DTD activities in ce11 extracts were determined in the presence of 35 pM
DCPIP in a buffer containing 25 mM Tris-HCl (pH 7.4), 0.23 rng/mL bovine serum albumin, 0.2 rnM NADPH, 0.01% Tween-20,4 pM FAD, with or without 25 pM DIC.
Protein concentration was measured using the Bradford method (1 976). 2.3.6 Western Blot Analysis
COS-1 cells were transfected with DTD expression vectors, grown to a density of
4 x 10' cells in 100 mm diameter tissue culture dishes, and harvested by scraping in 2
mL PBS. Half the ce11 suspension was used to determine DTD activity as described above
while the remaining half was used for Western blot analysis (Bumette, 198 1 ). Ce11 lysates
were prepared by resuspending ce11 pellets in 200 pL ce11 harvest buffer (0.1 M Tris-HCI,
1% SDS , 10 mM EDTA, 20 mM DTT) and incubation in a boiling water bath for two
minutes. Protein concentration was measured using the Bradford method (1976) and
protein (20 pg / lane) was separated by 12 % SDS polyacrylamide gel electrophoresis and eIecîro-transferred to nitrocellulose membranes. Following tramfer, membranes were
blocked in Tris-buffered saline (TBS) containing 5% skim milk powder and 1% heat-
inactivated FBS for 2h, and then incubated overnight with 15 mL of hybridoma supernatant containing a mixture of two of the anti-DTD monoclonal antibodies at 4°C.
Blots were washed in TBS containing 0.05% Tween-20 and incubated for 90 min with a
1 :4000 dilution of goat anti-mouse home radish peroxidase conjugated antibody in TBS containing 1% skim rnilk powder and 1% heat-inactivated FBS. Bands were visualized using an enhanced chemilurninescense detection kit (Amersharn Life Science, Oakville
ON. Canada) and autoradiography. Purified human recombinant DTD (20 ng) was included as a positive molecular weight control. Densitometric analysis was pedormed using a Computing Densitometer and ImageQuant v. 3.3 software package (Molecular
Dynarnics, Sunnyvale, CA). Band densities were quantified in ng relative to a positive molecular weight control. 2.3.7 Statistical Analysis
DTD activity was expressed as nmol/min/mg proteinhg hGH. Two-way analysis of variance (ANOVA) was used to compare the means of enzymatic activities in COS-1 ce1 1s transfected with either p~c/~MV.~~~m>Cor ~RCICMV.DTD~~. Data were evaluated as 2 treatments @Rc/CMV.DTDm or ~RC/CMV.DTD~~) each represented by 3 separate experirnents. Two-way ANOVA allows comparkon of means of treatments by separating the intra-experhental variation fiom inter-experimental variation.
2.4 Results
2.4.1 Expression of DTD- and DTD- in COS4 Cells
COS-1 cells were transiently transfected with eukaryotic expression vectors
(Figure 2.1) containing DTD cDNAs prepared from skin fibroblast strains homozygous for either the C or T nucleotide at position 609 (Kuehl et al., 1995). Untransfected COS-
1 cells displayed low levels of DTD activity (within the limit of detection of the assay) as did mock-transfected and vector control-transfected cells (mean 4 s.e.m. of three determinations: 3.6 f 1.3; 3.3 + 1.3; and 2.0 i 0.3 nmol/min/mg protein, respectively).
DTD activities of COS-1 cells transfected with pRc1CMV.DTDm were significantly higher than controls, ranging frorn 70- to 130- fold greater than vector control-transfected cells. Cells transfected with pRc/CMV.DTDmT , on the other hand, had DTD activities that were only 8-fold elevated relative to vector controls and were substantially lower t han PRCICMV.DTD- - transfected cells. When corrected for average background enzymatic activity, two-way ANOVA indicated that DTD activities in cells transfected wi th p~c/~~~.DTD- were significantly different (p< To control for the possibility that differences in DTD activities mise fiom differences in transfection efficiencies, cells were simultaneously transfected with the pXGHS plasmid, which provides for recombinant hGH expression from the rnouse metallothionein-1 promoter (mMP-1). Two-way ANOVA indicated that recombinant hGH levels were similar in ce11 strains transfected with either pRc/CMV.DTDMPCor p~~/~~~.~~~609T@ »o. 1). DTD activities were nomalized to recombinant hGH levels (transfection efficiency) and two-way ANOVA confirmed that DTD activities in ~RC/CMV.DTD- and ~RC/CMV.DTD- transfectants were significantly different @<<0.001) (Figure 2.3). COS- 1 cells transfected with ~RC/CMV.DID~displayed mean + s.e.m. DTD activities of 260 t 37 nmoWmin/mg proteinhg hGH, which were 10-fold greater than activities observed in ~RcICMV.DTD~"~~transfectants (25 + 15 nmol/min/mg proteidng hGH). These results confirmed that DTD cDNAs containing a T nucleotide at position 609 encode a DTD protein with reduced enzymatic activity. 2.4.2 Western Biot Aoalysis To examine whether the C to T nucleotide substitution Ieads to decreased protein stability, recombinant DTD protein levels in COS4 cells transfected with either the p~c/~~~.~~~W9Cor ~RCICMV.DTD- constructs were also exarnined. As show in Figure 2.4A, Western blot analysis of two independent ~RC/CMV.DTD- or I 609C normalized DTD activity 31 01 609T nonnalized DTD activity - 3169 f 93 9m - 40 f 3 - 2Of9 13f 3 = # Iz - I Experiment I Experiment 2 Experiment 3 Figure 2.3: Mean DTD activities in COS-1 cells transfected with plasmid constmcts ~RC/CMV.DTD- and ~RC/CMV.DTD~~.DT-diaphorase activities are normalized for transfected efficiencies and are expressed as nmol min" mg-'protein / ng m~-'hGH secretion. Values, given above error bars, represent the means * s.e.m. (p«0.001, two- way ANOVA for differences between 609C and 609T activity) in three independent experiments where each experiment consisted of three separate transfection dishes of COS- I cells infected with p~~/CM~.DTD609Cand three with pRc/CMV.DTDOWT. A rDTD 1 609C II 6ûQT 1 Vec Untrans I Figure 2.4: A) Western blot analysis of ce11 lysates of COS4 cells iransfected with ~RC/CMV.DTD~(609C), pRc/CMV.DTDmT (609T), pRc/CMV vector alone (Vec), and untransfected controls (Untrans) in experiment 1 of Figure 2.3. Blots were incubated with a mixture of DTD mAbs as described in Materials and Methods (section 2.3) and processed using an enhanced cherniluminescence detection kit and autoradiography. Purified hwnan DTD (20 ng) was included as a positive molecular weight control (rDTD). 8) Densitometry analysis of DTD band intensities. Values represent mean 9s.d. of five densitornetric readings for each lane standardized to rDTD. pR~/CMV.DTD6mtransfected ce11 extracts demonstrated that both mutant and wild- type recombinant DTD are expressed at high levels in COS-1 cells. The mutant DTD protein appears to run slightly faster than the wild-type DTD protein and densitometery (Figure 2.4B) indicated that ~Rc/CMV.DTD- transfectants contained approximately 3- fold greater DTD protein than ~RC/CMV.DTD~~transfectants. However, the rnean amounts of expressed recombinant DTD protein per 20 pg of total protein loaded fiom ~Rc/CMV.DTD- - transfected cells and ~RC/CMV.DTD~- transfected cells were estimated to be 240 and 80 ng, respectively. These results suggest that lower mutant DTD enzyme activities cannot be entirely accounted for by a decrease in protein stabilities, as discussed below. 2.5 Discussion DTD has been show to reduce quinone-containing chemotherapeutic dnigs such as MMC and the indoloquinone E09 to their hydroquinone forrns leading to the formation of DNA alkylating agents (Workrnan, 1994; Venviej et al., 1994). These dmgs rnay be used to target tumor cells that are rich in DTD. Elevated DTD activity has been observed in a nurnber of tumor ce11 lines (Robertson et al., 1992). Turnor biopsy materiai fiom patient lung, colon, and breast have also ken shown to contain elevated DTD activities compared to surrounding normal tissue (Koudstaal et al., 1975; Schlager and Powis, 1990). The actual role of DTD in controlling ce11 sensitivity to quinone-containing dmgs is, however, controversial since one electron reductases may also play important roles, especially in hypoxic cells (Rockwell et ol., 1993; Rauh et al., 1993). The recent work of Fitzsimmons et al (1996) showing a correlation between DTD enzymatic activity and aerobic sensitivity to MMC and E09 in the NCI human tumor ce11 line panel is currentiy the best evidence for this role. The one electron reductases NADPH:cytochrome P-450 reductase and NADH:cytochrome b, reductase failed to display such a correlation in this study. The alternative roie for DTD as a detoxifjring agent, by one step two-electron reduction of dietary and environmental quinones to redox active products remains a potentially important function for the enzyme (Powis, 1987). Cells that are homozygous for a C to T nucleotide transition at position 609 of the DTD cDNA were found to contain low to undetectable DTD enzymatic activities (Traver et al., 1992). It is estimated that approximately 40% of individuals are heterozygous for this nucleotide transition while 10% are homozygous (Kuehl et al., 1995). Limited data have shown that this point mutation is widespread (Rosvold et al., 1995; Rothman et al., 1996). occurs in both normal tissues and tumors of the same individuai (Eickelmann et al., 1994; Traver et al., 1997) and may occur with altered fiequencies in different ethnic groups (Rothman et al., 1996). Traver et al (1997) have reported that a recombinant serine 187 mutant DTD expressed in E. coli. exhibits only 2% of the specific enzymatic activity of wild type protein. Wu et al (1 997) have similarly expressed the mutant isoform in E.coli and have shown that relative to wild-type protein it has 34% specific activity with DCPIP as substrate. In addition to measuring catalytic properties they found that the dissociation constant for FAD of the mutant isoform was twenty times the wild-type enzyme and suggested the point mutation changes enzyme conformation. The identification of human cells that are homoygous for the DTD nucleotide 609T point mutation and do not express detectable DTD proteh (Marshall et al., 1991; Kuehl et al., 1995; Traver et al., 1997) raised the possibility that this mutation results in destabilization of the protein product. The results in Figure 2.4 indicate that there is a 3- fold reduction of mutant relative to wild-type protein in transfected COS4 cells. This value depends on the assumption that the mixture of monoclonal antibodies (B771 and A 180), used for Western blot analysis, binds with equal affinity to both the wild-type and mutant proteins. However, since one of the two antibodies binds to an epitope near the beginning of the region encoding by exon 6 of the DTD gene (Dr. David Ross, University of Colorado Health Sciences Center, personal communication), it is possible that the mutant isoform may not be as readily detected since the proline to serine change occurs in exon 6. Decreased mutant DTD protein stability has also ken reported by Pan et al (1 995) in studies of an arginine to tryptophan 139 substituted isoform in the HCT 1 16- R30A human colon cancer ce11 subline, which is resistant to MMC and homozygous for the point mutation encoding this isoform. This mutant DTD protein was detected by Western blot anaiysis in HCT 116-R30A cells but at 5% of the levels present in the parental MMC sensitive line. This mutant isoform was also expressed at detectable levels in E.coli and COS-7 cells (Hu er al., 1996). Therefore, stable expression of this DTD isoform may be ce11 type-specific. However, as the tryptophan 139 mutant has enzymatic activity comparable to the wild-type enzyme, the reduced activity displayed by HCT 1 16- R30A cells has been attributed to the low levels of mutant DTD protein. The present work suggests that the senne 187 mutation results in a reduction in both enzymatic activity and protein stability. Therefore, cells which express the senne 187 mutant would be predicted to be resistant to cinigs targeted for DTD activation. The proline to senne substitution at amino acid 187 in DTD may predispose individuals to cancer by removing an enzymatic defense mechanism againa carcinogenesis. However, the importance of DTD in cancer prevention is not clear, and factors such as the interplay of DTD with other enzymes acting on comrnon substrates need to be investigated Mer.The results of this study are consistent with the mode1 for a causal link between the C to T mutation at nucleotide 609 and predisposition to cancer based on ineffective xenobiotic detoxification. This mutation may also serve as a prognostic indicator for the effectiveness of chemotherapeutic dmgs activated by this enzyme. 2.6 References Begleiter A, Robotham E and Leith MK. Role of NAD(P)H:(quinone acceptor) oxidoreductase @T-diaphorase) in activation of mitomycin C under hypoxia. Mol Pham 41 : 677-682, 1992. Benson AM, Hunkler MJ and Talalay P. Increase NAD(P)H:quinone reductase by dietary antioxidants. Possible role in protection against carcinogenesis and toxicity. Biochem 77: 52 1 6-5220, 1980. Bradford MM. A rapid and sensitive method for the quantification of microgram quantities utiliùng the principle of protein-dye binding. Anal Biochem 72: 248-254, 1976. Bumette WN. "Western blotting": Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylarnide gels to unmodified nitrocellulose and radiographie detection with antibody and radioiodinated protein A. Anal Biochem 1 12: 195-203, 1981. Chesis PL, Levin DE, Smith MT, Emster L and Ames BN. Mutagenecity of quinones: pathways of metabolic activation and detoxification. Proc Natl Acad Sci USA 81: 1696- 1700, 1984, Eickelmann P, Schulz WA, Rohde B, Schitz-Drager B and Sies H. Loss of heterozygosity at the NAD(P)H:quinone oxidoreductase locus associated with increased resistame against mitornycin C in a human bladder ce11 line. Biol Chem Hoppe-Seyler 375: 439- 445. 1994. Ernster L. DT-diaphorase. Methodi Enzymol10: 309-3 17, 1967. Fitzsimmons SA, Workrnan P, Grever M, Paul1 K, Camalier R and Lewis AD. Reductase expression across the National Cancer Institute tumor ce11 line panel: correlation with sensitivity to mitomycin C and E09. J Nad Cancer htir 88: 259-269, 1996. Hone S. Advances in research on DT-diaphorase. Kitasaro Arch Exp Med 63: 11-30, 1990. Hu LT, Stamberg l and Pan SS. The NAD(P)H:quinone oxidoreductase locus in human colon carcinoma HCT 116 cells resistant to mitomycin C. Cancer Res 56: 5253-5259, 1996. lyenagi T. On the mechanism of one- and two-electron transfer by flavin enzymes. Chem Scr 27A: 3 1-36, 1987. Koster AS. Bioreductive activation of quinones: a mixed blessing. Pham Weekl [Sci] 13: 123-126, 1991. Koudstaal J, Makkink B and Overdip SH. Enzyme histochemicai pattern in human tumors - II. Oxidoreductases in the carcinoma of colon and hast. Eur J Cancer 11: 11 1-1 15, 1975. Kuehl BK, Paterson JWE, Peacock IW,Paterson MC and Rauth AM. Presence of a heterozygous substitution and its relationship to DT-diaphorase activity . Br J Cancer 72: 555-561,1995. Kuehl BK. The involvment of DT-diaphorase in mitomycin C sensitivity and in a cancer- prone phenotype. PhD thesis. University of Toronto. Dept Medical Biophysics. 1995. Lind C, Hochstein P and Emster L. DT-diaphorase as a quinoine reductase: a cellular control device against semiquinone and superoxide radicai formation. Arch Bioch Biophys 216: 178-185, 1992. Marshall RS, Paterson MC and Rauth AM. DT-diaphorase activity and mitomycin C sensitivity in non-transfonned ce11 strains denved Erom members of a cancer-prone family. Carcinogenesis 12: 1175-1180, 1991. Pan SS, Forrest GL, Akman SA and Hu LT. NAD(P)H:quinone oxidoreductase expression and mitomycin C resistance developed by human colon cancer HCT 116 cells. Cancer Res 55: 330-335 , 1995. Powis G. Metabolism and reactions of quinoid anticancer agents. Pharmacol Ther 35: 157-162, 1987. Rauth AM, Marshall RS and Kuehl BL. Cellular approaches to bioreductive drug mechanisms. Cancer Metastasis Rev 12: 153-164, 1993. Robertson N, Stratford IJ, Houlbrook S, Cannichael J and Adams GE. The sensitivity of human tumor cells to quinone bioreductive drugs. What role for DT-diaphorase? Biochem Pharm 44: 409-412,1992. Rockwell S, Sartorelli AC, Tomasz M and Kennedy KA (1993). Cellular pharmacology of quinone bioreductive alkylating agents. Cancer Metastasis Rev 12: 165-1 76, 1993. Ross DlSiegel D, Beall H, Prakash AS, Mulchay TM and Gibson NW. DT-diaphorase in activation and detoxification of quinones. Cancer Metustasis Rev 12: 83-101, 1993. Rosvold EA, McGlynn KA, Lustbader ED, Buetow RH. Identification of an NAD(P)H:quinone oxidoreductase polymorphism and its association with lung cancer and smoking. Pharmacogenetics 5: 1 99-206,1995. Rohan N, Traver RD, Smith MT, Hayes RB, Li G-L, Campleman S, Dosemeci M, Zhang L, Linet M, Wacholder S, Yin S-N and Ross D. Lack of NAD(P)H:quinone oxidoreductase activity (NQO,) is associated with increased benzene hematotoxicity. Proc Amer Assoc Can Res 37: 258, 1996. Schlager JJ and Powis G. Cytosolic NAD(P)H:(quinone acceptor) oxidoreductase in human normal and tumor tissue: effects of cigarette smoking and alcohol. Int J Cancer 45: 403-409, 1990. Selden RF, Howie KB, Rowe ME, Goodman HM and Moore DD. Human growth hormone as a reporter gene in regulation studies employing transient gene expression. Mol Cell Biol9: 3 173-3 179, 1986. Siegel D, Gibson NW, Preusch, PC and Ross D. Metabolism of NAD(P)H:(quinone acceptor) oxidoreductase (DT-diaphorase): Role in diequone-induced DNA damage and cytotoxicity in human colon carcinoma cells. Cancer Res 50: 7293-7300, 1990. Traver RD, Horikoshi T, Danenberg K, Stadlbauer THW, Danenberg PV, Ross D and Gibson NW. NAD(P)H:quinone acceptor oxidoreductase gene expression in human colon carcinoma cells: characterization of a mutation which modulates DT-diaphorase activity and mitomycin C sensitivity. Cancer Res 52: 797-802, 1992. Traver RD, Siegel D, Beall HD, Phillips KM, Gibson NW, Franklin WA and Gibson NW. Characterization of a polymorphism in NAD(P)H:quinone oxidoreductase (DT- diaphorase). Br J Cancer 75: 69-75, 1997. Verwiej J, Aamdai S, Schellens J, Koier 1 and Lund B. Clinical studies with E09, a new indoloquinone bioreductive alkylating cytotoxic agent. Oncol Res 6: 5 19-523, 1994. Walton MI, Sugget N and Workrnan . The role of human and rodent DT-diaphorase in the reductive metabolism of hypoxic ce11 cytotoxins. 1.J Radiat Oncol Biol Phys 22: 643- 647, 1 992. Workman P. Enzyme-directed bioreductive drug development revisited: A comrnentary on recent progress and future prospects with emphasis on quinone anticancer agents and quinone metaboliùng enzymes, particularly DT-diaphorase. Oncol Res 6: 46 1-475, 1994. Wu K, Deng PS-K and Chen S. Catalytic properties of a naturally occuring mutant of human NAD(P)H:quinone acceptor oxidoreductase (DT-diaphorase), Pro 187 to Ser, in "Pathophysiology of Lipid Peroxidases and Related Free Radicals" K.Yagi, ed. Japan Scientific Societies Press, Tokyo, 1997. Assessrnent of the Relationship Between Genotypic Status of a DT-dia phorsse Point Mutation and Enzymatic Activity 3.1 Abstract DT-diaphorase (DTD), primarily a cytosolic reductase, has been implicated as an activator of chemotherapeutic prodrugs and a detoxifier of certain potentially carcinogenic xenobiotics. A common C to T nucleotide 609 substitution in DTD cDNA has been associated with protein instability and reduced catalytic activity. The degree to which the allelic status of the substitution correlates with enzymatic activity was assessed in 45 normal human skin fibroblast strains using a PCR-RFLP assay. Included in this study was the 3437T strain which is unique in that it is heterozygous for the base change yet contains undetectable enzymatic activity. An ailele-specific RT-PCR-RFLP technique attributed this phenornenon to exclusive DTD mRNA expression fiom the variant allele. Overlap in activities was observed between individual strains homozygous for the wild- type allele and heterozygotes but the former group displayed enzymatic activity that was on average 2-fold higher. Western Blot analysis of the two strains in this panel that are homozygous for the variant allele revealed that they express relatively low amounts of DTD protein, consistent with the role of the substitution in protein instability. mis work confirms that genotypic status is a reliable initial estimate of DTD activity. 3.2 Introduction DTD [NAD(P)H:quinone oxidoreductase (NQO 1), Enzyme Commission No.1.6.99.21 is a homodimenc flavoenyme that acts on a wide range of cytosolic substrates including quinones and their derivatives (Ernster, 1967; Talalay et al., 1995). DTD is present in many tissues but is most abundant in liver (Jaiswal, 1994) and can also be elevated in colon, liver. and breast tumors relative to surrounding normal tissue (Schlager and Powis, 1990). The obligate two-electron reductase activity of DTD is the basis for its postulated role as a defense mechanism against the carcinogenic effects of quinone xenobiotics, which are reduced to hydroquinone products by DTD, thereby bypassing the formation of semiquinones and subsequent reactive oxygen intermediates (Lind et al., 1982; Emster and Navazio, 1987). A DTD gene knockout mouse exhibits increased quinone sensitivity, consistent with its role as a protective mechanism against xenobiotic toxicity (Radjendirane et al., 1998). A broad variety of structuraily unrelated compounds have been shown to induce DTD expression (Prochaska and Talalay, 1988; Begleiter el al., 1997). Many chemotherapeutic agents can be reductively activated by DTD (Workman and Stratford, 1993; Knox et al., 1993; Sartorelli et al., 1994; Winski et al., 1998), but one concern with the use of these prodrugs clinically is the variability in DTD enzyme activities observed in different tissue types (Marin et al., 1997) and turnors (Siegel et al., 1998). It has also ken suggested that the therapeutic index of drugs activated by DTD (Sreerama et al., 1995) can be increased through transcriptional activators (Yao et al., 1997) such as 1,2-dithiole-3-thiones which can selectively induce DTD activity in turnor cells (Doherty et al., 1998). DTD and related nitroreductases are also king studied as prodmg activating enzymes in vinis-directed enzyme-prodnig therapies (Fnedlos et al-? 1998; Searle et al., 1998; Warrington et al., 1998). Traver et al (1992) identified a C to T base change at position 609 of the DTD cDNA which would confer a proline to serine substitution at amino acid 187 of the DTD protein. This base change has been associated with low or absent DTD activity (Traver et al.. 1997; Kuehl et al., 1999, and transfection of eukaryotic cells with DTD expression vectors containing either a C or T nucleotide at position 609 demonstrated that the proline to serine substitution results in decreased catalytic activity and protein levels (Misra et al., 1998). Furthemore, mutant protein purified fiom an E.coli expression system exhibited 2% of the enzymatic activity of wild-type DTD (Traver et al., 1997). Allelic distribution of this common base change is in accordance with the Hardy-Weinberg equilibriurn with estimates of the fiequency of the 609T allele ranging fiom 0.16 to 0.49 depending on ethnic background (Kelsey et al, 1997; Traver et al., 1997; Gaedigk et al., 1998). Marshall et al. (1991) obsewed a relationship between DTD activity and mitomycin C sensitivity in a group of skin fibroblast strains donated by members of a cancer-prone farnily. However, in one heterozygotic ce11 strain, 3437T, DTD activity was undetectable (Kuehl et al., 1995), which conflicted with previous reports showing that the allelic status of the nucleotide 609 base change is a predictor for enzyme activity, with heterozygotes displaying activities intermediate to homozygous 609C (wild-type) and 609T (mutant) ce11 strains (Traver et al., 1992). DTD mRNA was present in this ce11 strain in similar quantities to cells exhibiting high DTD activities although the allelic origin of this rnRNA was not detennined. These results bring into question the reliability of predictions of DTD activity on the basis of the genomic status of the nucleotide 609 base change. In the present report, a panel of normal human skin fibroblast ce11 strains was assessed for the allelic distribution of the nucleotide 609 base change and the relationship between allelic status and DTD activity. As well, the allelic ongin(s) of DTD mRNA expression was determined in 3437T cells, as well as seven selected strains fiom the skin fibroblast panel. 3.3 Materials and Methods 3.3.1 Chernicals and Reagents Al1 reagents used for the DTD enzymatic assay were purchased fiom Sigma Chemicai Co. (St. Louis, MO, USA). Materials used for genomic and RT-PCR, including AmpliTaq polymerase were obtained fiom Roche Molecular Systems Inc. (Branchburg, NJ, USA). Hinf 1 and proteinase K were purchased fiom Gibco-BRL (Burlington, ON, Canada) and Nonident P-40, gelatin, and Tween-20 were obtained fiom Sigma Chernical Co. DTD monoclonal antibodies B77 t and A1 80 were kindly supplied by Dr. David Ross (University of Colorado Health Sciences Center, Denver CO, USA). Mouse anti-B- tubulin was obtained fiom Sigma. The protease inhibitor "cocktail" (cat no. 1697498) was purchased fiom Roche Bioscience (Palo Alto CA, USA). 3.3.2 CeU Culture Skin fibroblast ce11 strains used in this study were generously provided by Dr. Peter Ray and Ms. Nancy Cracknel of the Hospital for Sick Children, Toronto, Canada. 3437T cells were supplied by Dr. Malcom Paterson (Cross Cancer Institute, University of Alberta, Edmonton, Alberta, Canada). Al1 cells were grown in Alpha Minimal Essentiai Medium supplemented with 10% fetal bovine semm (Cansera, Rexdale ON, Canada) and maintained in a humidified atmosphere containing 5% CO2at 37°C. 3.3.3 Genomic PCR-Hinf 1 RFLP Assay Cells were grown in 175 cm' tissue culture flasks (Nalge Nunc International, Denmark), harvested by scraping and centrifuged at 250 x g. The pellet was resuspended in 1 mL lysis buffer [50 rnM KCI, 10 mM Tris-HC1 (pH 8.3), 2.5 mM MgCl,, 0.5 (rg proteinase K, 0.1 mg/d gelatin, 0.45% Nonident P-40, and 0.45% Tween-201 and heated at 55 OC for 1 hour followed by 95 OC for 10 min. To assess the status of the nucleotide 609 base change (Figure 3. l), pnmers were used to amplify the entire coding region of exon 6 producing a 405 bp product containing the nucleotide 609 base change as well as a native Hinf 1 site as described previously (Goldberg et al., 1998). Briefly, 5 pL of genomic extract was used in a 50 pL total PCR reaction containing reaction buffer (50 mM KCl, 10 mM Tris-HCI (pH 8.3), and 0.001% (w/v) gelatin), 4 mM MgCl,, 0.8 mM dNTP mix and 5 U AmpliTaq Polymerase with 100 ng of forward and reverse 24 mer prïmers: TGA I 1 DTD gene 118 165 131 114 102 - 1833 bp '=, 0 \ \ \ \ Mnf I * Hinf I 121 bp 1 151 bp 1 133 bp 405 bp Mnf l digest Allelic Status : C 1 C TIT CIT - Figure 3.1: Genomic PCR-RFLP schematic for DTD nucleotide 609 base change. Shaded boxes indicate exons of the DTD gene (- 20 kb) with number and length indicated above and below, respectively. The C to T base change, corresponding to nucleotide 609 of DTD cDNA results in the formation of a novel Hhf 1 site (shown with an asterisk). Assignment of allelic status of the base change based on banding pattern is shown. The PCR product contains a constitutive Hinf 1 site downstream of the base change which was used as a positive cutting control. The recognition sequence for H~nf1 is GANTC, where N represents any nucleotide. 5' GAG AAG CCC AGA CCA ACT TCT GTT 3' - forward primer (intron 5 of DTD gene) 5' CCA GGC TAA GGA ATC TCA TTT TCT 3' - reverse primer (exon 6 of DTD gene) A 5 min 94 OC denaturation step followed by 36 cycles of 1 min 20 sec at 94 OC, 1 min 20 sec at 60 OC (annealhg step), and 1 min 50 sec at 72 OC (polymerization step) and a final extension for 10 min at 72°C was performed using a PTC-100 thennocycler (MJ Research Inc. Wattertown, NJ). Forty percent of the PCR reaction was subjected to Kinf 1 digestion in a 24 pL mixture containing 5 mM Tris-HC1 (pH 8.0), 1 mM MgCl,, 5 mM NaCl and 10 U 15nf 1 for 1 hour at 37 OC. Fragments were separated on a 20% polyacrylarnide gel and visualized by ethidium bromide staining (10 pg / mL). Allelic status was assigned on the basis of band profile as previously described (Goldberg et al., 1998 and Figure 3.1). 3.3.4 DTD Assay DTD enzyme activity was measured using the substrate 2,6- dichlorophenolindophenol (DCPIP) as previously described (Misra et al., 1998, Chapter 2, Section 2.3.5). 3.3.5 mRNA Profile Assessrnent of the Nucleotide 609 Base Change Total RNA was isolated fiom selected fibroblast strains using the RNeasy Mini Extraction Kit (Qiagen Inc. Santa Clarita, CA). 5 pg of total RNA was reverse transcribed using the Superscript II Preamplification Kit (Gibco) and a DTD gene-specific primer, 5' TCC CAA CTG ACA ACC AGA TC 3' (nt 841-860 of DTD cDNA), according to the protocol of the manufacturer. 10% of the RT reaction was PCR amplified in a mixture (50 PL total volume) containing reaction buffer (50 mM KCl, 10 rnM Tris-HCl (pH 8.3), and 0.001% (wlv) gelatin), 1.25 mM MgCl,, 0.2 mM dNTP mix and 5 U AmpliTaq Polymerase with 100 ng each of DTD gene-specific forward and reverse primer: 5' GCC ATT CTG AAA GGC TGG TT 3' - forward primer (nt 38 1-400 DTD cDNA) 5' CCA TCA CTT GGG CAA GTC CA 3' - reverse primer (nt 82 1-840 DTD cDNA). The 460 bp fiagment was amplified (Figure 3.2) using a 'Touchdown' PCR protocol (Don et al., 1991) consisting of a 3 min denaturation at 94 OC, 2 cycles of 1 min denaturation at 94 OC, 1 min annealing at 60 OC, and a 2 min polymerization at 72 "C, 12 cycles in which the annealing temperature was decreased by 1 OC until a touchdown temperature of 48 OC was reached, an additional 17 cycles with the 48 OC annealing temperature, and a final extension for 10 min at 72 OC. The PCR product was digested with H~nfl,hctionated on o 1.2% agarose gel, and visualized by ethidium bromide (10 pg/mL) staining. The wild-type (609C) DTD allele generates a 377 bp fina fragment, whereas the variant allele (609T) produces two Hinfi fragments of 226 and 15 1 bp. A constitutive 83 bp Hinfl fiagment is produced in addition to the ones derived fiom the wild-type and variant alleles but not observed under these gel electrophoresis conditions and served as a positive cutting control. Total mRNA bases 841- 860 forward primer reverse primer b-381 - 400 bqes 821 840 0 9 - C .r, DTD cDNA 1 (singleatranded) ~infI* Hinf I 460 bp DTD cDNA 226 bp T' 151 bp 83 bp (double-stranded) La77bp- mRNA profile : C 1 C T/T Figure 3.2: RT-PCR-RFLPschernatic. Total cellular mRNA was reverse-transcribed by a DTD gene-specific primer into single-stranded cDNA followed by PCR of bases 38 1-840 of DTD cDNA to yield a 460 bp double-stranded product. Assessrnent of mRNA profile of the DTD nucleotide 609 base change based on the banding pattern of a Hinf I digest is as shown. The polymorphic nucleotide 609 Hinf' 1 site is indicated by an asterisk. A constitutive Hinf 1 site is located downstream of the polymorphic Hinf 1 site and was used as a positive cutting control, yielding an 83 bp hgment. This fragment was not seen with the agarose gel concentration used in these studies, therefore the presence of the 377 bp fragment was used as a positive cutting control. 3.3.6 Western Blot Analysis DTD protein expression was verified in selected ceIl strains. Fibroblast strains were grown to a density of 5 x IO6 cells in 175 cm2 tissue culture flasks, harvested by scraping. pelleted at 250 x g for 5 min, and resuspended in 200 pL lysis buffer (0.1 M Tris-HCI. 1% SDS, 10 rnM EDTA, 20 mM dithiothreitol) containing protease inhibitor cocktail. Cell extracts were placed in a boiling water bath for 2 min then centrifbged at 14000 x g in an Eppendorf (Westbury, NY, USA) 54 1SC centrifuge for 10 minutes at room temperature. The supernatant was recovered and protein concentration determined using the Bradford method (1976). A volume of cell extract containing 20 vg protein was mixed with 6X sample buffer (36% (v/v) glycerol, 5% P-mercaptoethanol, 10.3% (wh) SDS, 350 mM Tris-HC1 (pH 6.8), and 0.0 12% (w/v) bromphenol blue) heated to 100 OC for 2 min and fiactionated on a 14% Tris-Glycine polyacrylamide gel (Novex, San Diego CA) with 4% stacking gel at 2 mA/cm for 3 hours. Protein was electrotransferred to a 0.2 pm nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA) using a Mini Trans- Blot ce11 (Bio-Rad) at 6 mAkm in high ionic strength transfer buffer (25 mM Tris-HC1 (pH 8.0) and 144 mM glycine) for 10-12 hours. Membranes were Ponceau S (Sigma Chemical Co.) stained (Klein et ai 1995) to confirm transfer efficiency and blocked with Tris-buffered saline (TBS; 10 mM Tris-HC1 (pH 7.5) and 150 mM NaCI) containing 5% skim milk powder for 2h at RT. Blots were then incubated ovemight at 4°C with a mixture of mouse anti-human DTD monoclonal antibodies B771 and A180 or with a 1600 dilution of moue anti-tubulin in TBS containing 1% skim milk powder (antibody binding solution). Membranes were washed three times with TBS and incubated with a 1 :4000 dilution of goat anti-mouse horse radish peroxidaseconjugated antibody (Santa Cruz Biotechnology, CA) in antibody binding solution, washed as before, and treated with Chemilurninescence reagent (NEN, Boston, MA). Bands were visualized by exposure to NEN autoradiography film for 5 to 30s and quantified by densitometry within the linear range using a Molecular Dynamics (Sunnyvale, CA, USA) Computing Densitometer with ImageQunt version 3.3 software. 3.4 Resulb 3.4.1 The 609T Nucleotide Substitution Correlates with Reduced DTD Activity in a Panel of 45 Normal Human Fibroblast Cell Strains The C to T base change in nucleotide 609 of DTD cDNA, predicted to confer a proline to serine substitution in amino acid 187 of DTD protein, has been shown to impair catalytic activity and may result in protein instability (Misra et al., 1998, Chapter 2; Traver et al., 1997). The 609T nucleotide substitution produces a novel Hi 1 restriction site (RFLP) in the codon for amino acid 187, which can be used to screen genomic material (Figure 3.1 ) for this variant allele (Eickelmann et al., 1994; Goldberg et al., 1998). RFLP analysis of 45 human skin fibroblast ce11 strains, whose charactenstics are shown in Table 3.1 and Appendix 3.1, resulted in the following allelic distribution of the nucleotide 609 base change: 60% wild-type for both alleles (CIC fibroblasts); 35.5% heterozygous (Cm fibroblasts); and 4.5% homozygous for the variant ailele (T/T fibroblasts). Thirty-one of the strains were assessed independently 2-3 times with complete consistency of results. Number of Strains 45 Percent Male Donors 55 Percent Female Donors 45 Average Donor Age 3 yrs (one day - 16 yrs) Average Strain Passage Number 4 (2-11) Nwnber of Strains C/C at 609 allele 27 (60%) Number of Strains C/T at 609 allele 16 (35.5%) Number of Strains T/Tat 609 allele 2 (4.5%) Table 3.1: Characteristics of skin fibroblast strains used for allelic status and DT- diaphorase activity determination. An analysis of the relationship between allelic status and DTD activity is shown in Figure 3.3. Each assay for enzymatic activity was run using three different concentrations of ce11 lysate and the resulting data was averaged to give a value which is ploned in the Figure. Repeat independent measurements with five of the ceIl strains which were C/C and three which were Cm in genotype gave means with standard enors which were approximately plus or minus twenty percent of the mean (n = 3 - 4). C/C and C/T fibroblasts displayed a wide range of DTD activities but overall the group homozygous for the wild-type (609C) DTD allele had significantly greater enzymatic activities than heterozygous (609C/609T) ce11 strains (Mann-Whitney Rank Order Analysis, a < 0.01 for a two-sided test) with the former showing 2-fold greater mean DTD activities (754 f s.e.m. 78 nrnol/min/mg protein) than the latter (324 f s.e.m. 30 nmol/min/mg protein). In two fibroblasts strains that were homozygous for the variant allele, DTD activity was extremely low to undetectable. These results are consistent with the notion that the 609T substitution impairs the catalytic activity and/or stability of DTD, and that a correlation exists between 609T allelic status and DTD activity in the individual ce11 strains studied. 3.4.2 DTD mRNA Profile of 3437T Fibroblasts The 34371' ce11 strain is unusual in that it is heterozygous for the DTD nucleotide 609 base change but contains undetectable enzymatic activity (Marshall et al., 199 1; Kuehl et al., 1995). Genotypic status and lack of enzyme activity was confirmed in this ce11 strain in the present work (data not shown). These previous studies have show that I cm Genotype Figure 3.3: Relationship between allelic status of nucleotide 609 base change and DTD activity. Each point represents an individuai assessment of 1 polymorphism status (allelic status of nucleotide 609 base change) and DTD activity for a single ce11 strain in a group of 45 normal human skin fibroblast strains. There was some overlap of data points. Mean + s.e.m. DTD activity of the group homozygous for the wild-type allele and the heterozygote group are indicated. total DTD mRNA levels in 3437T cells are similar to other cell strains displaying detectable enzymatic activity. Because cells that are homozygous for the variant allele display very low to undetectable DTD activity, one possible explanation for the low DTD activity in 3437T cells is that they express the variant 609T allele exclusively. To examine this question, the relative levels of expression of the two alleles in 3437T cells were compared to seven selected fibroblast strains fiom Table 3.1 that are homozygous for the wild-type allele (3 strains). heterozygous (2 strains) or homozygous (2 strains) for the variant allele (Table 3 -2). Total RNA isolated fiom each ce11 strain was PCR-arnplified (Figure 3.2). As shown in Figure 3.4, the DTD mRNA expression profile in the group of fibroblasts fkom Table 3.2 was in agreement with the genomic status of the nucleotide 609 base change (with heterozygotes expressing DTD mRNA fiom both alleles). However, 3437T ceils displayed an RFLP pattern consistent with fibroblast strains that are homozygous for the variant (609T) allele, indicating that DTD mRNA expressed in 3437T cells arises exclusively fiorn the variant (609T) allele. Because expression of the 609T variant allele has been associated with reduced DTD protein levels (perhaps due to a decrease in protein stability), we also examined DTD protein levels in 3437T cells relative to this sarne panel of seven fibroblast ce11 strains from Table 3.2. As show in Figure 3.5, DTD protein levels in 3437T cells were reduced relative to homozygous wild-type and heterozygous ce11 strains to levels that were similar to that in the two fibroblast strains homozygous for the variant allele. This is consistent with the RT-PCR analysis pointing to exclusive expression of the variant 609T Cefl Strain Allelic Status DTD Activity (nmol/min/mg protein) 1151 1 C/C 621 * 361 (n=3) 9249 C/C 781 * 188 (n=3) Table 3.2: Allelic status and DTD activities of 7 skin fibroblast strains taken fiom panel and ce11 strain 3437T used for DTD mRNA profile studies (see Figure 3.4) and Western blot analysis (see Figure 3.5). Values represent mean DTD activities with number of independent assessments indicated. S.E.Ms are included for those values which have been assessed more than twice. Figure 3.4: DTD allelic mRN.4 expression profiles. Seven skin fibroblast strains fiom Table 3.1 and 3437T cells were exarnined for wild-type and variant DTD mRNA expression as descnbed in 'Materials and Methods.' The 377 bp band corresponds to wild type allele expression while the 226 and 151 bp bands correspond to variant allele expression. Lanes 1-3: fibroblasts homozygous for the wild-type DTD allele; Lanes 4 and 6: heteroygous fibroblasts; Lanes 7 and 8: fibroblasts homozygous for the variant allele; Lane 5: 3437T cells. L: 123 bp DNA size marker ladder. Figure 3.5: Western Blot analysis of DTD protein levels in normal human skin fibroblast strains and 3437T cells. Immunoblot analysis was performed on total protein extracts using a mixture of DTD monoclonal antibodies B771 and A180 as described in 'Materials and Methods.' Blots were also probed for P-tubulin expression as a protein loading control. Lanes 1-3: normal human fibroblasts homozygous for the wild-type (609C)DTD ailele; Lanes 4 and 6: heterozygous fibroblasts; Lanes 7 and 8: fibroblasts homozygous for the variant allele (609T); Lane 5: 3437T cells. Lanes 5, 7, and 8 were intentionally loaded with 2-fold the protein as the other lanes. Also show are DTD : B- tubulin expression for these strains as determined by densitometry. Values represent mean * s.d. of five densitometric readings for each lane standardized to 3437T cells. ailele in 3437T cells. The enzymatic activity in dl three of these ce11 strains was less than 20 nmoVmin/mg protein. 3.5 Discussion The probability that cells will fail to activate bioreductive prodrugs or inactivate xenobiotics that are substrates for DTD due to expression of the 609T variant isoform has been estimated by measuring the prevalence of the 609T allele in various populations. The fiequency of the base change varies between ethnic groups; with 2.5 - 25% of individuais homozygous for the variant 609T allele (Kelsey et al., 1997, see Table 5.1). The variant allele has also been examined in different tumor ce11 types, with reduced frequencies compared to disease free reference groups observed in lung (Chen er al., 1999), and increased fiequencies in certain infant leukemias (Wiemels et al., 1999). However, studies showing no significant differences in variant allele fiequencies in tumors have also been reported (Traver et al., 1997). A benzene exposed population stratified into hematological malignancy and case control groups displayed a higher occurrence of the variant allele in the former group (Rothman el al., 1996), consistent with the possibility that expression of the variant allele can impair the defense mechanism of the ce11 against carcinogenesis mediated by certain xenobiotics (Vasiliou et al., 1995). Estimating the magnitude of the response of individuais or ce11 populations to xenobiotics or prodrugs that are DTD substrates on the basis of the allelic status of the nucleotide 609 base change will depend on the extent to which allelic status reflects enzyme activity. The variant 609T allele has been implicated as a marker for some diseases associated with insufficient detoxification of xenobiotics by DTD, and occurs at a higher fkequency in some tumors. However, the absence of significant levels (Marshall et al.. 1991; Kuehl et al., 1995) of DTD activity in at least one human ce11 swn heterozygous for the variant allele, 3437T, brought into question the relationsh@ between genotype and phenotype. In this study we have re-examined the prevalence of the DTD nucleotide 609T substitution in the normal population using 45 normal fibroblast strains denved fiom skin biopsies (Table 3.1), and have used this information to investigate the correlation between genotypic status and DTD activity. The variant 609T allele occurred with a fiequency of 0.21 in our panel of 45 normal human skin fibroblast ce11 strains, with a distribution that is in accordance with Hardy-Weinberg equilibrium and is consistent with previous reports (Kelsey et al., 1997, Rosvold et al.! 1995, and Schulz et al., 1997). However, the aforementioned variation in frequency based on ethnicity may be relevant as a risk factor for disease associated with environmental and dietary exposures as hypothesized for other drug metaboliring enzymes that display genetic polymorphisms (Nebert er al., 1999). The ethnicity of al1 the donors in the present skin fibroblast panel was not revealed. Limited information indicated they represented a mixture of ethnic backgrounds (see Appendix 3.1 ). Fibroblasts homozygous for the wild-type (609C) allele displayed on average 2- fold greater DTD activities as compared to heterozygous fibroblast strains in this panel. This observation is consistent with a gene dosage effect arising fkom expression of the single wild-type 609C allele in heterozygous strains. The two homozy gous 609T fibroblast strains contained very low to undetectable enzyme activities in agreement with dl reports of other groups for the effect of this genotype. However, considerable overlap in enzyme activity was observed between the individuai homozygous wild-type 609C and heterozygous fibroblast ce11 strains. A likely explanation for the broad range of enyme activities within these two groups is a variation in the levels of DTD mRNA expressed in each ce11 strain. Traver et al (1 992) observed a strong correlation between DTD mRNA levels and enzyme activity in a colon cancer ce11 panel using a semi-quantitative RT-PCR technique. Siegel et al. (1999) also showed that DTD protein levels in human saliva were directly correlated mith the DTD nucleotide 609 genotype with C/C individuais showing twice the protein level as C/T individuals and Tm individuals having undetectable levels. A similar overlap between C/C individuals and C/T individuals was seen in this work. The 3437T ce11 strain is unique among ce11 strains in the present work and in previous studies in that it is genetically heterozygous but displays very low to undetectable DTD activity. Western blot analysis indicates that these cells express DTD at lower levels than cells homozygous or heterozygous for the wild-type allele, perhaps due to instability of the variant isoform. The level of expression of DTD protein is similar in 3437T cells and the two homozygous mutant allele ce11 strains (Figure 3.5, lanes 5,7, and 8). The levels of protein in homozygous wild-type ailele ceIl strains are greater than those in the heterozygotes (ANOVA, p < 0.01). RT-PCR analysis confirmed that the variant allele is expressed exclusively in 3437T cells. Therefore, although 3437T cells represent an extreme case, they illustrate a potential problem in the use of allelic status of the nucleotide 609 base change to predict DTD activity in heterozygotes. The overlap in enzymatic activities observed between homo ygous wild-type 609C and heterozygous fibroblasts also illustrates the dificulty in distinguishing C/Cfiom C/T cells on the basis of enzyme activity, although such cet1 strains are clearly distinguishable fiom T/T ce11 strains. The absence of measurable DTD mRNA expression fiom the C allele in 3437T cells may be due to methylation leading to matemal or paternal gene silencing (Rarin, 1998) or to defects in the upstream promoter region affecting basal transcription. Similarly, the variation in DTD activities observed within groups of C/C and C/T celis may reflect individual differences in transcriptional responses given the complexity of pathways involving the xenobiotic response element and the antioxidant response element (Jaiswal, 1991; Favreau and Pickett, 1991) and their corresponding cis- and trans- acting factors. In conclusion, prediction of cellular responses to prodrugs and xenobiotics that depend on DTD activity cm be based to a first approximation on genotype status, which can distinguish between significant levels of enzyme activity in homozygous wild-type and heterozygous cells and very low to undetectable levels of enzyme activity in homozygous variant cells. Although heterozygotes have on average half the activity of wild-type cells, the ranges are overlapping. The existence of a heterozygous ce11 strain with no detectable DTD activity, due to allele specific expression, was a unique occurrence in the present work and is unlikely to play a large role in predictions based on geno typing. 3.6 References Begleiter A, Leith MK, Curphey TJ and Doherty GP. Induction of DT-diaphorase in cancer chemoprevention and chemotherapy. Oncol Res 9: 37 1-382, 1997. Bradford MM. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the p~cipleof protein-dye binding. Anal Biochem 72: 248- 254, 1976. Chen H, Lum A, Seifiied A, Wilken LR and Le Marchand L. Association of the NAD(P)H:quinone oxidoreductase 609C to T polymorphism with a decreased lung cancer risk. Cancer Res 59: 3045-3048, 1999. Doherty GP, Leith MK, Wang X, Curphey TJ and Begleiter A. Induction of DT- diaphorase by 1,S-dithiole-3-thiones in human turnour and normal cells and effect on anti-tumour activity of bioreductive agents. Br J Cancer 77: 124 1 - 1252, 1998. Don RH, Cox PT, Wainwright BJ, Baker K and Mattick JS. 'Touchdown' PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res 19: 4008, 1991. Eickelmann P, Schulz WA, Rohde D, Schrnitz-Drager B and Sies H. Loss of heterozygosity at the NAD(P)H:quinone oxidoreductase locus associated with increased resistance against mitomycin C in a human bladder carcinoma ce11 line. Biol Chem Hoppe Seyler 375: 439-445, 1994. Emster L. DT-diaphorase. Merhods E-ymol 10: 309-3 17, 1967. Emster L and Navazio F. DT-diaphorase: A historical review. Chernica Scripta 27A: I- 13, 1987. Favreau LV and Pickett CB. Transcriptional regulation of the rat NAD(P)H: quinone oxidoreductase gene. Identification of regdatory elements controlling basal level expression and inducible expression by planar aromatic compounds and phenolic antioxidants. J Bi01 Chem 266: 4556-456 1, 1991. Friedlos F, Court S, Ford M, Demy WA and Springer C. Gene-directed enzyme prodnig therapy: quantitative bystander cytotoxicity and DNA darnage induced by CB1954 in cells expressing bacterial nitroreductase. Gene mer 5: 105-1 12, 1998. Gaedigk A, Tyndale RF, Juïima-Romet M, Sellers EM, Grant DM and Leeder JS NAD(P)H:quinone oxidoreductase: polymorphisms and allele fiequencies in Caucasian, Chinese and Canadian Native Indian and Inuit populations. Pharmacogenetics 8: 305- 313. 1998. Goldberg ZL, Cummings BJ, Chapman WB, Klamut HJ and Rauth AM. Role of a DT- diaphorase mutation in the response of anal canal carcinoma to radiation, 5-fluorounicil and mitomycin C. Int J Radiation Oncology 42: 33 1-334, 1998. Jaiswal AK. Human NAD(P)H:quinone oxidoreductase (NQO1) gene structure and induction by dioxin. Biochemistry 30: 10647-10653, 1991. Jaiswal AK. Human NAD(P)H:quinone oxidoreductase2. Gene structure, activity, and tissue-speci fic expression. J Biol Chem 269: 14502-1 4508, 1994. Kelsey KT, Ross D, Traver RD, Christiani DC, Zuo ZF, Spitz MR, Wang M, Xu X, Lee BK, Schwartz BS and Wiencke K. Ethnic variation in the prevalence of a common NAD(P)H quinone oxidoreductase polymorphism and its implications for anti-cancer chernotherapy. Br J Cancer 76: 852-854, 1997. Klein D, Kem Rh4 and Sokol RZ. A method for quantification and correction of proteins after transfer to immobilization membranes. Biochem Mol Biol Int 36: 59-66, 1995. Knox RJ, Friedlos F, Biggs PJ, Flitter WD, Gaskell M, Goddard P, Davies L and Jarman M. Identification, synthesis and properties of 5-(aziridin- 1-yl)-2-nitro-4 nitrosobenzamide, a novel DNA crosslinking agent derived fiom CB 1954. Biochem Pharmacol 46: 797-803, 1993. Kuehi BL, Paterson JW, Peacock JW, Paterson MC and Rauth AM. Presence of a heterozygous substitution and its relationship to DT-diaphorase activity. Br J Cancer 72: 555-561, 1995. Lind C' Hochstein P and Emster L. DT-diaphorase as a quinone reductase: a cellular control device against semiquinone and superoxide radical formation. Arch Biochem Biophys 216: 178-185, 1982. Marin A, Lopez de Cerain A, Hamilton E, Lewis AD, Martinez-Penuela IM,Idoate MA and Bello J. DT-diaphorase and cytochrome BS reductase in human lung and breast turnours. Br J Cancer 76: 923-929, 1997. Marshall RS, Paterson MC and Rauth AM. DT-diaphorase activity and mitomycin C sensitivity in non-transfonned ce11 strains derived fkom members of a cancer-prone farnily. Carcinogenesis 12: 1 175-1 180, 1991. Misra V, Klamut HJ and Rauth AM. Transfection of COS-1 cells with DT- diaphorase cDNA: role of a base change at position 609. Br J Cancer 77: 1236-1 240, 1998. Nebert DW, Ingelman-Sundberg M and Daly M. Genetic epidemiology of environmental toxicity and cancer susceptibility: human allelic pol ymorphisms in drug metabolizing enzyme genes, their functional importance, and nomenclature issues. Dmg Metab Rev 2: 467487,1999. Prochaska HJ and Taialay P. Regdatory mechanisms of monofunctional and bibctional anticarcinogenic enzyme inducers in murine liver. Cancer Res 48: 4776-4782, 1988. Radjendirme V, Joseph P, Lee YH, Kimura S. Klein-Szanto AJ, Gonzalez FJ and Jaiswal AK. Disruption of the DT Diaphorase (NQO1) gene in mice leads to increased menadione toxicity. J Biol Chem 273: 7382-7389, 1998. Razin A. CpG methylation, chromatin structure and gene silencing-a three-way connection. EMBO J 17: 49054908,1998. Rosvold EA, McGlynn KA, Lustbader ED and Buetow KH. Identification of an NAD(P)H:quinone oxidoreductase polymorphism and its association with lung cancer and smoking. Pharmacogenetics 5: 199-206, 1995. Rothman N, Traver RD, Smith MT, Hayes RB, Li G-L, Campleman S, Dosemeci M, Zhang L, Linet M, Wacholder S, Yin S-N and Ross D. Lack of NAD(P)H:quinone oxidoreductase activity (NQO1) is associated with increased risk of benzene hematotoxicity. Proc Amer Assoc Cancer Res 37: 258, 1996. Sartorelli AC, Hodnick WF, Belcourt MF, Tomasz M, Haf33y B, Fischer JJ and Rockwell S. Mitomycin C: a prototype bioreductive agent. Oncol Res 6: 50 1-508, 1994. Schlager JJ and Powis G. Cytosolic NAD(P)H:(quinone-acceptor)oxidoreductase in human normal and turnor tissue: effects of cigarette smoking and alcohol. Int J Cancer 45: 403-409, 1990. Schulz WA, Krummeck A, Rosinger 1, Eickelmann P, Neuhaus C, Ebert T, Schmitz- Drager BJ and Sies H. Increased fiequency of a null-allele for NAD(P)H: quinone oxidoreductase in patients with urological mdignancies. Pharmacogenetics 7: 235-239, 1997. Searle PF, Weedon SJ, McNeish IA, Gilligan MG, Ford MJ, Friedlos F, Springer CJ, Young LS and Ken DJ. Sensitisation of human ovarian cancer celts to killing by the prodrug CB 1954 foIlowing retrovirai or adenoviral transfer of the E. coli nitroreductase gene. Adv Exp MedBiol451:107-113, 1998. Siegel D, Franklin WA and Ross D. Immunohistochemical detection of NAD(P)H: quinone oxidoreductase in human lung and human tumors. Clinicd Cancer Res 4: 2065- 2070, 1998. Siegel D, McGuinness SM, Winski SL and Ross D. Genotype-phenotype relûtionships in studies of a pol y morphism in N AD(P)H :quinone oxidoreductase 1. Pharmacogenefics 9: 113-121, 1999. Sreerama L, Hedge MW and Sladek NE. Identification of a class 3 aidehyde dehydrogenase in hurnan saliva and increased levels of this enzyme, glutathione S- transferases, and DT-diaphorase in the saliva of subjects who continually ingest large quantities of coffee or broccoli. Clin Cancer Res 1: 1 153-1 163, 1995. Talalay P, Fahey JW, Holtzclaw WD, Prestera T and Zhang Y. Chernoprotection against cancer by phase 2 enzyme induction. Toxicol Leu 82-83: 173- 1 79, 1995. Traver RD, Horikoshi T, Danenberg KD, Stadlbauer TH, Danenberg PV, Ross D and Gibson NW. NAD(P)H:quinone oxidoreductase gene expression in hwnan colon carcinoma cells: characterization of a mutation which modulates DT-diaphorase activity and mitomycin sensitivity. Cancer Res 52: 797-802, 1992. Traver RD, Siegel D, Beall HD, Phillips RM, Gibson NW, Franklin WA and Ross D. Characterization of a polymorphism in NAD(P)H:quinone oxidoreductase (DT- diaphorase). Br J Cancer 75: 69-75, 1997. Vasiliou V, Shertzer HG, Liu RM, Sainsbury M and Nebert DW. Response of [Ah] battery genes to compounds that protect against menadione toxicity. Biochem Pharrn 50: 1885-1891: 1995. Warrington KH Jr. Teschendorf C, Cao L, Muzyczka N and Siemann DW. Developing VDEPT for DT-diaphorase (NQO1) using an AAV vector plasmid. Int J Radiaf Oncol Biol Phys 42: 909-912, 1998. Wiemels JL, Pagnamenta A, Taylor GM, Eden OB, Alexander FE and Greaves MF. A lack of a fûnctionai NAD(P)H:quinone oxidoreductase allele is selectively associated with pediatrïc leukemias that have MLL fusions. United Kingdom Childhood Cancer Study Investigators. Cancer Res 59: 4095-4059, 1999. Winski SL, Hargreaves RH and Ross D. A new screening system for NAD(P)H:quinone oxidoreductase (NQO1)-directed antitumour quinones: identification of a new aziridiny 1benzoquinone, RH 1, as a NQO1 -directed antitumor agent. Clinicul Cancer Res 4: 3083-3088, 1998. Workman P and Stratford IJ. The experimental development of bioreductive dmgs and their role in cancer therapy. Cancer Metastasis Rev 12: 73-82, 1993. Yao KS, Hageboutros A, Ford P and O'Dwyer PI. lnvolvement of activator protein-l and nuclear factor-kappaB transcription factors in the control of the DT-diaphorase expression induced by mitomycin C treatment. Mol Pharmacol 51: 422-430, 1997. Characteristics of the Human Skin Fibroblast Panel Used For DTD Nucleotide 609 Base Change Frequency Analysis Table A3.1.1: Characteristics of the human skin fibroblast strallis used for DTD nul1 allele fiequency and expression studies discussed in Chapter 3. Cell Shah Allelic Status of Mean s.e.m. Gender Age of Ethnie & Passage Nucleotide 609 DTD Activity of Donor Background ~urnber' Base Change- (nmoUmin/mg Donor (years) protein) : 11511,4 CIC 621 * 180 F 6.41 9249,5 CIC 781 * 109 F 2.78 11623,4 CA' 275 * 110 F 0.74 9152,s C/C 412 *45 M 0.52 1 1686,2 C/T 247 48 F 1 0.98 11754,3 CIC 1092 M 1 NIA' 1 11816,S C/T 563 (774,353) M NIA 3692.7 CIC 974 M 2.04 7859,4 CIC 528 (155,902) F 8.00 East Indian 6825,4 CIC 1916 (1593,2240) M 6.55 5875,4 CIC 757 F 0.75 I 1 1 1 I 5436,3 1 C/T 1 50 1 1 M 1 22.70 1 Caucasian 1 1 1 1840,s CIC 428 F 0.05 11496,3 CIC 237 (404.71) F 0.55 Mexican I 1 1 1 . - 11525, 10 ] C/T 1 272 1 M 1 NIA 1 I 1 1 1548,9 C/C 1559 (1350, 1768) M NIA 11520, 11 C/T 265 F NIA Finish 11528, 10 CIC 501 M 10.60 1 11512, 11 C/T 70 (32, 109) F 1.26 11829,4 T/T 19 (17,21) M 3 .O6 11871,2 C/T 400 (343,458) F 0.92 8813,2 CIC 364 NIA NIA 11723,7 CIC 1117 (512, 1722) NIA 0.58 11804,2 C/C 349 M NIA 11446,5 C/C 785 M 0.772 11456, 11 C/T 347 (141,554) NIA NIA 11720,8 C/C 987 (391,1582) F 7.96 11760.7 C/C 488 F 0.709 Table A3.1.1 continued Cell Strain Aiielic Status of Mean r s.e.m. Etbnic & Passage Nucleotide 609 DTD Activity Background Number' Base Change (nmoumidmg protein) : 11845,3 Cm 333 8573,2 CIC 43 1 Fr. Canadian 1 11864,S CIT 271 (527, 15) 11571,3 C/T 306 (376,237) East Indian 1 11849,2 C/T 424 (776,73) 11911,3 Cm 495 NIA 8635,2 C/C 640 (167, 1120) N/A N/A Fr. Canadian 11798.4 CIC 873 (1 14, 1632) 7 ' Passage number includes two passages afler obtaining fiom ce11 repository. : DTD activity values with standard errors represent three or four independent deteiminations. Those followed by values in parentheses represent two determinations while single values represent one determination. " NIA and blanks indicates characteristics that were not available due to donor confidentiality by Hospital for Sick Children, Toronto. b Ce11 strain added after initial genotype and DTD activity detemination to panel of 45 strains for RT-PCR-RFLPanalysis. Exogenous Expression of the Prodrug-Activating Enzyme DT- Diaphorase Via AD5 Delivery 4.1 Abstract Mitomycin C (MMC), a prototypical bioreductive prodnig, can be activated by DT-diaphorase (DTD) hto a DNA alkylating agent. Intra-tumoral injection of recombinant adenoviral type 5 vectoa (Ad5) that carry prodmg-activating enzymes like DTD could be used to selectively target tumor cells for chemotherapy. To demonstrate the feasibility of this approach, Ad5 vectors were constmcted that express D?D minigenes for both wild-type and mutant (C to T change in nucleotide 609 in DTD cDNA) DTD under the control of the cytomegalovirus (CMV) promoter. HT29 human colon carcinoma cells express wild-type DTD, whereas BE human colon carcinoma cells have low to undetectable DTD activity due to the nucleotide 609C to T mutation and are 6-fold more resistant to MMC than HT29 cells. A test of the ability of Ad5 to infect these cells using a B-gaiactosidase CMY driven minigene indicated that BE cells were 90- 100% infected at a multiplicity of infection (MOI) of 100 while HT29 cells are only 15- 40% infected at this MOI. Infection of BE cells in vitro with recombinant Ad5 carrying a minigene for wild-type DTD resulted in high levels of expression of exogenous DTD and increased MMC sensitivity. Infection of HT29 cells with recombinant Ad5 expressing a non-functional DTD mutant produced little change in endogenous DTD activity, and tfiere was no evidence of a dominant negative effect. Infection of BE cells at MOIS of 3 to 100 resulted in a progressive increase in DTD activity and enhanced sensitization to MMC. BE cells infected at an MOI of 100 and expressing wild-type DTD were 8-fold more sensitive to MMC as rneasured by a colony forming assay. HTî9 cells were sensitized 2-3 fo1d following treatment with AdS.DTD at a MOI of 100. These results indicate that adenovinis-mediated gene transfer and expression of wild-type DTD can sensitize tumors cells to MMC and that in vivo testing of this approach is warranted. 3.2 Introduction DT-diaphorase [NAD(P)H (reduced nicotinarnide adenine dinucleotide, with or without phosphate):quinone oxidoreductase (NQO,), Enzyme Commission No. 1 -6-99.21 is a homodimenc, cytosolic flavoenzyme that acts on its substrates by a concened two electron reduction process (Emster, 1967). This enzyme acts on a wide variety of substrates (Talalay et al., 1999, but most attention has been given to naturally occurring (Keyes et al., 1984) and synthetic quinone-based compounds (Phillips et al., 1999). DTD acts as a quinone reductase capable of detoxifying a variety of xenobiotics including quinones, azo-dyes, and nitro-compounds (Emster. 1987; Talalay, 1989). MMC (Figure I -3) is a naturally occurring quinone that is considered to be a prototypical bioreductive agent (Sartorelli et al., 1994). Metabolic reduction of MMC by DTD converts this prodmg into a cytotoxic DNA alkylating agent (Figure 1.3). Reaction with nucleophilic DNA bases, preferentially by guanine and cytosine (Iyer and Szybalski, 1964) towards two electrophilic centers (carbons 1 and 10) results in intra- or interstrand crosslinking, the latter of which is believed to result in cytotoxicity (Workman and Stratford, 1993). A definitive physiological role for DTD has not yet been established. The liver contains the most abundant levels of DTD (Jaiswal, 1994) but other tissues, including heart, lung, and kidney also exhibit varying DTD activities (Maritus, 1963). Elevated DTD activities have been reported in tumor relative to normal tissue (Marin et al., 1997) and in tumor ce11 lines (Kepa and Ross, 1999). It has been suggested that this difference be exploited towards enzyme-prodrug therapy using compounds that are DTD substrates (Workman, 1994). However, a high degree of heterogeneity in DTD enzymatic activities (Traver et al., 1992) and protein levels (Belcourt et al., 1998) have been observed both within and across normal and tunor tissue types. This is due to a nurnber of factors including the inducibility of the enzyme (Begleiter et al., 1997) and to a high fiequency C to T base change at nucleotide 609 of the DTD mRNA resulting in heterozygotes with approximately hdf the DTD activity as compared to wild-type cells (Chapter 3). This base transition results in a proline to serine change in arnino acid 187, impaired DTD catalytic activity, and protein instability (Traver et al., 1992; Siegel et al., 2000). To circumvent the problem of heterogeneity in DTD activity in tumor cells, a first generation Ad5 vector was created that expresses wild-type human DTD under the control of the CMV promoter. Ad5 vectors have been previously used with good success to deliver minigenes encoding prodrug activating enzymes to tumor cells (Bndgewater et al.?1995). The results demonstrate that adenovims-mediated gene transfer and expression of DTD can sensitize human colon carcinoma cells, homozygous for either wild-type or variant alleles (Traver et al., 1992) of the DTD gene, to MMC. 4.3 Materials and Methods 4.3.1 Chernicals and Reagents Al1 reagents used for the DTD enzymatic assay were purchased fiom Sigma Chernical Co. (St. Louis, MO, USA). Materials used for PCR, including AmpliTaq polymerase were obtained fiom Roche Molecular Systems Inc. (Branchburg, NJ, USA). DTD monocIonal antibodies B771 and A1 80 were supplied by Dr. David Ross (University of Colorado Healîh Sciences Center, Denver CO, USA). Ant i-human integrin a, (MAB 1960) was purchased from Chernicon (Temecula, CA, USA). MMC (catalog no. M4287) was obtained fiom Sigma. X-gal, potassium femcyanide, and potassium ferrocyanide, used for histochemical detection of D-galactosidase, were also purchased from Sigma. 4.3.2 Ceil Culture 293T human embryonic kidney cells were obtained fiom Microbix Biosystems (Toronto, ON, Canada) for the production of El-defective recombinant Ad5 (Chapter 1.4.4). Hela hurnan cervical carcinoma cells, used to test for wild-type recombinant virus, were purchased from American Type Culture Collection, Manassas, VA, USA. BE and HT29 hwnan colon carcinoma cells were chosen for this work because of previous work that showed that their 6-fold difference in MMC sensitivity was due to low or absent DTD activation in BE cells (Traver et al., 1992) and that both cells can be infected by Ad5 (Dr. Henry Klamut, University of Toronto, personal communication). BE and HT29 hurnan cells were obtained fiom Dr. T. Mulcahy, Wisconsin Clinical Cancer Centre. Madison, WI, USA. Al1 ce11 lines were maintained as monolayer cultures in growth medium consisting of a-minimal essential medium supplemented with 10Y0 fetal bovine serum (Gansera, Rexdale ON, Canada) and maintained in a hurnidified atmosphere containing 5% CO, at 37OC in 175 cm2 polystyrene tissue culture flasks (Nunc, Life Technologies, Burlington, ON, Canada). 4.3.3 Preparation of the Ad5 CM-Bascd Shuttk Vector An Ad5 shuttle vector containing the CMV promoter and a bovine growth hormone (BGH) poly A signal was constructed by Dr. HJ Klamut, Ontario Cancer Institute. as shown in Figure 4.1. A 953 bp Sal 1 -%a 1 fragment (fragment 1) containing the CMV promoter and the entire polylinker except for the Apa I site of the pRc/CMV (Invitrogen Co., San Diego, CA, USA) mammaiian expression vector and a 299 bp Xba I - Barn HI fragment (fragment 2) containing the BGH poly A region of pRc/CMV, were inserted into the corresponding restriction sites in the polylinker of pAE 1sp 1B (Microbix Biosystems) to generate pAE 1sp 1BCMV. pAEl sp1 BCMV is a modified Ad5 shuttie vector that can also serve as a eukaryotic expression vector. 4.3.4 Preparation of Ad5 CMV-Based Shuttle Vectors Containing DTD cDNA Kind II1 - Xba 1 fragments containing DTD cDNA with a C or T nucleotide at position 609 were isolated fiom ~Rc/CMV.DTD- and ~RC/CMV.DTD~~~,respective1 y (see section 2.3.2 for a detailed description of construction of these vectors) and inserted into the corresponding restriction sites of pAE 1sp 1 BCMV to generate pAE 1 sp 1 B.DTDm and pAE 1 spl B.DTD- (Figure 4.2). Vector sequences were verified by automated DNA sequencing using florescent-labelled dideoxy nucleotides (Wilson et al., 1990) by the Core Molecular Biology Facility (York University, Toronto, Canada). Fragment 1 (953 bp) , Fragment 2 (299 bp) Cut with Sa1 I & Xba I. ,.' ' Cut with Xba I & BamH I b' I '4 -9 9 --s-s-- '-9- œ zEug gg~5;5ggng*za0 rru,mu,xm~erumzx~)~xXbaI BamH I ------Cut with Sol I and Xba I * and Ligated with Fragment 1 Shuttle Vector ------* Cut with Xba I and BrmH I and Ligated with Fragment 2 ------,continued on next page Figure 4.1: Scheme for construction of pAEl spl BCMV, a CMV promoter-based Ad5 shuttle vector. Summary is on next page. ' \ \ ' \ ' \ '4 continued from pmvious page Fragment 1 Fragment 2 CMV - Based Shuttle Vecbr pA~i~p~~~~~ 7625 bp Figure 4.1 continued: Two pRc/CMV fragments, a Sa1 1 -%a 1 fiagment (fiagment 1) containing the CMV promoter and some of the polylinker region and a Xba 1 - Barn H 1 fiagment (fragment 2) containing the BGH polyA sequence, was inserted into the Ad5 pAElsplB shuttIe vector. The modified Ad5 shuttle vector, pAElsplBCMV, contains a polylinker with some sites fiom pAElsplB and some sites fiom pRc/CMV, with the latter sites shown in italics. niis ailows rninigene expression under the control of the CMV promoter. Hind III Xbal / 0 ' Cut with 5' ,fO Hind III & Xba I b' Hind III Xbal DTD \ DTD609C or T 609C or T 5' 3' 'N, Ligate into Hind III & Xba I ".\ of PAN-1 BCMV * 3' \ ' xba 1 1-1 1-1 BGHpolyA BamH I Bgl il AEI(1.0 - 9J rn.u.) Shuttie Vector pA~i~piB.DTD~O~C Figure 4.2: Cloning scheme for p~~lsp 1B.DTD- and ~AE~S~~B-DTD~~. 43.5 Large Scak Production and Purification of AdS.DTD Vectors AdS.DTD vectors were produced according to a modification of a method developed by McGrory et al. (1988, Figure 1.6). 2 x 10' 293T cells were seeded in each well of a 6 well tissue culture dish (Nunc) in 5 mL of growth medium and incubated ovemight in a hurnidified atmosphere containing 5% CO2at 37 OC to reach a confluency of approximately 70%. Cells were cotransfected using a CaPO, CO-precipitationmethod (Graham and Van der Eb, 1973) using 5 pg of AdS.DTD shuttle vector (pAE 1spl B-DTD- or pAE 1sp 1 B.DTDmT) and 15 pg of pJM 1 7 (Microbix Biosystems, Figure 1.6). Transfection medium was removed by aspiration after 24 hours followed by a gentle wash with growth medium. Agarose overlay solution was prepared by melting a 1.2 % stock solution of SeaPlaqueTMagarose (Mandel Scientific Co., Guelph, ON, Canada) in water using a microwave to mise the solution to its boiling point, cooling to 50 OC in an isothermic water bath, and mixing with an equal volume of room temperature 2X a-minimal essential medium without phenol red and supplemented with 20% fetal bovine sem. 3 mL of agarose overlay solution was added dropwise to each well and placed undisturbed at room temperature for 20 minutes foliowed by incubation in a humidified atmosphere containing 5% CO2at 37 OC. After 5 and 10 days, a Mer2 mL of agarose overlay solution was added to each well as before to refeed the cells. 3-4 days later, pIaque formation was observed. Plaques were isolated by aspiration using a Gilson 1000 pL Pipetman (Mandel Scientific Co.) with a 1 mL tip, transferred to 2 mL microtubes (DiaMed Lab Supplies Inc., Mississauga, ON, Canada) and diluted with 1 mL of PBS. The agarose was mixed into solution by gentle resuspension with a 1000 PL Pipemian and the mixture was centnfuged at 14000 x g in an Eppendorf (Westbury, NY, USA) 5415C cenhfuge for 5 minutes at 4 OC . The pellet was diswded and the supematant was subjected to 1-2 rounds of fieeze thaw (1 minute in a dry ice / ethanol bath followed by 1 minute in a 37 OC water bath to disrupt host cells). 0.2 mi, of the supernatant was added to 24 well dishes (Nunc) that were seeded with 2 x IO4 293T cells the night before. Afier 30 minutes to ailow viral absorption to monolayer, each well was topped off with 1.8 rnL of growth medium. Afier 2-3 days, cytopathic effect (CPE) mediated by virus was observed. At this point, dishes were placed at room temperature for 10-1 5 minutes to promote host ce11 detachment. Cells were harvested by gentle scraping and resuspension with a 5 mL plastic pipette (Costar Corp., Cambridge, MA, USA), transferred to 2 mL polypropylene cryogenic vials (Costar), and centrifuged at 14000 x g for 5 minutes at 4 OC. 1 mL of supematant was used to infect a 25 cm2 tissue culture flask (Nunc) that was seeded with 4 x IO' 293T cells and containing 5 mL of growth medium the night before. After 2-3 days, viral CPE was observed and cells were harvested as before, transferred to 15 mL polypropylene tubes (Becton Dickinson Labware, Franklin Lanes, NJ, USA). and centrifuged at 250 x g for 5 minutes at 4 OC. 2 mL of supernatant was used to infect an 80 cm' tissue culture flask (Nunc) that was seeded the night before with 10' 293T cells and containing 20 mL of growth medium. After 2 days, viral CPE was observed and cells were harvested as before, transferred to 50 mL polypropylene tubes (Becton Dickinson), and centrifuged at 250 x g for 5 minutes at 4 OC. 10 mL of supernatant was used to infect a 175 cm2 tissue culture flask that was seeded the night before with 4 x 10' 2931 cells and containing 30 mL of growth medium. After 2-3 days, upon viral mediated CPE, this procedure was repeated to infect 4 x 175 cm' tissue culture flasks and scaled up once more to infect 12-16 flasks. Mer 2-3, days, when complete vual CPE was observed, flasks were placed at room temperature for 20 minutes to promote host ce11 detatchment. Cells were gently scraped and pooled into 500 mL centrifuge bottles (Nunc) and spun for 10 minutes at 4 OC in a Sorvall (Dupont, Wilmington, DE, USA) GS3 rotor at 6000 x g. Viral supematants were stored at 4 OC for up to one month for use in starter cultures. Ce11 pellets were pooled and resuspended in 5 rnL of 10 rnM Tris-HCI (pH 8), subjected to 3 rounds of fieeze-thaw (2 minutes in a dry ice 1 ethanol bath followed by 2 minutes in a 37 OC water bath), and 1.2 rnL of Tris buffer was added to top off volume to 6.2 mL. 0.62 mL of 5% sodium deoxycholate (Sigma) was added and the mixture was incubated for 30 minutes on ice at which point 70 pL of 1M MgSO, (Sigma) was added (for a final concentration of 10 mM) as well as 20 PL (200 U) of DNase 1 (Life Technologies) and incubated at room temperature for 20 minutes with intermittent gentle shaking. 4 rnL of CsCl (Life Technologies) saturated water (-1.8 g 1 mL) was added and the mixture was transferred to a Beckman (Palo Alto, CA, USA) 16 x 76 mm Quick-Seal tube, topped with a solution containing 10 rnM Tris-HC1 (pH 8), 5% sodium deoxycholate, and CsCl saturated water (approximately 65 g 1 100 mL), and centrifuged in a Ti-50 rotor (Beckman) at 43,000 x g for 18 hours at 15 OC. The viral band, located 0.5 - 1 cm above the ce11 pellet, was removed by an 18 gauge needle puncture, transferred to dialysis tubing with a molecular weight cut-off of 12,000 - 14,000 Da (Spectxum Laboratory Products, Houston. TX' USA) and dialyzed against 2L of 10 mM Tris-HCI (pH 8) for 24 hours at 4 OC to dilute the CsCI, with transfer to fiesh dialysis solution at the 3rd and 6th hours. Viral stocks were aliquoted into 200 pL volumes and stored at -70 OC. Total viral stock volume was between 1-5 to 2 mL. 4.3.6 Determination of AdS.DTD Viral Titres Ad5.DTD viral particle concentration was determined by 0D2, measurement (1 OD,, = 50 pg 1 mL double-stranded DNA, Sambrook et al., 1989) of samples of viral solution (dialyzed viral solution) and viral particle concentration was calculated with the assumption of a genomic molecular weight of 2.4 x IO7 g / mol. Ad5.DTD viral titres were determined to be: 5.4 x 109 particles 1 pL - A~S.DTD- 4.4 x 109 particles 1 PL - A~s.DTD~~~ Ad5.DTD infectious titres were determined by measurement of plaque-forming units (PFU) per unit volume (Graham and Prevec, 1991). For each vira1 solution (pst- dialysis solution) two 6-well dishes were plated with 3 x 10' 293T cells containing growth medium the night before. 10-fold senal virai dilutions were prepared by diluting 25 pL of viral stock with 225 pL of senun-fiee a-minimal essential medium to an end point of 10''~fold dilution. Growth medium was gently asphted fiom each well and replaced with 200 pL of serum-fiee a-minimal essential medium followed by 100 pL of each dilution. Dishes were incubated for 1 hour in a humidified atmosphere containing 5% CO2 at 37 OC at which point 3 rnL of agarose overlay solution was applied to each well as described in section 4.3.5. Plaque formation was assessed 5-8 days later and number of plaques / 100 PL / dilution factor was averaged and converted to PFU / pL for an estimate of viral stock concentration. The AdS-DTD infectious titres were determined to be: 1o8 PFU !pL - A~~.DTD- 108 PFU / pL - A~S.DTD~~ Thus, 1/5O viral particles were infectious. 4.3.7 Assay for Wild-Type Ad5 Contamination To check for wild-type Ad5 variants arising through recombination with the El region of 293T cells, 2 x 10' Hela cells, which do not contain Ad5 El sequences and therefore can only support replication of wild-type virus, were seeded in 6 well dishes with 5 mL of growth medium and incubated ovemight in a humidified atmosphere containing 5% CO, at 37 OC. Medium was gently aspirated and replaced with 5 mL of growth medium containing 10' viral particles of A~S.DTD- or A~s.DTD~'~~. Monoiayers were monitored for up to 10 days for potential CPE. 4.3.8 PCR Diagnostic for AdS.DTD Viral DNA for PCR was prepared by mixing 10" viral particles with 10 rnM Tris- HCI to a final volume of 50 PL. An equal volume of phenol:chlorofom:isoamyl alcohol (25:24:1) was added. mixed vigorously for 30 seconds using a Genie 2 Vortex (Fisher Scientific, Nepean, ON, Canada), and spun at 14000 x g for 30 seconds at room temperature. The pellet containing protein was discarded and the top (aqueous) phase containing Ad5 DNA was added to an equal volume of ch1oroform:isoamyl alcohol (24: 1) and mixed and separated by centrifugation as before. The top phase was retained and this step was repeated once more to remove as much phenol as possible. The aqueous phase (- 300 PL) containing Ad5 DNA was mixed with 0.1 volumes of 3M sodium acetate (pH 5.2) and 2.5 volumes of ethanol, vortexed and DNA was precipitated for 3h at -20 OC. DNA was pelleted by centrifugation at 14000 x g for 10 minutes at 4 OC. The supernatant was gently aspirated leaving a small volume (20 - 50 PL) behind to prevent disturbance of the pellet. The pellet was washed by resuspension in 0.5 rnL of 80% ethanol. centrifugation at 14000 x g for 1 O minutes at 4 OC. The supernatant was removed and a smali volume of solvent was left behind as before. The residual volume was removed by evaporation using a Savant AS290 SpeedVac Concentrator (Gibson Air Ltd., Toronto, ON, Canada) for 3 - 5 minutes in a chamber temperature of 45 OC. Pellets were soaked for 10 minutes in 50 PL of TE buffer containing 10 mM Tris-HCl (pH 8) and 1 mM EDTA followed by gentle resuspension and washing of walls of tube to ensure recovery of as much DNA as possible. 2 pL of DNA template extract in TE buffer (250 ng DNA / mL dilutions of pAE 1sp 1B .DTD-, pAE 1sp 1B.DTD~'~~, and pJM 17 used as positive molecular weight controls, or stock solutions of A~S.DTD- or A~s.DTD~~)or 2 PL of TE without DNA (negative contamination control) was PCR amplified in a mixture (50 pL total volume) containhg reaction buffer [50 mM KCI, 10 rnM Tris-HCl @H 8.3)' and 0.001% (w/v) gelatin)], 1.25 rnM MgC12, 0.2 mM dNTP mix and 5 U AmpliTaq Polymerase with 100 ng each of DTD gene-specific or Ad5 fiber protein forward and reverse primer: 5' GAG TCT GTT CTG GCT TAT AA 3' - forward primer (nt 26 1-280 DTD cDNA) 5' GTC AAA GAG GCT GCT TGG AG 3' - reverse primer (nt 702-72 1 DTD cDNA) 5' AGC ACG MGGAG GCT AAG TG 3' - forward primer (nt 5861-5880 of Ad5 genome) 5' TGA ATA TCA AAT CCT CCT CG 3' - reverse primer (nt 6132 - 6151 of Ad5 genome) The 461 (DTD cDNA amplification) and 291 (Ad5 genome amplification) bp fragments were amplified using a 'Touchdown' PCR protocol @on et al 1991) consisting of a 3 min denaturation at 94 OC. 2 cycles of 1 min denaturation at 94 OC, 1 min annealing at 60 OC, and a 2 min polymenzation at 72 OC, 12 cycles in which the annealing temperature was decreased by 1 OC until a touchdown temperature of 48 OC was reached, an additional 20 cycles with the 48 OC annealing temperature, and a final extension for 10 min at 72 OC. The PCR products were fiactionated on a 1.2% agarose gel, and visualized by ethidium bromide (1 0 pg/mL) staining. 4.3.9 Infection of BE and HT29 Cells with Ad5.O-Galactosidase IO5 BE and HT29 cells were seeded in each well of 6 well tissue culture dishes (Nunc) in 5 mi, of growth medium and incubated ovemight in a hurnidified atmosphere containhg 5% CO, at 37 OC. Growth medium was gently aspirated from each well and replaced with 0.5 of senim-fiee a-minimal essential medium with a recombinant Ad5 vector carrying the E.coli O-galactosidase (B-gal) gene under the control of the CMV promoter (provided by Dr. Henry Klamut, University of Toronto) at a MOI (plaque forming units / cell) of 3, 10, 30, 60, or 100. AAer 30 minutes of incubation, 5 mL of growth medium was added to each well. Histochemical detection of Bgal positive cells was performed 48 hours later (Sanes et al., 1986). Infection eficiency of Ad5.O-gal was assessed by determining mean s.d. fraction O-gal positive cells in 5 random 4.3 mm2 fieIds at 80X magnification. Positive cells were determined based on blue staining independent of variations in color intensity. 4.3.10 Western Blot Analysis of HT29 and BE Cells or BE Cells Infected with AdS.DTD 5 x 10' BE or HT29 cells were seeded in 10 cm diameter tissue culture dishes (Nunc) containing 10 mL of growth medium overnight in a humidified atmosphere containing 5% CO, at 37°C. Medium was gently aspirated fiom BE cells and replaced with 1 mL of sem-fiee a-minimal essential medium containing a MOI of 10 or 100 of A~S.DTD~~'or A~S.DTD~~ and incubated in a humidified atmosphere containing 5% CO, at 37OC for 30 minutes at which point 10 mL of growth medium was added to each dish. 48 h later, Ad5-DTD infected or uninfected BE and HT29 cells were harvested by scraping. Total ce11 protein extracts were prepared, protein concentrations determined, and Western Blot and densitometry analysis carried out as described in Chapter 3.3.6 with the exception that blots were probed for integrin a, expression (using a 1:1000 dilution of anti-human integrin a,) as a loading control and protease inhibitors were not used. 4.3.1 1 DTD Assay DTD enzyme activity was measured using the substrate 2,6- dichiorophenolindophenol (DCPIP) as previously described (Misra et al., 1 998, Chapter 2.3 S). 4.3.12 A~s.DTD- Infection of HT29 Ceils 5 x 10' HT29 cells were seeded in six IO cm diameter tissue culture dishes containing 10 mL of growth medium overnight in a humidified atmosphere containing 5% CO, at 37'C. Medium was gently aspirated fiom three of the dishes and replaced with serum-free a-minimal essential medium containing a MOI of 100 of A~S.DTD~~.Cells were incubated in a humidified atmosphere containing 5% CO, at 37OC for 30 minutes, afier which 10 mL of growth medium was added to each dish. The remaining three uninfected dishes were used as controls. 48 hours later, cells were harvested and DTD activities were deterrnined. 4.3.13 DTD Enzyme Activity in AdS-DTD-infected BE Celfs 5 x 10' BE cells were seeded in 10 cm diarneter tissue culture dishes containing 10 mL of growth medium overnight in a humidified atmosphere containing 5% CO2 at 37°C. Medium was gently aspirated and replaced with 1 rnL of serum-fiee a-minimal essential medium containing a MOI of 10 or 100 of A~s-DTD-. Cells were incubated in a humidified atmosphere containing 5% CO, at 37OC for 30 minutes, afler which 10 mL of growth medium was added to each dish. DTD enzyme activity was detennined 24,48, and 72 hours pst-infection. In one experiment, BE cells iïifected with A~S.DTD- at an MOI of 100 were trypsinized and half the cells were reseeded at the 48 hour time point and DTD activity was determined 48 hours later (96 hour time point). 4.3.14 Clonogenic Survival of Cells Exposed to MMC IO5 BE or Hl29 cells were seeded in 10 cm diameter tissue culture dishes containing growth medium. After ovemight incubation in a humidified atmosphere containing 5% COz at 37 OC, medium was gently aspirated and replaced with 1 mL of sem-ffee a-rninimal essential medium containing the desired MOI of A~~.DTD-or A~S.DTD~~.For experiments examining the effect of viral titration on MMC sensitivity, dilutions were perfomed starting at 100 MOI / mL A~S-DTD- stock and carried out dotvn to 3 MOI / mi, A~s.DTD- in senun-fiee a-minimal essential medium. 1 mL was added to each dish. Dishes were incubated with virus for 1 hour in a hurnidified atmosphere containing 5% CO, at 37 OC, after which 9 mL of growth medium was added to each dish and merincubated for 48 hours before MMC expusure. For MMC sensitivity determination of uninfected controls, cells were similarly plated ovemight prior to MMC exposure. MMC stocks were prepared by reconstitution with PBS at 1 mg I rnL and stored for up to 1 month at 4 OC in the dark. Stock concentrations were verified by measuring absorbance at 363 nrn (E = 21,840 cm-' M-'). In experiments requiring MMC treatrnent. medium (with or without virus) was replaced with IO mL of growth medium containing MMC at the desired concentrations and cells were incubated with drug for 1 hour in a humidified atmosphere containing 5% CO, at 37 OC. Mer MMC exposure, medium containing drug was decanted and residual dmg was removed by washing cells with 5 mL PBS and cells were harvested by trypsinization and pelleted by centrifugation at 250 x g for 5 minutes at 4 OC. Pellets were resuspended in 10 mL of growth medium and ce11 density was assessed by 0.2% trypan blue (Life Technologies) exclusion using a hemocytometer. Ce11 dilutions were prepared at IO4, IO3, and 10~ells/ rnL and 1 mL of each dilution was plated in 60 x 15 mm dishes (Nunc) in duplicate (for drug treated cells) or in triplicate (untreated controls) and incubated as before. Colonies were counted 12-14 days later afier staining with 0.5 % methylene blue (Sigma) dissolved in equal parts ethanol to water. Plating efficiency was assessed by dividing the number of observed colonies by the total number of cells plated. The mean * s.e.m. (n = 5) plating esciencies of uninfected HT29 and BE control cells were 0.84 * 0.09 and 0.69 + 0.12, respectively. Fold sensitivities of BE cells infected at various MOIS of A~~.DTD- to MMC (Figure 4.12) were determined by comparing the ratio of dose (pg/mL) required to achieve 1 log of ce11 kill relative to the initial plating efficiency to that of BE cells infected with ~d5.DTD~'~~. 4.3.15 Statistics One-way analysis of variance (ANOVA) was used to compare mean DTD activities of uninfected and Ad5 .DTD- - infected HT29 ceiis. This procedure involves calculation of 1) each activity variance fiom the overall mean; 2) variance within group (uninfected or A~S.DTD~~- infected cells) means; and 3) variance between group means. The Student's t - test was used to compare the mean survivd of BE cells infected with A~S.DTD- at a MOI of 100 for 48 hours and uninfected controls. ANOVA of regression was used to compare the MMC sensitivities of HT29 cells infected with A~s.DTD~'~'and uninfected controls. 4.4 Results 4.4.1 Ceneration of AdS.DTD to Tmnsduce Cells Expressing Low Levels of DTD The Ad5 shuttle vector, pAElsplB, was modified by insertion of a CMV promoter, a multiple cloning site, and a BGH poly A sequence derived fiom the plasmid pRc/CMV (Invitrogen, Figure 4.1). DTD cDNA encoding the DTD 609C or 609T isoforms was incorporated into these modified shuttle vecton (Figure 4.2) to generate pAE 1sp 1B .DTD- or pAE 1sp l B .DTD~'~?Each piasmid was individually cotransfected with the Ad5 packaging vector, pJM17, into 293T cells to produce A~S-DTD- or A~S.DTD~~.PCR diagnostic analysis (Figure 4.3) of the A~S.DTD- and A~S-DTD- vectors confirmed the presence of both DTD cDNA (bp 261 -721) and Ad5 sequences corresponding to the fiber protein coding region (bp 5861 - 6 15 1 of Ad5 genome, fiom Figure 4.3: PCR diagnostic analysis of A~S.DTD- and A~S.DTD-. Total genomic DNA was isolated from A~S-DTD- and A~s.DTD~~and subjected to PCR amplification as described in 'Materials and Methods.' PCR reactions were run on a 1% agarose gel. A) PCR reactions using 20 mer primers complementary to DTD cDNA (forward and reverse primers start at bp 261 and 721 of DTD cDNA, respectively). Lane 1; 123 bp ladder, Lanes 2 and 3; AdS-DTD shuttle vectors used to construct A~S.DTD- and A~S.DTD-. respectively, and included as a positive molecular weight control, Lane 4: A~S.DTD~,Lane 5; A~~.DTD~~,Lane 6; no DNA template (negative contamination control). B) PCR reactions using 20 mer pnmers complementary to Ad5 genome encoding for a fiber protein (forward and reverse primers start at bp 5861 and 615 1 of Ad5 genome, respectively). Lane 1; 123 bp ladder, Lane 2; pJM17 Ad5 packaging vector used as a positive molecular weight control, Lane 3; A~S.DTD-, Lane 4; A~~.DTD~"~~,Lane 5; no DNA template (negative contamination control). PCR reactions represented in Lanes 3 and 4 used the same AdS.DTD genomic samples as in A) (Lanes 4 and 5) indicating the presence of both DTD cDNA and Ad5 fiber region in the sme sample and suggesting successful recombination between Ad5.DTD shuttle vecton and the pJM 17 packaging vector. pJM 17), indicating successful recombination between these vectors. The absence of contamination of viral stocks by wild-type recombinant Ad5 (produced by gaining host El sequences) was confirmed by failure of each of the AdS-DTD clones to generate a CPE in Hela cells (less than one wild-type recombinant per 6 x IO4 particles of viral stock), which are devoid of El sequences. A single base transition corresponding to nucleotide 609 of DTD cDNA (C to T) has been show to result in irnpaired DTD activity and decreased protein stability (Traver et al.. 1992; Misra et al., 1998; Chapter 2). HT29 cells (homozygous for the wild-t)p DTD dlele) displayed mean * s.e.m. (n=5) DTD activity of 1060 * 21 5 nmol / min / mg protein whereas BE cells (hornozygous for the variant DTD allele) contained very low to undetectable DTD activity (less than 20 nmol I min / mg protein). Consistent with the notion that the variant allele promotes decreased DTD protein stability, HT29 cells display Western blot reactivity whereas BE cells express very low to undetectable levels of DTD protein (Figure 4.4). In this blot. DTD protein was not detectable. However, low levels of the 609T isoform can be detected by increasing antibody incubation time and / or protein concentration (data not shown). 4.4.2 Infection of HT29 and BE Cells with Ad5.B-ga1 Previous studies of Ad5.B-gal infection of HT29 and BE cells demonstrated 40% and 100% infection efficiency, respectively (Dr. Henry Klamut, University of Toronto, persona1 communication). This work verified that HT29 cells are not as good hosts for Ad5 infection as BE cells (Figure 4.5). At a MOI of 100, only 15% of HT29 cells were infected with Ad5.D-gal whereas BE cells exhibited an infection efficiency of 90%. Figure 4.4: Western Blot analysis of DTD protein levels in HT29 (lane 1) and BE (lane 2) human colon carcinoma cells. Blots were probed with a mixture of DTD monoclonal antibodies B77 1 and A 180 as descnbed in 'Materials and Methods' and were also probed for a, integrin expression as a protein loading control. MOI Figure 4.5: Infection efficiency of HT29 and BE cells by Ad5.B-gal. HT29 and BE cells were transduced with the E. coli P-gal gene under the control of the CMV promoter via Adj-P-gal infection at various MOIS. B-gal positive cells were stained 48 hours later as described in 'Materials and Methods.' Examples of A) HT29 and B) BE cells infected with Adj$-gal at a MOI of 100 are shown at 80X magnification. C) Infection efficiency of Ad5.B-gai was assessed by detemination of fiaction of P-gal positive cells. Each bar represent mean * s.d. of fraction of P-gal positive cells in five randomly chosen 4.3 mm' fields. Therefore, BE cells are capable of hi& infection eficiencies with recombinant Ad5 vectors. 4.43 Protein Expression of DTD Isoforms in Human Cells Homozygous for the Variant DTD Aiiele by Ad5.DTD Transduction The ability of each recombinant AdS.DTD to express DTD protein in human cells containing low to undetectable DTD activity was assessed by Western blot analysis (Figures 4.6 and 4.7). Monolayers of BE cells were infected with A~s-DTD- and Western blot reactivity was determined 48 hours pst-infection (Figure 4.6A). As shown in Figure 4.68, densitometry analysis indicated that BE cells infected at a MOI of 100 contained approximately 3-4 times greater DTD- protein levels than cells infected at a MOI of 10 when normalized to integrin a, loading controls. A similar fold increase in expression of the 609T isoform, mediated by A~~.DTD~'~~infection of BE cells at a MOI of 100 compared to a MOI of 10, was obse~ed48 hours pst-infection (Figure 4.7). In this blot, the presence of the lower molecular weight band (0.5 - 1 kDa smaller than wild- type DTD) may be a result of proteolytic processing or chemical degradation of the 609T isoform (Dr. David Ross, University of Colorado Health Sciences Center, personal communication) and was included in the densitometry analysis. 4.4.4 A~s.DTD- Infection of Human Cells Containing the Wild-Type DTD Allele To assess whether exogenous expression of the 609T DTD isofonn modulates Integrin a, (150 kDa) Figure 4.6: A) Western blot analysis of BE cells afier 48h infection with A~S-DTD-. Blots were probed with DTD monoclonal antibodies as described in 'Materials and Methods' and were also probed for a, integrin expression as a protein loading control. Lane 1; uninfected cells, Lanes 2 and 3; A~SDTD- infected cells at a MOI of 10 and 100, respectively. B) Densitometry analysis of DTD band intensities normalized for integrin a, levels. Values represent mean I s.d. of five densitometric readings for each lane standardized to uninfected cells. DT-Diap horase (30 kDa) Integrin a, (150 kDa) Figure 4.7: A) Western blot analysis of BE cells after 48h infection with A~S.DTD~~. Blots were probed with DTD monoclonal antibodies as described in 'Materials and Methods' and were also probed for a, integrin expression as a protein loading control. Lanes 1 and 2; Ad5DTD609Tinfected cells at a MOI of 10 and 100, respectively. B) Densitometry analysis of DTD band intensities nomalized for integrin a, levels. Values represent mean * s.d. of five densitometric readings for each lane standardized to cells infected at a MOI of 10. enzyme activity in human cells that contain the wild-type DTD allele, such as through a dominant negative effect (Levine, 1990), three dishes of HT29 monolayer cultures were infected with A~S-DTD- at a MOI of 100 and DTD activity was determined 48 hours later and compared with 3 dishes containing unidected controls (Figure 4.8). Dm enzymatic activities did not significantly differ between A~.DTD- infected HT29 cells and uninfected cells (p > 0.1, one-way ANOVA). This result is qualified by the fact that the Ad5 infection efficiency of HT29 cells is 15-40%. 4.4.5 Activity of Exogenous Wüd-Type DTD Mediated by A~S.DTD- in BE Cells The ability of A~~.DTD-to express fùnctional exogenous DTD was tested by determining DTD activity in BE cells at different times post-infection. In particular, this kinetic study was used to assess whether sufficient exogenous DTD activity can be achieved in hurnan cells for in vitro MMC sensitization- 106 BE cells were infected with Ad5 .DTDmCat a MOI of 10 or 100 and DTD activity was assessed at various time points as indicated in Figure 4.9. BE cells infected at 100 MOI with A~S.DTD- displayed a linear increase in DTD enzyme activity up to 72 hours pst-infection. Activity peaked at around 2400 mol DCPIP reduced / min / mg protein. Beyond this point, ceIIs began to lifi (monolayers reached confluency at 48 hours pst-infection) and did not survive. In one experiment, BE cells infected with A~S.DTD- at a MOI of 100 were trypsinized and re-seeded at a 1 :2 dilution and activity was determined 48 hours later (96 hour time point). Enzyme activity at this point remained high (around 2000 nmol DCPIP HT29 control HT29 + AdSDTD 609T Figure 4.8: DTD activities of uninfected HT29 cells or cells infected with ~d5DTD~"~*at a MOI of 100. Each bar represents the mean k s.d. (expressed as nrnol min-' mg-'protein) of three independent activity determinations of a ce11 lysate. mock T MOI 10 Time (h) Figure 4.9: DTD enzyme activity kinetics in BE cells. 106 BE cells were infected with A~S.DTD~~'at a MOI of 10 or 100 or mock infected and DTD enzyme activity was determined at the times shown pst-infection. Each point indicates the mean * s.e.m. of three independent experiments. In addition, in one experiment, BE cells infected with A~S.DTD~~'at 100 MOI were trypsinized and haif the cells were re-seeded at the 48h time point and DTD activity was determined 48h later (trypsin, 96h time point, n = 1). reduced / min / mg protein), indicating that exogenous DTD continued to be expressed afier trypsinization. However, cells infected at a MOI of 10 had a lower rate of increase of Dm activity than cells infected with a MOI of 100, which p!ateaued after 24 hours at around 250 nmol DCPIP reduced I min / mg protein. At 48 hours pst-infection, cells infected with a MOI of 100 displayed DTD activities of around 1320 * 70 nmol DCPIP reduced / min / mg protein, which is similar to that observed for HT29 cells. Therefore, A~S.DTD- at a MOI of 100 provides for linear increases in DTD activity in BE cells with time and within the range of DTD activity in HT29 cells. Exogenous DTD activity continues to increase for at lest 96 hours. However, optimal ce11 viability for MMC exposure studies is maintained up to 48 hours pst-infection for BE cells. 4.4.6 Clonogenic Suwival of AdS.DTD Infected Cells Exposed to MMC The HT29 / BE human colon carcinoma pair has ken previously used to demonstrate the association of the variant DTD allele with MMC resistance (Traver et al., 1992; Winski et al., 1998). In accordance with previous acute 1 hour MMC exposure studies comparing àrug sensitivity of human cells containing wild-type and variant DTD alleles (Marshall et al., 1991), BE cells displayed a 56 fold resistance to MMC compared to HT29 cells when assessed for colony forming ability 12- 14 days after exposure (Figure 4.1 OA). To examine whether exogenous expression of wild-type DTD could sensitize resistant cells to MMC, BE ceIls were infected with A~s.DTD- at a MOI of 100 and MMC exposure was carried out 48 hours pst-infection. BET 100 Figure 4.10: Clonogenic survivai of cells exposed to MMC for 1 hour as described in 'Materials and Methods.' Fractional colony swival (plating efficiency) upon MMC treatrnent of A) uninfected HT29 and BE celis and B) HT29 cells infected with A~~DTD~'~'(HT29C100) and BE cells infeçted with A~SDTD~~or A~SDTD- (BET 1 O0 and BEC 100, respectively). The dashed curves show in B) are taken fiorn A) for cornparison of MMC toxicity of infected cells to uninfected controls. C) Mean * s.e.m. DTD activities of cells in B) at time of MMC exposure. Each value in A), B), and C) represents the mean I s.e.m. of three to five independent experiments. As shown in Figure 4.10B, BE cells infected with A~S.DTD- at a MOI of 100 were 8-fold more sensitive to MMC compared to cells infected with A~S-DTD-, as determined by ratio of the doses of MMC required to achieve 1 log ce11 kill from the initial PE. However, untreated BE cells infected with A~S.DTD- had a reduced plathg efficiency compared to uninfected controls (mean PE * s.e.m., n=3. was 0.45 * 0.1 1 versus 0.69 * 0.12, respectively, p < 0.05, Student's t - test) indicating toxicity due to expression of exogenous wild-type DTD. This toxicity is most likely not due to the vector itself since untreated BE cells infected with A~~.DTD~'~~at a MOI of 100 had a pplating eficiency of 0.8 1 * 0.14 (Figure 4.1OB). BE cells infected with A~S.DTD- exhibited no major increase in MMC sensitivity, consistent with previous observations of impaired catalytic activity and protein instability of the DTD 609T isoform (Traver et al., 1997; Misra et al., 1998; Chapter 2). As shown in Figure 4.10C, exogenous expression of DTD was confirmed in cells infected with A~S-DTD- by assessing DTD activities in cells that were infected in parallel with the MMC sensitivity studies described above. HT29 cells infected with A~~.DTD~'~'at a MOI of 100 were 2-fold more sensitive to MMC than uninfected controls in terms of ratio of doses required to achieve 1 log ce11 kill from their initial PEs (Figure 4.10B), despite exhibiting similar DTD activities as wiinfected cells (Figure 4.1 OC). ANOVA of regsession indicated that HT29 cells infected with A~s.DTD- at a MOI of 100 were signficantly more sensitive to MMC than uninfected controls (p < 0.0 1). These results show that A~~.DTD-cm be used to sensitize BE cells to MMC in vitro by providing DTD activity similar to HT29 cells after 48 hours infection at a MOI of 100. 4.4.7 Effeet of Titration of MOI of A~S.DTD- on MMC Sensitinty of BE Cells To assess the effect that different levels of exogenous wild-type DTD expression would have on MMC sensitivity, BE cells were infected with A~~.DTD- at MOIS ranging from 3 to 100. DTD activities 48 hours pst-infection increased linearly with MOI (Figure 4.1 1A). As before (Figure 4.1 OB), BE cells infected with ~d5.DTD- at a MOI of LOO displayed reduced plating efficiency (mean PE * s.e.m. was 0.42 * 0.06) compared to uninfected controls. Also, as shown in Figure 4.1 IB, drug sensitivity was dependent on MOI. Similar acute MMC toxicities were observed between BE cells infected with A~~.DTD-at a MOI of 60 and uninfected HT29 cells. However, these cells displayed less than 50% of the DTD activity (mean * s.e.m. was 495 * 1 3 nmol DCPIP reduced / min / mg protein, Figure 4.1 1A) of Hl29 cells (1060 * 2 15 nmoI DCPIP reduced / min / mg protein) In this experirnent, BE cells infected with Ad5.DTDm at a MOI of 100 exhibited 4-5 tirnes increased toxicity due to MMC relative to a MOI of 60. This can be seen by plotting plating efficiencies in Figure 4.1 1B as fold increases in MMC sensitivity relative to BE cells infected with A~S.DTD~~(in terms of ratio of doses corresponding to 1 log ce11 kill from initial PE) at a MOI of 100 as a Cunction of MOI, as shown in Figure 4.12. Since DTD activity determinations in HT29 cells can vqby as much as 2-fold fiom one experiment to another, it is possible that MOIs greater than 60 are required to achieve equivalent MMC toxicity in BE cells. These results show that transduction of BE ceils with A~~.DTD- provides for a continual increase in the sensitization to MMC, from MOIs of 60 to 100. O 20 40 60 80 100 MOI MOI r3 10 A 30 .60 100 Figure 4.1 1: Clonogenic survival of BE cells infected with Ad5DTDm for 48 hours and exposed to MMC for 1 hour as a îùnction of viral titration as described in 'Materials and Methods.' A) Mean r s.e.m. DTD activities of cells infected with A~SDTD- at various MOIS. B) Fractional colony survival of A~sDTD--infected BE cells 12-14 days afler MMC exposure. The dashed curves are taken fiom Figure 4.10 A) for cornparison of MMC toxicity of infected cells to unuifected controls. Each value in A) and B) represents the mean f s.e.m. of three independent experiments. O 20 40 60 80 1O0 MOI Figure 4.12: Fold increases of BE cells to MMC sensitization via A~s.DTD- transduction. Each point represents the mean @s.e.m.folci increase in MMC sensitivity as a function of MOI. Values were taken fiom 4.1 1 B). 4.5 Discussion Bioreductive enzyme - prodmg therapy involves the use of compounds that undergo metabolic reduction to cytotoxic metabolites (Workman, 1993). One bder to this approach is the presence of insufficient levels of activating enzymes in tumor tissue (Belcourt et al., 1998). DTD is a bioreductive enzyme that can activate a variety of prodrugs, including quinone-based dmgs, to DNA alkylating agents. DTD activity has been reported to be elevated in a range of tumor types (Fitzsimmons, 1996) such as in hwnan non-small ce1 lung cancer (Kepa and Ross, 1999; Marin et al., 1997), breast, liver (Belinsky and Jaiswal, 1993), and colon turnors relative to surrounding normal tissue. It has been proposed that various prodrugs that are substrates for DTD could be usefid for the treatment of tumors with high DTD activity (Beall et al., 1995). However, not al1 tumors express elevated levels of DTD. For example, heterogeneity in DTD expression has beeiï reported in the National Cancer lnstitute (USA) human tumor ce11 line panel, which includes ce11 lines denved fiom leukemic, lung, colon and breast turnors (Fitzsimmons et al., 1996). This conclusion was based on both DTD mRNA expression levels and Western blot analysis. Heterogeneity was observed both within and across tumor ce11 types but a correlation was observed between DTD expression and MMC sensitivity in this panel. DTD enzymatic activity has also been show to be comparable in breast and lung turnor samples relative to corresponding normal tissue (Marin et al., 1997). Further complicating the heterogeneous nature of DTD expression across tissue types is the role of a single base change, a C to T transition in nucleotide 609 of DTD cDNA, confemng a predicted proline to serine change in arnino acid 187. This base change is frequent in the population (Chapter 1, Section 1.3.2; Chapter 5, Table 5.1) and has been associated with reduced catalytic activity, protein instability, and MMC resistance (Traver et al., 1992, Misra et al., 1998; Chapter 2). One way to address this problem is to introduce a bioreductive enzyme like DTD into tumor cells using gene transfer techniques (Searle et ai., 1998). Such a gene therapy approach to deliver activating enzymes to sensitize human tumors to prodrugs has been most widely used with non-marnmalian enzymes, such as herpes simplex virus - thymidine kinase (Esandi et al., 1997) and E. coli cytosine deaminase (Block et ai., 2000) in combination with the nucleoside analogues gancyclovir and 5-fluorocytosine, respectively. In the curent study, human DTD coding sequences were delivered using a fvst generation (El and E3 deleted) replication incompetent adenoviral vector to increase DTD activity and sensitize tumor cells to MMC. To test the MMC sensitization effect mediated by AdS.DTD in viîro, recombinant adenoviral vectors were constructed that express either the DTD wild-type 609C or mutant 609T isoforms under the control of a CMV promoter (A~s.DTD- and A~S.DTD-). BE cells express low to undetectable levels of DTD activity and protein (Figure 4.4) and are good hosts for Ad5 infection (Figure 4.5). Infection of BE cells, which are homo ygous for the variant allele, with A~s.DTD- at a MOI of 10 or 100 resulted in high levels of recombinant DTD expression that increased with increasing MOI (Figure 3.6). Sirnilarly, expression of the DTD 609T isoform could be demonstrated 48 hours post-infection of BE cells with A~S.DTD~~at a MOI of 10 or 100 in BE cells (Figure 4.7). These results demonstrate, qualitatively, that increased levels of recombinant DTD expression occur with increasing MOI. The DTD protein levels in BE cells infected with A~S-DTD- at a MOI of 100 versus 10 was consistent with their ratio of DTD activities at 48 hours post-infection (Figure 4.9). Relating recombinant DTD protein levels to enzymatic activity may be complicated by the possibility of a dominant negative effect (Levine, 1990) on DTD activity in cells expressing the variant allele. This also poses a potentiai barrier for the sensitization by AdS.DTDMPCof MMC resistant tumor cells which express the variant allele. To test this possibility, HT29 cells, which are homozygous for the wild-type DTD allele, were infected with the mutant Ad5.DTD609Tat a MOI of 100 and cellular DTD activity was assessed 48 hours pst-infection. Since expression of the 609T allele did not have an appreciable effect on endogenous wild-type enzymatic activity (Figure 4.8), it is predicted that the mutant allele does not exert a dominant-negative effect. These results are qualified by the low infection efficiency of Ad5 towards HT29 cells (Figure 4.5) and by the fact that the level of mutant DTD protein was not assessed. However, over-expression of the 609T isoform in cells homozygous for the wild-type DTD allele has recently been shown by others to not affect cellular DTD activity despite being present at levels up to 5-fold in excess of wild-type protein (Leith et al., 2000). Therefore, it is unlikely that the 609T isoform can mediate a dominant-negative effect on wild-type DTD which allows tumor cells expressing the variant allele to be effectively transduced by A~S-DTD-. The ability of A~~.DTD-to provide for fùnctional exogenous DTD expression in tumor cells was demonstrated by the linear increase in DTD activity with time following infection at a MOI of 100, where activity peaked at 2400 nmoI / min / mg protein at 72 hours pst-infection (Figure 4.9). This is consistent with previous work dernonstrating that wild-type DTD protein has a half life that is greater than 24 houn (Siegel et al., 2000). The culture conditions did not allow activity determinations beyond this time point, since the cells lif'ted off shortly after achieving confluency. However, expression of exogenous wild-type DTD in BE cells likely extends beyond 72 hours post- infection. This was demonstrated by infection of BE cells with A~S.DTD- at a MOI of 100, followed by trypsinization and re-seeding of half the cells at 48 hours pst-infection. DTD activity assessrnent 48 hours later hdicated greater than expected activity (2000 nmol / min / mg protein) when accounting for ce11 dilution. At a MOI of 10, BE cells infected with A~S.DTD- also displayed an increase in DTD activity with time, although at a lower initial rate than at a MOI of 100. However, activity appeared to level off after 24 hours. This effect cannot be explained by differences in ce11 growth rates since uninfected controls and BE cells infected with A~S-DTD~at a MOI of 10 and 100 reached confluency at the same time, between 48- 72 hours. The levelling-off of DTD enzyme activities after 24 hours in BE cells infected with A~~DTD-at a MOI of 10 may reflect a balance between the rates of recombinant DTD production and degradation which may level off to 10-fold higher leveis in BE cells infected at a MOI of 100. The levelling off points rnay represent maximal production of recombinant DTD protein from the fraction of cells that were infected at a MOI of 10. These results demonstrate that high expression of recombinant DTD protein occurs over 3-4 days post-infection with A~s.DTD- and that a MOI of 100 provides an advantage over a MOI of 10 for achieving higher tevels of DTD activity in BE cells. In MMC sensitization studies, BE cells were treated with MMC 48 hours post- infection with A~S.DTD- at a MOI of 100. Of the two MOIS and various times pst- infection of A~S.DTD- tested, a MOI of 100 for 48 hours was chosen because it results in maximal DTD activity without toxicity and optimal cell viability (Figure 4.9). Untreated BE cells display a 5-6 fold increase in MMC resistance compared to HT29 cells (Figure 4. LOA). However, treatment with A~S.DTD- at a MOI of 100 resulted in a 8-fold increase in MMC sensitivity relative to BE cells infected with A~S.DTD- at a MOI of 100 when assessed by clonogenic survival 12-14 days subsequent to a 1 hour exposure to MMC (Figure 4.10B). Transduction of BE cells with A~S-DTD- at a MOI of 100 resulted in a 1.5-fold reduction in plating efficiency. Treatment with A~S.DTD- at a MOI of 100 had no effect on BE ce11 swival, indicating that increased sensitivity to MMC can be attributed to expression of exogenous wild-type DTD, and not to the Ad5 vector itself. The presence of BE cells that are MMC resistant due to faihre of A~S.DTD~~to tramduce and / or infect these cells is unlikely since drug sensitization by this vector appeared to be present down to 4 to 5 logs of cell kill. This indicates that complete A~S.DTD- transduction of BE cells cm be achieved at a MOI of 100 in vitro. Interestingly, HT29 cells infected with A~S.DTD- were 2-fold more sensitive to MMC compared to uninfected controls (Figure 4.10B) despite exhibiting similar DTD activities (Figure 4.8 and Figure 4.1 OC). Even when accounting for the observation that DTD activity determinations can Vary up to 2-fold with HT29 ce11 extracts, the reason for the increased sensitivity of Ad5.DTDm - infected cells is unclear since there was no apparent effect on their plating efficiency compared to uninfected controls, as seen with similar infection of BE cells. One possibility for the increased sensitivity of A~~.DTD- - infected HT29 cells is more efficient metabolism of MMC by exogenous DTD. The small degree of additional MMC sensitization may be due to infection of 1540% of these ce1 1s by ~d5.DTD-. Another possi bility is that cellular compartmentalization of DTD may determine the degree of MMC reduction. However. no precedent for these possibilities has been found. DTD activity increased linearly in BE cells as a fûnction of MO1 of A~S.DTD- up to a MOI of 100. (Figure 4.1 IA). Also, MMC sensitization of BE cells was shown to be dependent on MOI of A~S.DTD- since a senes of viral titrations (Figure 4.1 1B) at a MOI of 3: 10, 30, 60, 100 resulted in a 0.7, 1 .O, 1.1, 3, and 8-fold increase in MMC sensit ivity compared to uninfected controls, respective1y (Figure 4.1 2). Therefore, exogenous DTD activities and MMC sensitivities of BE cells infected with A~~.DTD- did not saturate at MOIS up to 100. The inability of currently available gene therapy vectors to sensitize al1 cells in a turnor population may be partially obviated by two approaches utilizing 1) the bystander effect observed with CB 1954 (Vile et al., 2000); and 2) use of enzyme 1 prodrug systems that are more efficient than DTD / MMC to increase the effective dose of active metabolite capable of generating a bystander effect towards untransduced cells. Another human DTD isofonn, containing a glutamine to tyrosine substitution in arnino acid 104 (Q 104Y), displays a 6-fold greater catalytic activity towards CB 1954, similar to rat DTD (Chen et al., 1997). In summary, first generation recombinant Ad5 vectors have been constmcted that can stably express both the wild-type (609C) and 609T DTD isoforms in hurnan cells. Treatrnent of MMC-resistant BE cells with A~S.DTD- was shown to increase MMC sensitivity by 8-fold. This result raises the possibility that a gene therapy approach based on DTD expression in tumor cells resistant to MMC can be developed to treat tumors in vivo which, in general, are heterogeneous in DTD activity. Incorporation of DTD isoforms that are more cataiytically efficient than wild-type DTD, and the use of dmgs that exhibit a substantial bystander effect, wouid significantly improve the thetapeutic potential of this strategy. 4.6 References Beall HD, Murphy AM, Siegel D, Hargreaves RH, Butler J and Ross D. Nicotinarnide adenine dinucleotide (phosphate): quinone oxidoreductase (Df-diaphorase) as a target for bioreductive antitumor quinones: quinone cytotoxicity and selectivity in hurnan lung and breast cancer ce11 lines. Mol Pharmacol 48: 499-504,1995. Begleiter A, Leith MK, Curphey TJ and Doherty GP. Induction of DTD in cancer chemoprevention and chemotherapy. Oncol Res 9: 371-82,1997 Belcourt MF, Hodnick WF, Rockwell S and Sartorelli AC. Explorhg the mechanistic aspects of mitomycin antibiotic bioactivation in Chinese hamster ovary cells overexpressing NADPH:cytochrome C (P-450) reductase and DT-diaphorase. Ah, Enzyme Regul38: 1 11-133, 1998. Belinsky M and Jaiswai AK. NADp)H:quinone oxidoreductase 1 (DT-diaphorase) expression in normal and tumor tissues. Cancer Metastasis Rev 12: 103-1 1 7, 1993. Block A, Freund CT, Chen SH, Nguyen KP, Finegold M, Windler E and Woo SL. Gene therapy of metastatic colon carcinoma: regression of multiple hepatic metastases by adenoviral expression of bacterial cytosine deaminase. Cancer Gene Ther 7: 438-445, 2000. Bridgewater JA, Springer CJ, Knox RJ, Minton NP, Michael NP and Collins MK. Expression of the bacterial nitroreductase enzyme in mamrnalian cells renders them selectively sensitive to killing by the prodrug CB1954. Eur J Cancer 31A: 2362-2370, 1995. Chen S, Knox R, Wu K, Deng PSK, Zhou D, Bianchet MA and Amzel LM. Molecular bais of the catalytic differences among DT-diaphorase of human, rat, and mouse. J Biol Chem 272: 1437-1439, 1997. Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS. 'Touchdown' PCR to circurnvent spurious pnrning during gene amplification. Nucleic Acids Res 19: 4008, 199 1. Emster L. DT-diaphorase. Methods Enzymol10: 309-31 7, 1967. Ernster L. DT-diaphorase: A historical review. Chemica Scripta 27A: 1 - 13, 1987. Esandi MC, van Someren GD, Vincent AJ, van Beklcum DW, Valerio D, Bout A and Noteboom JL,. Gene therapy of experirnental malignant mesothelioma using adenovim vectors encoding the HSVtk gene. Gene mer 4: 280-287, 1997. Fitzsimmons SA, Workman P, Grever M, Pauil K, Carnalier R and Lewis AD. Reductase enzyme expression across the National Cancer Institute Tumor ce11 line panel: correlation with sensitivity to mitomycin C and E09. JNctrl Cancer lnst 88: 259-269, 1996. Graham FL and Van der Eb AJ. A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52: 456467,1973- Graham FL and Prevec L. Manipulation of adenovim vectors. Methods Mol Biol7: 109- 128, 1991. Iyer VN and Szybalski W. Mitomycins and Porifomycin: chernical mechanism of activation and cross-linking of DNA. Science 145: 55-58, 1964. Jaiswal AK. Human NAD(P)H:quinone oxidoreductase2. Gene structure, activity, and tissue-specific expression. J Biol Chem 269: 14502- 14508, 1994. Johnson LG, PickIes RI, Boyles SE, Moms JC, Ye H, Zhou 2, Olsen JC and Boucher RC. In vitro assessrnent of variables affecting the efficiency and efficacy of adenovirus- mediated gene transfer to cystic fibrosis ainvay epithelia Hum Gene Ther 7: 5 1-59, 1996. Kepa JK and Ross D. DT-diaphorase activity in NSCLC and SCLC ce11 lines: a role for fos/jun regulation. Br J Cancer 79: 1679- 1684, 1999. Keyes SR, Fracasso PM, Heimbrook DC, Rockwell S. Sligar SG and Sartorelli AC. Role of NADPH:cytochrome c reductase and DT-diaphorase in the biotransformation of mitomycin C 1. Cancer Res 44: 5638-5643, 1984. Leith MK, Doherty GP, Pan S and Begleiter A. Mutant DT-diaphorase does not act as a dominant-negative in heterozygous human cells. Proc Amer Assoc Cancer Res 41: 767, 2000. Levine AJ. Tumor suppressor genes. Biuessays 12: 60-66, 1990. Marin A, Lopez de Cerain A, Hamilton E, Lewis AD, Martinez-Penuela JM, Idoate MA and Bello J. DT-diaphorase and cytochrome B5 reductase in hwnan lung and breast tumours. Br J Cancer 76: 923-929, 1997. Maritus C. The Enzymes, ed. Boyer PD, Lardy H and Myrback K, vol VII, pp. 5 17-532. Academic Press, New York, 1963. Marshall RS, Paterson MC and Rauth AM. DTD activity and mitomycin C sensitivity in non-transformed ceIl strains denved fiom members of a cancer-prone family. Carcinogenesis 12: 1 175-80, 1991. McGtory WJ, Bautista DS and Graham FL. A simple technique for the rescue of early region 1 mutations into infectious human adenovirus type 5. Virology 163: 614-6 17, 1988. Misra V, Klamut HJ and Rauth AM. Transfection of COS-1 cells with DT- diaphorase cDNA: role of a base change at position 609. Br J Cancer 77: 1236-1240, 1998. Phillips RM, Naylor MA, Jaffa. Ml Doughty SW,Everett SA, Breen AG, Choudry GA and Stratford IJ. Bioreductive activation of a series of indotequinones by human DT- diaphorase: structure-activity relationships. J Med Chem 42: 407 1-4080, 1999. Sartorelli AC, Hodnick WF, Belcourt MF, Tomasz M, HeB, Fischer JJ and Rockwell S. Mitomycin C: a prototype bioreductive agent. Oncol Res 6: 501-508, 1994. Sambrook J, Fritsch EF and Maniatis T. Molecdar Cloning. A Laboratory Manual. 2nd Ed., Cold Spring Harbor Laboratory Press, Chapter 1 1, p.28, 1 989. Sanes JR Rubenstein JL and Nicolas JF. Use of a recombinant retrovirus to study post- implantation ce11 lineage in mouse ernbryos. EMBO J 5: 3 133-3 142, 1986. Searle PF, Weedon SJ, McNeish IA, Gilligan MG, Ford MJ, Friedlos F, Springer CJ, Young LS and Kerr DJ. Sensitisation of human ovarian cancer cells to killing by the prodrug CB1954 following retrovirai or adenoviral transfer of the E. coli nitroreductase gene. Adv Ejrp Med Biol451: 107-1 13, 1998. Siegel D, Anwar A, Winski S, Kepa JK and Ross D. The NQ01*2 polymorphism in NAD(P)H: quinone oxidoreductase 1 targets the protein for proteosomal degradation. Proc Amer Assoc Cancer Res 41: 35,2000. Talalay P. Mechanisms of induction of enzymes that protect against chernical carcinogenesis. Adv Enz Reg 28: 149-1 59, 1989. Talalây P, Fahey JW, Holtzclaw WD, Prestera T and Zhang Y. Chernoprotection against cancer by phase 2 enzyme induction. Toxicol Leti 82-83: 173- 1 79, 1995. Traver RD, Horikoshi T, Danenberg KD, Stadlbauer TH, Danenberg PV, Ross D and Gibson NW. NAD(P)H:quinone oxidoreductase gene expression in human colon carcinoma cells: charactenzation of a mutation which modulates DTD activity and mitomycin sensitivity. Cancer Res 52: 797-802,1992. Traver RD, Siegel D, Beall HD, Phillips RM, Gibson NW, Franklin WA and Ross D. Characterization of a polymorphism in NAD(P)H: quinone oxidoreductase @T- diaphorase). Br J Cancer 75: 69-75, 1997. Vile RG, Russell SJ and Lemoine NR. Cancer gene therapy: hard lessons and new courses. Gene Ther 7: 2-8,2000. Wilson RK, Chen C, Avdalovic N, Burns J and Hood L. Development of an automated procedure for fluorescent DNA sequencing. Genomics 6: 626-634, 1990. Winski SL, Hargreaves RH and Ross D. A new screening system for NAD(P)H:quinone oxidoreductase (NQO1)-directed antitumour quinones: identification of a new aziridinylbemquinone, RH 1, as a NQO 1-directed antitumor agent. Chical Cancer Res 4: 3083-3088, 1998. Workman P and Stratford IJ. The experimental development of bioreductive drugs and their role in cancer therapy. Cancer Metastusis Rev 12: 73-82, 1993. Workman P. Enzyme-directed bioreductive dmg development revisited: a commentary on recent progress and future prospects with emphasis on quinone anticancer agents and quinone metabolizing enzymes, particularly DT-diaphorase. Oncol Res 6: 46 1-475, 1994. CHAPTER 5 Discussion and Future Directions An important requirement for the treatment of tumors with bioreductively activated prodnigs is the presence of enzymes capable of carrying out the process of activation (Stratford and Workrnan, 1998). Treatment of tumors with specitic dmgs such as mitomycin C (MMC) that are activated by DT-diaphorase @TD) illustrates the potential for use of enzyme-directed prodrug therapy (Workman and Stratford, 1993). The heterogeneous nature of DTD expression in twnor cells (Marin et al., 1997) has limited the eficacy of bioreductive dmgs that are substrates for this enzyme- A fiequent point mutation that results in decreased DTD activity and protein stability (Traver et al., 1992) has prompted interest in how it affects the role of this enzyme not only as an activator of bioreductive dmgs, but also as a xenobiotic detoxifier leading to the prevention of carcinogenesis (Ross et al., 1993). The following section discusses the implication of the work presented in the previous three chapters in relation to recent studies of the nature of the DTD 609C to T point mutation and developments in the field of gene-directed enzyme prodrug therapy. The final sections of this chapter describe future directions and conclude with an overall summary. 5.1 Role of the 609C to T Nucleotide Substitution in DTD cDNA Studies of the possibility of targeting MMC to radiation-resistant hypoxic turnor cells (reviewed by Rockwell, 1992) led to the identification of ceIl lines and strains that are resistant to this drug under aerobic conditions (Marshall et al., 1989). This resistance was lost under hypoxic conditions and these cells were found to be deficient in DTD activity (Marshall et al., 1991). It was subsequently shown that one electron reductases control this differential aerobic / hypoxic cytotoxicity towards MMC, whereas DTD is important for its aerobic toxicity (Belcourt et al., 1996). As discussed in Chapter 1.3.1, work by Ross and CO-workers (1992) with cells exhibiting a range of sensitivities to MMC, in particular with HT29 and BE human colon carcinoma cells, led to the association of the DTD 609T allele with impaired catalytic activity and MMC resistance. This association raised four main questions: 1) Does this base change result in a fùnctionally mutant DTD protein? 2) If so, how? 3) How does this mutation affect DTD function in terms of its role as a detoxification enzyme and as a dmg activating enzyme? 4) Can genotypic status predict for phenotype? 5.1.1 Relationship of C to T Base Change with DTD Enzymatic Activity and Protein Stability Direct evidence that the serine 187 DTD isoform is functionally mutant for enzyme activity was provided by examination of expressed and purified recombinant protein fiom E. coli (Traver et al.. 1997; Wu et al., 1998). These products exhibited 2 - 10% of the catalytic activity of wild-type DTD depending on the substrate used to test enzyme activity. Similarly, mammalian ce11 transfectants expressing the serine 187 isoform displayed 10-fold lower cellular DTD activity compared to wild-type DTD transfectants (Chapter 2; Misra et al., 1998). It has recently been shown that the serine 187 DTD isoform is more susceptible to proteasome - mediated degradation compared to wild-type DTD (Siegel et al., 2000). This is consistent with the lower levels of DTD protein in the 609T compared to 609C COS4 ce11 transfectants and very low to undetectable Western blot reactivity in ce11 lines and strains that are hornozygous for the 609T allele (Figures 3.5 and 4.4). Therefore, the C to T base change is a mutation that impairs enzymatic activity and results in decreased protein stability. 5.1.2 Nature of the C to T Mutation Detailed examination of serine 187 isofonn expression by Siegel et al. (2000) revealed that not only do mRNA tranxripts of wild-type and mutant DTD have similar half-lives. they are translated with equal efficiency. "s-methionine labeling and cycloheximide treatment of HT29 and BE cells revealed that the wild-type DTD isofonn was not degraded over a 24 hour period while mutant DTD was completely degraded in 2 hours. This effect was also seen with purified DTD isoforms that were expressed in a rabbit reticulocyte lysate (RRL) system. Furthemore, general protease inhibitors, such as cysteine or serine protease inhibitors, did not prevent degradation of mutant DTD. Degradation of mutant DTD was, however, prevented by the proteasorne inhibitor clasto- lactacystin p-lactone. Degradation was also ATP - dependent and mutant DTD purified from the RRL system reveaied proteins of increased molecular weight which were immunoreactive for ubiquitin. Rapid degradation via the ubiquitin - directed proteasomal pathway could account for the presence of only trace amounts of mutant DTD protein in cells that are homozygous for the DTD 609T mutation. Overexpression of mutant DTD in BE cells via A~s.DTD~"~infection resulted in the presence of DTD protein of a molecular weight that is 0.5 - 1 kDa smaller than wild- type DTD (Figure 4.6). Densitometric analysis demonstrated that the amount of protein corresponding to the lower band was approximately 3 and 0.25 times the amount relative to the upper band, at a multiplicity of idection (MOI) of 10 and 100, respectively. The lower band has also been observed in immunoblots of mutant protein punfied nom E. coli (Dr. David Ross, personal communication, University of Colorado Health Sciences Center), but only at relative levels to the upper band closer to that seen with BE cells infected with A~S.DTD- at a MOI of 10 (Le. present at 3-fold greater levels than the upper band) Beigleiter (personai communication, Manitoba Institute of Cell Biology) and co- workers have also detected both bands in heterozygotes by Western blot analysis under stringent conditions involving longer length gel separation and low current. In the analysis of heterozygotes by Beigleiter and CO-workers,the relative levels of the lower to upper bands varied. It is possible that high levels of mutant DTD present in BE cells infected at high MOIS may saturate the degradative processing of this isofom, resulting in a 0.25: 1 ratio of the lower to the upper band. Recently, the crystai structure of hurnan wild-type DTD was detemined to 1.7 A resolution (Faig et al., 2000). Sirnilar to the crystal structure of the rat enzyme (Figure 13, discussed previously in Chapter 1, Section 1.2.2, amino acid 187 of the hurnan DTD protein does not appear to contribute to dimerization or the active site. It has been proposed that a serine substitution at amino acid 187 results in a lower *nity towards FAD and impaired catalytic activity (Wu er al., 1998). Altematively, the proline to serine 187 substitution could result in a change in native conformation that makes the mutant serine 187 isoform more susceptible to degradation via the ubiquitination - directed proteasomal pathway . The degradative product corresponding to the lower band is unlikely to be proteasorne - mediated since no higher molecular weight bands indicative of ubiquitination. Also, the proline to serine change does not appear to resdt in a novel cleavage site for proteases that would generate a 0.5 - 1 kDa fragment. This is consistent with the inability of general protease inhibitors to protect against degradation of mutant DTD. However, it is possible that changes in conformation of the mutant isoforrn may create a proteolytic cleavage recognition site other than within amino acid 187. If a high concentration of purified mutant DTD is dissolved in a protease - free buffer solution at room temperature or at 37 OC, both the upper and lower bands are seen by Western blot analysis (Dr. David Ross, personai communication, University of Colorado Health Sciences Center). With increasing incubation time, the concentration of protein corresponding to the lower band increases relative to the upper band. This observation suggests that the lower band represents a chernical degradation product of the mutant isoform. Having established that the 609C to T base change is a point mutation that impairs DTD activity and results in protein instability, studies of the serine 187 isoform with the aim of providing a structure - function basis for this mutation are warranted. 5.1.3 Genotype vs. Pbenotype 1: Frequency Analysis of the Variant Allele and the Role for DTD in Xenobiotic Detoxification Since the C to T mutation results in a DTD protein that is defective for enzymatic activity and has decreased stability, it has raised the issue of how it affects physioiogicai DTD function. This was first addressed by estimating its prevalence in the population. The possibility that penetrance of this DTD mutation is associated with impairment of xenobiotic detoxification (in tems of its relevance as a defense mechanism against disease) and chemotherapeutic dnig activation has prompted a number of mutational fiequency studies (surnmarized in Table 5.1). These studies compared groups fiom disease-fiee populations to those exhibiting various types of cancer. They also include correlative studies of DTD activity relative to the presence of the nucleotide 609C to T change in ce11 lines and strains (Traver et al., 1992; Traver er al., 1995; Misra et al., 1998; Chapter 2). Ethnic variation (Table 5.1) has ken observed in the fiequency distributions of the wild-type and variant alleles of DTD (Wiencke et al., 1997; Kelsey et al. 1997; Gaedigk et al., 1998; Chen et al., 1999), with the mutant allele occurring at fiequencies as high as 0.49 in a Chinese population. The fiequency of the variant allele may reflect regional environmental pressures relating to metabolism of dietary products or xenobiotic detoxitication as hypothesized for other polymorphic drug metabolism enzymes such as the cytochrorne P450s (for review see Nebert et al. 1999). A number of studies have associated increased fiequency of the variant allele with various chernical toxicities and malignancies (Table 54, such as benzene-related hematotoxicity (Rothrnan et al., 1996; Moran et al., 1999), acute leukaemia (Wiemels et al.. 1999). rend cancer and urothelial carcinoma (Schulz et al., 1997), as well as in benign prostatic hyperplasia (Steiner et al., 1999). However, in the case of lung cancer, some reports have shown no relationship between lung cancer and the variant allele (Traver et al., 1997; Bouchardy et al., 1999). No change was seen in the fiequency of the variant allele in lung tumor biopses compared to surrounding normal tissue (Siegel et al., 1999), and one study even found an association between the variant ailele and decreased Table 5.1: Summary of some reports describing A) the occurrence of the variant in human populations; B) its association with either toxin exposure; or C) cancer incidence. Al Freauencv of the variant ailele in various human - ,pulations Observation Detection Reference Svstem 4% of 628 British unrelated individuals Gel-protein Edwards et al. (1 980) lack DTD protein. staining Fust identification of nucleotide 609 C to T change RT-PCRand Traver et al. (1992) in the BE colon carcinoma ce11 line and hypothesis cDNA that it is associated with low DTD enzymatic sequencing activity. The base change results in the formation of a novel Eickelmann et al. (1 994) K~nf1 site. First report of the PCR-Kiinf 1 assay and its use to demonstrate the association of loss of the wild-type allele in a human bladder carcinoma ce11 line with MMC resistance. 9% of 44 normal bone marrow donors were Kuehl et al. (1995) homozygous for the variant allele while 40% were carriers. Identification of a non-small ce11 lung cancer ce11 Enzymatic Assay Traver et al. (1 995) line, H596, that, like BE cells, is homozygous for and cDNA the variant allele which contains undetectable DTD sequencing activity and protein. 13% of 82 individuals fiom a reference farnily were PCR-SSCP Rosvold et al. (1 995) homozygous for the variant allele. An Asian population (n=l18) displayed allele PCR-Hinf 1 Kelsey et al. (1 997) frequencies of OS6 and 0.44 for the wt and variant alleies, respectively, and represented the highest fiequency of the variant DTD allele among four ethnic groups. Frequency of the variant allele in Chinese (n=86), Gaedigk et al. (1998) Inuit (n=83), Native Indian (n= 152), and Caucasian populations (575) were 0.49,0.46,0.40, and 0.16, respective1y. DTD protein is detected in biopsy material Siegel et al. (1999) including lung tissue and bone marrow as well as in saliva sarnples but not in tissues that contain ceils that are homozygous for the variant allele Table 5.1 continued BI Association of the variant allele with disease related to toxin exmsure Detection Reference Sy stem -- PCR-Kinf I Rothman et al. ( 1996) controls (n=35) revealed that individuals homozygous for the variant allele had a 3.2-fold 1 increased risk of benzene-related disease. 1 COS- 1 cells transfected with eukaryotic expression Enzymatic Assay Misra et al. (1 998) vectors containing wild-type DTD cDNA had 10- fold higher DTD activity and 3-fold higher DTD protein levels than transfectants receiving cDNA derived fiom the variant allele. DTD failed to be induced in bone marrow cells that Enzyrnatic Assay Moran er al. (1 999) were homozygous for the variant allele upon And Western exposure to a benzene hydroquinone metabolite biot analysis whereas high and intermediate induction was observed in cells that were wild-type and heterozygous, respectively . C) Incidence of variant allele in relation to cancer Observation Detection Reference System Allele fiequencies were 0.58 wt and 0.42 mutant in PCR-ffinf 1 Wiencke et al. (1 997) a Mexican-American population (n= 1 6 1) and 0.78 wt and 0.22 mutant in an Afican-American (n=136) population and the wild-type allele was associated with lung cancer cases. Frequency analyses of hurnan lung tissue samples from various ethnic backgrounds consisting of 45 matched sets of turnor and uninvolved tissue revealed a 7% incidence of individuals that were homozygous mutant and 42% were heterozygotes with no evidence of elevated frequency of the variant allele in tumor tissue. Increased fiequencies of the variant allele were O bserved with rend cancer (0.1 9 1, n= 13 1) and urothelial carcinoma (0.1 82, n=99) patients compared to a normal population (0.133, n=260). Table 5.1 C) continued Observation Detection Reference S ystem Notable differences in the variant allele fiequencies PCR-Hinf 1 Chen et al. (1 999) were seen among ethnic groups: Japanese (0.34, n=276), Caucasians (0.17, n=306), and Hawaiians (0.1 9, n= 185). The variant allele was associated with decreased lung cancer risk. Variant allele fiequency was slightly higher in PCR-Huif 1 Steiner et al. ( 1999) benign prostatic hyperplasia patients (0.235, n=49) than case controIs (0.175, n=100)- Study failed to show any significant relationship PCR-Hinf I Bouchardy et al. (1 999) between presence of the variant allele and iung cancer risk in Caucasian smokers (n=322) Variant allele fiequency was higher in acute PCROfinf 1 Wiemels et al. (1 999) leukaemia patients (0.45) that contain translocations involving the MLL gene versus its fiequency in a control group (0.17). It was not significantly higher in groups with other forms of acute leukaemia not involving the MLL translocation (n=36) incidence of lung cancer (Wiencke et al., 1997). Therefore, the relationship between the frequency of the variant allele and lung cancer remains uncertain. A definitive conclusion as to whether impairment of DTD function contributed to the disease outcomes described in Table 5.1 cannot be presently made. The range of cancers examined was not oniy broad in nature but there was in~~cientevidence for their cause. A role for DTD in cancer prevention cannot be established until these causal factors are identified and the contribution of impaired DTD activity to carcinogenesis can be assessed. The establishment that DTD knockout mice are healthy but are highly sensitive to menadione toxicity (Radjendirane et al-, 1998) points to an anti-cancer role for DTD involving xenobiotic detoxification. However, these mice did not display an increased incidence of cancer. DTD has also ken show to protect against benzene related hematotoxicity (Rothman et al., 1996) through its ability to metabolize cytotoxic benzenes to inactive species (Sarlauskas et al., 1997). Furthemore, the degree of cellular sensitivity to benzene exposure was related to the allelic status of the variant allele (Moran et al. 1999; Table 5.1); i-e. cells homozygous for the wild-type allele had the lowest sensitivity, while heterozygotes displayed intermediate sensitivity and cells homozygous for the variant allele had the greatest sensitivities. These results implicate the variant allele as a genetic marker for benzene toxicity. 5.1.4 Genotype vs. Phenotype 2: The C to T Mutation as a Predictor of Dnig Response At present, a more definitive argument can be made for a role for the variant allele in encoding for a DTD mutant with reduced catalytic activity and protein stability than for its use as a genetic marker for cellular responses that depend on DTD activit~.As discussed in Chapter 3, there is considerable overlap in DTD activities between cells that are homo ygous or heteroygous for the wild-type allele (Figure 3-3). nierefore, genotypic analysis can only provide a first approximation at the cellular level for response to dmgs that are DTD substrates. The low to undetectable levels of DTD protein associated with the variant allele observed in Chapters 2 and 3 is consistent with recent work by Siegel et al. (1 999), where DTD protein levels were quantified in human saliva samples by irnrnunoblot analysis relative to a recombinant DTD standard. This method provides for a non-invasive approach to relating wild-type DTD protein levels and DTD activity. As discussed above, the apparent lower molecular weight of the mutant DTD isoform is a useful property for identification and subsequent quantification of wild-type DTD in protein sampIes. In addition. PCR-based analysis would be usefùl for identification of heterozygotes since the upper band may potentially consist of a mixture of wild-type DTD and unprocessed mutant DTD, although the latter would be expected to be present in trace amounts. Therefore, these analyses may provide for a reliable prediction of the response of tumor cells to drug activation mediated by DTD in vivo. 5.1.5 A~S.DTD- as a Gene Therapy Vector Cwent cancer gene therapy vectors, including adenovinses, are unable to deliver transgenes to al1 target cells in a tumor (Brenner, 1999). In addition, individual tumon are not equaily permissive to adenoviral infection due to differences in coxsackievirus- adenovirus and integrin a, receptor expression (Pearson et al., 1999). In this study, heterogeneous infection eficiencies were observed in BE and HT29 cells. BE cells were 2-3 fold more receptive to Ad5 infection than HTî9 cells (Figure 4.5). 4-5 logs of ce11 killing by MMC of BE cells infected with A~S.DTDM" at a MOI of 100 (Figure 4.1 1 ) is consistent with a transduction efficiency of 90-95% (this work, Figure 4.5) or 100% (Dr. Henry Klamut, University of Toronto, personal communication) at this MOI. However, the degree of ce11 killing may be enhanced by a minor bystander effect. Such a bystander effect. calculated as a 1.3-fold reduction in MMC concentration required to achieve equivalent cell kill in mixed cultures consisting of equal numbers of BE and HT29 cells compared to pure cultures of BE cells, has been observed (Dr. Hugo Zhang, Ontario Cancer Institute, personal communication). This result illustrates the advantage of clonogenic assays to detect high levels of ce11 Ming compared to other less sensitive methods used to assess ce11 viability, such as the use of 3-[4,S-dimethylthiazol-2-YI]-2,s- diphenyltetrazolium bromide (MTT, Bridgewater et al., 1997). Infection of HT29 cells with AdCMV.Dga1 at a MOI of 100 resulted in only 15% (this work) to 40% (Dr. Henry Klamut, University of Toronto, personal communication) of the cells testing positive for Dgal expression (Figure 4.5). Furthemore, even though a high level infection of BE cells can be achieved in vitro, it is likely that a smaller fraction will be infected in vivo where a ratio of 100 infectious viral particles per ce11 will be difficult to attain since the virus will not be equally accessible to al1 cells. This limitation was seen by Lax (2000), where inira-tumoral injection of AdCMV.Dga1 at a MOI of 100 in nasopharyngeal carcinoma xenografts resulted in transduction of only 25% of the turnor mass. The problem of heterogeneous infection of tumors by Ad5 vectors that deliver prodrug-activating enzymes may be partially offset by the use of drugs that can mediate a bystander effect (Vile et al., 2000), i.e. killing of cells exposed to activated prodmgs originating fiom cells able to carry out the activation. Gancyclovir is widely used for gene therapy applications with its activating enzyme, herpes simplex virus thymidine kinase (as discussed in Chapter 1, Section 1.4.5). Since its activated metabolite, gancyclovir-triphosphate, is transferred by gap junctional intercellular communication (GJIC, Bi et al., 1993), another level of heterogeneity cornes into play involving variation amongst tumor cells in their ability to participate in GJIC (Princen et al., !999). Another combination that is widely used in gene-directed enzyme-prodrug therapy is the E. coli cytosine deaminase (CD) gene and the nucleoside analogue 5-fluoroc ytosine (5-FC, Lawrence et al., 1998). Unlike gancykovir-triphosphate, the activated metabolite of 5- FC, 5-fluorouracil(5-FU), does not rely on GJIC for its bystander effect. Although MMC may mediate a minor bystander effect, it is well documented that CB1954 displays strong bystander effects as high as 1000-fold (Searle et al., 1998) calculated as fold reduction of CB1954 concentration resulting in 50% ce11 kill in mixed cultures consisting of equal numbea of resistant and sensitive cells compared to pure cultures of resistant cells). As discussed in Chapter 1, Section 1.4.5, the Chydroxylamine derivative of CB 1954 (Figue 1 -7)can rnediate a bystander effect via aqueous solution and ce11 uptake by passive diffusion. CB1954 provides an advantage over 5-FU in that its cytotoxicity is not ce11 proliferation-dependant and therefore may be capable of eliminating non-dividing tumor cells (Nishihara et al., 1998). As discussed in Chapter 1, Section 1.4.5, although the bacterial nitroreductase is even more catalytically efficient towards CB 1954 activation than the Q 104Y isoform of DTD (90-fold more active than wild-type DTD, Anlezark et al., 1992), the Q104Y enzyme could provide an advantage over non-mammalian enzymes since it rnay be relatively non-immunogenic. It has been shown that breast cancer lines that stably express DTD Q104Y cDNA are more sensitive to CB 1954 than those expressing wild- type DTD (Wu and Chen, 1999). Another attractive candidate for viral directed enzyme-prodrug therapy is human NQ02 mAD(P)H:(quinone acceptor) oxidoreductase 21, since it is a dihydronicotinamide riboside (NRH) - dependent oxidoreductase that is 3000 times more effective than wild- type DTD in activating CB 1954 (Wu et al., 1997). However, although this enzyme rnay be non-immunogenic, its use for gene therapy requires NRH to be supplied exogenously since there may be insufficient amounts of NADH oxidation required to generate NRH in the host ce11 (Friedlos and Knox, 1992). 'Virtual cofactors' have been developed that have advantageous properties compared to naturally-occurring prodnig-activating cofactors. For example, nicotinic acid riboside reduced, a synthetic derivative of NADPH. can be supplied with bacterial NTR for enzyme-prodrug therapy since this virtual cofactor has an enhanced stability towards metabolic breakdown by serum plasma enzymes compared to NADPH (Knox et al., 1995). Therefore, the requirement for activation by a endogenous cofactor that is present in insufficient quantities may not be a limitation for the use of bioreductive enzymes in cancer gene therapy. An alternative to the utilization of enzymes which have superior hgactivating capability than DTD is the use of dmgs that can be efficientiy metabolized by this enzyme. Analogues of compounds which themselves are considered better DTD substrates than MMC are in development. For example, Ross and CO-workers have synthesized an analogue of the aziridinylbenzoquinone, methyldiaziridinequinone (MeDZQ, Figure 5.1, in itself an analogue of MMC that can be reduced at a 2-3 fold greater rate by DTD) that is reduced by DTD at a 3-fold greater rate than MeDZQ (Winski et al., 1998). The compound, RH1 (Figure SA), is more water-soluble than MeDZQ, thereby overcoming formulation problems encountered with MeDZQ, and is being considered for clinical trials. In addition, Wilson, Demy, Siim, and colleagues have been developing new bioreductively activated drugs that are efficiently reduced by bacterial nitroreductases (Siirn et al., 1997; Demy and Wilson, 19%). Figure 5.1: Structures of MeDZQ and its analogue, RHl. Adapted fiom (Winski et al., 1998). 5.2 Future Directions A better understanding of the nature of the DTD 609C to T mutation at a molecular level will involve detailed structural cornparisons of the wild-type and mutant isoforms. Also, with the limitation of curent viral gene therapy vectors to infect every ce11 in a tumor. it is necessary to explore other DTD isoform - prodrug combinations that are more efficient in dnig activation and in mediating a bystander effect than the wild- type DTD - MMC system. 5.2.1 Structure and Properties of the Serine 187 Isoform X-ray diffraction studies with human wild-type DTD were used to demonstrate how conformational differences between rat and human DTD accounted for their di fferent binding affinities and catalytic activities towards CB 1954 (Faig et al., 2000). The ability to express and purie the wild-type and mutant isofonns from mammalian cells (Siegel et al., 2000) provides an opportunity to use x-ray difhction to compare their structures when complexed with quinone substrates (Li et al., 1995). Since the location of amino acid 187 does not contribute to a fiuictional domain, x-ray difiaction analysis may reveal possible conformational differences between these DTD isoforms to explain their different activities and stabilities. Also, direct amino acid sequencing of the 0.5 - 1 kDa truncated mutant DTD or cornparison of fragmentation patterns obtained by mass spectrometry with wild-type DTD may dlow determination of the region that is susceptible to proteolytic or chernical degradation in mutant DTD. Byron et al. (1997) have used analytical ultracenûifigation to determine various States of oligomerization of DTD based on its sedimentation equilibrium observed at 275 nrn. This method can be applied to detect possible dimerization differences between the wild-type and mutant isofoms since dimer association can be disrupted by potassium isothiocy ante and monomer reassociation constants (kJ can be determined through equilibrium sedimentation after dilution of the destabilizing agent. Due to the similarity of their molecular weights, this method cannot be used to study the possible assembly of wild-type and mutant DTD monomers into heterodimers. However, conformational stabilities of dimers can also be measured by heat denaturation or in denaturing solvents such as urea and guanidine hydrochloride (Neet and Timm, 1994). This ailows determination of the temperature at which a monomer - dimer transition state occurs (Tm) as well as calculation of the dimer fkee energy of association. Isolation of heterodimers for structural detemination or evaluation of conformation kinetics can involve CO-expressionof wild-type and mutant DTD cDNAs; each containing separate labels, such as histidine (Wu et al.. 1998) and FLAG tags (C hubet and Brizzard. 1996). In this case, purification and separation of heterodimers fiom homodimers would involve use of afinity columns containing nickel nitrilotnacetic acid and antibodies for the FLAG epitope, respectively. Histidine tagging of both wild- type (Cui et al., 1995) and mutant (Wu et al., 1998) DTD has ken previously shown not to affect enzyme activity. Therefore, such studies can provide direct information conceming the structural consequences of the proline to senne 187 mutation. 5.2.2 Evaluation of the Bystander Effect of CBl954 In Vitro and In Vivo A nurnber of in vilro assays have been developed to estimate dmg bystander effects in vivo. The HT29 / BE human colon carcinoma pair is usefûl for these studies since they both form isolated colonies with morphologies that can be easily distinguished from each other. Preliminary experiments involving acute exposure of MMC to co- cultures of HT29 and BE cells have been used to detect minor bystander effects (around 1 -3-fold) associated with this dmg (data not shown). This may be due to activation of MMC in DTD-rich HT29 cells which then affects DTD-deficient BE cells. Co-cultures of CB 1954-resistant NIH 3T3 mouse embryo fibroblasts with cells that were rendered sensitive via retroviral mediated transduction with a NTR minigene have been used to demonstrate bystander effects (Bridgewater et al., 1997). This approach can be applied for mixed cultures of CB1954-resistant BE cells with cells transduced with DTD isoform minigenes via recombinant Ad5 delivery. A frequently used method to assess bystander effects of active metabolites that are water soluble is to transfer growth medium firom cultures of cells that can activate the drug to resistant cells (Bndgewater et al., 1997). This method has shown that the diffusable activated metabolite of CE3 1954 (Figure 1.7) has high stability, since exposure to the conditioned medium demonstrated cytotoxicity for at least 48 hours. Another candidate dnig for assessrnent of bystander effects using these approaches is RH1. Initial work has shown that RH1 is stably reduced by DTD since activation did not result in detectable oxygen consumption associated with auto-oxidation and inactivation (Dr. David Ross, personal communication, University of Colorado Health Sciences Center). Therefore, the studies discussed above cm be used to compare the relative therapeutic effects of various DTD isoform - prodrug combinations in ternis of both in vitro clonogenic survival and the production of active difisable metabolite. Of particular interest is the NQO, - CB1954 combination due to the high activation efficiency of NQO, towards this dmg as well as the potential advantage of using a relatively non-imrnunogenic hurnan procimg-activating enzyme for gene therapy (Brenner, 1999). Tumors denved fiom cells stably transfected with minigenes for prodnig- activating enzymes can provide an in vivo assessrnent of the relative efficacy of various enzyme - prodrug combinations. This approach has been previously used to establish a line of BE cells that were stably transfected with wild-type DTD (Winski er al., 1998) to screen for drugs that are DTD substrates. BE cells are normally resistant to these dmgs due to the variant DTD allele, ailowing this method to be used to test the efficacy of other DTD isoforms since they have low to undetectable background DTD activity. HT29 and BE cells have aiready ken show to form tumor xenografts in severe combined immunodeficiency mice (Curnmings et al., 1998) and they have been used as an in vivo screen for chemotherapeutic effects of dmgs that rely on DTD activation. The requirement of a dnig that mediates a bystander effect for adenovirus-directed enzyme-prodrug therapy is apparent when considering that adenoviruses penetrate tumors poorly. The effective depth of penetration of recombinant Ad5 is around 1 to 10 ce11 layers from the tumor surface (Grace et al., 1999). Poor penetration of adenoviral vectors also limits their use for targeting hypoxic tumor cells in the intenor regions of tumors. A further possible limitation for the application of therapeutic Ad5 vectors to target hypoxic tumor cells is poor transgene expression of Ad5 infected cells under hypoxia (see Appendix 5.1). This could also be due to inadequate infection efficiency of Ad5 under hypoxic conditions. Poor accessibility of Ad5 vectors also limits the use of hypoxia specific transcriptional enhancers to target tumor cells by selective transgene expression in hypoxic turnor cells. Therefore, activation of prodmgs in transduced aerobic cells in a tumor might be used to treat hypoxic regions via a bystander effect. 5.3 Summary The stuclies described in this thesis have shown that the DTD 609C to T base change is a mutation that impairs DTD activity and others have directly shown that it also results in decreased protein stability. Correlative studies of the allelic statu of the base change have shown that there is overlap in DTD activities between cells that are homozygous for the wild-type allele and heterozygotes. Therefore, from a therapeutic standpoint, quantitative assessrnent of individual DTD protein levels by immunoblot analysis provides for a more accurate method of predicting cellular responses that depend on DTD activity than fjïnfl-RFLP genotypic analysis. This work also demonstrates the feasibility of the use of Ad5 vectors to deliver wild-type DTD by rendering dmg-resistant human tumor cells hg-sensitive in vitro. It has also shed light ont0 the prospect for the use of DTD in an Ad5-directed gene therapy approach by addressing the application of DTD isofoms that are more catalytically efficient than wild-type human DTD and dmgs that are better substrates for this enzyme than MMC. This becomes important when considering differences in infection eficiency of Ad5 between hurnan tumor cells, such as HT29 and BE cells. Drugs that exert a bystander effect may offset the inability of current viral gene therapy vectors to infect or express transgenes in al1 cells in a tumor, in particular, radiation-resistant hypoxic tumor cells. 5.4 References Anlezark GM. Melon RG, Sherwwd RF, Coles B, Friedlos F and Knox RJ. The bioactivation of 5-(aziridin- 1 -YI)-2,4dinitrobenzamide (CB 1954)-1. Purification and properties of a nitroreductase enzyme fiom Escherichia coli-a potential enzyme for antibody-directed enzyme prodrug therapy (ADEPT). Biochem Pharmacol 44: 2289- 2295, 1992. Belcourt MF, Hodnick WF, Rockwell S and Sartorelli AC. Bioactivation of mitomycin antibiotics by aerobic and hypoxic Chinese hamster ovary cells overexpressing DT- diaphorase. Biochem Phamcol 51 : 1669-1678, 1996. Bi WL, Parysek LM, Waniick R and Starnbrook P.J. In vitro evidence that metabolic cooperation is responsible for the bystander efZect observed with HSV tk retroviral gene therapy. Hum Gene Ther 4: 725-731, 1993. Bouchardy C, Mitrunen K, Dayer P and Hirvonen A. NAD(P)H: quinone oxidoreductase 1 (NQO,) genetic polymorphism and lung cancer risk in caucasian smokers. Proc Amer Assoc Cancer Res 40: 248, 1999. Brenner MK. "The end of the beginning": molecular and cellular biology of gene therapy keystone. 14-20 Januaryl999. Biochim Biophys Acta 1424: RS-R9,1999. Bridgewater JA, Knox RI, Pitts JD, Collins MK and Springer CJ. The bystander effect of the nitroreductase/CB 1954 enzyme/prodnig system is due to a cell-penneable metabolite. Hum Gene Ther 8: 709-717, 1997. Byron O, Mistry P, Suter D and Skelly J. DT diaphorase exists as a dimer-tetramer equilibrium in solution. Eur Biophys J25: 423-430, 1997. Chen H, Lum A, Seifned A, Wilken LU and Le Marchand L. Association of the NAD(P)H:quinone oxidoreductase 609C to T polymorphism with a decreased lung cancer risk. Cancer Res 59: 3045-3048,1 999. Chubet RG and Brizzard BL. Vectors for expression and secretion of FLAG epitope- tagged proteins in mammalian cells. Biotechniques 20: 136- 14 1, 1996. Cui K, Lu AY, and Yang CS. Subunit functionai studies of NAD(P)H:quinone oxidoreductase with a heterodimer approach. Proc Natl Acod Sci USA 92: 1043-1047, 1995. Curnrnings J, Spanswick VJ, Gardiner J, Ritchie A and Smyth JF. Phannacological and biochemical determinants of the antitumour activity of the indoloquinone E09. Biochem Pharmacol 55: 253-260,1998. Denny WA and Wilson WR. The design of selectively-activated anti-cancer prodmgs for use in antibody-directed and gene-directed enzyme-prodrug therapies. J Pharm Pharmacol 50: 387-394. 1998. Edwards YH, Potter J and Hopicinson DA. Hurnan FAD-dependent NAD(P)H diaphorase. Biochem J 187: 429436,1980. Eickelmann P, Schulz WA, Rohde D, Schmitz-Drager B and Sies H. Loss of heterozygosity at the NAD(P)H:quinme oxidoreductase locus associated with increased resistance against mitornycin C in a human bladder carcinoma ce11 line. Bol Chern Hoppe Seyler 375: 439-445, 1994. Faig M, Bianchet MA, Talalay P, Chen S, Winski S, Ross D and Amzel LM. Structures of recombinant human and mouse NAD(P)H:quinone oxidoreductases: species cornparison and structural changes with substrate binding and release. Proc Natl Acad Sci USA 97: 3 177-3 182,2000. Freernan SM, Abboud CN, Whartenby KA, Packman CH, Koeplin DS, Moolten FL and Abraham GN. The "bystander effect": tumor regression when a fiaction of the tumor rnass is genetically modified. Cancer Res 53: 5274-5283, 1993. Friedlos F and Knox RJ. Metabolism of NAD(P)H by blood components. Relevance to bioreductivel y activated prodmgs in a targeted enzyme therapy s ystem. Biochem Pharmacol 44: 63 1-635, 1992. Gaedigk A, Tyndale RF, Jurima-Romet M, Sellers EM, Grant DM and Leeder JS. NAD(P)H:quinone oxidoreductase: polymorphisms and allele frequencies in Caucasian' Chinese and Canadian Native lndian and Inuit populations. Pharrnacogenetics 8: 305- 313,1998. Grace MJ, Xie L, Musco ML, Cui S, Gurnani M, DiGiacomo R, Chang A, lndelicato S, Syed J. Johnson R and Nielsen LL. The use of laser scanning cytometry to assess depth of penetration of adenovirus p53 gene therapy in human xenografl biopsies. Am J Parhol 155: 1 869- 1 878, 1999. Li R, Bianchet MA, Talalay P and Amzel LM. The three-dimensionai structure of NAD(P)H:quinone reductase, a flavoprotein involved in cancer chernoprotection and chemotherapy: Mechanism of the two-electron reduction. Proc Natl Acad Sei USA 92: 8846-8850, 1995. Kelsey KT. Ross D, Traver RD, Christiani DC, Zuo ZF, Spitz MR, Wang M, Xu X, Lee BK, Schwartz BS and Wiencke K. Ethnic variation in the prevalence of a cornmon NAD(P)H quinone oxidoreductase polymorphism and its implications for anti-cancer chemotherapy. Br J Cancer 76: 852-854, 1997. Knox RJ, Friedios F, Jarman M, Davies LC, Goddard P, AnIezark GM, Melton RG and Sherwood RF. Virtual cofactors for an Escherichia coIi nitroreductase enzyme: relevance to reductively activated prodrugs in antibody directed enzyme prodrug therapy (ADEPT). Biochem Pharmacol 49: 164 1- 1647,1995. Kuehl BL, Paterson JW, Peacock JW, Paterson MC and Rauth AM. Presence of a heterozygous substitution and its relationship to DTD activity. Br J Cancer 72: 555-561, 1995. Lawrence TS, Rehemtulla A, Ng EY, Wilson M, Trosko JE and Stetson PL. Preferentiai cytotoxicity of cells transduced with cytosine deaminase compared to bystander cells after treatrnent with 5-flucytosine. Cancer Res 58: 2588-2593,1998. Lax SA. Adenovhs-pS3 gene therapy in nasopharyngeal carcinoma xenografts. MsC thesis. University of Toronto. Dept Medical Biophysics. 2000. Marin A, Lopez de Cerain A, Hamilton E, Lewis AD, Martinez-Penuela JM, Idoate MA and Bello J. DT-diaphorase and cytochrome B5 reductase in hurnan lung and breast tumours. Br J Cancer 76: 923-929,1997. Marshall RS, Paterson MC and Rauth AM. Deficient activation by a human ce11 strain leads to mitomycin resistance under aerobic but not hypoxic conditions. Br J Cancer 59: 34 1-346, 1989. Marshall RS, Paterson MC and Rauth AM. Studies on the mechanism of resistance to mitomycin C and porfiromycin in a human ce11 strain derived fiom a cancer-prone individual. Biochem Pharmacol 4 1: 1 35 1 - 1 3 60, 199 1. Misra V, Klamut HJ and Rauth AM. Transfection of COS- 1 cells with DTD cDNA: Role of a base change at position 609. Br J Cancer 77: 1236-40, 1998- Mora. JL, Siegel D and Ross D. A potential mechanism underlying the increased susceptibility of individuals with a polymorphism in NAD(P)H:quinone oxidoreductase 1 (NQO1 ) to benzene toxicity. Proc Nafl Acad Sci USA 96: 8 1 50-81 55, 1 999. Nebert DW, Ingelman-Sundberg M and Daly AK. Genetic epidemiology of environmental toxicity and cancer susceptibility: Human allelic polymorphisms in dnig- metabolizing enzyme genes, their hctional importance, and nomenclature issues. hg Metab Rev 31: 467-487,1999. Neet KE and Tirnm DE. Conformational stability of dimeric proteins: quantitative studies by equilibrium denaturation. Protein Sci 3 : 2 167-21 74, 1 994. Nishihara E, Nagayama Y, Narimatsu M, Namba H, Watanabe M, Niwa M andyamashita S. Treatment of thyroid carcinoma cells with four different suicide gene/prodmg combinations in vitro. Anficancer Res 18: 1 52 1 - 1525, 1998. Pearson AS, Koch PE, Atkinson N, Xiong M, Finberg RW, Roth JA and Fang B. Factors limiting adenovirus-mediated gene transfer into human lung and pancreatic cancer ceil lines. Clin Cancer Res 5: 4208-42 1 3, 1 999. Princen F, Robe P, Lechanteur C, Mesnil M, Rigo JM, Gielen J, Merville MP and Bours V. A ce11 type-specific and gap junction-independent mechanism for the herpes simplex virus- 1 thymidine kinase gene/ganciclovir-mediated bystander effect. Clin Cancer Res 5: 3639-3644, 1999. Radjendirane V, Joseph P, Lee YH, Kimura S, Klein-Szanto AJ, Gonzalez FJ and Jaiswd PX. Disruption of the DT diaphorase (NQO,) gene in mice leads to increased menadione toxicity. J Biol Chem 273: 7382-7389, 1998. Rockwell S. Use of hypoxia-directed dmgs in the therapy of solid tumors. Semin Oncol 19 (Suppl11): 29-40, 1992. Ross D. Siegel D, Beall H, Prakash AS, Mulcahy RT and Gibson NW. DT-diaphorase in activation and detoxification of quinones. Bioreductive activation of mitomycin C. Cancer Met Rev 12: 83-101, 1993. Rosvold EA, McGly~KA, Lustbader ED and Buetow KH. Identification of an NAD(P)H:quinone oxidoreductase polymorphism and its association with lung cancer and smoking. Pharmacogenetics 5: 199-206, 1995. Rothrnan N, Traver RD, Smith MT, Hayes RB, Li G-L, Campleman S, Dosemeci M, Zhang L, Linet M, Wacholder S, Yin S-N and Ross D. Lack of NAD(P)H:quinone oxidoreductase activity (NQO1) is associated with increased risk of benzene hematotoxicity. Proc Amer Assoc Cancer Res 37: 258, 1996. Sarlauskas J, Dickancaite E, Nemeikaite A, Anusevicius 2, Nivinskas H, Segura-Aguilar J and Cenas N. Nitrobenzimidazoles as substrates for DT-diaphorase and redox cycling compounds: their enzymatic reactions and cytotoxicity. Arch Biochem Biophys 346: 2 19- 229, 1997. Schulz WA, Knunmeck A, Rosinger 1, Eickelmann P, Neuhaus C, Ebert T, Schmitz- Drager BJ and Sies H. Increased fiequency of a variant-allele for NAD(P)H: quinone oxidoreductase in patients with urological malignancies. Pharmacogene~ics7: 235-239, 1997. Searle PF, Weedon SJ, McNeish IA, Gilligan MG, Ford MI, Frîedlos F, Springer CJ, Young LS and Kerr DJ. Sensitisation of human ovarian cancer cells to killing by the prodrug CB1954 following retroviral or adenoviral transfer of the E. coli nitroreductase gene. Adv Exp Med Biol451: 107-1 13, 1998. Siegel D, McGuinness SM, Winski SL and Ross D. Genotype-phenotype relationships in studies of a polymorphism in NAD(P)H:quinone oxidoreductase 1. Pharmacogenetics 9: 113-121, 1999. Siegel D, Anwar A, Winski S, Kepa JK and Ross D. The NQO1.2 polymorphism in NAD(P)H: quinone oxidoreductase 1 targets the protein for proteosomal degradation. Proc Amer Assoc Cancer Res 41: 35,2000. Siim BG, Denny WA and Wilson WR. Nitro reduction as an electronic switch for bioreductive drug activation. Oncol Res 9: 3 57-369, 1997. Steiner M, Hillenbrand M, Borkowski M, Seiter H and Schuff-Werner P. 609 C + T polymorphisrn in NAD(P)H:quinone oxidoreductase gene in patients with prostatic adenocarcinorna or benign prostatic hyperplasia. Cancer Letters 135: 67-71, 1 999 Stratford 1J and Workrnan P. Bioreductive drugs into the next millennium. Anticancer Drug Des 13: 5 19-528, 1998. Traver RD, Horikoshi T, Danenberg KD, Stadlbauer TH, Danenberg PV, Ross D and Gibson NW. NAD(P)H:quinone oxidoreductase gene expression in hurnan colon carcinoma cells: characterization of a mutation which modulates DTD activity and mitomycin sensitivity. Cancer Res 52: 797-802, 1992. Traver RD, Phillips RM, Gibson NW and Ross D. A point mutation in both human and lung carcinoma ce11 lines leading to a loss of DT-diaphorase activity. froc Am Assoc Cancer Res 36: 525, 1995. Traver RD, Siegel D, Beall HD, Phillips RM, Gibson NW, Franklin WA and Ross D. Characterization of a polymorphism in NAD(P)H: quinone oxidoreductase (DT- diaphorase). Br J Cancer 75: 69-75, 1 997. Vile RG, Russell SJ and Lemoine NR. Cancer gene therapy: hard lessons and new courses. Gene Ther 7: 2-8,2000. Wiemels JL, Pagnamenta A, Taylor GM, Eden OB, Alexander FE and Greaves MF. A lack of a fùnctional NAD(P)H:quinone oxidoreductase allele is selectively associated with pediatric leukemias that have MLL fusions. Cancer Res 59: 4095-4099, 1999. Wiencke JK, Spin MR, McMillan A and Kelsey KT. Lung cancer Mexican-American and Afncan-Amencan is associated with the wild-type genotype of the NAD(P): quinone oxidoreductase polymorphisrn. Cancer Epidemiology Biomarkers and Prevention 6: 87- 92, 1997. Winski SL, Hargreaves HJ, Butler J and Ross D. A new screening system for NAD(P)H:quinone oxidoreductase (NQO 1)-directed antitumor quinones: Identification of a new azindiny lbenzoquinone, RH 1, as a NQO 1 directed antitumor agent. Clin Cancer Res 4: 3083-3088, 1998. Workman P and Straaord II. The experimental development of bioreductive drugs and their role in cancer therapy. Cancer Metastasis Rev 12: 73-82, 1993. Wu K, Knox R, Sun XZ, Joseph P, laiswal AK, Zhang D, Deng PS and Chen S. Catalytic properties of NAD(P)H:quinone oxidoreductase-2 (NQ02), a dihydronicotinarnide nboside dependent oxidoreductase. Arch Biochem Biophys 347: 22 1-228, 1997. Wu K, Deng PS, and Chen S. Catalytic properties of a naturaily occurring mutant of human NAD(P)H:quinone acceptor oxidoreductase @T-diaphorase), pro- 187 to ser. Pathophysiology of Lipid Peroxides and Related Free Radicak pp. 1 35- 148, Japan Sci Soc Press, Tokyo, Japan, 1998. Wu K and Chen S. Direct demonstration of the activation and cytotoxic effect of CB 1954 using DT-diaphorase (NQO 1 ) transfected MDA-MB-23 1 breast cancer ce11 Iines. Proc Amer Assoc Cancer Res 40: 3 84, 1999. APPENDCY 5.2 Transduction Efficieacy of Ad5 in Hypoxic versus Aerobic Human Tumor Cells AS. 1.1 Introduction The application of recombinant adenoviral type 5 vectors (Ad5) to deliver prodnig- activating enzymes to radiation-resistant hypoxic tumor cells relies on infection of a sufficient fraction of these cells. It also relies on sufficient amounts of gene expression in transduced hypoxic cells to sensitize them to their prodrug substrates. The possibility that hypoxic conditions modulate the infection efficiency of Ad5 was addressed by testing the relative transduction efficiency of Ad5 in aerobic versus hypoxic turnor cells. AS. 1.2 Materials and Methods 2 x 1o6 BE hurnan colon carcinoma and MCF-7 human breast carcinoma ce11 lines were seeded overnight in 8 oz glass prescription bottles (Fisher Scientific, Nepean, ON, Canada) with 20 mL of growth medium consisting of a-minimal essential medium containing 10% fetal calf serum (Cansera. Rexdale ON, Canada). Afier a 3 hour exposure to gas containing either 5% CO2; 95% air or 5% CO, ; balance N, at 37 "C, medium was removed using 25 mL glass syringes (Fisher Scientific) to maintain gassing statu followed by a 30 minute exposure to an Ad5 vector carrying a firefly Iuciferase minigene (suppIied by Dr. Henry Klamut, University of Toronto) under the control of the cytomegalovirus (CMV) promoter at a multiplicity of infection (MOI) of 100 in 1 mL of PBS @re-equilibrated with 5% CO, ; balance N, for hypoxic exposure). Previously removed conditioned medium, left in syringe, was repIaced followed by a Mer24 hour incubation under hypoxic or aerobic conditions, at which time cells were harvested for luciferase activity detemination using an assay kit (Promega, Madison, WI, USA). To ensure linearity for the luciferase assay, titrations of ce11 lysate volumes were performed. AS. 1.3 Results As shown in Table AS. 1.1, aerobic BE and MCF-7 cells displayed 3-fold and 25-fold higher luciferase activities, respectively, relative to hypoxic conditions. Table A5. 1.1 : Relative transduction efficiency of Ad5 .CMV.luciferase under aerobic vs. hypoxic conditions. Ce11 Type Mean * s.d. I 1 107 Relative Light Units / L r BE aerobic I1 21 1: 1 1 BE hypoxic I8*1 MCF-7 aerobic 98*21 MCF-7 hypoxic 4* 1 L - 1 1 * based on three independent experiments, one bottle per treatment :background luciferase activity did not exceed 1/IO00 of each value AS. 1.4 Conclusions and Discussion Lower luciferase activities in hypoxic versus aerobic conditions in BE and MCF-7 cells upon Ad5.CMV.luciferase infection suggests that Ad5 may infect permissive cells less eficiently under hypoxia. It is also possible that hypoxic and aerobic cells do not differ in their Ad5 uptake, but the former may provide for lower levels of transgene expression. BE and MCF-7 cells are equally permissive to Ad5 infection under aerobic conditions (Dr. Henry Klamut, University of Toronto, personal communication). The possibility of decreased Ad5 receptor expression in these cells under hypoxic conditions can be addressed by Western blot analysis of primary and CO-receptor levels after hypoxic exposure. Decreased viral uptake anaor transgene expression must be considered when targeting hypoxic tumor cells in a AdS-directed enzy me-prodrug therapy approach including possible use of hypoxia-specific tranxriptional enhancers