UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Enhancing drug sensitivity in melanoma cells through pro-apoptotic pathways that subvert chemoresistance… Bush, Jason Allan 2002

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

Item Metadata

Download

Media
831-ubc_2002-750019.pdf [ 15.12MB ]
Metadata
JSON: 831-1.0090502.json
JSON-LD: 831-1.0090502-ld.json
RDF/XML (Pretty): 831-1.0090502-rdf.xml
RDF/JSON: 831-1.0090502-rdf.json
Turtle: 831-1.0090502-turtle.txt
N-Triples: 831-1.0090502-rdf-ntriples.txt
Original Record: 831-1.0090502-source.json
Full Text
831-1.0090502-fulltext.txt
Citation
831-1.0090502.ris

Full Text

ENHANCING DRUG SENSITIVITY IN MELANOMA CELLS THROUGH PRO-APOPTOTIC PATHWAYS THAT SUBVERT CHEMORESISTANCE MECHANISMS by J A S O N A L L A N B U S H B . S c , The University of British Columbia, 1992 M.Sc . , The University of British Columbia, 1997 A thesis submitted in partial fulfilment o f the requirements for the degree of D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Experimental Medicine Program, Department of Medicine) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A September 2002 © Jason Al l an Bush, 2002 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, 1 agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ( U - f ^ \ C ( frj The University of British Columbia Vancouver, Canada DE-6 (2788) ABSTRACT The general ineffectiveness of current chemotherapeutic agents for malignant melanoma warrants the investigation into biological approaches that increase sensitivity to the cytotoxic effects of anticancer drugs. The majority of anticancer agents induce cancer cell death by means o f apoptosis. A n escape by cancer cells from apoptosis seems to be closely associated with the development of anticancer drug resistance; therefore, determining what mechanisms melanoma cells utilize for survival is our objective. Using three distinct strategies, we demonstrate the importance of apoptotic pathways for melanoma chemoresistance after drug treatment. In the first approach, we found that the mutational status of p53 can dictate several key phenotypes of a cell population. A dominant negative p53 stimulated faster growth in melanoma cells compared to the parental cell line and after treatment with the DNA-damaging drug, camptothecin, cell survival was significantly greater in the mutant /?53-bearing cell line attributable to a general decrease in apoptosis. We established that chemosensitivity of melanoma cells with normal p53 can occur through selective suppression of anti-apoptotic proteins like Bcl -2 and of drug transporters like P-glycoprotein while dysfunction of p53 inhibits the suppression leading to up-regulation of these proteins and increasing chemoresistance. We further evaluated the relationship between p53 and P-glycoprotein in vivo using a mouse model to confirm our in vitro results that dysfunction of p53 causes increased expression of P-glycoprotein. Since dysfunction of p53 contributes to chemoresistance, overcoming the p53 block through the overexpression of other pro-apoptotic genes was explored and validated in a second strategy. Our previous data has suggested that the mitochondria-mediated apoptosis pathway plays a critical role in drug-induced apoptosis for melanoma. Using ectopic overexpression of upstream pro-apoptotic genes such as Bax and the newly identified P U M A and Noxa that are involved in pore formation within the outer membrane of mitochondria, we could reduce the threshold of melanoma resistance to i i drug treatment. B y several distinct methods, we determined that an increase in apoptosis was the mechanism for increased sensitivity to drug treatment. A further analysis of genes acting downstream of the mitochondria, such as Apaf-1 and caspase-3, that play a more terminal role in executing apoptogenic signals from the mitochondria also proved useful for the sensitization of melanoma cells to anticancer agents. We observed a p53-dependent activation of some candidate genes as they were more effective in a cell line with wild-type p53, function than a mutant p53 cell line. In the third approach, we focussed on the antitumour properties of the powerful antioxidant, curcumin. Our endeavour to find alternative compounds that may improve today's therapy regimes for melanoma yielded useful biochemical data for this compound. Contrary to other reports in some tumour cell lines demonstrating a p53-dependence for curcumin action, we showed that curcumin-treated melanoma cells undergo apoptosis independently of the p53 pathway. Our results also suggest that curcumin-induced apoptosis is not modulated through the mitochondria but rather the Fas death receptor-mediated pathway is responsible for curcumin activity on melanoma cells since inhibition of Fas activation could increase cell survival after curcumin treatment. In this thesis, we have employed specific approaches targeting particular molecular abnormalities for the induction of apoptosis by up-regulating apoptotic signals. Although based on different targets and delivery methods, our strategies have the common goal of eliminating melanoma cells by restoration of the apoptotic function. Our hope is that the mechanistic understanding of cell death w i l l have profound impacts upon the practice of medical oncology and outlook for many patients. in TABLE OF CONTENTS Abstract Table of Contents List o f Tables List o f Figures Abbreviations Acknowledgements C H A P T E R 1 G E N E R A L I N T R O D U C T I O N 1.1 Melanoma 1.1.1 The biology of melanocytes 1.1.2 Melanoma etiology 1.1.3 Malignant melanoma 1.2 P53, the cellular gatekeeper 1.2.1 Gene and protein structure 1.2.2 Roles in transcriptional regulation 1.2.3 Roles in cell cycle arrest 1.2.4 Roles in D N A repair 1.2.5 Roles in apoptosis 1.2.6 P53 mutation and the implication for cancer 1.3 Drug-resistance of cancer 1.3.1 The M D R phenotype 1.3.2 Decreased intracellular drug concentration 1.3.3 Reduced drug activation or increased detoxification 1.3.4 Alterations of drug target via increased damage repair 1.3.5 Abrogation of apoptosis 1.3.5.1 Caspase-family, the effector pathway 1.3.5.2 Apoptosis, the mitochondria pathway 1.3.5.3 Apoptosis, the death receptor pathway 1.4 Antioxidants in cancer 1.4.1 Chemoprevention mechanisms 1.4.2 Chemoprevention in melanoma C H A P T E R 2 M A T E R I A L S A N D M E T H O D S 2.1 Materials 2.1.1 Cel l lines and cell culture 2.1.2 Reagents, enzymes, and chemicals 2.1.3 Plasmids and c D N A s 2.2 Methods 2.2.1 Expression of constructs by transfection iv 2.2.2 Determining transfection efficiency with G F P vectors 33 2.2.3 Cytotoxicity (cell survival) assay 37 2.2.4 Enzyme-linked immunosorbent assay (ELISA) of apoptosis 37 2.2.5 Flow cytometry 38 2.2.6 D N A fragmentation by gel electrophoresis 38 2.2.7 Annexin-V staining 39 2.2.8 Fluorescence microscopy 39 2.2.9 MitoCapture™ mitochondrial transmembrane potential detection 40 2.2.10 Caspase inhibition 40 2.2.11 Low-temperature culture inhibiting receptor aggregation 40 2.2.12 Northern blot analysis 41 2.2.13 Reverse Transcriptase (RT)-PCR 41 2.2.14 Western blot analysis 42 2.2.15 Chloramphenicol acetyltransferase ( C A T ) E L I S A 43 2.2.16 Breeding, genotyping, dissection, and isolation of cells from mice 44 2.2.17 Immunoblotting and immunohistochemistry on tissues 45 2.2.18 Luciferase and P-galactosidase assays 47 C H A P T E R 3 R O L E S OF P53 I N M E L A N O M A C H E M O S E N S I T I V I T Y 48 3.1 Rationale and Hypothesis 48 3.2 Results 50 3.2.1 Mutant p53 can increase the growth rate 50 3.2.2 CPT-induced apoptosis is p53-dependent 52 3.2.3 Antisense Bcl-2 enhances the p53-dependent effect of C P T treatment 58 3.3 Discussion 58 3.3.1 P53 mutation correlates with reduced chemosensitivity 58 3.3.2 CPT-induced cell death is G l -independent 62 3.3.3 P53 regulates expression of Bax/Bcl-2 63 3.3.4 P53 down-regulates P-gp 65 C H A P T E R 4 C H E M O S E N S I T I Z A T I O N OF M E L A N O M A C E L L S B Y 66 P R O - A P O P T O T I C G E N E S 4.1 Rationale and Hypothesis 66 4.2 Results 69 4.2.1 Bax sensitizes melanoma cells to anticancer drug treatment 69 4.2.1.1 Overexpression of Bax sensitizes melanoma cells to drug 69 treatment 4.2.1.2 Inducible Bax decreases N F - K B 69 4.2.1.3 A tyrosinase promoter can specifically target Bax expression 72 in melanoma cells 4.2.2 PUMA sensitizes melanoma cells to anticancer drug treatment 75 4.2.2.1 PUMA enhances the sensitivity of human melanoma cells to 75 C P T treatment 4.2.2.2 PUMA overexpression induces apoptosis not necrosis 75 4.2.3 Noxa sensitizes mouse melanoma cells to anticancer drug treatment 77 v 4.2.4 Downstream mediators of the mitochondrial apoptosis pathway can 80 sensitize melanoma cells to anticancer drug treatment 4.2.4.1 Apaf-1 sensitizes melanoma cells to anticancer drug treatment 80 A.l A.l Caspase-3 sensitizes melanoma cells independently o f p53 83 4.3 Discussion 83 4.3.1 Pro-apoptotic genes can subvert chemoresistance mechanisms in melanoma 83 4.3.2 Bax overexpression is a proven candidate 85 4.3.3 Drug-induced apoptosis is enhanced by overexpression of PUMA and Noxa 86 4.3.4 Drug-induced apoptosis is enhanced by overexpression of Apaf-1 and 88 Procaspase-3 4.3.5 Overcoming the anti-apoptotic block in melanoma improves drug efficacy 89 C H A P T E R 5 T H E U S E O F D I E T A R Y A N T I - O X L D A N T S T O SENSITIZE 91 M E L A N O M A C E L L S T O A P O P T O S I S 5.1 Rationale and Hypothesis 91 5.2 Results 93 5.2.1 Curcumin induces apoptosis in melanoma cells 93 5.2.2 Curcumin-induced apoptosis is p53-independent 96 5.2.3 Curcumin-induced apoptosis can inhibit N F - K B and X I A P 96 5.2.4 Curcumin induces apoptosis by activation of caspases 99 5.2.5 Curcumin induces Fas expression and aggregation 101 5.3 Discussion 104 5.3.1 Curcumin induces p53-independent apoptosis in melanoma 104 5.3.2 Curcumin inhibits the pro-survival pathway 110 5.3.3 Curcumin activates the caspase cascade 111 5.3.4 Curcumin induces membrane-mediated apoptosis 112 C H A P T E R 6 G E N E R A L C O N C L U S I O N S 115 6.1 Summary 115 6.2 Future Directions 117 R E F E R E N C E S 119 A P P E N D I X I R E L A T I O N S H I P B E T W E E N P53 A N D P - G L Y C O P R O T E L N 144 A . l Rationale and Hypothesis 144 A.2 Results 148 A.2.1 Dysfunction of p53 increases expression of P-gp 148 A.2.2 The functional significance of elevated P-gp 150 A.2.3 P-gp expression is increased in specific tissues 153 A.3 Discussion 155 A.3.1 The validity of the in vivo observations versus the in vitro system 155 A.3.2 The role of p53 in trans-repressing other apoptosis genes 158 A.3.3 The absence of p53 increases P-gp isoforms in some tissues 158 A.3.4 The importance of in vivo model systems for human cancers 159 v i LIST OF TABLES Table 1.1 Common p5 3-regulated genes and their cellular roles 11 Table 1.2 A B C transporters involved in multidrug resistance 21 Table 2.1 Relative transfection efficiencies of our commonly used cell lines 35 Table 3.1 Cel l cycle analysis of CPT-treated M M AN-vector and M M A N - p E D l 59 cells by flow cytometry Vll LIST OF FIGURES Figure 1.1 Schematic representation of the p53 protein 9 Figure 1.2 Diagrammatic representation of the most common configuration for full 19 and half A B C transporters Figure 2.1 Fluorescent evaluation of transfection efficiency 34 Figure 2.2 Plasmid map of the pCIneoIRES-GFP vector 36 Figure 2.3 Genotypic analysis ofp53 littermates 46 Figure 3.1 Overexpression of mutant p53 on growth rate of the melanoma cell line 51 M M A N Figure 3.2 Overexpression of mutant p53 on the sensitivity of M M A N cells to C P T 53 treatment Figure 3.3 Pre-treatment of M M A N - v e c t o r and M M A N - p E D l cells with the p53- 54 stabilizing compound, CP-31398 Figure 3.4 Overexpression of mutant p53 on CPT-induced apoptosis and D N A 56 fragmentation Figure 3.5 The effect of mutant p53 on CPT-induced apoptosis by flow cytometry 57 Figure 3.6 The effect of mutant p53 on Bax, Bcl-2 , p 2 1 w a f l , and P-gp protein expression 60 Figure 3.7 Pre-treatment of M M A N - v e c t o r and M M A N - p E D l cells with an antisense 61 Bcl-2 O D N Figure 4.1 Protein expression of the pro-apoptotic factor, Apaf-1, in melanoma cell lines 67 Figure 4.2 Overexpression of Bax sensitizes melanoma cells to C P T treatment 70 Figure 4.3 Bax induction reduces protein expression of the small subunit (p50) 71 of N F - K B Figure 4.4 A tyrosinase promoter/enhancer vector can target expression of Bax 73 specifically to melanoma Figure 4.5 Overexpression of PUMA sensitizes melanoma cells to C P T treatment 76 Figure 4.6 Overexpression of Noxa sensitizes mouse melanoma cells to C P T treatment 78 Figure 4.7 The stable expression of Noxa in B16 mouse melanoma cells induces 79 apoptosis v i i i Figure 4.8 The overexpression of Apaf-1 sensitizes melanoma cells to anticancer drug 81 treatment Figure 4.9 Procaspase-3 overexpression reduces cell survival in melanoma 84 Figure 5.1 Curcumin decreases cell survival in the melanoma cell line M M R U 94 Figure 5.2 Curcumin induces apoptosis in M M R U 95 Figure 5.3 Curcumin-induced apoptosis in melanoma cells is p53-independent 97 Figure 5.4 The effects of curcumin on pro- and anti-apoptotic gene expression 100 Figure 5.5 Curcumin-induced apoptosis in melanoma cell lines is reduced with caspase 102 inhibitors Figure 5.6 Curcumin induces cell death by activating Fas receptor aggregation 105 Figure 5.7 Blockage of Fas aggregation by low-temperature incubation on curcumin- 107 induced apoptosis of M M R U cells Figure 5.8 Northern blot analysis of Fas ligand m R N A levels after curcumin treatment 108 Figure 5.9 Molecular pathways of curcumin action 114 Figure A . l The classic hypothetical model for p53 regulation of the Mdrl promoter 145 Figure A . 2 Increased expression of P-gp in p53 mutant melanoma cell lines 147 Figure A .3 Increased Mdrla/b and P-gp inp53'A cells 149 Figure A .4 Activation of the Mdrlb promoter in the absence of p53 151 Figure A.5 Comparison of drug sensitivity between p53+/+ and p53''~ mFbs 152 Figure A .6 Correlative increase in expression levels of P K C isozymes 154 Figure A .7 Organ-specific expression of P-gp protein and Mdrl a or Mdrlb m R N A 156 ix L I S T O F A B B R E V I A T I O N S Abbreviation Definition 6-4PP 6-4 photoproducts A B C ATP-binding cassette A C T H adrenocorticotropic hormone A F P ct-Fetoprotein a - M S H a-melanocyte stimulating hormone Apaf-1 apoptotic protease activating factor-1 AP-1 activator protein-1 A R E / p E R E anti-oxidant/electrophile responsive element A T F 2 activating transcription factor 2 A T P adenosine triphosphate B C N U l,3-bis(2-chloroethyl)-l -nitrosourea Bcrp breast cancer resistance protein B E R base excision repair b F G F basic fibroblast growth factor B H 3 Bcl-2 homology domain 3 bZIP basic region leucine zipper C A R D caspase recruitment domain C D C 2 cyclin B(2)/p34 Cdk cyclin-dependent kinase cFLJP F L I C E inhibitory protein CIS cisplatin C M M cutaneous malignant melanoma C O X - 2 cyclooxygenase-2 C P D cyclobutane pyrimidine dimers CPP32 caspase-3 C P T camptothecin C S B Cockayne syndrome B DFF45 D N A fragmentation factor 45 DISC death-inducing signaling complex D M S O dimethylsulfoxide D O X doxorubicin D R 5 death receptor 5 D T I C D T I C - D O M E (dacarbazine) E G F R epidermal growth factor receptor E L I S A enzyme-linked immunosorbent assay F A D D Fas-associating death domain FITC fluoroscein isothiocyanate F L I C E F A D D - l i k e interleukin-lb eta-converting enzyme Gadd45 growth arrest and D N A damage45 G G R global genome repair G S H reduced glutathione G S S G oxidized glutathione G S T glutathione 5-transferase H B V hepatitis B virus X h T E R T human telomerase reverse transcriptase TAP inhibitor of apoptosis I C A D inhibitor of caspase-activated deoxyribonuclease TL-2 interleukin-2 I N F a interferon-a K L F 4 kruppel-box transcription factor-4 L R P lung cancer resistance related protein M A G E melanoma-associated antigens M A P 4 microtubule-associated protein 4 M A P K mitogen-activated protein kinase M C K muscle creatine kinase Mdm2 mouse double minute-2 M D R multidrug resistance M E F mouse embryonic fibroblasts mFb mouse fibroblasts M G M T 0 6-methylguanine-DNA methyltransferase M I T F microphthalmia-associated transcription factor M M malignant melanoma M M P matrix metalloproteinases M R P MDR-related protein M V P major vault protein M X R mitoxantrone resistance N E R nucleotide excision repair N L S nuclear localization signal N u M A nuclear mitotic apparatus protein P A G E polyacrylamide gel electrophoresis P A K 2 p21 -activated kinase-2 P A R P poly(ADP-ribose) polymerase P C N A proliferating cell nuclear antigen P C R polymerase chain reaction P-gP P-glycoprotein Pidd p5 3-induced D N A damage PTP permeability transition pore P U M A p53-upregulated modulator of apoptosis P V D F polyvinylidene difluoride R G P radial growth phase R O S reactive oxygen species R P A replication protein A S A G E serial analysis of gene expression S A P K stress-activated protein kinase SDS sodium dodecyl sulphate S M A C second mitochondria derived activator of caspase T C R transcription-coupled repair T G F - a transforming growth factor alpha T N F a tumour necrosis factor alpha T N F R tumour necrosis factor receptor T R A D D TNFR-associated death domain protein T R A F 2 TNFR)-associated factors T R A I L TNF-related apoptosis-inducing ligand XI TRP-1 Tyrosinase related protein 1 TSP-1 thrombospondin-1 U V R ultraviolet radiation V E R verapamil V G P vertical growth phase V I N vincristine VP16 VP-16-213 (etoposide) X P A Xeroderma pigmentosum complementation group A x i i ACKNOWLEDGEMENTS I would like to graciously thank my supervisor, Dr. Gang L i , without whom this work would not be possible. We tried many things...many things didn't work, but because of your philosophy about science, things have worked. Y o u have my gratitude and admiration. I want to thank my friends and colleagues in the lab K-John, Yvonne, Eric and the new 'recruits' for their help, support and friendship, you w i l l be missed. I would like to express my sincerest thanks to my supervisory committee, Drs. Vincent Ho, B i l l Salh, Vince Duronio, and Shoukat Dedhar for their intellectual discussions and personal recommendations. To my family, I give thanks for their constant support and encouragement over the years. Lastly, to my soulmate, Wendy, I can't imagine being here without you. x m CHAPTER 1 . GENERAL INTRODUCTION 1.1 M E L A N O M A 1.1.1 The Biology of Melanocytes Melanocytes are pigment-synthesizing cells found in almost all organisms. In humans, they originate from the neural crest cells of the neuroectoderm primordium. They are found in several locations including the skin, retina of the eye, mucosal epithelia of the inner ear and the leptomeninges. In skin, dendritic cells migrate to the basal layer of the epidermis and differentiate into functional melanocytes. They make dendritic contacts primarily with keratinocytes in a ratio of about thirty-six keratinocytes to one melanocyte (Scott and Haake, 1991); therefore, melanocytes constitute 2-4% of the total epidermal cell population (Szabo et al., 1988; Vancoill ie et al, 1999). Melanocytes produce melanins, light absorbing pigments, formed in specialized organelles called melanosomes that appear to resemble lysosomal secretory vesicles. Several factors appear to regulate melanocytes and skin pigmentation including the locally produced peptide hormones a-melanocyte stimulating hormone (ct-MSH) or adrenocorticotropic hormone ( A C T H ) acting through a .G-protein coupled receptor which stimulates melanin production by inducing maturation or switching of melanin type (Vancoillie et al., 1999). The effects of a - M S H / A C T H on melanogenesis are mediated principally via tyrosinase, the rate-limiting enzyme in the melanogenesis pathway. Binding of the peptide hormones to their receptor increases tyrosinase activity and produces light red/yellowish pheomelanin within the pheomelanosome, whereas the eumelanosome produces darker eumelanin via induction of additional enzymes that subsequently accounts for the skin-darkening effect. 1 The current model of melanin transport involves melanosomal maturation through a number of distinct morphological changes within the melanosome followed by retrograde transport via dendrites of the melanocyte similar to neurotransmitter transport within neurons. Transfer of the melanin to a keratinocyte most likely occurs through a regulated exocytosis mechanism following stimulation (possibly by ultraviolet irradiation) and then proceeds by reciprocal phagocytosis from the keratinocytes (Tsatmali et al., 2002). The pigment ultimately forms a cap over the nucleus of keratinocytes protecting chromatin from the exposure of hazardous ultraviolet ( U V ) solar light. Interestingly, recent evidence suggests that melanocytes have other functions in the skin in addition to their ability to produce melanin. They are able to secrete a wide range o f signaling molecules, including cytokines, proopiomelanocortin peptides, catecholamines, and nitric oxide in response to U V irradiation and other stimuli (Tsatmali et al, 2002). Potential targets of these secretory products are keratinocytes, lymphocytes, fibroblasts, mast cells, and endothelial cells, all o f which express receptors for these signaling molecules. Melanocytes may therefore act as important local regulators of a range of skin cells. 1.1.2 Melanoma Etiology The primary etiological factor in melanoma skin cancer is U V radiation. The U V spectrum is functionally separated in ranges from the high-energy U V C (wavelengths below 280 nm) to the mid-range o f U V B (280 - 320 nm) to the weaker and longer wavelengths o f U V A (320 - 400 nm). In the outer regions of the earth's atmosphere, molecular oxygen (O2) absorbs high-energy U V C breaking-down into (O) which can recombine to form ozone (O3). Ozone absorbs wavelengths up to about 310 nm thus excluding U V C and most of the U V B from reaching the earth's surface. However, with continuing diminution of the protective ozone (a rarified layer which extends from 12 to 50 km in the stratosphere) the residual U V B can still have deleterious effects on macromolecules of the skin (de Gruijl , 2000). 2 Studies from the 1960s (Beukers and Berends, 1960; Setlow and Carrier, 1966) identified that both U V C and U V B cause specific modifications of D N A . They found two UV-induced changes occurring almost exclusively at sites of neighbouring thymine bases referred to as cyclobutane pyrimidine dimers (CPD) and 6-4 photoproducts (6-4PP). These damaged sites are removed through a process of nucleotide excision repair (NER) which involves removal of bulky D N A adducts created by U V light and is fundamentally different from base excision repair (BER) which removes damage induced by ionizing radiation and alkylating agents (Friedberg, 1995; Balajee and Bohr, 2000). The B E R pathway is mediated by N-glycosylases, enzymes which remove the modified bases from the D N A (usually as a stretch of 5-7 nucleotides) by hydrolysis of the N-glycosidic bond between the deoxyribose sugar and the affected base. This site is then repaired by specific endonucleases. The N E R pathway is more versatile and complex with more than 20 different proteins involved and requires these basic steps: (i) recognition of a D N A lesion, (ii) single strand incision at both sides of the lesion, (iii) excision of the lesion containing single stranded D N A fragment, (iv) D N A repair synthesis to replace the excised nucleotides, and (v) ligation of the remaining single-stranded nick. N E R can be further subdivided into the fast transcription-coupled repair (TCR) (Mellon et al., 1987) that removes damage from the actively transcribed genes and the global genome repair (GGR) that removes damage from the remaining chromatin at a slower rate (Mullenders and Berneburg, 2001). These multiple mechanisms demonstrate the evolutionary importance of D N A damage and the repair processes that have been developed by all organisms. 1.1.3 Mal ignan t Melanoma Melanomas can arise in any tissue where melanocytes exist. Clearly, cutaneous malignant melanoma ( C M M ) is the most prevalent particularly in those sun-rich regions such as Australia, southern Europe, South Africa and the southern sunbelt states in the U .S that light-skinned 3 people populate (Dummer et al, 2001). The second most common site for primary melanoma is intraocular or uveal melanoma that is of growing concern because of its increasing incidence (Wol l etal, 1999). Several factors are believed to correlate with increased risk for development of melanoma. The preneoplastic state for melanoma is referred to as a dysplastic nevus or atypical mole. Individuals with multiple dysplastic nevi have an increased risk for developing melanoma in their lifetimes (Marks, 2000). Congenital nevi appear to be a predisposing condition as well (Marks, 2000). Familial melanoma results form germline mutations in the cyclin-dependent kinase inhibitor, pl6 (CDKN2A) locus (Pollack and Trent, 2000). Interestingly, unlike non-melanoma skin cancers, which are associated with chronic sun exposure, melanoma development correlates better with irregular acute U V radiation exposure causing severe sunburns particularly during childhood or adolescence (Whiteman et al, 2001). Melanoma is the second most rapidly increasing cancer type among all malignancies in the Caucasian population (Howe et al, 2001). Malignant melanoma is associated with one of the highest mortality rates particularly for advanced disease and of an estimated 53,600 new patients diagnosed with the disease for 2001 within the U.S . , there were an estimated 7,400 deaths (Howe et al, 2001). In Canada, approximately 3,800 new diagnoses of melanoma were made for 2001 and an estimated 820 patients died from this disease (National Cancer institute o f Canada: Canadian Cancer Statistics 2001, Toronto, Canada, 2001). Although the sheer numbers are not as high as the U .S . , melanoma incidence is in eighth position on the incidence chart for Canada and addressing the mortality rate is equally challenging particularly when considering that the ratio of cases to deaths is higher in Canada at 0.22 compared to the U . S . at 0.14 (National Cancer Institute of Canada: Canadian Cancer Statistics 2001, Toronto, Canada, 2001; Howe et al, 2001). 4 Melanoma can be a histologically difficult disease to diagnose. It resembles several other benign conditions such as congenital nevi, Spitz nevi, and some more obscure lesions such as Spindle cell nevi of Reed to list a few (Balch et al, 1998; Reed, 1999). This difficulty, particularly for non-dermatopathologists, is compounded by the fact that although most melanomas present with pigmentation, amelanotic melanomas are also found. Melanoma growth generally occurs in two dimensions, the radial growth phase (RGP), which is confined to the epidermis, and therefore easier to treat by surgical resection, and the vertical growth phase (VGP) in which breaching of the basement membrane and invasion into the dermis occurs. The V G P is associated with increased malignancy because of the potential for metastasis. The standard classification and clinical and pathological staging of melanoma, the T N M system (T = primary tumour, N = regional lymph nodes, M = distant metastasis), was last adopted in 1997 by the American Joint Committee on Cancer. This essentially takes into consideration two variables, the thickness of the primary tumour (Breslow depth) and the level or depth of invasion into the underlying dermis and subcutaneous fat (Clark's level I-V). However, as of 2002, there have been some major revisions to the classification system that now includes other prognostic indicators such as primary tumour ulceration, number of lymph nodes involved, anatomic site of metastases and the replacement of Clark's levels by tumour thickness as a better predictor of survival (Slominski et al., 2001; K i m et al., 2002). Current treatment strategies for melanoma are generally dictated by whether the disease is local or whether metastases have occurred. The single greatest treatment as with most solid tumours is surgical resection, followed by adjuvant therapy. The best chemotherapy treatment for metastatic melanoma appears to be dacarbazine (DTIC), a methylating drug that shows clinical responses in the order of 20% of patients (Cohen and Falkson, 1998). Other single agents include platinum analogues like cisplatin, nitrosoureas or other alkylating agents, and microtubule toxins such as the vinca alkaloid, vindesine. Many of the same types of drugs are 5 used for treating metastatic disease such as Temozolamide, an analogue of the active dacarbazine metabolite that has shown some success (Olhoffer and Bolognia, 1998; Bonaccorsi et al, 2001). Some combination treatments of drug cocktails are also showing some promise but usually at the expense of systemic toxicity (Olhoffer and Bolognia, 1998; Slominski et al., 2001; Eton et al, 2002). Another regimen in use for metastatic melanoma is immunotherapy using cytokines particularly interferon-a (INFa) and to a lesser extent interleukin-2 (IL-2) (Olhoffer and Bolognia, 1998; Bonaccorsi et al, 2001). LNFa as adjuvant therapy administered intravenously seems to have encouraging results for advanced disease with improvements in up to 40% of individuals (Bonaccorsi et al., 2001; Eton et al., 2002). These are still poor survival numbers and do not dramatically increase quality-of-life for these victims. Other modalities take advantage of the fact that melanomas are very antigenic tumours that express several kinds of specific surface markers distinguishing them from surrounding normal tissue. Several vaccines utilizing these melanoma-associated antigens as well as humanized monoclonal antibodies targeting these markers are both in development and clinical trials with the hope that enhanced cellular and humoural immune responses w i l l be activated toward the melanoma cells (Cohen and Falkson, 1998; Olhoffer and Bolognia, 1998). In most cases, radiotherapy treatment is only used for patients with multiple metastatic melanoma and then only as a palliative measure for the associated pains of the disseminated disease. Because melanoma is notoriously radioresistant, this is used infrequently on malignant tumours (Jenrette, 1996). The aforementioned facts bring us to the overwhelming problem of advanced malignant melanoma. This cancer has a poor prognosis because chemotherapy is largely ineffective due to the resistance of melanoma cells to systemic treatment with anticancer agents. The basis for the chemoresistant phenotype, whether a result of inherent drug resistance or acquired drug resistance, is undefined. What seems clear from the literature is that most anticancer drugs k i l l susceptible cells through induction o f apoptosis or programmed cell death (see section 1.3.5 for 6 detailed description). Therefore, it appears that alterations in the apoptotic pathways that lead to apoptotic escape may account for the ability of some tumour cells to resist cytotoxic drug therapy. Cancers such as melanoma are prone to having distorted apoptotic pathways with fewer pro-apoptotic molecules and thus two broad mechanisms for subverting the effects of drugs would be to (1) reduce or eliminate the drug concentration within the cell by actively pumping the drugs out or (2) alter or bypass the regulatory checkpoints of the apoptotic machinery to avoid its induction. We have focussed on these two mechanisms in an attempt to clarify the resistance mechanisms of melanoma and to overcome biochemical blocks that permit melanoma survival. When combined with the fact that melanomas have a tendency to metastasize quickly to sites such as the lung, liver, brain, G l tract, and bone, then the urgency to decipher melanomagenesis seems essential (Reintgen et al., 2001). The complexity of the molecular apoptotic pathways suggests that the melanoma cell may have a variety of possibilities in which to select for deregulated apoptosis. Since biochemical redundancy is a recurring property of any cell, compensation whether at a single point or at multiple levels could be sufficient to generate apoptotic deficiency; therefore, determining the biochemistry of this process is essential to understanding how melanoma progresses. 1.2 P53, THE CELLULAR GATEKEEPER 1.2.1 Gene and Protein Structure Stress to the cell in any form is evaluated by a limited number of key factors. These are often referred to as 'gatekeepers' or 'nodes'. Unquestionably, the best-studied gatekeeper is the tumour-suppressor gene p53. The human p53 gene is located on chromosome 17pl3.1. It is composed of 11 exons and spans a genomic region of over 20 kb. The first intron separates exons 1 and 2 by over 10 kb and contains the second or downstream promoter (Lamb and 7 Crawford, 1986; Reisman et al, 1988). The gene product of p53 is a 393-residue nuclear phosphoprotein and transcription factor. Alternative splicing of intron 9 in the m R N A leads to the production of a 341 amino acid protein truncated at the C-terminus. The protein has essentially 6 functional domains. From the N-terminus: two transactivation domains (residues 1-42 and 43-63) used for transcriptional activation of target genes; a proline-rich region (residues 64-90) most likely involved with protein-protein interactions; a sequence-specific D N A binding domain (residues 100-300); a nuclear localization signal (residues 316-325); an oligomerization domain (residues 334-356) required for the tetramerization function o f p53; and a basic C -terminal regulatory domain (residues 363-393) (Figure 1.1). Since p53 has very profound effects on the status of the cell, it is kept at very low basal levels and has a very short half-life of about 5-20 min in most cell types studied. It is tightly regulated in turn by M D M 2 , an E3 ligase that targets both p53 and itself for degradation through the ubiquitin complex in a negative feedback manner (Barak et al., 1993; W u et al, 1993). This occurs by several different mechanisms that include binding to the transactivation domain of p53 (Momand et al., 1992), targeting p53 for ubiquitination (Haupt et al., 1997), inhibiting acetylation of p53 (Grossman et al., 1998), and by shuttling p53 to the cytoplasm (Stommel et al., 1999). Following D N A damage, the half-life of p53 increases several-fold (Price and Calderwood, 1993; M a k i and Howley, 1997; Bean and Stark, 2001) and this is generally mediated by perturbing M D M 2 function either by phosphorylation of both p53 and M D M 2 which consequently blocks their interaction or selective down-regulation of M D M 2 expression (Balint and Vousden, 2001). 1.2.2 Roles in Transcr ipt ional Regulation The list o f signals and stressors that p53 can respond to appears to grow on a daily basis. Suffice it to say that some o f the more classical genotoxic and nongenotoxic stressors are D N A damage, oncogene activation, hypoxia, growth factor deprivation, and detachment from the substratum or 8 Figure 1.1 Schematic representation o f the p53 protein. The human p53 protein has 393 amino acid residues. Numbers under each domain indicate the length o f the region. Vertical bars represent mutational 'hotspots' and the relative frequency in the population is denoted by the height o f each bar. 9 loss of cell adhesion (Balint and Vousden, 2001; Bargonetti and Manfredi, 2002; Blagosklonny, 2002). Stabilization and regulation o f p53 occurs by several post-translational modifications, primarily phosphorylation/dephosphorylation and acetylation/deacetylation to specific sites that appears stimulus-dependent. The addition or removal of glycosyl or other carbohydrate moieties can also modulate p53 function (Blagosklonny, 2002). The modifications to p53 affect its tertiary folding and subsequently its interaction with both other proteins and DNA-bind ing regions. For example, since the carboxy terminus o f p53 normally folds back and inhibits the DNA-bind ing domain in the central part of the protein, acetylation of lysine or phosphorylation of serine at the carboxy terminus can enhance p53 binding to D N A (Grossman et al, 1998). This may also help mediate p53 tetramerization which is believed to be the activated structure for p53 in solution. To exert its functional activities, p53 translocates to the nucleus via its nuclear localization signal (NLS) where it can recognize sequence-specific D N A regions. A consensus D N A binding site was determined to be two copies of a 10 bp motif 5'-PuPuPuC(A/T)(T/A)GPyPyPy-3 ' separated by 0-13 base pairs (El-Deiry et al, 1992). Since the definition of this sequence, numerous sequence searches have identified new p53 targets and confirmed previous expressional studies of p5 3-regulated candidates by the presence of this consensus site in promoter regions (Wang et al, 2001; X u and Raafat el-Gewely, 2001). Although transcriptional activation by p53 mediated through sequence specific D N A binding is well known, the molecular basis of transcriptional repression by p53 is poorly understood. This is simply because most genes reportedly repressed by p53 have no classical p53-binding sites in their promoter. In the absence of direct D N A binding, the mechanism is usually attributed to sequestration of components of the basal transcription machinery by p53 through protein-protein interactions. However, p53-mediated repression of target gene expression by direct D N A binding has been demonstrated in several instances (Ori et al, 1998; Lee et al, 1999; Budhram-Mahadeo 10 Table 1.1 Common p53-regulated genes and their cellular roles. Genes Funct ion Regulation Reference TSP-1 angiogenesis activated Dameron, 1994 C O X - 2 angiogenesis repressed Subbaramaiah, 1999 Bax apoptosis activated Miyashita, 1995 pasApol/CD95 apoptosis activated Owen-Schaub, 1995 D R 5 apoptosis activated W u , 1997 Noxa apoptosis activated Oda, 2000 P U M A apoptosis activated Y u , 2001;Nakano, 2001 Pidd apoptosis activated L i n , 2000 Presenilin-1 apoptosis repressed Roperch, 1998 M M P - 2 cell adhesion activated Bian, 1997 Fibronectin cell adhesion repressed Iotsova, 1996 E G F R cell proliferation activated Deb, 1994 T G F - a cell proliferation activated Shin, 1995 P C N A cell proliferation activated Morris , 1996 Gadd45 D N A repair/growth arrest activated Smith, 1994 M G M T D N A repair induction repressed Harris, 1996 XPE/p48 D N A repair activated Hwang, 1999 p l 4 A R F feedback control repressed Bates, 1998 p2^ w a f i ' c'pi G l arrest activated El-Deiry, 1993 K L F 4 G l arrest activated Zhang, 2000 14-3-3o G2 growth arrest activated Hermeking, 1997 C D C 2 G2 growth arrest repressed Taylor, 1999 Cycl in G growth-promoting activated Reimer, 1999 A F P hepatic transformation repressed Lee, 1999 H B V - X hepatic transformation repressed Ori , 1998 Tyrosinase melanin synthesis activated Nylander, 2000 M A P 4 microtubule-binding repressed Murphy, 1996 M C K myogenesis activated Zambetti, 1992 Mdm-2 negative feedback activated Barak, 1993 h T E R T telomerase catalytic subunit repressed Kanaya, 2000 b F G F tumour progression repressed Ueba, 1994 11 et al, 1999). For these genes, p53 binds to the DNA-binding site, which overlaps the binding site of another transactivator protein, and so the repression by p53 results from displacement of the activator binding. Recent evidence suggests that the activated and repressed genes of p53 display quite different temporal patterns (Wang et al, 2001). Nonetheless, whether up-regulating or down-regulating, as a cellular gatekeeper, p53 has a dual role in maintaining genomic integrity by effecting two global functions of the cell: (1) to initiate either cell cycle and growth arrest or (2) to induce apoptosis. The more common targets are presented in Table 1.1. 1.2.3 Roles in Cell Cycle Arrest If D N A damage occurs, the ability of p53 to transactivate downstream target genes results in cell-cycle arrest at specific checkpoints in the cell cycle. A s a consequence, the cell cycle stops either before D N A replication in G l , or before mitosis in G2, accompanied by a severe decrease in the proportion of cells at S phase. P53-dependent G l arrest is able to interfere with cell-cycle progression by blocking cyclin-dependent kinases (Cdk) via increases in transcription of the primary Cdk inhibitor, p 2 1 w a f l (El-Deiry et al, 1993). Elevated p 2 1 w a f l levels inactivate accumulated Cdk2/cyclinE complexes and this prevents the Cdk complex from phosphorylating substrates such as the retinoblastoma tumour suppressor protein (Rb) (Xiong et al, 1993). Consequently, cell cycle progression is inhibited which blocks the transition from G l into S phase so that cells collect in late G l (El-Deiry et al, 1993). Another checkpoint response pathway operates in S phase to slow down the rate of D N A replication. Increased p 2 1 w a f l also binds to the proliferating cell nuclear antigen ( P C N A ) , titrating it from the nucleus and creating replication arrest by blocking the elongation step of D N A replication (Waga et al, 1994; Morris et al, 1998). P C N A is considered another important molecular switch in the decisions of cell fate and interacts with a number of D N A mismatch repair components (Paunesku et al, 2001). 12 The p53 protein can also promote a G 2 / M checkpoint by ^raws-activating the 14-3-3a protein. The 14-3-3a protein sequesters the Cdc25C phosphatase in the cytoplasm preventing it from activating Cdc2, the Cdk that promotes the C d k l / c y c l i n B l complex that determines the timing o f mitosis (Ford et al., 1994, Hermeking et al., 1997). Therefore, the p53-dependent induction of 14-3-3cr connects D N A damage with the Cdk-driven G 2 / M progression in the same way that the p53-dependent p21 induction connects D N A damage with the Cdks required for G l / S progression. 1.2.4 Roles in D N A Repai r The p53 protein can induce specific enzymes involved in different aspects of D N A repair depending upon the extent and the nature o f the D N A damage. This has usually been demonstrated by either direct interaction of p53 with repair pathways or p53 ^raws-activation of downstream genes. This list includes Gadd45; members of the Xeroderma pigmentosum complementation group ( X P A through X P G ) but in particular, X P A and X P E for DNA-damage recognition, and both X P B and X P D for repair through N E R ; the p62 subunit of the multisubunit TFLTH protein complex required for basal transcription/repair; the Cockayne syndrome B (CSB) protein; and other replication-dependent components such as replication protein A (RPA) and Ref-1 (Balint and Vousden, 2001; L i u and Kulesz-Martin, 2001). The C-terminus o f p53 directly binds damaged D N A in multiple forms including single-stranded D N A , ends of double-strand breaks and D N A 'bulges' resulting from insertion/deletion mismatches (Lindahl and Wood, 1999; L i u and Kulesz-Martin, 2001). There is some controversy with the view that p53 directly participates in repair processes since it has a relatively short half-life which may not give it enough time to be involved in extensive genomic repair directly (Liu and Kulesz-Martin, 2001). L i et al. (1996) demonstrated that mouse keratinocytes with mutant p53 still have some residual D N A repair capacity suggesting alternative mechanisms. The alternative is that other 13 biochemical activities of p53, such as D N A reannealing, D N A strand transfer and 3 ' -5 ' -exonuclease activity might also play a role in its repair function (Albrechtsen et al, 1999). 1.2.5 Roles in Apoptosis In the event that a stimulus has created too much damage to the cell, p53 w i l l initiate the apoptotic machinery that leads to apoptotic cell death. This is the concerted process of cell suicide that occurs through characteristic morphological changes and is a completely different process to necrotic cell death—these w i l l be discussed in more detail in section 1.3.5. The commitment to such a pathway is irreversible and as such, the control has to be exquisite. Trans-activation of apoptotic components is one of the fundamental functions of p53. The number of activated genes is considerable and is exemplified by the largest class of p53-responsive genes in Table 1.1. p53 activates two distinct branches of apoptosis: (i) the mitochondria-mediated mechanism which include subsets of the Bcl-2 family members such as anti-apoptotic Bcl -2 (Shimizu et al., 1999) or the pro-apoptotic Bax and Bak (Shimizu et al., 1999), Noxa (Oda et al., 2000), and the p53-upregulated modulator of apoptosis ( P U M A ) (Nakano and Vousden, 2001; Y u et al, 2001a) that accelerate the opening of mitochondrial porin channels for release of cytochrome c, or (ii) the membrane-mediated death receptor mechanism such as Fas /Apol /CD95 (Owen-Schaub et al, 1995) and DR5/Ki l l e r (Wu et al, 1997). Death receptors can modulate apoptosis in a mitochondria-dependent manner (see section 1.3.5.3 below for details) which also leads to pore formation, altered membrane potential (Zamzami et al, 1995), and cytochrome c release from the mitochondria that complexes with the apoptotic protease activating factor-1 (Apaf-1) to form the apoptosome (the functional apoptotic unit). The apoptosome activates the cysteine-aspartic acid specific proteases or caspase family of proteolytic enzymes, the so-called 'executioners' o f apoptosis (reviewed in Zornig et al, 2001). Death receptors can also activate certain caspases independently of the mitochondria (Ashkenazi and Dixi t , 1998). L ike the cell 14 cycle arrest pathway, p53 can activate a number of different components within the apoptotic cascade operating at different and multiple points. Another competing model for p53-mediated apoptosis suggests that cell death may result from the inhibition of expression of survival factors which function to maintain cell viability. This model is currently undergoing more scrutiny but it does imply the alternative that rather than trans-activation, p53 may induce apoptosis through generalized transcriptional repression (Colman et al., 2000). 1.2.6 P53 Muta t ion and the Implication for Cancer The significance of p53 studies in cancer comes to light when considering that p53 is the most frequently mutated gene in human cancers with mutations occurring in over 50% of all tumours (Soussi and Beroud, 2001). A s with many gene mutations, alterations can lead to inactivation. Because p53 is involved in so many diverse cellular processes, a mutation in p53 diminishing one of its functions may provide a cell with a particular selective advantage that could in turn lead to secondary p53 mutations as a byproduct. The accumulation of mutations is part of the cancer phenomenon and is responsible for its stepwise progression as outlined by Hanahan and Weinberg (2000). On the other hand, rendering p53 completely non-functional can give a tumour cell an enormous growth advantage, the defining feature of a neoplasm. Despite the statistic that 74% of tumour-derived p53 mutants are missense mutations (Soussi and Beroud, 2001), which can still produce full-length proteins, they are generally transcriptionally inactive, but some mutants still retain the ability to transactivate a subset o f p53 target genes (Sigal and Rotter, 2000). In addition to simple loss of function, some p53 mutants may be oncogenic or even carcinogenic through a dominant negative mechanism (Sigal and Rotter, 2000; Soussi and Beroud, 2001). The majority of p53 mutations occur in the DNA-bind ing core region of exons 5-8 (residues 102-292 in Figure 1.1) thereby disrupting expression/repression of downstream targets. Mutational analyses of the p53 locus have determined that of the over 100 codons 15 identified with alterations, several 'hotspots' or nucleotide positions with a high incidence of mutation exist (Figure 1.1). Even these common mutations have a range of characteristics on the p53 protein and can significantly vary its oncogenicity despite occurring in the same domain (Sigal and Rotter, 2000). Another interesting common mutation in the p53 locus is a polymorphism at position 72 that leads to a variation in the protein sequence (Arg/Pro variation). Several reports have correlated pathological consequences of this polymorphism but these are controversial claims (Storey et al., 1998). More pertinent to this work is the variation in the population frequency such that about 17% o f people in Sweden and Finland have the Pro72 version while 63% of black Africans of Nigeria have the Pro72 raising the speculation that this polymorphism may protect against skin cancer (Beckman et al., 1994). Generally speaking, p53 mutations correlate with more aggressive cancers and clinical drug resistance. In vitro experiments with lymphoma cells lines (Fan et al., 1994) and astrocytic tumour cells (Iwadate et al., 1998) have validated this hypothesis. Furthermore, p53 status is linked to drug resistance in several other lymphoid malignancies. Often in these tumour types such as myelodysplastic syndrome, and both chronic and acute lymphocytic leukemias, p53 mutations are rare but associated with disease progression and poor prognosis (Wattel et al., 1994). When patients are analyzed according to the p53 status, tumour response and survival decrease with p53 alterations. Individuals having cancers that rarely exhibit p53 mutations and who relapse after treatment have a higher percentage o f tumours harbouring mutant p53 (Wattel et al., 1994). Moreover, patients with p53 mutant tumours are less likely to undergo remission compared to patients with relapsed tumours bearing normal p53 (Diccianni et al., 1994). One may predict that enrichment for cells with p53 mutations would confer a survival advantage during chemo/radiotherapy regimes. Similar findings are linked with breast carcinomas where p53 mutations are compelling predictors of treatment failure, relapse and even death (Clahsen et al, 1998). In human melanoma, p53 mutation occurs in about 20-30% of assessed samples 16 (Weiss et al, 1995; Sparrow et al., 1995). Unfortunately, it appears that most reports investigating the p53 sequence in melanomas have focussed on the DNA-bind ing region or exons 5-8 (Soussi and Beroud, 2001). Therefore, these figures may underestimate the actual p53 mutation rate in melanoma. The significance of p53 in melanoma is validated by the fact that p53 mutation often correlates with disease progression and metastatic potential (Stretch et al., 1991; Weiss et al, 1995; Kanoko et al, 1996; Grant et al, 1998). These facts educe what additional aspects of p53 biology may contribute to p53 mutations and drug resistance. 1.3 D R U G - R E S I S T A N C E O F C A N C E R 1.3.1 The M D R Phenotype When cancer cells become resistant, they tend to manifest simultaneous resistance to a number of structurally and functionally distinct chemotherapeutic drugs. This is known as the multidrug resistance ( M D R ) phenotype. The main mechanisms for chemoresistance of tumour cells can be generalized in four schemes for the cytostatic action of drugs: 1) decrease of drug concentration in the cell due to activation of transporter proteins, 2) reduced drug activation or increased detoxification of the drug within the cell, 3) alterations of drug target and increased repair of the damaged target, and 4) abrogation of apoptosis or cell cycle arrest, for example, due to mutation of the p53 gene. From the literature, it is clear that steps (1) and (4) represent the most common and feasible approaches to understanding the biological phenomenon of M D R but each of these w i l l be discussed in detail. 1.3.2 Decreased Intracellular Drug Concentration The primary cause of the M D R phenotype is attributed to the overexpression of some members of a highly conserved family (from bacteria and barley to man) of transmembrane proteins 17 characterized by an ATP-binding cassette or domain, and therefore called the A B C superfamily of transporters. Wi th most of the coding sequence o f the human genome completed, current predictions bring the total number of the A B C superfamily to 48 members that have been classified into seven groups (subfamilies A - G ) according to phylogenetic analyses. The structural and functional roles of this family in cancer have recently been reviewed (Dean et al., 2001; Tan et al., 2000). The A B C superfamily can loosely be divided into full- or half-transporters based on structural analyses. Essentially, a full transporter has two hydrophobic domains comprised of six membrane-spanning sequences (some transporters have 3 of these domains) and two ATP-binding cassettes while the half-transporters have one and one, respectively (Figure 1.2), but other arrangements of the transmembrane and A B C domains are present. Not all members of this diverse family are specifically involved i n drug resistance. However, members from three families, subfamily B , C, and G which do contain known multidrug transporters, w i l l be emphasized for this thesis. The characteristics of these subfamilies including transporter type, substrates, and tissue localization are summarized in Table 1.2. The prototypical representative of this family, A B C B 1 (P-glycoprotein), was first characterized in Chinese hamster ovary cells and identified as being encoded by the ABCB1 (Mdrl) gene in humans (Juliano and Ling, 1976; Roninson et al, 1986). A B C B 1 can transport a great variety of different molecules including drugs and dyes (Table 1.2). Furthermore, the same gene is conserved in rodents as well (Dhir et al., 1990). Isoforms of ABCB1 (ABCB4 and ABCB11) have been found in humans but a biological function for these has not been thoroughly ascribed to the M D R phenotype in cancers (Turton et al., 2001; Childs et al., 1998). Another subfamily of the transporters, the A B C C (MRP/CFTR) family includes at least five members (ABCC1-5) that are Mr/>-related genes and have an established role in multidrug transport, particularly the glutathione-conjugated derivatives o f several toxic compounds 18 Figure 1.2 Diagrammatic representation of the most common configuration for full and half A B C transporters. The transmembrane domain generally consists o f 6 a-helices (grey bars) and the ATP-binding cassette ( A B C ) . Varieties of this arrangement are present among different family members. Ha l f transporters such as M X R may homo or heterodimerize with other family members. Some of these glycophosphoproteins are relatively large with molecular masses around 200 kDa such as M R P 1 . l1) (Ishikawa et al, 2000; Hopper et al., 2001). The most recent A B C transporter identified with a role in the M D R phenotype is ABCG2 (also known as Bcrp or breast cancer resistance related protein and MXR or mitoxantrone resistance protein) (Doyle et al., 1998). This is the only half transporter recognized in drug resistance and whether it homo- or heterodimerizes to form a fully functional A B C transporter has not been ascertained (Kowalski et ai, 2001). A s a group, the multidrug transporters have a diverse number of substrates including drugs, antibiotics, dyes, stains, and lipophilic compounds that are actively effluxed out of cells through an energy-dependent mechanism involving A T P hydrolysis. The key characteristic of A B C transporters is the simultaneous cross-resistance to structurally and functionally unrelated compounds. Recently, a new multidrug transporter was found to cause drug resistance in lung cancers and therefore named the lung cancer resistance related protein (LRP) (Scheffer et al., 1995). L R P is the major vault protein ( M V P ) and unlike the A B C transporters does not contain an A T P -binding cassette and as such has a much different cellular function. Vault proteins are very large ribonucleoprotein complexes first detected as artefacts in vesicle preparations from rat liver (Kedersha and Rome, 1986). Their unique hollow, barrel-shaped structure is thought to compartmentalize drugs and xenobiotics away from intracellular targets then extrude these molecules either through a vesicle-mediated exocytosis or efflux by conventional A B C transporters (Scheffer et al., 2000). The regulation of multidrug transporters through detailed analyses of their respective genetic loci and/or control regions has been performed for seven of the ten putative members (nine ABC subfamily members listed in Table 1.2 and LRP). The most extensively characterized promoter appears to be that of the ABCB1 locus. The elucidated structure of the ABCB1 promoter identified two regions, an upper or 5' region and lower or 3' promoter region that are differentially regulated (Ueda et al, 1987; Madden et al, 1993; Strauss and Haas, 1995). The 3' 20 Table 1.2 A B C transporters involved in multidrug resistance. V P 16, etoposide; G S H , reduced glutathione; G S S G , oxidized glutathione; CPT , camptothecin. Nomenclature Common Name Trans-porter Type Substrates Tissue Localization MDR/TAP A B C B 1 M D R 1 Ful l Hydrophobic compounds, Adrenal, subfamily B P-GP steroids, colchicine, digoxin, brain, doxorubicin, vinblastine, dyes, kidney, liver saquinivir, paclitaxel, V P 16 A B C B 4 M D R 3 Ful l Phosphatidylcholine, Liver P G Y 3 doxorubicin? A B C B 1 1 B S E P Ful l Bi le salts, taxol? Liver SPGP MRP/CFTR A B C C 1 M R P 1 Fu l l Anionic conjugates + G S H , + Lung, subfamily C sulphate, + G S S G , rhodamine, P B M C , doxorubicin, daunorubicin, Testes V P 16, colchicine, vincristine A B C C 2 M R P 2 Ful l Anionic conjugates + G S H , + Intestine, C M O A T sulphate, vinblastine, liver, kidney sulfinpyrazone A B C C 3 M R P 3 Ful l Anionic conjugates, bile salts, Intestine, methotrexate, V P 16 liver, lung A B C C 4 M R P 4 Ful l Nucleoside monophosphates Prostate A B C C 5 M R P 5 Fu l l Organic anions, nucleotide Ubiquitous analogs, cyclic nucleotides White A B C G 2 B C R P 1 Ha l f C P T analogs, daunorubicin, Placenta, subfamily G A B C P doxorubicin, mitoxantrone, breast, liver rhodamine 21 region is known to control basal transcription but mutations in this promoter may be responsible for some o f the regulatory inconsistencies seen in some cancer types (Madden et al, 1993; Stein et al., 1994). Like most promoter analyses, many different motifs and response elements within the human ABCB1 promoter have now been identified that are responsible for its regulation. Several of these motifs are present in the homologous ABCB1-like promoters from rats and mice. Both cis- and frans-regulatory mechanisms for ABCB1 promoters have been suggested (Ueda et al., 1987). Recently, methylation of the promoter region was also suggested to be a factor controlling expression (Kantharidis et al., 1997). In comparison, the regulation and promoter analysis of other multidrug transporters has not been studied to the same extent although characterization of some common positive and negative regulatory elements have been identified for the ABCB4 (Smit et al., 1995), ABCB11 (Ananthanarayanan et al, 2001), ABCC1 (Zhu and Center, 1994), ABCC2 (Tanaka et al, 1999; Stockel et al, 2000), ABCC3 (Stockel et al, 2000; Takada et al, 2000), and LRP promoter regions (Lange et al, 2000). In contrast, no analysis of promoter regulation for the ABCC4, ABCC5 or ABCG2 loci has been reported. To fully understand the regulation of these loci , further analyses on the control regions need to be completed. We should also acknowledge that of these aforementioned transporters only the ABCB1, ABCC1 and LRP promoters contain the consensus p53 binding sequences. 1.3.3 Reduced Drug Activation or Increased Detoxification Intracellular detoxification of drugs via glutathione metabolism is accompanied by both an increase in glutathione levels ( G S S G is the oxidized form and G S H is the reduced form) and activity of the glutathione •S-transferase (GST) in tumour cells (Zhang et al, 1998). The G S T / G S H system is involved in adaptation to oxygenic stress (by-products of reactive oxygen species activity) through conjugation of glutathione to electrophiles. Resistance to a wide variety 22 of anticancer agents including cisplatin, l,3-bis(2-chloroethyl)-l-nitrosourea ( B C N U ) , and doxorubicin has been associated with an increase in G S T or the related isoenzymes in several cell lines derived from ovarian and colonic cancers. Although high glutathione levels were measured in some melanoma cells, no correlation was found between alterations in G S T / G S H metabolism and drug sensitivity in melanoma (Benathan et al., 1992; Schadendorf et al., 1995). 1.3.4 Alterations of Drug Target via Increased Damage Repair The specificity of cytotoxic drugs can often be compensated or offset by modulation of corresponding D N A repair mechanisms thereby evading apoptotic signals. For example, in drug-resistant melanoma cell lines, decreasing the capacity for DNA-mismatch repair can cause drug resistance by interfering with the normal D N A damage repair system that may otherwise be invoked after specific drug effects (Fink et al., 1998; Lage et al., 1999). Furthermore, increases in both repair activity and the repair enzymes were found in human melanoma cell lines with established resistance to different drugs (Lage et al., 1999). A similar study showing an increase in 0 6 -methylguanine-DNA methyltransferase ( M G M T ) , an enzyme involved in repair caused by alkylation, was associated with fotemustine-resistant melanoma cells (Runger et al., 2000). In contrast, improved survival was seen in melanoma patients expressing low levels of M G M T after treatment with temozolamide (Middleton et al., 2000). Evidently, this relationship needs to be clarified further. 1.3.5 Abrogation of Apoptosis Apoptosis was first acknowledged based on its characteristic morphological effects (Kerr et al., 1972). The changes include shrinkage of the cell volume, dilatation o f the endoplasmic reticulum and alterations of the plasma membrane. Apoptosing cells break-up into a series of membrane-bounded bodies, the 'apoptotic bodies', which contain normal but compacted 23 organelles. The nucleus undergoes extensive changes that include dissolution of the nuclear lamina, chromatin condensation, and nucleolar degradation. The apoptotic bodies are generally phagocytosed by adjacent cells without an acute inflammatory response. This coordinated transformation is in stark contrast to necrosis, the disruption of cells that is observed in acute, high-dose toxicological exposure or severe hypoxia. These injured cells swell and their plasma membranes burst releasing proinflammatory material into the extracellular space (Wyllie, 1997). Apoptosis plays a critical role as a physiologic event that regulates cell number and eliminates damaged cells during embryogenesis, tissue remodelling, and, most importantly from our perspective, tumour growth. With intense study of the programmed cell death mechanisms, it has become clearer in the past few years that two generalized pathways exist. Apoptosis can be mediated by an intrinsic, mitochondrial-based release o f signals to stimulate/activate the terminal effectors of the process or an extrinsic, membrane receptor-mediated cascade can initiate the program. The complexity of this pathway accepts that molecular cross-talk exists and therefore these two mechanisms are surely not mutually exclusive. 1.3.5.1 Caspase-family, the effector pathway A l l members of this ICE-family of cysteine proteases (now called caspases) exhibit a preference for cleavage adjacent to aspartic acid residues. The mammalian caspase family contains 14 members, a subset of which drives the effector process of apoptosis, with the remainder likely to be involved in the processing of pro-inflammatory cytokines (Creagh and Martin, 2001). When added to prepared nuclei from non-dying cells in vitro, they swiftly induce all the structural changes characteristic of authentic apoptotic cells (Lazebnik et al., 1993). Considerable evidence suggests that caspases are essential components of the mammalian cell death machinery. Because the proteolytic cleavage of proteins is largely irreversible, activation of these enzymes may represent a rate-limiting step in apoptosis. Affinity labelling techniques 24 revealed that the caspases are activated in a branched protease cascade (Hirata et al, 1998). Caspase-8 activates caspase-3 and - 7 , and caspase-3, in turn, activates caspase-6 (Hirata et al., 1998; Creagh and Martin, 2001). O f the caspases identified to date, caspase-3 has been shown to be one of the major activated caspases present in apoptotic cells, suggesting that it plays a prominent role in the cell death process (Faleiro et al., 1997). Caspase-3 (CPP32) cleaves nuclear mitotic apparatus protein ( N u M A ) and mediates D N A fragmentation and chromatin condensation (Hirata et al., 1998; L i u et al, 1997; Rudel and Bokoch, 1997). It is also involved in extranuclear apoptotic events: cleavage of p21-activated kinase-2 (PAK2) , formation of apoptotic bodies and exposure of phosphatidylserine on the cell surface. Caspase-3 is also responsible for the proteolytic processing of poly(ADP-ribose) polymerase (PARP) . Conversely, caspase-6 cleaves N u M A at sites distinct from caspase-3, and mediates the shrinkage and fragmentation of nuclei (Hirata et al, 1998; Creagh and Martin, 2001). Using cells from a caspase-3 knockout mouse model, Woo et al. (1998) showed a reduction in chemosensitivity of mouse embryonic fibroblasts (MEFs) in the absence of CPP32 and a ' restoration of chemosensitivity by reintroducing wild-type CPP32. 1.3.5.2 Apoptosis, the mitochondria pathway Bcl-2 is the founding member of a multi-gene family characterized by four highly conserved domains, B H 1 , B H 2 , B H 3 , and B H 4 , the Bcl-2 homology domains required for dimerization. Some Bcl-2 family proteins function as inhibitors of apoptosis (Bcl-2, M c l - 1 , B C 1 - X L , Bcl -w) , while others promote cell death. The pro-apoptotic members are further divided into two groups, the multidomain (Bax, Bak, Bok, Bc l -Xs) and the 'BH3 domain only' proteins (Bad, B i d , B i k , Bim) . Protein interactions among this family, with formation of both homo- and heterodimers, regulate the sensitivity of cells to apoptotic stimuli (Sato et al, 1994). Homodimers of Bax, a 21 kDa protein with sequence homologous to Bcl-2 , actively induce or promote cell death and the 25 ratio of Bax to Bcl-2 is thought to determine the degree of heterodimer formation, and thus the sensitivity of a cell to apoptosis (Oltvai et al., 1993). M c l - 1 , another Bcl -2 homolog, has been shown to delay c-myc induced apoptosis in Chinese hamster ovary cells and functions like Bcl -2 to neutralize Bax-mediated cell death in a yeast system (Reynolds et al., 1994). B c l - X is another important apoptosis regulator: alternative splicing results in both a long ( B c l - X L ) and a short form (Bcl-Xs) of this protein. B C 1 - X L blocks cell death while B c l - X s promotes apoptosis (Boise etal, 1993). When cells are exposed to apoptotic stimulation, pro-apoptotic proteins are activated through post-translational modifications or changes in their conformation. The main site of action for Bax and Bax-like multidomain proteins is the mitochondria. Successful dimerization can induce permeabilization by stabilizing pore formation in the mitochondrial outer membrane formed by a protein complex referred to as the permeability transition pore (PTP) (Martinou and Green, 2001). Permeability transition causes uncoupling of the respiratory chain with collapse of the electrochemical proton gradient (A^m) and cessation of A T P synthesis, matrix C a 2 + outflow, depletion of reduced glutathione, depletion of N A D P H , and hypergeneration of superoxide anion, leading to cytochrome c release from the mitochondrial intermembrane space into the cytosol (Green and Reed, 1998; Martinou and Green, 2001). Free cytochrome c combines with Apaf-1 to form a functional apoptosome that activates procaspase-9 via interaction with a conserved caspase recruitment domain ( C A R D ) . The apoptosome can then recruit procaspase-3, which is cleaved and activated by the active caspase-9, releasing it to mediate apoptosis (Zimmermann et al, 2001). 1.3.5.3 Apoptosis, the death receptor pathway The roles of membrane death receptor-mediated apoptosis are wel l established both in vitro and in vivo (Ashikenazi and Dixi t , 1998; Timmer et al., 2002). In particular, the mechanisms for 26 Fas /APO- l (CD95) , a type-I transmembrane protein that belongs to the tumour necrosis factor (TNF) and nerve growth factor (NGF) receptor superfamily (Oehm et al., 1992; Nagata and Golstein, 1995). Ligation of Fas by the Fas ligand (FasL) or a cross-linking anti-Fas antibody forces receptor trimerization followed by the recruitment and binding of the intracellular adaptor molecule F A D D (Fas-associating death domain) to the cytoplasmic domain o f the receptor (Chinnaiyan et al., 1995). The NH 2 -terminal region of F A D D recruits and activates the formation of a death-inducing signaling complex (DISC) that binds procaspase-8 (Muzio et al., 1996), and thus triggers a terminal apoptotic cascade. Caspase-8 then cleaves procaspase-3 and -7, and active caspase-3 cleaves procaspase-6. Similar pathways appear to be activated for other death receptors, e.g., T N F R 1 and death receptors D R 3 , D R 4 , D R 5 , and D R 6 , although the ligands and adaptor molecules are different (Ashikenazi and Dixi t , 1998; Earnshaw et al, 1999). Fas mutations are associated with its down-regulation and have been demonstrated in human melanomas (Shin et al., 1999), while FasL expression is absent on human melanoma (Chappell et al., 1999). These results are confirmed by the work by Owen-Schaub et al (1998), demonstrating that Fas-FasL interactions can suppress metastasis and that tumour Fas loss-of-function may be causally linked to metastatic progression. TNFct is a cytotoxic cytokine to many malignant cell lines (Munker et al., 1987), and is produced primarily by activated macrophages but plays an important role in cancer, inflammation, and differentiation. T N F exerts these multiple effects via binding to two specific receptors on target cells, T N F R 1 (primarily for cytotoxicity) and T N F R 2 (accessory role). Ligand-bound T N F receptors can interact with the cytoplasmic adaptor protein T R A D D through the death domain after oligomerizing. This interaction can elicit two general responses mediated by further protein-protein binding (Liu et al, 1996): F A D D activates the caspase protease cascade while T R A F 2 is involved in activation of stress-activated protein kinases (SAPKs) . T N F sensitivity can be induced in virtually all types o f cells by chemically blocking de novo 27 protein synthesis, which suggests that some form of translational processing must occur to escape cell death. A recent study demonstrated that N F - K B inhibitors could cause melanoma cells to become sensitized to TNFoc treatment (Bakker et al., 1999). A third member of the death receptor family, TNF-related apoptosis-inducing ligand ( T R A I L / A P O - 2 L ) , promotes apoptosis by binding to the transmembrane receptors T R A I L -R1 /DR4 and T R A I L - R 2 / D R 5 . Its cytotoxic activity is relatively selective to human tumour cell lines with little effect on normal cells. T R A I L exerts antitumor activity without causing toxicity as is apparent by studies with several in vivo models (Ashkenazi et ah, 1999; Nagane et al., 2000). The major pathway of its action also proceeds through the formation o f DISC and activation of caspase-8. The apoptotic processes, therefore, follow two signaling pathways, that is the mitochondrial-independent activation of caspase-3, and mitochondrial-dependent apoptosis due to cleavage o f B i d by caspase-8, the formation o f apoptosomes, and activation o f caspase-9 and the downstream caspases (Ruiz-Ruiz et al., 2002). Surprisingly, Bcl-2 and B c l - X L have no effect on TRAIL-induced apoptosis in lymphoid cells whereas these genes suppress apoptosis in nonlymphoid cancer cells (Fulda et al., 2002). T R A I L participates in cytotoxicity mediated by activated N K cells, monocytes, and some cytotoxic T cells (Srivastava, 2001). 1.4 A N T I O X I D A N T S I N C A N C E R 1.4.1 Chemoprevention Mechanisms The recent advances in molecular pharmacology have increased the likelihood that cancer prevention w i l l rely increasingly on interventions collectively termed 'chemoprevention'. Cancer chemoprevention is the use of agents to inhibit, delay or reverse carcinogenesis. Several classes of agents including antioxidants and other diet-derived compounds have shown great potential for chemoprevention of cancers (Galati et al., 2000; Salganik, 2001; Yang et al., 2001). The 28 basic rationale is that these agents act as scavengers of oxidative damage induced by free radicals and other reactive oxygen species (ROS), since they are regarded as having carcinogenic potential and are associated with tumour promotion. Convincing results have been obtained for specific compounds on certain types of cancer such as green and black tea components on skin and oral carcinogenesis (Bickers Alhar, 2000; Hsu et al, 2002); lycopenes (naturally occurring terpenes) from tomatoes on lung cancer ( K i m et al., 2000); vitamin E and selenium in prostate cancer (Neuhouser et al., 2001); and multiple compounds including carotenoids on colorectal carcinogenesis (Wargovich et al., 2000). The molecular action of chemopreventives is plainly not universal but rather is dependent on the molecular species of the compound and the nature of the tumour type. For instance, because U V radiation can cause excessive generation of R O S thereby resulting in an oxidative stress condition, approaches designed to counteract R O S production (in combination with sunscreens) could be fruitful for the prevention of skin cancer (Afaq et al, 2002). Some mechanisms by which cancer chemopreventive agents increase the expression of detoxification and antioxidant genes have been identified. A number of agents appear capable of activating a gene battery that includes NAD(P)H:quinone oxidoreductase, aldo-keto reductases, glutathione S-transferases, gamma-glutamylcysteine synthetase, glutathione synthetase and heme oxygenase (Hayes and McMahon, 2001). Transcriptional induction occurs via the antioxidant/electrophile responsive element ( A R E / E p R E ) within certain promoters that are bound by basic region leucine zipper (bZIP) transcription factors such as the Nuclear Factor-Erythroid 2p45-related factors, N r f l and Nrf2. Upon challenge with inducing agents, they translocate from the cytoplasm to the nucleus where N r f l and Nrf2 are recruited to the A R E as heterodimers with either small M a f proteins, FosB, c-Jun, JunD, activating transcription factor 2 (ATF2) or A T F 4 (Hayes and McMahon , 2001). Induction of these detoxifying enzymes in general leads to protection of cells/tissues against exogenous and/or endogenous carcinogenic 29 intermediates. However, some phenolic compounds such as resveratrol (derived from grape seed extract) can inhibit both phorbol ester- or UV-induced AP-1-mediated activity through the inhibition of non-receptor tyrosine kinase and mitogen-activated protein kinase ( M A P K ) pathways (Yu, 2001b). It is possible that in proliferating or stimulated cells, these chemopreventive compounds may block proliferation by inhibiting these signaling kinases. Alternatively, in non-proliferating or quiescent cells, some of these compounds may activate these signaling kinases leading to gene expression of the cellular defensive enzymes such as detoxifying enzymes. Therefore, a modified mechanism is suggested by which these agents activate the caspase pathways, leading to apoptosis, a potential beneficial effect i f occurring within preneoplastic/neoplastic tissues, but a potential cytotoxic response i f occurring within normal tissues (Klaunig and Kamendulis, 1999; Galati et al., 2000; Kong et al., 2001). 1.4.2 Chemoprevention in Melanoma In melanoma, several studies report the beneficial role of chemopreventives for therapeutic suppression of melanoma progression. Lung metastases were inhibited in mice induced with B16 melanoma cells and treated with polyphenolic compounds (Menon et al, 1995; Caltagirone et al, 2000). Vitamin E could decrease cell proliferation, increase apoptosis, and reduce melanoma growth in mice (Malafa et al., 2002). Furthermore, increases in the apoptotic rate of melanoma cells were found with naturally derived isoprenoids and lignans (Mo and Elson, 1999; Lambert and Meyers, 2001). The additive and potentially synergistic actions of these compounds in the suppression of tumour cell proliferation and initiation of apoptosis provides strong evidence that their use is warranted and that their mechanisms be investigated. 30 CHAPTER 2. MATERIAL AND METHODS 2.1 M A T E R I A L S 2.1.1 C e l l Lines and C e l l Cul ture Nine melanoma cell lines with known p53 status were used: The M M A N , M M R U , R P E P and P M W K cell lines were kind gifts of Dr. H.R. Byers (Boston University School of Medicine, Boston, M A ) (Byers et al, 1991); the SK-mel-2, SK-mel-5, and SK-mel-28 cell lines were obtained from the Tissue Bank at the National Institute o f Health, U S A ; the M e W o and SK-mel -110 cell lines were kind gifts of Dr. A . P . Albino (Memorial Sloan-Kettering Cancer Center, New York, N Y ) . The p53 status in the melanoma cell lines has been determined by D N A sequencing: M M R U , M M A N , R P E P and SK-mel-5 contain wild-type (wt) p53 (L i et al, 1995), while P M W K , SK-mel-2, SK-mel-28, SK-mel-110 and M e W o contain mutant (mut) p53 (Bae et al, 1996). H E K 2 9 3 and HeLa were kind gifts of Dr. S. Dedhar (Biochemistry, U B C ) . B16 and H460 were kind gifts of Dr. W. Jia (Surgery, U B C ) . NTH3T3 mouse fibroblasts were obtained from the American Type Culture Collection (Manassas, V A ) . A l l cell lines were grown at 37°C with 5% CO2 (v/v) in Dulbecco's modified Eagle medium ( D M E M ) containing 100 U / m L penicillin G , 100 u.g/mL streptomycin, and 0.25 u.g/mL amphotericin B and supplemented with 5-10% fetal bovine serum (FBS) (Invitrogen, Burlington, ON). 2.1.2 Reagents, Enzymes, and Chemicals G418, hygromycin B , ampicillin, kanamycin, tetracycline, proteinase K , agarose, restriction enzymes, and T4 ligase were obtained from Invitrogen (Mississauga, ON) . Chemicals such as camptothecin (CPT), vincristine (VIN), doxorubicin (DOX) , cisplatin (CIS), verapamil ( V E R ) , curcumin were obtained from Sigma-Aldrich (Mississauga, ON) . Stock solutions were prepared 31 in dimethylsulfoxide (DMSO) for C P T , curcumin and D O X ; methanol for V L N ; ethanol for CIS and all were stored at -20°C. The CP-31398 compound (Pfizer, Groton, CT) was dissolved in deionized distilled water (ddH^O) and stored at 4°C. A l l other chemicals were purchased from Sigma-Aldrich. 2.1.3 Plasmids and cDNAs Plasmids and c D N A s were generously provided by the following suppliers: . p E G F P - N l control plasmid (BD Clontech, Palo Alto , C A ) • p G L 3 control plasmid (Promega, Madison, WT) • Mouse p53 plasmids ppw ip53WT) and pMR53 (p53MUT, an NH2-terminal deletion mutant) and p E D l (human p53 containing a point mutation that changes cysteine-135 to serine courtesy of Dr. S. Benchimol, University of Toronto). • Human FADD dominant negative construct (FADD-DN) (Dr. V . Dixi t , Genentech, South San Francisco, C A ) . C A T control plasmid, WT-Mdrlb::CAT, MUT-Mdrlb-CAT (Dr. M . Raymond, M c G i l l University) • pHA-PUMA (Dr. B . Vogelstein, Johns Hopkins University, Baltimore, M D ) • pHA-mNoxa and pHA-miVoxa-AB (Dr. T. Shibue, University of Tokyo, Tokyo) • pCmeoIRES-GFP (Dr. J. Eggermont, Catholic University of Leuven, Belgium) • procaspase-3 (Dr. T. Tenev, Imperial College, London) . pcDNA3-y4pa/-l (Dr. K . Bhalla, H . Lee Moffitt Cancer Center, Tampa, F L ) . pTyrex-2 (Dr. D . Bartlett, N C I , Bethesda, M D ) • pLacZ and human Bax c D N A (Dr. W . Jia, Surgery, U B C ) • Bcl-2 antisense oligodeoxynucleotides (Dr. P Rennie, Prostate Group, V G H H S C ) 32 2.2 M E T H O D S 2.2.1 Expression of Constructs by Transfection Cells were grown to 50% confluency and transfected by Effectene (Qiagen, Mississauga, ON) as per the manufacturer's instructions. The amount of D N A used corresponded with the plate size: 0.2 pg for 24-well plate, 0.3 ug for 12-well, 0.4 u.g for 6-well/35 mm plate, and 1 pg for 60 mm dish. Stable clones were selected after 48 h using either G418 (neomycin) or hygromycin B in the range of 100 ug/mL to 800 pg/mL (depending on the cell line) for 3-4 weeks then colonies were pooled. The Bcl-2 antisense oligodeoxynucleotides (ODN) sequence is 5'-T C T C C C G G C T T G C G C C A T - 3 ' . This was diluted in 10 m M Tris (pH 7.4) as a stock solution of 50 u M and stored at -20°C. 2.2.2 Determination of Transfection Efficiency with GFP Vectors We primarily used commercially available non-liposomal reagents for our transfections. To assess the transfection efficiency of a particular cell type, we would introduce the equivalent amount o f GFP-bearing control plasmids in parallel with experimental plasmid conditions (as indicated above in section 2.2.1) and determine the relative number o f cells expressing G F P after 24 h. Cells were visualized using fluorescence microscopy using a fluoroscein isothiocyanate (FITC) channel (488 nm filter). For several of the melanoma cell lines we use, routine transfection efficiencies approaching 70% is possible (Figure 2.1 and Table 2.1). For a more efficient method of determining exactly which cells express the 'gene-of-interest', we are currently using more elaborate bicistronic vectors coupled with G F P to analyze some of our cell systems (Figure 2.2). This vector contains an internal ribosome entry site (IRES) sequence from the encephalomyocarditis virus ( E C M V ) that permits the translation of two open reading frames from a single messenger R N A (Jackson et al., 1990). 33 Figure 2.1 Fluorescent evaluation o f transfection efficiency. We use both a standard p E G F P -N l ( B D Clontech) and a bicistronic vector (Figure 2.2) to assess G F P fluorescence in transfected cell lines. (A) FIEK293 cells transfected with a G F P plasmid. Cells were viewed and photographed under 100X magnification using a fluorescence and white light microscope. The number o f stained and unstained cells is estimated within a given area and an approximate percentage obtained. (B) Image of dendritic M e W o melanoma cells at 400X magnification exhibiting G F P fluorescence. B 34 Table 2.1 Relative transfection efficiencies of our commonly used cell lines. The cell line, its origin, known pigment expression, p53 status, position of mutation and percentage of transfection by G F P plasmid is indicated as described in the Figure 2.1 legend and Materials and Methods, section 2.2.2. T Y R , tyrosinase expression; TRP-1 , TRP-1 expression. Cell Line Origin TYR a (+/-) TRP-13 (+/-) P53 Status Mutation Position Transfection Rate (%)b M M A N Melanotic melanoma + + Wt N A 40 M M R U Melanotic melanoma + + Wt N A 65 R P E P Radial phase melanoma + + Wt N A 35 SK-mel-2 Thigh skin metastasis + + Mut 245 G/S 25 SK-mel-5 Axi l la ry node metastasis + + Wt N A 25 SK-mel-28 Amelanotic melanoma - - Mut 145 C / V 40 SK-mel-110 Amelanotic melanoma - - Mut 9 missense 30 P M W K Primary in situ melanoma + + Mut 244 G / A 50 M e W o Lymph node metastasis + + Mut 258 G / A 45 B16 Mouse melanotic melanoma + + Wt N A 55 H E K 2 9 3 Ad5 -transformed embryonic kidney - - Wt N A 75 H e L a Cervical carcinoma - - Wt N A 60 H460 Lung carcinoma - - Wt N A 40 NIH3T3 Mouse embryonic fibroblasts - - Wt N A 55 Mouse Fb Dermal fibroblasts - - Wt/Nul l N A 30 a = Predicted expression b = Approximate transfection rate 35 Figure 2.2 Plasmid map o f the pCIneoIRES-GFP vector. We also use a bicistronic vector to assess G F P fluorescence in transfected cell lines. Intron NeoR 36 2.2.3 Cytotoxicity (Cell Survival) Assay Cells were plated into 96-well microtitre plates at l x l 0 4 / w e l l , grown for 24 h and treated with different concentrations of anti-cancer drugs for 48 h. Relative toxicity due to growth inhibition was determined by the sulphorhodamine B (SRB) (Sigma-Aldrich) assay as described elsewhere (Skehan et al., 1990; L i et al., 1998). S R B is a bright pink aminoxanthene dye with two sulfonic groups. Under mildly acidic conditions, S R B binds to protein basic amino acid residues in trichloroacetate (TCA)-fixed cells to provide a sensitive index o f cellular protein content. Briefly, the cells were fixed with 10% T C A for 1 h at 4°C, rinsed 5 times with water, and air-- dried. Cells were then stained with 0.4% S R B in 1% acetic acid for 30 min. After rinsing 4 times with 1% acetic acid and air dried, 100 u L of 10 m M Tris (pH 10.5) was added to the wells for 30 min. The colourimetric reading was carried out in a microplate autoreader (Bio-Tek Instruments Inc., Winooski , V T ) at 550 nm. 2.2.4 Enzyme-Linked Immunosorbent Assay (ELISA) of Apoptosis Cells were seeded into 96-well microtitre plates at a density of 2 x l 0 4 per well , grown for 24 h, and exposed to anti-cancer drugs for 24 h. The microtitre plates were centrifuged at 2000xg for 10 min. The E L I S A was performed using a Cel l Death Detection E l i s a p l u s kit (Roche, Laval , QC) according to the manufacturer's protocol. Briefly, the cells were resuspended in 200 u L lysis buffer and incubated for 30 min at room temperature. The lysate was centrifuged at 2000xg for 10 min and 150 u L of supernatant was collected. Then 20 u L of the supernatant was incubated with anti-histone-biotin and ant i -DNA peroxidase at room temperature for 2 h. After washing the wells three times with incubation buffer, 100 u L o f substrate solution (2,2'-azino-bis[3-ethylbenzthiazoline-sulphonic acid]) was added to each well for 15-30 min. The colourimetric analysis was carried out in a microplate autoreader at 405 nm. 37 2.2.5 D N A Content Analysis by Flow Cytometry Confluent cells ( ~ l x l 0 6 ) were harvested by trypsinization, neutralized in D M E M containing 5% F B S , then washed twice in cold phosphate-buffered saline (PBS) / 5 m M E D T A (PBSE). Ce l l pellets were resuspended after centrifugation (lOOOxg for 3 min) in 250 u.L cold P B S E followed by slow addition o f 750 u L cold 100% ethanol. Cells were mixed well and fixed for a minimum of 1 h at 4°C. After fixation, cells were washed twice in cold P B S then stained in the dark with 1 m L P B S containing 50 pg/mL propidium iodide (PI) (Sigma- Aldrich) and 20 pg/mL RNase A (Invitrogen) for 20 min at 37°C. Cel l suspensions were then run on a Coulter Epics Elite flow cytometer (Beckman Coulter, Mississauga, ON) and analyzed with multicycle (Phoenix software). 2.2.6 D N A Fragmentation by Gel Electrophoresis Cells were seeded at a density of 2x10 6 / l 00mm dishes for 24 h, and exposed to the cytotoxic agent for the indicated time (CPT for 24 h; P U M A for 24 h; curcumin for 12, 24, or 48 h). Both the detached and the attached cells were collected and pelleted by centrifugation. The cells were incubated in 500 u L o f a lysis buffer containing 100 m M N a C l , 5 m M E D T A , 10 m M T r i s - H C l (pH 8.0), 0.5% sodium dodecyl sulphate (SDS) and 1 mg/mL proteinase K at 54°C for 3 h. D N A was extracted with phenol-chloroform and ethanol precipitation. R N A was removed by digestion with pancreatic RNaseA. The D N A was analyzed on 2% agarose gel for D N A fragmentation. The gels were stained with 0.5 pg/mL ethidium bromide (EtBr) and photographed under ultraviolet light with a Gel Print System (Biophotonics Co. , A n n Arbor, MI) . The intensity of the bands was quantified with a phosphoimager (Bio-Rad, Mississauga, ON) . 38 2.2.7 Annexin-V Staining by Flow Cytometry Cells were grown in 60-mm plates until 90% confluent then treated with 60 u M curcumin for 0, 6, or 12 h and trypsinized and counted. Cells were stained with Annexin-V-FITC as per manufacturer's directions (BD Pharmingen, Mississauga, ON). Briefly, cells were washed 2 X in cold P B S and resuspended in binding buffer at l x l 0 5 cells/100 uL. Cells were stained with 5 u L Annexin-V and 10 p L PI then incubated for 15 min at room temperature in the dark. About 400 u L of binding buffer was added before the cells were analyzed by flow cytometry. 2.2.8 Fluorescence Microscopy Cells were seeded to 35-mm plates with coverslips at l x l 0 5 cells, grown for 24 h followed by treatment with 60 p M curcumin for 0, 12, 24, or 48 h and stained with PI. Cells were washed 3 times with P B S and fixed in 1:1 acetone:methanol at -20°C for 10 min. They were air-dried then treated with 50 ug/mL PI containing 20 p.g/mL RNaseA for 20 min at room temperature. They were washed 3 times with P B S then air-dried and visualized using an epifluorescent inverted microscope fitted with a 510-560 nm wide-green excitation filter set (Nikon, Tokyo, Japan). For PI cell counts, 3 random fields under 400X magnification (approximately 300 cells) were scored in triplicate for apoptotic cells and expressed as % apoptotic cells/total cells. For Fas aggregation, cells were seeded as above but treated with 60 u M curcumin for 0 or 6 h before fixing then blocked for 15 min in 1% bovine serum albumin (BSA) /PBS . Cells were incubated with either Fas- (BD Pharmingen) or integrin p i - F I T C (Transduction Laboratories, Mississauga, ON) for 1 h at R T , washed and air-dried as above. Cells were then viewed using an epifluorescent inverted microscope fitted with a 420-490 nm blue excitation filter set and images taken using a digital camera (Minolta, Richmond, B C ) . 39 2.2.9 M i t o C a p t u r e ™ Mitochondrial Transmembrane Potential Detection Disruption o f the mitochondrial transmembrane potential was detected using a MitoCapture™ Apoptosis Detection K i t (Calbiochem, San Diego, C A ) . The assay was performed according to the manufacturer's specifications. Briefly, cells were grown in 35-mm plates and treated at 80% confluency with 500 n M D O X for 24 h. Following treatment, the medium was removed and the cells were incubated with 1 m L of diluted MitoCapture™ reagent at 37°C in a 5% C O 2 incubator for 15 min. After incubation, the dye solution was removed and the cells were washed twice with 1 m L of the pre-warmed incubation buffer. The cells were then observed immediately under a fluorescence microscope and images taken using a digital camera. 2.2.10 Caspase Inhibition Cells were seeded to 96-well microtitre plates at l x l 0 4 / w e l l , grown for 24 h and pre-treated for 3 h with 50 u M of the caspase-3 inhibitor, A c - D M Q D - C H O , caspase -8 inhibitor, Z - I E T D - F M K , caspase-9 inhibitor, Z - L E H D - F M K , or broad-based caspase inhibitor, Z - V A D - F M K (Calbiochem) followed by the addition of 60 u M curcumin for 24 h. Ce l l survival was determined by the S R B assay as described on section 2.2.3. 2.2.11 Low-Temperature Culture Inhibiting Receptor Aggregation Cells were seeded on to 96-well microtitre plates at l x l 0 4 / w e l l , grown for 24 h and placed at 4°C for 30 min. They were treated with curcumin at 0, 30, 60, 100, or 300 u M for 1 hr at 4°C or 37°C, washed twice with P B S and incubated with fresh medium for 24 h. Alternatively, cells were seeded as above, placed at 4°C for 30 min and treated with curcumin at 0, 30, 60, or 100 p M at either 4°C or 37°C for 24 h. Cel l death was determined by S R B assay as described in section 2.2.3. 40 2.2.12 Northern Blot Analysis Total R N A was extracted with Trizol reagent (Invitrogen) and the concentrations were determined by U V spectrophotometry. Samples were heated to 65°C and run on 1% agarose gels containing formaldehyde and 0.5 pg/mL EtBr. After separation, capillary transfer to nylon membrane was performed overnight at room temperature and its efficiency assessed by U V light. The blot was then cross-linked with the U V cross-linker ( U L T R A - L U M Inc, Paramount, C A ) . Pre-hybridization was carried out by incubating the blot with the Rapid-Hyb buffer (Amersham-Pharmacia, Montreal, QC) at 42°C for 15 min. The 340 bp Fas ligand (FasL) probe was generated by P C R (polymerase chain reaction) amplification of exon 1 fragment from M M R U genomic D N A (forward primer 5' - A T G C A G C A G C C C T T C A A T T A C C - 3 ' and reverse primer 5 ' - G T T C T G C C A G C T C C T T C T G T A G - 3 ' ) , and then labeled with oc - 3 2 P[dCTP] using the Random Prime Labeling System-Redi prime II according to the manufacturer's protocol (Amersham-Pharmacia, Montreal, QC). Hybridization was performed by incubating the blot with the labeled probe at 42°C for 16-24 h. Filters were washed with 2 X SSC/0 .1% SDS twice for 5 min and then 0.1X SSC/0.1% SDS twice for 15 min at 42°C, and exposed to Kodak M R single emulsion X-ray films (Kodak Canada, Toronto, ON) . 2.2.13 Reverse Transcriptase ( R T ) - P C R Total R N A was extracted with Trizol reagent (Invitrogen) and quantitated by U V spectrophotometry. First-strand c D N A was prepared from 5 pg of total R N A reverse-transcribed into c D N A in the presence of 10 Units /pL o f S U P E R S C R I P T II Rnase H" Reverse Transcriptase (Invitrogen), 5 X first strand buffer (250 m M Tr i s -HCl (pH 8.3), 375 m M K C L , 15 m M M g C l 2 ) , 100 m M D T T , 10 m M dNTP M i x (10 m M each of d A T P , dGTP, dCTP and dTTP at p H 7.0), and 2 pmol of oligo-d(T) N primer (Sigma-Aldrich) in a total volume of 20 pL . The R T mix was 41 then incubated at 42°C for 1 hr. The reaction was inactivated by heating at 70°C for 15 min. For semi-quantitative P C R , equal amounts of c D N A samples (2 u L or 10% of the first strand reaction) were used for the 20 u L of P C R reaction contained, 10X P C R Buffer (200 m M Tris-HC1 (pH 8.4) and 500 m M KC1), 50 m M M g C l 2 , 10 m M dNTP M i x , 5 Units /uL of the Tag D N A polymerase (Qiagen), and 25 pmol/uL of the specific primers: Mdrl a forward, 5'-C C C A T C A T T G C G A T A G C T G G - 3 ' , Mdrl a reverse, 5 ' - T C C A A C A T A T T C G G C T T T A G G C - 3 ' (500 bp); Mdrlb forward, 5 ' - T G C T T A T G G A T C C C A G A G T G A C - 3 ' , Mdrlb reverse, 5'-T T G G T G A G G A T C T C T C C G G C T - 3 ' (450 bp); GAPDH forward, 5'-C T C A T G A C C A C A G T C C A T G C C A T C - 3 ' , GAPDH reverse, 5'-C T G C T T C A C C A C C T T C T T G A T G T C - 3 ' (270 bp). Amplification was carried out as follows: 1) initial denaturation at 94°C for 3 min, 2) denaturation at 95°C for 45 sec, 3) annealing at 55°C for 45 sec, 4) polymerization at 72°C for 45 sec, 5) repeat of step 2 to step 4 for the indicated cycles (25, 30, or 40), and 6) final polymerization at 72°C for 5 min. Samples were then electrophoresed on 1% agarose gels containing 0.5 pg/mL EtBr and visualized under U V light as described in section 2.2.6. 2.2.14 Western Blot Analysis Cells were washed 3 times with P B S , lysed and sonicated in triple-detergent buffer (50 m M Tris-HC1 (pH 8.0), 150 m M N a C l , 0.02% N a N 3 , 0.1% SDS, 1% NP-40, 0.5% sodium deoxycholate, 100 pg/mL phenylmethylsulfonyl fluoride, 1 ug/mL aprotinin, 1 pg/mL leupeptin, and 1 pg/mL pepstatin A ) for 20 min on ice. The lysate was centrifuged at 12,000xg for 10 min and the supernatant collected. The protein concentration was determined with Bio-Rad Protein Assay Reagent (Bio-Rad). Protein (100 pg) was separated by SDS-polyacrylamide gel electrophoresis ( P A G E ) and electroblotted on to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). 42 Membranes were incubated with primary anti-sera overnight at 4°C with shaking. Blots were washed 3 times in P B S - T (PBS and 0.04% Tween) for 5 min then incubated with horseradish peroxidase (HRP)-conjugated secondary anti-sera for 1 h at room temperature. Signals were detected with SuperSignal E C L (Pierce, Rockford, IL). Antibodies used for Western blotting were p53 (CM-1) which recognizes both wild-type and mutant p53 (Dimension, Mississauga, ON) , Bax and Bcl-2 (StressGen, Victoria, B C ) , Bak, P-gp, p 2 1 w a f l , P K C a , P K C 0 , P K C ^ P K C s , p50, and p65 (Santa Cruz Biotechnology, Santa Cruz, C A ) , X I A P , caspase-3 and -8, P-actin, Apaf-1 (BD Pharmingen, Mississauga, ON) , caspase-9 and P A R P (Calbiochem), and Fas (Oncogene, Cambridge, M A ) . A l l densitometric values were obtained using the Quantity One software (Bio-Rad). For the respective blot, identical regions were assessed in triplicate and the control lanes were assigned a basal value of 1 with respect to the p-actin levels. The relative fold induction/suppression was then calculated with repect to the control lanes. 2.2.15 Chloramphenicol Acetyltransferase ( C A T ) E L I S A We used a colourimetric method for the quantitative determination of C A T expression (Roche). The C A T E L I S A is based on the sandwich E L I S A principle in which ant i -CAT antibodies are pre-bound to the surface of the microplate. Briefly, cells were seeded to 35mm/6-well plates and transfected for 24 h with either a C A T control plasmid, WT-Mdrlb::CAT, or MXJT-Mdrlb-CAT. After treatment, cells were washed 3 times with ice-cold I X P B S then lysed in 0.5 m L of I X lysis buffer for 30 min at room temperature. The cell lysate was transferred to a 1.5 m L eppendorf and centrifuged at 12,000xg for 10 min at 4°C. The supernatant was removed and an aliquot (10 uL) was used to determine protein concentration as for Western analysis as described in section 2.2.12. Approximately 150 pg of protein were used in a 200 p L volume and incubated in a well o f the microplate for 1 h at 37°C. Wells were washed 5 times with 250 p L of washing 43 buffer, followed by the addition of 200 u L of digoxigenin (DIG)-labeled ant i -CAT and incubated for 1 h at 37°C. The wash step was repeated with 250 uL of washing buffer for 5 times and then 200 u L of anti-DIG-conjugated-peroxidase was added to the wells and incubated for 1 h at 37°C. Again, the wash step was repeated with 250 u L of washing buffer for 5 times, followed by the addition of 200 u L of peroxidase substrate that was incubated for 15 min at room temperature with gentle shaking on an orbital shaker (250 rpm). The absorbance was read at 405 nm with a microplate autoreader and plotted against a calibration curve of diluted C A T standards as per the manufacturer's instructions. 2.2.16 Breeding, Genotyping, Dissection, and Isolation of Cells from Mice Both C57BL/6 and syngeneic p53-knockout mice were initially purchased from Taconic (Germantown, N Y ) . Wild-type C57BL/6 males and wt C57BL/6 females (6-week old) were mated together to maintain the strain. P53'A males and p53+/~ females were mated together to produce a mixed genotype as p53'A females have very low fecundity. This mating yields approximately 50% homozygous and 50% heterozygous /j>53-defcient offspring. The p53~ /~ mice carry a disrupted, non-functional p53 gene, created by homologous recombination in an embryonic stem cell line and by microinjection of the stem cells into 3.5-day old C57BL/6 blastocysts (Donehower et al, 1992). The litters were weaned at 3 weeks of age and tail tips were removed for D N A extraction and genotyping. Briefly, tail ends were incubated in 500 p i of a lysis buffer containing 100 m M N a C l , 5 m M E D T A , 10 m M T r i s - H C l (pH 8.0), 0.5% SDS and 1 mg/mL proteinase K at 54°C for 3 h. D N A was extracted with phenol-chloroform, ethanol precipitated, and resuspended in 20 p L of T E buffer p H 7.5. P C R was performed using 1-2 u L of genomic D N A and amplified for 30 cycles at 54°C using primer sets specific to either the wild-type p53 gene or the introduced knockout cassette. The following primer set was used for 44 wt p53, forward primer - 5 ' - G T G T T C A T T A G T T C C C C A C C T T G A C - 3 ' and reverse primer -5 ' - A T G G G A G G C T G C C A G T C C T A A C C C - 3 ' that generates a 320 bp band; the primer set for mut p53, forward primer - 5 ' - G T G G G A G G G A C A A A A G T T C G A G G C C - 3 ' and reverse primer - 5 ' - T T T A C G G A G C C C T G G C G C T C G A T G T - 3 ' that generates a 150 bp band (Figure 2.1). Dermal fibroblasts of p53+/+ and p53' /' mice were isolated from 4-week old mice. The mice were sacrificed by cervical dislocation and a 2 x 2 cm skin biopsy were dissected from the dorsal area. The hair was removed and the skin biopsy was disinfected with 2.5% betadine for 1 min, followed by 1 min in 70% ethanol, and washed with phosphate buffered saline (PBS) twice. The skin tissue was then minced and incubated in D M E M containing 200 Units /mL collagenase (Sigma-Aldrich) at 37°C for 6 h. The digested skin tissue was centrifuged at 1000 rpm for 10 min and the pellet washed with pre-warmed D M E M twice. The cells were resuspended in D M E M containing 10% F B S and incubated at 37°C in a 5% C 0 2 atmosphere. 2.2.17 Immunoblotting and Immunohistochemistry on Tissues Twelve-week old p53+/+ and p53'A male mice were sacrificed and the organs removed for protein extraction using a microhomogenizer and lysis buffer with protease inhibitors as previously described (Cheung et al., 2000). Total protein was then separated by S D S - P A G E . Immunohistochemistry was performed on 6-micron sections cut from paraffin-embedded blocks of biopsies from 12-week old male mice. Briefly, tissues were dewaxed by heating at 55°C for 30 min followed by three washes 5 min each in xylenes. Samples were re-hydrated by washing for 5 min each in 100%, 90% and 70% ethanol and 30 min in P B S . Endogenous peroxidase activity was quenched in a 0.3% H 2 0 2 solution for 10 min and unmasking performed by microwaving samples for 4.5 min in a sodium citrate solution (pH 6.0). Samples were blocked for 20 min in non-specific rabbit serum and immunolabeled using the polyclonal anti-P-gp rabbit antibody with the ImmunoCruz staining kit (Santa Cruz Biotechnology). Signals were 45 Figure 2.3 Genotypic analysis of p53 littermates. Tai l tips were removed for genomic D N A extraction as described in Materials & Methods, section 2.2.16. A litter of four animals was analyzed to determine which alleles they carry. The wild-type p53 (WT) and mutant p53 ( M U T ) primers sets were used in separate reactions. Animals 1 and 2 are heterozygous while animals 3 and 4 are homozygous mutant. L = 100 bp ladder; W = wtp53 control (320 bp); N = no D N A control; M = homozygous mut p53 control (150 bp). L W N M 1 2 L 3 4 46 developed using a D A B substrate in a hydrogen peroxide buffer (Vector Laboratories, Burlington, ON) . 2.2.18 Luciferase and P-Galactosidase Assays Both the luciferase and p-galactosidase (P-gal) activity assay kits were obtained from Sigma-Aldr ich. Briefly, cells were seeded to 6-well plates and co-transfected with either the pTyrex-2 plasmid or a control pGL3 plasmid and a basic LacZ-containing plasmid for 24 h. After treatment, cells were washed 3 times with ice-cold P B S then lysed in 200 u L of I X lysis buffer for 15 min at room temperature. Cells were scraped into 1.5 m L eppendorfs and centrifuged at 12000xg for 1 min at 4°C, and then 20 u L o f supernatant was removed and stored in wells o f a microtitre plate on ice for analysis. To the wells was added 100 u L of luciferase assay substrate (coenzyme A , A T P , and luciferin). Plates were immediately analyzed in a TR717 microplate luminometer (Applied Biosystems, Foster City, C A ) for 30 sec integration times. Fluorescence values were standardized to the P-gal control. For p-gal measurement, 150 u L of the cell lysate was removed and an equal volume of 2 X assay buffer (200 m M N a H P 0 4 , 2 m M M g C l 2 , 100 m M p-mercaptoethanol, 1.33 mg/mL o-nitrophenyl p-D-galactopyranoside [ONPG]) was added. This was incubated at 37°C for 30 min or until a yellow colour developed then the reaction was stopped with 500 u L stop solution (1 M NaCOa). Absorbance was read at 405 nm on a microplate autoreader and the activity calculated. 47 CHAPTER 3. ROLES OF P 5 3 IN MELANOMA CHEMOSENSITIVITY 3.1 R A T I O N A L E A N D H Y P O T H E S I S The p53 gene is one of the most frequently mutated loci in human cancers. Overexpression of the p53 tumour suppressor is frequently observed in metastatic malignant melanoma ( M M ) but to a lesser degree in primary melanoma (Lassam et al, 1993; Cristofolini et al., 1993; L i et al., 1995; Hartmann et al., 1996). D N A sequencing studies have been variable but suggest overall that mutations occur within the p53 gene in about 20-30% of metastatic melanomas (Weiss et al., 1995). In primary melanoma this rate appears reduced (Levin et al., 1995), suggesting that mutation of the p53 gene may play a role in the progression of M M (Sparrow et al., 1995; Hartmann et al., 1996). To support this statement, recent studies have shown that p53 overexpression is associated with increased depth o f invasion o f melanoma (both Breslow thickness and Clark's levels), tumour ulceration, and poor prognosis (Korabiowska et al., 1995; Yamamoto et al, 1995; Weiss et al, 1995). Malignant melanoma is a life-threatening skin cancer that metastasizes very rapidly. Metastatic M M is usually incurable, with a 5-year survival <10% and a poor response to chemotherapy (Koh, 1991; Roses et al, 1991). A s stated previously, the reason for the cellular resistance mechanisms contributing to melanoma chemoresistance after anticancer drug treatment have not yet been well-defined. Recently, accumulated evidence supports the model that anticancer drugs exert their cytotoxicity predominantly by inducing apoptosis in tumour cells (Kamesaki, 1998; Sekiguchi et al, 1998). Therefore, it seems likely that dysregulation of the genes involved in the process of apoptosis may contribute to the chemoresistance. P53 can mediate apoptosis through the transcriptional regulation of target genes, in particular, the Bcl-2 family. Because the ratio of anti-apoptotic Bcl-2/pro-apoptotic Bax can 48 dictate cell survival/death, this suggests that an imbalance between Bax/Bcl-2 proteins may contribute to chemo- and radioresistance. Several studies now describe how p53 controls the Bax/Bcl-2 ratio by up-regulating Bax and suppressing Bcl-2 (Raisova et al., 2001). Bcl-2 protein is thought to prevent many types of apoptotic cell death, whereas Bax protein heterodimerizes with Bcl-2 through its various B H domains and promotes apoptosis (Oltvai et al., 1993). The relationship of p53 protein as a regulator of both Bcl-2 and Bax expression has been determined both in vitro and in vivo (Miyashita et al., 1994). A number of groups have now determined the significance o f Bcl-2 overexpression in melanoma (Grover and Wilson, 1996). Moreover, this supports the development of many clinical trials underway using Bcl-2 antisense strategies to treat advanced melanoma disease (Jansen et al., 1998; Olie et al., 2002). Thus, the dysregulation of Bax/Bcl-2 by mutant p53 protein may lead to changes in the apoptotic rate in melanoma after chemotherapy. We feel this warrants investigations into improved therapy regimes using the general hypothesis that pro-apoptotic mechanisms (e.g. p53) can mediate the inherent chemosensitivity in melanoma cells. Our group has previously compared the response rate to anticancer drugs of the melanoma cell lines that have different p53 status (Li et al, 1998). It was found that melanoma cell lines, which carry a wild-type p53 gene have much higher response rates to chemotherapy drugs than cell lines with a mutant p53 gene (Li et al., 1998). To further confirm the role of p53 in melanoma chemosensitivity, we asked the question; does p53 mutational status play a role in the chemoresistance of melanoma cells? We hypothesized that mutation of p53 interferes with the normal cytostatic mechanisms in melanoma cells after anticancer drug treatment. We determined that the best system to address our hypothesis would be a stable, dominant-negative p53 melanoma cell line. We treated these cells with the anticancer drug, camptothecin (CPT) (Figure 3.2A), and then compared the apoptosis rate and the expression of apoptotic genes between the pair of cell 49 lines that either carry a wild-type or overexpress a mutant p53. Camptothecin is a naturally occurring alkaloid that was identified in the 1960s by Wani and Wal l in a screen of plant extracts for antineoplastic drugs (reviewed in Wal l and Wani, 1995). Camptothecins are a class of D N A -damaging agents that bind irreversibly to the DNA-topoisomerase I complex, inhibiting the reassociation of D N A after cleavage by topoisomerase I and traps the enzyme in a covalent linkage with D N A . The enzyme complex is ubiquinated and destroyed by the 26S proteosome, thus depleting cellular topoisomerase I (Desai et al., 1997). We chose C P T as a chemotherapeutic agent because of its relative novelty and the fact that some of its derivatives (topotecan and irinotecan) are FDA-approved and showing potential in clinical trials. A s a class o f drugs, they also possess some o f the best anticancer/toxicity ratios among experimental drugs (Saleem et al, 2000). 3.2 R E S U L T S 3.2.1 M u t a n t p53 C a n Increase the Growth Rate Previously, it was shown that four wild-type p53 melanoma cell lines have higher response rates to chemotherapy than four mutant p53 melanoma cell lines (L i et al, 1998). To further evaluate that the difference in chemosensitivity between wild-type p53 and mutant p53 melanoma cell lines is due to wild-type p53-induced apoptosis, we transfected the wild-type p53 melanoma cell line M M A N (Li et al, 1995) with an expression vector carrying a DNA-bind ing region mutant p53 gene called, p E D l . After selection with G418, the transfectants that carry mutant p53 genes were expanded into a cell line, designated as M M A N - p E D l . M M A N cells transfected with vector alone were established as controls and are designated as M M A N - v e c t o r . Western blot analysis, using a polyclonal p53 antibody C M 1 , that recognizes both wild-type and mutant conformations of p53, showed that M M A N - p E D l cells have higher levels of p53 protein 50 Figure 3.1 Overexpression of mutant p53 on growth rate of the melanoma cell line M M A N . (A) p53 level in M M A N - p E D l and MMAN-vector cells by Western blotting using the CM1 antibody. (B) M M A N cells containing either pEDl or vector alone were seeded at 6x l0 4 cells/35 mm petri dishes, and cultured for 4 days. The cells were counted with a hemocytometer at 24-hour intervals. vector pED1 51 compared to M M A N alone, indicating that the mutant p53 gene is expressed in the transfectants (Figure 3.1A). To investigate whether overexpression of the mutant p53 protein has any effect on cell growth, we compared the growth rate of M M A N - p E D l and M M A N - v e c t o r cells. Figure 3.IB shows that the number o f M M A N - p E D l cells increased by 37% compared to M M A N -vector cells after a 4-day culture (P < 0.01, Student's paired Mest). 3.2.2 CPT- Induced Apoptosis Is p53-Dependent We treated the M M A N - p E D l and M M A N - v e c t o r cells with the anticancer drug C P T (Figure 3.2A) and compared the chemosensitivity between these two cell lines. The S R B cytotoxicity assay indicates that cells bverexpressing mutant p53 are significantly more resistant to C P T (Figure 3.2B). After treatment with 30 n M , 100 n M , or 300 n M of C P T for 48 h, the survival rates o f M M A N - p E D l cells were 98%, 64%, and 28%, compared to 67%, 29%, and 11% in the controls (P = 0.004, 0.00002, and 0.004, respectively). The I C 5 0 is increased by three-fold in M M A N - p E D l cells compared to that of M M A N - v e c t o r cells (180 n M vs 60 nM) . To confirm the results of S R B assay, we used trypan blue exclusion to determine the percentage of dead cells after C P T treatment in these two cell types. Figure 3.2C demonstrates that C P T induced significantly more cell death in vector than p E D l cells (P = 0.0005). Because the M M A N - p E D l overexpresses the dominant negativep53 mutant, we treated both cell lines with a new compound, CP-31398 which is able to stabilize the D N A binding domain o f p53 in the active conformation to enhance its transcriptional activity and possibly revert p53 mutants to wild-type by maintaining an active conformation (Foster et al, 1999; Luu et al, 2002). We pre-treated M M A N - v e c t o r and M M A N - p E D l cells for 1 h with CP-31398 before the addition of 300 n M C P T for 24 h. M M A N - p E D l survival was significantly reduced compared to the controls by 20% and this difference was amplified upon drug treatment (Figure 3.3). 52 Figure 3.2 Overexpression of mutant p53 on the sensitivity of M M A N cells to C P T treatment. (A) Structure of CPT . (B) Cel l survival by S R B assay. Cells were seeded into 96-well plates at l x l 0 4 / w e l l for 24 h, and treated with 0, 30, 100, or 300 n M of C P T for 48 h (see Material and Methods, section 2.2.3). (C) Trypan blue exclusion assay. Cells were seeded into 60 mm dishes at 5x l0 5 /d i sh for 24 h, and treated with 100 n M C P T for 48 h. The floating and attached cells were pooled and resuspended in 0.4% trypan blue/PBS. Both viable and dead cells were counted with a hemocytometer. Figure 3 . 3 Pre-treatment o f M M A N - v e c t o r and M M A N - p E D l cells with the p53-stabilizing compound, CP-31398. Cells were seeded into a 24-well plate at l x l 0 5 / w e l l , treated for 1 h with 8 pg/mL CP-31398 before the addition of 300 n M C P T for 24 h. Ce l l survival was determined by S R B assay. 54 To investigate whether the resistance of M M A N - p E D l cells to chemotherapy is due to a reduced apoptotic rate, we used E L I S A and D N A fragmentation assays to examine the anticancer drug-induced apoptosis in M M A N - p E D l and M M A N - v e c t o r cells. Figure 3.4A shows that the apoptosis rate was reduced by 52% and 57% in M M A N - p E D l cells at 30 n M and 100 n M , respectively (P = 0.005). To further confirm that anticancer drug-induced apoptosis is reduced in M M A N - p E D l cells, M M A N - v e c t o r and M M A N - p E D l cells were treated with 0, 30, and 100 n M of C P T for 24 h. Attached and floating cells were pooled and the D N A samples were analyzed by gel electrophoresis. In Figure 3.4B, we demonstrate that D N A fragmentation is reduced by 3.5-fold in M M A N - p E D l cells (phosphoimager scanning). In addition, we performed flow cytometry to determine i f mutant p53 renders resistance to CPT-induced apoptosis. Figure 3.5 demonstrates that mutant p5J-containing M M A N - p E D l cells have significantly less apoptosis than M M A N - v e c t o r cells. A t 30 n M and 100 n M , the apoptosis rate is 2.5 and 3.8-fold lower in p E D l cells compared to M M A N - v e c t o r cells. To establish the relationship between CPT-induced apoptosis and cell cycle arrest, we performed cell cycle analysis on CPT-treated M M A N - v e c t o r and M M A N - p E D l cells. Table 3.1 demonstrates that C P T induced S/G2 arrest in both M M A N - v e c t o r and M M A N - p E D l cells, indicating that the S/G2 arrest is p53-independent. Since p53 is a transcriptional regulator, we examined which genes are regulated by p53 after anticancer drug treatment. We assessed the expression of p 2 1 w a f l , Bax, Bc l -2 , Bak, and P-glycoprotein in M M A N - v e c t o r and M M A N - p E D l cells before and after 100 n M of C P T treatment. We found that p53 expression was dramatically increased after C P T treatment in both cell lines while Bax expression was only modestly induced in the M M A N - v e c t o r cells. The expression of Bcl-2 and P-gp was significantly reduced in M M A N - v e c t o r but not in M M A N -p E D l cells after C P T treatment, while the expression of Bak and p 2 1 w a f l did not significantly change (Figure 3.6). 55 Figure 3.4 Overexpression o f mutant p53 on CPT-induced apoptosis and D N A fragmentation. (A) Cells were seeded into a 96-well plate at 2 x 10 4/well, treated with 30 or 100 n M C P T for 24 h, and the apoptosis rate was determined by colorimetric analysis using a Ce l l Death Detection E L I S A p l u s kit (Material and Methods, section 2.2.4). (B) Cells were seeded at a density o f 2 x 106/100 mm dishes for 24 h, and exposed to 0, 30, 100 n M of C P T for 24 h. The D N A was extracted from the cells and electrophoresized on a 2% agarose gel. (0 "5 o a. o o. < c o 3 73 33 O control C P T B vector pED1 MW 100 nM 56 Figure 3 . 5 The effect o f mutant p53 on CPT-induced apoptosis by flow cytometry. Cells were treated with C P T for 24 h and analyzed by flow cytometry as described in Material and Methods, section 2.2.5. The percentage of the apoptotic cells with hypodiploid D N A content is shown by the brackets over the sub-Go/Gi peak. Cel l number is shown on the y axis. PI, propidium iodide. VO, V30 , V100: M M A N - v e c t o r cells treated with C P T at 0, 30, or 100 n M . P0, P30, P100: M M A N - p E D l cells treated with C P T at 0, 30, or 100 n M . 57 3.2.3 Antisense Bcl-2 Enhances the P53-Dependent Effect of C P T Treatment In the previous section, we found that Bcl-2 was suppressed in the M M A N - v e c t o r cells. To further evaluate this mechanism, we used an alternative approach to enhance C P T treatment. Pre-treatment of M M A N - v e c t o r and M M A N - p E D l cells by transfection with 1500 n M of a Bcl-2 antisense O D N before addition o f C P T could significantly enhance the cell death associated with C P T and reduce the cell survival in the M M A N - v e c t o r cells (Figure 3.7). After Bcl-2 O D N treatment, survival was 74% without C P T and 39% with 100 n M C P T in M M A N - v e c t o r cells after 48 h (P = 0.009 and P = 0.007, respectively). In M M A N - p E D l cells, survival was 94% without C P T and 78% with 100 n M C P T (P = 0.06 and P > 0.05, Student's Mest). 3.3 D I S C U S S I O N 3.3.1 P53 Mutation Correlates with Reduced Chemosensitivity Melanoma is a life-threatening skin cancer that has high potential for metastasis. Unfortunately, there is still no cure for metastatic melanoma. One major obstacle in melanoma treatment is the resistance of metastatic melanoma to chemotherapy. Many studies have indicated that mutation of the p53 gene is associated with increased depth o f invasion, low response rate of chemotherapy, and poor prognosis in M M (Korabiowska et al., 1995; Yamamoto et al., 1995; Weiss et al., 1995). Previously, we have demonstrated that melanoma cell lines carrying a wi ld-type p53 gene have greater response rates to various types of anticancer drugs than mutant p53 melanoma cell lines (L i et al., 1998). The objective of this study was to further investigate the importance o f the p53 function in chemotherapy-induced apoptosis o f melanoma cells. We chose C P T as the chemotherapeutic agent in this study because C P T is known to induce apoptosis in various types of tumour cells and C P T has shown potential in clinical trials (Pantazis, 1995; Stewart et al, 1996; Muggia et al, 1996; Rothenberg et al, 1997). In addition, 58 Table 3.1 Ce l l cycle analysis of CPT-treated M M A N - v e c t o r and M M A N - p E D l cells by flow cytometry. Cells were treated with C P T for 24 h. Data represent mean ± SD from triplicates. C e l l Type C P T Dose (nM) % G l % S % G 2 / M M M A N - v e c t o r 0 69.3 ± 4 . 2 26.1 ± 3 . 0 4.6 ± 0 . 9 30 14.2 ± 1.1 57.5 ± 5 . 6 28.3 ± 3 . 1 100 10.7 ± 1.3 40.9 ± 3 . 5 48.4 ± 5 . 0 M M A N - p E D l 0 68.8 ± 5 . 6 29.0 ± 2 . 8 2.2 ± 0 . 3 30 10 ± 0 . 9 44.8 ± 4 . 1 45.2 ± 5 . 2 100 6.1 ± 0 . 8 55.6 ± 4 . 7 38.3 ± 3 . 6 59 Figure 3.6 The effect of mutant p53 on Bax, Bcl-2 , p 2 1 w a , and P-gp protein expression. M M A N - v e c t o r and M M A N - p E D l cells were treated with or without 100 n M C P T for 24 h. Proteins were extracted and subjected to Western blot analysis. Densitometry values were obtained using the Quantity One software (Bio-Rad) and the fold induction/suppression is indicated with respect to the control lane (see Materials and Methods for details, section 2.2.14). Vector pED1 0 100 0 100 nM 1 63.7 6.4 29.3 1 1.8 0.9 0.8 1 0.3 1.1 1.1 1 1.2 0.7 0.8 1 1 0.8 0.8 1 1 0.2 0.5 0.4 p53 Bax Bcl-2 Bak p21 waf1 P-gp p-act in 60 Figure 3.7 Pre-treatment of M M A N - v e c t o r and M M A N - p E D l cells with an antisense Bcl-2 O D N . Cells were seeded into a 24-well plate at l x l 0 5 / w e l l and transfected with a Bcl-2 antisense O D N (1500 nM) for 24 h as described in Materials and Methods, section 2.2.1. This was followed by 100 n M C P T treatment for 24 h. Cel l survival was determined by S R B assay. 120 61 C P T demonstrated a broad spectrum of antitumour activity due to the lack o f clinical cross-resistance with existing antineoplastic compounds (Rothenberg et al., 1997). Our results indicate that overexpression of a mutant p53 gene increased the growth rate of M M A N cells (Figure 3.IB), reduced the response rate to chemotherapy (Figures 3.2 and 3.3), and reduced the apoptosis rate after C P T treatment (Figures 3.4 and 3.5). Reduced chemosensitivity was also observed in mutant p53 cells after treatment with another anticancer drug vincristine (data not shown). The addition of a p53-stabilizing compound could enhance the chemosensitivity in the M M A N - p E D l cells more so than the parental M M A N perhaps through rescue o f the mutant p53 function leading to increased cell death (Figure 3.3). Furthermore, overexpression of a dominant negative mutant p53 abolished the down-regulation of Bcl-2 and P-gp in response to C P T treatment (Figure 3.6). However, the dominant negative mutant was not significantly affected by suppressing Bcl-2 expression with an antisense O D N (Figure 3.7). 3.3.2 CPT- induced Cell Death Is Gl-Independent It is known that p53 transcriptionally up-regulates p 2 1 w a f l , which is a cyclin-dependent kinase inhibitor, and therefore arrests cells in G l phase o f the cell cycle (Haffner and Oren, 1995). However, there is no significant change in p 2 1 w a f l expression between wild-type and mutant p53 cells before or after C P T treatment (Figure 3.6), indicating that CPT-induced cell death is not related to p53-induced G l arrest. Cel l cycle analysis clearly demonstrates that C P T induced a p53-independent S/G2 phase arrest, not G l arrest (Table 3.1). Our data support the findings that: (1) C P T induces break points in late S and G2 phases (Bassi et al., 1998); (2) increased sensitivity to C P T is not associated with the loss of a G l checkpoint (Slichenmyer et al., 1993); and, (3) C P T treatment results in an accumulation of cells in G 2 / M phase (Jones et al, 1997). 62 3.3.3 P53 Regulates Expression of Bax/Bcl-2 Recent research has indicated that apoptosis is the primary mechanism of the cytotoxicity for various types of anticancer drugs (Kamesaki, 1998; Sekiguchi et al, 1998). Therefore, the expression of the apoptotic regulators may influence the outcome of chemotherapy. It is well known that p53 regulates the process of apoptosis. In our system, a melanoma cell line overexpressing a mutant p53 gene had a lower rate of apoptosis and supports the model that mutant p53 binds and inactivates the wild-type p53 protein and impairs the apoptotic process. The exact mechanisms as to how p53 triggers apoptosis in melanoma cells after chemotherapy is unknown, but it appears that p53 down-regulates Bcl-2 in melanoma cells to trigger apoptosis. Bcl-2 and Bax are a pair of pivotal genes that control programmed cell death, or apoptosis, with Bax being the apoptosis promoter and Bcl-2 the apoptosis inhibitor. Bax and Bcl-2 form protein dimers with each other and the relative ratio of Bax/Bcl-2 is believed to be a determinant of the balance between life and death within a cell (Oltvai et al., 1993). Studies indicate that unbalanced Bax/Bcl-2 expression alters the responses of tumour cells to chemotherapy. For instance, elevated Bcl-2/Bax is a consistent feature of apoptotic resistance in B-cel l chronic lymphocytic leukaemia and is correlated with in vivo chemoresistance, with the most resistant cells expressing elevated Bcl-2 levels and lower Bax protein (Pepper et al, 1998). Conversely, overexpression of Bax gene sensitizes K562 erythroleukemia cells to apoptosis induced by selective chemotherapeutic agents (Kobayashi et al., 1998) while overexpression of Bcl -2 resulted in protection from drug-induced apoptosis of ovarian carcinoma cells (Eliopoulos et al, 1995). Bcl-2 overexpression and Bax down-regulation was associated with p53 mutant immunophenotype in neuroendocrine lung tumours (Brambilla et al, 1996). The expression of Bcl-2 is p53-dependent in acute lymphoblastic leukemia, and in most wild-type p53 cells ionizing radiation induces down-regulation of Bcl-2 and up-regulation of Bax (Findley et al, 63 1997). However, other studies demonstrate that p53 does not regulate Bcl-2 or Bax expression. In UV-induced apoptotic mouse keratinocytes, the expression of Bcl -2 and Bax is independent of p53 status (L i et al., 1996; Tron et al., 1998). In IL-7 treated pro-T apoptotic cells, increased Bcl-2 and decreased Bax expression were also p53-independent ( K i m et al, 1998). Therefore, it appears that Bax/Bcl-2 regulation by p53 is tissue- and stimulus-specific. We observed a reduced Bcl-2 expression associated with increased p53 in M M A N - v e c t o r cells, indicating that wild-type p53 down-regulates Bcl-2 and eliminates its apoptosis-suppressing effect (Figure 3.6). The down-regulation o f Bcl-2 with an antisense O D N could further enhance the effect o f C P T in the M M A N - v e c t o r cells (Figure 3.7). Overexpression of mutant p53 did not down-regulate B c l -2 (Figure 3.6), was not significantly affected by Bcl-2 elimination (Figure 3.7) and resulted in protection from drug-induced apoptosis (Figures 3.4 and 3.5). This result agrees with the observations by Elipoulos et al. (1995) that overexpression of both Bcl-2 and mutant p53 protected the ovarian cancer cells from drug-induced apoptosis and by Jansen et al. (1998) that Bcl -2 antisense O D N treatment enhances the chemosensitivity o f human melanoma grown in severe combined immunodeficient (SCID) mice. Surprisingly, we did not see a dramatic induction o f Bax despite high levels of p53 expression (Figure 3.6). These drug-treated cells are clearly undergoing apoptosis and it appears to be p53-dependent as there is no change in Bax in the M M A N - p E D l cells; therefore, this suggests several possibilities for the role of Bax: (1) the levels of Bax expression are sufficient to activate apoptosis, (2) Bax induction is not as important for apoptosis in this melanoma cell line as Bcl -2 suppression, (3) other factors or proteins not examined here may be required to help stabilize Bax and influence its function, or (4) the primary mechanism for CPT-induced apoptosis is Bax independent. We favour the latter alternative as Bax-independent apoptosis has been demonstrated in other CPT-treated tumour cells (Chatterjee et al., 1996; Goldwasser et al., 1996). 64 3.3.4 P53 Down-Regulates P-gp Another factor that may influence the outcome of chemotherapy is the expression level o f P-glycoprotein. Recent work suggests that p53 may regulate the expression of Mdrl (reviewed in Bush and L i , 2002). Overexpression of the Mdrl gene correlates with mutant p53 expression in human non-small cell lung cancer (Galimberti et al, 1998), breast cancer (Linn et al, 1996), and colorectal cancer metastases (de Kant et al., 1996). Overexpression of a tows-dominant negative p53 into rodent H35 hepatoma cells produced markedly elevated levels of P-gp and Mdrl a m R N A (Thottassery et al., 1997). While these studies suggest that mutant p53 up-regulates P-gp expression, others indicate that wild-type p53 also up-regulates the promoter function and the endogenous expression of the rat Mdrlb (Zhou and Kuo , 1998). In addition, a p53 consensus binding site has been identified in the Mdrlb promoter region (Zhou and Kuo , 1998). These conflicting reports imply a complex interaction between the p53 and Mdrl -related genes. Our results demonstrate that upon exposure to CPT , a reduction of P-gp expression is associated with increased wild-type p53, but not mutant p53 expression, implying a down-regulation of P-gp by p53. The reduction of P-gp in wild-type p53 cells agrees with our observation that anticancer drugs induced more apoptosis in the wild-type p53 cells (Figures 3.4 and 3.5). Unchanged P-gp expression in M M A N - p E D l cells suggests that the mutant p53 protein itself is unable to bind to the P-gp promoter, but may interfere with the binding of the wild-type p53 to the P-gp promoter (Figure 3.6). Taken together, our results indicate that p53 mutational status can be a determinant of melanoma chemosensitivity. We show that p53 may down-regulate both the apoptotic protector Bcl-2 and the Mdrl gene to trigger apoptosis in melanoma cells after exposure to the anticancer drug, C P T . Mutation of the p53 gene in the DNA-binding domain can interfere with this process to suppress the normal apoptotic mechanisms. 65 CHAPTER 4. CHEMOSENSITIZATION OF MELANOMA CELLS BY PRO-APOPTOTIC GENES 4.1 R A T I O N A L E A N D H Y P O T H E S I S A paradox with melanoma is that although p53 mutation does occur and is associated with increasing stages of melanoma, practically, it is only mutated in 20-30% of melanoma, which clearly suggests that other genes must be involved. A s a means to avoiding apoptosis, suppression of the p53 cell death mechanisms would be a growth advantage for developing melanoma cells. Recent results by Soengas et al. (2001) have established that the p53-mediated pathway is perturbed in metastatic melanoma through the inactivation of the pro-apoptotic Apaf-1 gene. Cytochrome c release from the mitochondria permits the formation of an apoptosome complex with Apaf-1 and the subsequent activation of caspase-9. Inactivation of Apaf-1 may primarily occur by epigenetic mechanisms acting to silence the gene since methylation inhibitors such as 5-aza-2'-deoxycytidine can restore the p53-induced cell death pathway in these cells (Soengas et al, 2001). Interestingly, in eight melanoma cell lines that we examined, we did not find any significant difference in Apaf-1 levels among the cell lines (Figure 4.1 A ) even though half of the melanoma cell lines have p53 mutations and Apaf-1 is a transcriptional target of p53 (Moroni et al., 2001). In comparison to human dermal fibroblasts as control tissue, we did not see any appreciable difference in the melanoma cell lines suggesting that in these cell lines Apaf-1 may not be repressed or inactivated. Furthermore, as shown in Figure 4 . IB , after treatment with D O X , which is known to induce p53 expression, we did not observe significant changes in the expression of Apaf-1. This suggests that the role of Apaf-1 in melanoma may not be as crucial as other factors or that perhaps other factors are required for its function. 66 Figure 4.1 Protein expression of the pro-apoptotic factor, Apaf-1, in melanoma cell lines. (A) Human fibroblasts and multiple melanoma cell lines were assessed for endogenous Apaf-1 expression. (B) Cells were treated with 1000 n M of D O X for 24 h and the expression levels of Apaf-1 and p53 were determined. Cel l lines marked with an asterisk indicate the presence of a p53 mutation, p-actin was used as a control for loading. H Fb M M R U Sk-mel-5 P M W K * Sk-mel-110* M M A N R P E P I MeWo* I S k-mel-2* I Apaf-1 P-actin B H F b M M R U S k - m e l - 5 M e W o * P M W K * S k - m e l - 2 * - + - + - + - + - + - + D O X Apaf-1 p53 P-actin 67 Another point in the apoptotic network that melanomas have exploited is the overexpression of apoptosis inhibitors. The newly characterized apoptotic inhibitor, Survivin, a member of the growing inhibitor-of-apoptosis (IAP) family has reportedly been able to bind to and inhibit the active forms of caspase-3 and -7 (Grossman et al., 2001). In addition, another candidate, L i v i n , was identified that can also physically interact with the processed forms of caspases-3, -7, and -9 independently of Survivin (Vucic et al., 2000; Kasof and Gomes, 2001). To this end, expression of the pro-apoptotic second mitochondria derived activator of caspase ( S M A C / D I A B L O ) is reduced in melanoma cells (Hersey and Zhang, 2001). Escape from TRAIL-induced apoptosis, a member of the T N F superfamily of cytokines that can activate cell death in a range of malignant cell lines, particularly melanoma, but not in normal tissues (Zhang et al., 1999; Thomas et al., 2000) represents yet another level at which melanoma cells promote self-survival. Like other related ligands, T R A I L activates apoptosis through its cognate death receptors, D R 4 and D R 5 , and is dependent on recruitment of mitochondrial pathways by caspase-8. In cultured melanoma cells, T R A I L but not FasL or T N F induces changes in the mitochondrial membrane potential that correlates with apoptosis (Hersey and Zhang, 2001). Overexpression of Bcl-2 and perhaps other members of the anti-apoptotic branch of the Bcl -2 family can block Trail-induced apoptosis (Thomas et al., 2000). Previously Tang et al. (1998) showed that anti-apoptotic regulators B c l - X L and Mcl-1 are overexpressed in M M but not in benign nevi. Based on these findings, we proposed to use a gene therapy or gene transfer approach to sensitize melanoma cells to anticancer drugs. Although alteration of p53 itself may only be part of the mechanism for melanoma chemoresistance, inactivating the upstream or downstream components within the pathways that p53 regulates seems crucial for melanomagenesis. Since p53 can activate a mitochondrial pathway to apoptosis, we hypothesized that overexpression of pro-apoptotic genes acting immediately upstream of the mitochondrion (e.g. Bax, Noxa, P U M A ) or further downstream 68 effectors in the apoptotic pathway (e.g. Apaf-1 and caspase-3) could effectively sensitize M M cells to drug treatment and overcome endogenous resistance. 4.2 R E S U L T S 4.2.1 Bax Sensitizes Melanoma Cells to Anticancer Drug Treatment 4.2.1.1 The overexpression of Bax sensitizes melanoma cells to drug treatment From our previous work, we show that the drug C P T can activate endogenous p53 expression to very high levels (Figure 4.2A). The fact that p53 up-regulates pro-apoptotic Bax suggests that Bax should be a valid candidate to use for exogenous activation of cell death. Our initial studies focused on using Bax expressed in a vector under the control o f a C M V promoter to sensitize melanoma cells such as M M R U that has normal p53. After C P T treatment, there was a massive decrease in cell survival, 28% (P = 0.0003, Student's paired Mest) in those melanoma cells that overexpressed Bax compared to 67% (P < 0.01) in the baseline GFP-transfected control cells (Figure 4.2B). Surprisingly, Bax alone induced only a small decrease in cell survival of about 10%) (P = 0.05) suggesting that additional cofactors or pathways are involved after drug treatment to amplify the Bax signal. 4.2.1.2 Inducible Bax decreases N F - K B The fidelity of cells that transiently express Bax is suitable for short-term studies but these cells are unsuitable for continuous cultures since they are expected to undergo apoptosis at a higher rate. Therefore, we wanted to refine the expression of Bax by using a tetracycline-inducible vector, p S T A R (Zeng et al, 1998), which is a self-contained, single-plasmid expression system. Using this system in the M M R U cell line, we found that increasing concentrations o f tetracycline 69 Figure 4.2 Overexpression of Bax sensitizes melanoma cells to C P T treatment. (A) Western blot of M M R U cells treated with C P T for 24 h. (B) Quantification of cell survival by S R B assay of M M R U transiently transfected with either Bax or G F P expression plasmids for 24 h then treated with 200 n M C P T for 24 h. Note, cell survival values were corrected according to the transfection efficiencies estimated by G F P expression in each control experiment. control CPT B a x 70 Figure 4.3 Bax induction reduces protein expresston of the small subumt (p50) of OT^ffl. Human Bax was subcloned into the p S T A R vector and transiently transfected into the M M A N ceU line Twenty-four hours post-transfection, cells were treated with increasing doses of tetocycline for 48 h then cells were harvested for protein extraction and Western blot analysis. 0 .02 .2 2 20 ug/ml Tet MB - Bax 1 2.1 2.3 3.5 5.3 Fold induction — m mm -p50 1 .81 .85 .55 .23 Fold reduction f ** - Actin 71 lead to a gradual increase in Bax expression that was coincident with a down-regulation of the small subunit (p50) of the pro-survival factor, N F - K B (Figure 4.3). This strong correlation needs to be evaluated further but it does suggest that a general mechanism involving cross-talk between the pro-apoptotic signals for cell death exists that is able to direct a concomitant decrease in pro-survival signals. 4.2.1.3 A tyrosinase promoter can specifically target Bax expression in melanoma cells The use of generic promoters based on the C M V or SV40 viral control regions is useful for high-expression but lacks the targeting capabilities that are essential for improved therapy in cancers. Consequently, we wanted to better target overexpression of Bax to melanoma cells and chose to use a melanocyte/melanoma-specific promoter based on the tyrosinase gene to preferentially treat melanoma cells versus non-melanoma cells (Park, et al., 1999). We subcloned the human Bax c D N A into the pTyrex2 vector creating the pTy-Bax construct (Figure 4.4A). The original plasmid had a luciferase reporter gene downstream of the Tyrosinase promoter so we were able to measure the relative luciferase activity from the melanoma cell lines. We compared the expression of luciferase fluorescence in two non-melanoma cell lines, HeLa and H460, with three melanoma cell lines, M M A N , M M R U and mouse B16 (Figure 4.4B). Luciferase values were dramatically higher in the melanoma cell lines ranging from 29- to 101-fold higher than baseline autofluorescence and this corresponded well with the empirical observations that the respective cell lines had increasing pigmentation M M A N < M M R U < B16 (data not shown). A s can also be seen, strong luciferase activity was detectable in all cell lines with the control luciferase reporter, pGL3 (Figure 4.4B). We transfected either a G F P or the pTy-Bax plasmid into both the M M R U and the HeLa cell lines to demonstrate that we were getting specific expression o f Bax in the melanoma cells and not in the cervical cells as the H e L a were less sensitive to drug treatment (75%, P < 0.01) compared to the M M R U which were significantly 72 Figure 4.4 A tyrosinase promoter/enhancer vector can target expression of Bax specifically to melanoma. (A) Plasmid map of the pTyrex-2 construct into which the luciferase or human Bax gene has been subcloned yielding the pTy-Bax plasmid. (B) Specificity of the tyrosinase promoter (closed bars) for melanoma cell lines versus epithelial cell lines, H460 (lung carcinoma) and HeLa (cervical carcinoma). Relative luciferase activity was measured and the fold induction between cell lines plotted. A control luciferase plasmid (open bars) was used to indicate relative activity. (C) The pTy-Bax or G F P vectors were introduced into both H e L a (open bars) and M M R U (closed bars) cell lines and the cell survival determined after 200 n M C P T treatment for 48 h. 73 A Amp F1 ori LacZ ColE1 ori v Tandem \ \ WISE Tyr promoter B o (A 2 u 3 O c g +5 O 3 73 C 73 o Li-control HeLa H460 MMAN MMRU B16 100 80 A 75 60 > E 3 (0 40 A 20 • HeLa T • MMRU T - • - 1 GFP BAX 74 more sensitive to drug treatment as cell survival was reduced to about 49% (P < 0.005) after standardizing to controls (Figure 4.4C). 4.2.2 P U M A Sensitizes Melanoma Cells to Anticancer Drug Treatment 4.2.2.1 P U M A enhances the sensitivity of human melanoma cells to C P T treatment Because several recent studies have now validated using Bax to activate apoptosis ectopically (Kobayashi et al, 1998; Tai et al., 1999; L i et al, 2001), we attempted to use other p53-induced pro-apoptotic genes to further evaluate our hypothesis. P U M A is a BH3-only domain containing protein that is believed to bind to Bcl-2 and localize to the mitochondria where it functions in the permeabilization o f the outer membrane affecting release of cytochrome c from the mitochondria (Zou et al, 1997; Y u et al, 2001a; Nakano and Vousden, 2001). We demonstrate that overexpression of PUMA can sensitize melanoma cells to drug treatment. A s shown in Figure 4.5A, PUMA was transiently transfected into M M R U cells then treated with camptothecin and the cell survival was quantified. Clearly, introduction of this gene could sensitize the transfected cells by decreasing cell survival to 64% alone or nearly 34% after C P T treatment (P < 0.0001) compared to the GFP-transfected control cells where survival was 99% and 70% (P < 0.0001), respectively. Therefore the combination of P U M A and C P T has an additive effect to reduce cell survival. 4.2.2.2 P U M A overexpression induces apoptosis not necrosis Determining cell survival is a strong indication that these cells may be undergoing programmed cell death but we needed to confirm this specifically. We used an ELISA-based protocol which quantifies the number of histone/DNA particles that are cleaved during the apoptotic process. Wi th this assay, we found that apoptosis was significantly increased in those M M R U (52- and 75 Figure 4.5 Overexpression o f PUMA sensitizes melanoma cells to C P T treatment. (A) Ce l l survival after transient transfection o f either the control G F P (open bars) or P U M A (closed bars) c D N A s into M M R U cells with or without 200 n M C P T treatment for 24 h. (B) M M R U and M e W o cell lines were tranfected with P U M A in 24-well plates for 24 h then treated with 200 n M C P T for a further 24 h before lysing the cells for the E L I S A assay o f apoptosis (see Materials and Methods, section 2.2.4). (C) D N A ladder of PUMA-tranfected M M R U cell line after 48 h. M W , 100 bp molecular weight marker. no drug CPT B 120 (0 O 100 O- 80 MW GFP PUMA • G F P • PUMA no drug C P T no drug C P T MMRU MeWo 76 94-fold) or M e W o (17- and 20-fold) cells overexpressing PUMA combined with drug treatment (Figure 4.5B). Interestingly, there is a distinct difference between the two cell lines. Since M M R U is p 5 3 W T we expect that the endogenous p53 would augment the overexpression of PUMA because it is a p53-inducible gene. To confirm the previous result with another assay, we show that cells overexpressing PUMA exhibit a DNA-laddering effect—the hallmark of apoptosis—not seen in GFP-expressing control cells (Figure 4.5C). 4.2.3 Noxa Sensitizes Mouse Melanoma Cells to Anticancer Drug Treatment The other B H 3 domain-only containing protein Noxa, which is the mouse homolog of the human A P R gene (Hijikata et al., 1990) was overexpressed in the mouse melanoma cell line B16. We initially used transient transfection into the B16 cells and compared the cell survival to G F P -transfected controls. B16 cells that overexpress Noxa were more sensitive to C P T treatment, comparing 88% (P = 0.007) cell survival in G F P cells to 62% (P = 0.002) for Noxa cells (Figure 4.6A). Then we determined the apoptotic rate by E L I S A and observed an obvious induction of cell death for the Noxa-expressing cells since apoptosis was increased by almost 60-fold in the drug-treated cells compared to the untransfected controls (Figure 4.6B). " Next, we established stable populations by neomycin-resistance for Noxa and a Noxa deletion mutant missing the B H 3 domain (Oda et al., 2000). We assessed the relative cell survival after D O X treatment and found that there was a significant decrease in survival between those cells that overexpress wt Noxa compared to the normal B16 control cells, 49% (P < 0.003) versus 74% (P = 0.0007). Constitutive overexpression of mutant Noxa had a small rescue effect and could enhance the inherent chemoresistance of B16 cells to slightly above control levels, 81% (P < 0.05) (Figure 4.7A). We used an alternative assay to look for both changes in mitochondrial membrane permeabilization and apoptosis induction. This procedure utilizes a cationic dye by which we could track the influx of the stain into the mitochondria. In healthy cells, the reagent aggregates 77 Figure 4.6 Overexpression of Noxa sensitizes mouse melanoma cells to C P T treatment. (A) S R B cell survival after transient transfection of mouse Noxa or G F P into B16 cells seeded to a 24-well followed by treatment with diluted D M S O (open bars) or 300 n M C P T (closed bars) for 24 h. (B) E L I S A assay for apoptosis (see Materials and Methods, section 2.2.4) under the same conditions indicated in (A). B control G F P Noxa control G F P Noxa 78 Figure 4.7 The stable expression o f Noxa in B16 mouse melanoma cells induces apoptosis. (A) Decreased cell survival in a stable B16 cell line overexpressing Noxa but improved survival in a Noxa-mut cell line. A l l three cell lines were treated with either diluted D M S O (open bars) or 500 n M D O X (closed bars) for 24 h. (B) Increased mitochondria permeability and apoptosis in the stable Noxa-expressing cells as measured by the MitoCapture procedure (see Materials and Methods, section 2.2.9). Both B16 (left) and Noxa (right) cells were treated with diluted D M S O or 500 n M D O X for 24 h. Cells were viewed and photographed under 400X magnification using a fluorescence microscope. 79 in the mitochondria and generates a bright red-orange fluorescence. In apoptotic cells, the reagent cannot penetrate the mitochondria outer membrane and remains in the cytoplasm where it emits a green fluorescence (Figure 4.7B). The control B16 cells have significantly more red-staining cells than the Noxa-expressing cells (Figure 4.7B, top). After D O X treatment for 24 h, the Noxa cells have dramatically more green (apoptotic) cells than the B16 parental cell line (Figure 4.7B, bottom right and bottom left, respectively). 4.2.4 Downstream Mediators of the Mitochondrial Apoptosis Pathway can Sensitize Melanoma Cells to Anticancer Drug Treatment 4.2.4.1 Apaf-1 sensitizes melanoma cells to anticancer drug treatment In the previous sections, our results show that several genes upstream of the mitochondria, Bax, P U M A , and Noxa were very capable of inducing cell death. We then asked whether downstream apoptosis effectors could be as effective at enhancing cell death in melanoma. We found that Apaf-1 was also a potent inducer of decreased survival (Figure 4.8). In a transient transfection experiment, Apaf-1 could decrease cell survival to 73% (P < 0.05) alone or to 39% (P < 0.005) after C P T treatment (Figure 4.8A). To determine the extent of apoptosis, we used the E L I S A -based protocol and found similar results to P U M A although to a lesser degree. A s shown in Figure 4.8B, addition oi Apaf-1 to M M R U cells could increase apoptosis alone by 25-fold over the GFP-transfected cells while this value more than doubled after drug treatment. The net amount o f change was significant but not as dramatic in the M e W o cell line. Without drug treatment, apoptosis induction only increased by approximately 14-fold in M e W o cells compared to 26-fold after C P T treatment. We next selected for Apaf-1 stable colonies in M M R U cells using the same procedure outlined in section 4.2.3. The pooled population of Apaf-1 overexpressing cells (Figure 4.8C, right) was more sensitive than M M R U to a number of 80 Figure 4.8 The overexpression of Apaf-1 sensitizes melanoma cells to anticancer drug treatment. (A) Ce l l survival after transient transfection of either the control G F P (open bars) or Apaf-1 (closed bars) c D N A s into M M R U cells with or without 200 n M C P T treatment for 24 h. (B) M M R U and M e W o cell lines were tranfected with G F P or Apaf-1 in 24-well plates for 24 h then treated with 200 n M C P T for a further 24 h before lysing the cells for the E L I S A assay of apoptosis. (C) Left panel, decreased cell survival in a stable M M R U cell line overexpressing Apaf-1 {left). Both cell lines were treated with various drugs: D O X (500 nM) , V I N (800 n M ) , or C P T (200 n M ) for 24 h. Right panel, Western blot showing endogenous levels of Apaf-1 in the parental M M R U and the Apaf-1-expressing stable cell line. 81 B 120 £ , 80 "ro • | 60 | 40 0) O 20 120 V) "35 2 100 Q. O Q. O c o o 3 "D C o 80 60 40 20 no drug CPT MMRU • G F P T • Apaf-1 1 i no drug C P T i • G F P • Apaf-1 I J J no drug CPT MeWo 120 o O 40 M M R U Apaf-1 control DOX VIN CPT Drug Treatment Apaf-1 82 different drugs as both the parental M M R U and the Apaf-1 lines were treated with either D O X (500 nM) , V I N (800 nM) , or C P T (200 nM) for 48 h (Figure 4.8C, left). C P T was the most effective drug at reducing cell survival with the concentrations used for the experiment. 4.2.4.2 Caspase-3 sensitizes melanoma cells independently of p53 Caspase family members are the terminal effectors of apoptosis. We speculated that overexpression of a proenzyme form of a caspase would activate apoptosis to a greater extent when combined with a pro-apoptotic stimulus such as an anticancer drug than the drug alone. In Figure 4.9A, we demonstrate that addition of C P T to the melanoma cell line M M R U is able to induce cleavage of caspase-3. Therefore, we used a procaspase-3 expression plasmid to transiently transfect both a p 5 3 W T ( M M R U ) and p 5 3 M U T (MeWo) melanoma cell line. In Figure 4.9B, as we would predict, M M R U cells are more sensitive to drug treatment alone than M e W o cells and have reduced cell survival (69%, P = 0.005 versus 92%, P = 0.026). Interestingly however, we observed an approximately equal decrease in cell survival in both cell lines compared to the controls after procaspase-3 transfection (49%, P < 0.002 versus 53%, P = 0.01). This suggests that procaspase-3 can enhance cell death in the presence of an anticancer drug and that the effect on cell survival is independent ofp53 mutation. 4.3 DISCUSSION 4.3.1 Pro-Apoptotic Genes Can Subvert Chemoresistance Mechanisms in Melanoma Since chemotherapy is not sufficiently effective, an alternative strategy for the treatment of advanced melanoma could be the gene therapy approach. Our studies suggest that direct mutation of p53 although important for melanoma progression may only play a limited role. However, more recent studies propose that alteration of other components within the p53 83 Figure 4.9 Procaspase-3 overexpression reduces cell survival in melanoma. (A) Western blot showing induction of cleaved caspase-3 in melanoma cells after 200 n M C P T treatment. (B) Quantification of cell survival for M M R U and M e W o transiently transfected with either a G F P or procaspase-3 expression plasmid for 24 h then treated with either diluted D M S O or 200 n M C P T for 48 h. A control CPT B GFP caspase-3 GFP caspase-3 M M R U MeWo S4 pathway such as the upstream checkpoint kinase, Chk2, (Satyamoorthy et al., 2000) or the downstream apoptosis regulator, Apaf-1 (Soengas et al., 2001) are the important modes for the resistance seen in melanoma. We undertook a pro-active course to sensitize melanoma cells by introducing pro-apoptotic genes. Activation of mitochondria-mediated apoptosis represents a key antitumour response by p53. Induction o f this pathway by p53 occurs through direct transactivation. We surmised that some members of the Bcl-2 family such as Bax, P U M A , and Noxa, which are known to be direct targets of p53-mediated apoptosis, would be ideal candidates for these experiments. The Bax protein belongs to the multidomain Bcl-2 family, while Noxa and P U M A are BH3-domain-only proteins. We also wanted to compare whether increasing downstream effectors of the activated mitochondria, in particular, Apaf-1 and the terminal executioner caspase, caspase-3, would be as effective as the upstream regulators. This was a proof-of-principle study for specific genes that may represent promising therapeutic approaches. 4.3.2 Bax Overexpression Is a Proven Candidate Initially our focus was Bax since this is one of the primary initiators of mitochondria-mediated cell death. We found that overexpression of Bax could significantly decrease cell survival in a melanoma cell line and that survival could be further reduced in combination with the drug, C P T (Figure 4.2). We refined the expression of Bax by controlling it with a tetracycline-inducible promoter and found that increasing Bax caused a concomitant decrease in N F - K B (Figure 4.3). Whether this decrease was mediated by cross-talk from p53, other mitochondrial factors, or an up-regulation of the N F - K B inhibitors is not clear. We suggest that induction o f Bax most likely causes the release of S M A C / D I A B L O from the mitochondria preventing IAPs from activating the N F - K B cascade (Hersey and Zhang, 2001). We were able to authenticate our system by demonstrating high transfection efficiency with our reagents (see Materials and Methods, Table 2.1). 85 Another strategy adopted a more targeted approach to melanoma cells. For this approach, accurate delivery of the therapeutic gene required cell-specific gene expression. Since melanocyte cells are characterized by their pigmentation, and since tyrosinase is the key enzyme involved in melanogenesis, we first studied the expression of a luciferase reporter gene which is under the control of the Tyrosinase promoter/enhancer (Park et al., 1999) in three melanoma and two non-melanoma cell lines. Reporter gene expression was up-regulated over 100-fold compared to a control G F P plasmid (Figure 4.4B). Intriguingly, the gene expression was strongly associated with the capacity of melanoma cells for melanin synthesis. For example, the B16 cells are dramatically pigmented and exhibit the highest luciferase activity while M M A N is far less pigmented. The functional efficacy of the promoter was tested using Bax introduced into HeLa and M M R U cells where we could clearly see a decrease in M M R U cell survival by over 50% in combination with drug treatment while HeLa cells showed no Bax-related decrease in cell survival (Figure 4.4C). These results confirm that the use of tissue-specific gene regulatory elements might provide a new opportunity for targeting therapeutic genes to melanoma cells. 4.3.3 Drug-Induced Apoptosis Is Enhanced by Overexpression of PUMA and Noxa The utility o f Bax as a mode o f inducing apoptosis through gene therapy has now been developed by other groups in human cancers such as ovarian (Tai et al, 1999; Arafat et al, 2000; Tsuruta et al., 2001), prostate (L i et al., 2001; Lowe et al., 2001), glioma (Lee et al., 2000), cervix (Huh et al., 2001), and lung (Kagawa et al, 2000). Therefore, we modified our focus to other less characterized genes. The c D N A s , PUMA and Noxa, were introduced into human and mouse melanoma cell lines and were treated with various DNA-damaging drugs (Figures 4.5-4.7). While PUMA has only been identified in humans ( Y u et al., 2001a; Nakano and Vousden, 2001), the Noxa gene is the mouse homologue o f a human phorbol-12-myristate-13-acetate-responsive gene that was highly expressed in the Jurkat cell line (Hijikata et al., 86 1990). Noxa was identified through a differential screen of X-ray-irradiated p53"'" mouse embryonic fibroblasts (Oda et al., 2000). We show that introduction of PUMA and Noxa transgenes lead to significantly decreased cell survival that was magnified in an additive manner with drug treatment. This was followed by analyses confirming apoptosis and the establishment of B16 cell lines that constitutively express either wild-type Noxa or a mutant Noxa that lacks the important B H 3 domain necessary for protein-protein interactions (Figure 4.5-4.7). The N o x a W T cells had an overall lower threshold of survival compared to the parental B16 and these were more sensitive to several different drug treatments suggesting that Noxa alone could increase cell death (Figure 4.7). Also , the N o x a M U T cells were able to partially rescue the drug-sensitized phenotype of the N o x a W T cells. It is noteworthy to reiterate that P U M A appeared to be better at inducing apoptosis in M M R U than the M e W o cell line. Although consistent with our previous results showing that M e W o is more resistant to drug-induced apoptosis (L i et al., 1998), this could be due to a combination of several factors. Upon a DNA-damaging stimulus, p53 transcriptionally activates target genes. Since PUMA is a transcriptional target of p53 (Yu et al., 2001a; Nakano and Vousden, 2001), perhaps the fact that p53 is mutated in M e W o cells means that apoptosis is only due to the exogenous levels of P U M A while p53 can also induce endogenous P U M A in M M R U cells thereby amplifying the response. This needs to be clarified further with PUMA-spec i f i c antisera; however, we do have correlative data demonstrating a functional role by the apoptosis E L I S A assay and D N A laddering in . which we were not able to detect PUMA-induced fragmentation as easily as in M e W o cells (Figure 4.5C and data not shown). Furthermore, these were transient transfections and from Table 2.1, it is clear that the transfection efficiency is significantly higher in M M R U than M e W o cells. A s well , L i et al. (1998) demonstrate that the doubling time of M M R U cells is faster than that of M e W o cells and therefore, the initial number 87 of cells may also contribute to the overall tranfection differences. We are currently establishing stable melanoma cell lines that express P U M A to extend these studies. 4.3.4 Drug-Induced Apoptosis Is Enhanced by Overexpression of Apaf-1 and Procaspase-3 Formation of the functional apoptosome between Apaf-1 and caspase-9 may be one of the rate-limiting steps for the activation of apoptosis (Green and Reed, 1998). Some studies have revealed that upon activation of apoptosis, massive amounts of cytochrome c are released into the cytoplasm (Hajek et al, 2001). Since Apaf-1 appears to be at a general steady state level within melanoma cell lines (Figure 4.1), we surmised that increasing the expression of Apaf-1 could potentially amplify the formation of the apoptosome through 'scavenging' o f the cytochrome c. We introduced the Apaf-1 transgene into M M R U cells and assessed the cell survival (Figure 4.8). Undoubtedly, there was a substantial decrease in survival and this too was amplified in combination with drug treatment. The decreased cell survival was mediated by a 60-fold increase of apoptosis (Figure 4.8B). Surprisingly, we observed a similar difference between the apoptotic levels in the M M R U and M e W o cell lines that we observed with P U M A suggesting that the transfection efficiency is an important factor determining the extent of apoptosis. B y creating a stable M M R U cell line overexpressing Apaf-1 in an identical process to the B16/Noxa clones, we could get a generalized sensitivity of the cells to diverse D N A -damaging agents (Figure 4.8C). New studies confirm our observations that overexpression of Apaf-1 promotes apoptosis in other tumour cell types (Perkins et al, 1998; Kamarajan et al, 2001; Shinoura et al, 2002); however, malignant melanoma is the only neoplasm for which significant Apaf-1 inactivation has been reported (Soengas et al, 2001) Activation of caspases, particularly the terminal effectors such as caspase-3, appears to be an important step of apoptosis although not absolutely essential since some forms of programmed cell death occur independently of caspase-3 activity (Shinomiya, 2001). Once 88 activated, the effector caspases are responsible for the proteolytic cleavage of a broad spectrum of cellular targets, leading ultimately to cell death. The known cellular substrates include structural components (such as actin and nuclear lamin), regulatory proteins (such as D N A -dependent protein kinase), inhibitors of deoxyribonuclease such as D N A fragmentation factor 45 (DFF45) or inhibitor o f caspase-activated deoxyribonuclease ( ICAD) , and other proapoptotic proteins and caspases. Cleavage of the DFF45 leads to removal of its inhibition of C A D , which degrades the chromosomes into nucleosomal fragments during apoptosis (Hirata et al., 1998; see section 1.3.5.1 for further details). We demonstrate that caspase-3 can sensitize both the M M R U (p53 W T ) and the M e W o (p53 M U T ) cell lines only after drug treatment. Previously, we established that C P T could induce the active form of caspase-3 in melanoma cells (Figure 4.9A). Therefore, we introduced the zymogen form of the proteolytic enzyme, procaspase-3, into both M M R U and M e W o cells. Intriguingly, in the absence of a death-inducing stimulus, we did not detect any significant change in cell survival. Only upon the addition of C P T was there a dramatic reduction (~ 50%) in survival for both cell lines (Figure 4.9B). This indicates a p53-independent mechanism. A recent report verified a similar mechanism in a drug-resistant breast cancer cell line (Friedrich et al., 2001). These results suggest that using a factor such as procaspase-3, a pivotal downstream executioner of apoptotic pathways, would be promising for any type of tumour cell line in which upstream mutations affecting the apoptosis mechanism occur. 4.3.5 Overcoming the Anti-Apoptotic Block in Melanoma Improves Drug Efficacy Our strategy to investigate both upstream and downstream components of the mitochondrial-induced apoptosis pathway has been encouraging. During our studies, several of the candidate genes particularly Bax, Apaf-1 and caspase-3 have been reported as potential targets to exploit for the apoptotic machinery. In our hands, each transgene could successfully enhance the sensitivity of the cells to an apoptogenic stimulus including CPT , D O X , or V I N . However, 89 examination of the survival percentages after drug treatment suggests that PUMA is a particularly good target for overexpression in melanoma. The additivity of the PUMA/dmg combination significantly exceeded the effects of drug treatment alone. Little is known about the functional roles of P U M A and the other Bax family of pro-apoptotic factors. The relative novelty of this gene as well as Noxa makes them excellent choices for further analysis. Understanding their contribution to mitochondrion permeabilization is critical to elucidating the complex mechanism o f apoptosis in cancer treatment. In summary, the efficacy of conventional chemotherapeutic drugs is due to their ability to induce apoptosis. The data presented here further endorses the therapeutic value of sensitizing melanoma cells to anticancer drugs by up-regulating apoptotic signals through ectopic expression of genes as a practical and functional means of treatment. We continue to work in this area to confirm that overexpression of these pro-apoptotic factors is promoting cell death via apoptosis. The next stage is to prove that these are feasible for in vivo tumour reduction. 90 CHAPTER 5. THE USE OF DIETARY ANTIOXIDANTS TO SENSITIZE MELANOMA CELLS TO APOPTOSIS 5.1 R A T I O N A L E A N D H Y P O T H E S I S The general ineffectiveness of current chemotherapeutic agents for malignant melanoma warrants the investigations into alternative compounds to improve today's therapy regimes or as a means of chemoprevention. One such alternative compound that exhibits tremendous antineoplastic effects is curcumin. Curcumin (diferuloylmethane), a polyphenolic phytochemical, is the primary component of the spice turmeric (Figure 5.1 A ) . The pharmacological safety of curcumin is well demonstrated by the fact that people in certain countries have consumed curcumin as a dietary spice for centuries in amounts in excess of 100 mg/day without any side effects (Amnion and Wahl, 1991). Ample evidence exists to support its use in cancer prevention for its antiproliferative and anticarcinogenic properties or as an adjunct in overall cancer treatment (Nagabhushan and Bhide, 1992; Huang et al, 199"'a; Chan et al., 1998) . Curcumin is a potent inhibitor of the initiation and promotion of chemical carcinogen-induced skin carcinogenesis in mice (Lu et al., 1994; Huang et al., 1997a; Kawamori et al., 1999) . Topical application of curcumin on mouse skin inhibited chemically-induced skin carcinogenesis (Azuine and Bhide, 1992; Huang et al, 1997b). The anticarcinogenic mechanisms of curcumin action are not fully understood. It appears that curcumin can suppress the cell growth pathway by inhibiting the cellular protein kinases like P K C , c-Jun NH(2)-terminal kinase (JNK), and the epidermal growth factor (EGF) receptor kinase leading to growth inhibition (Jiang et al., 1996; Chen and Tan, 1998; Korutla et al, 1995). Curcumin also has the profound ability to block the N F - K B cell survival pathway (Singh and 91 Aggarwal, 1995). Moreover, curcumin is known to activate apoptosis pathways. Recent evidence suggested that curcumin could inhibit growth and activate apoptosis of a human basal cell carcinoma cell line in a p53-dependent manner. P53 can mediate apoptosis under many stress conditions, and its downstream targets, p 2 i w a f l / c , p l and G A D D 4 5 are overexpressed during curcumin-induced apoptosis in a basal cell carcinoma (Jee et al., 1998). Also , p53 and c-myc were up-regulated in a human hepatoblastoma cell line after curcumin treatment (Jiang et al., 1996). However, p53 was decreased, with a concomitant increase of the heat-shock protein, HSP70, after curcumin treatment in colorectal carcinoma cells (Chen et al., 1996; Kato et al., 1998). Therefore, the role of p53 tumour suppressor in curcumin-induced apoptosis appears to be tissue-specific. The anticarcinogenic effect of curcumin in human melanoma is not well-defined. We investigated the molecular pathways targeted by curcumin during cell death of human melanoma cell lines to determine the utility of such a dietary supplement for possible prevention and treatment of melanoma. We assessed the dependence of p53 by analyzing the effects of curcumin in eight melanoma cell lines; four with p 5 3 W T and four with p 5 3 M U T . There was no dramatic difference in the survival levels among melanoma cell lines after increasing concentrations of curcumin suggesting that curcumin acts independently of the tumour-suppressor gene, p53. We hypothesized that curcumin must activate other apoptosis pathways independently of p53 in melanoma cells. 92 5.2 R E S U L T S 5.2.1 C u r c u m i n Induces Apoptosis in Melanoma Cells The ability o f curcumin to induce apoptosis has been shown in a number o f cell types. M u c h o f the variable results cited in the literature appear dependent on the cell type. The effect of curcumin in human melanoma cell lines has not been investigated in detail. We have found that curcumin induces cell death of a malignant melanoma cell line M M R U , which contains wi ld-type p53, in a dose- and time-dependent manner (Figure 5.IB and C). A t a curcumin concentration above 30 u M , significant cell death is induced. The D N A ladder assay indicated that curcumin-induced cell death proceeds through an apoptotic pathway not necrosis (Figure 5.1C). A t a dose of 100 u M , curcumin induced apoptosis in virtually all the cells after 48 h treatment (Figure 5.IB). These concentrations agree with studies from other cell types illustrating the apoptotic potential of curcumin (Chen and Huang, 1998; Jee et al., 1998; Bhaumik et al, 1999; Piwocka et al, 1999). We further confirmed that apoptosis was occurring in M M R U cells by staining with Annexin-V and PI after 60 u M curcumin treatment (Figure 5.2A and B) . At 6 h and 12 h, approximately 40% and 64% of cells stained positive for Annexin-V representing cells undergoing apoptosis. In an effort to analyze which particular apoptotic pathway curcumin may be working through, we looked at different melanoma cell lines that either have a normal or mutant p53 locus. We compared 8 different cell lines (4 wild-type for p53 including M M R U and 4 mutant) and found that there was no significant difference in their sensitivity to curcumin (Figure 5.3A). A n d again, the cell death being initiated occurs via an apoptotic process as demonstrated by the presence of apoptotic bodies in two melanoma cell lines (Figure 5.3B and C) . This suggests that curcumin-induced apoptosis in these cell lines is p53-independent. Our results contradict recent work by Jee et. al. (1998) showing a p53-dependence in a single basal cell carcinoma cell line. 93 Figure 5.1 Curcumin decreases cell survival in the melanoma cell line M M R U . (A) Structure of curcumin. (B) Dose-dependent cell death by curcumin treatment. Cells were treated with curcumin at various doses for 48 h and cell survival was determined by S R B assay. Experiment was repeated three times and similar results obtained. (C) Time-course o f curcumin-induced cell death. M M R U cells were treated with 60 u M curcumin for 0, 12, 24, or 48 h. D N A was extracted and analyzed on a 2% agarose gel. A B Curcumin Dose (uM) 12 24 48 h 94 Figure 5.2 Curcumin induces apoptosis in M M R U . (A) Flow cytometry of Annexin V-positve staining M M R U cells after curcumin treatment. Cells were either treated with 60 p M curcumin for 6 and 12 h or left untreated. (B) The percentage of cells staining positive for Annexin-V was quantified. Data represent mean + SD from three independent experiments. 95 Because our melanoma cell lines are derived from both radial phase and metastatic phase melanomas, curcumin's apoptotic activity demonstrated in all cell lines regardless of the stage of melanoma growth emphasizes its universal antineoplastic properties against melanomas. 5.2.2 Curcumin-induced Apoptosis Is P53-Independent Next, we confirmed the expression levels of various pro-apoptotic or pro-survival gene products. Our results show that p53 expression is unchanged in either a representative wild-type p53 cell line M M R U or a mutant-bearing p53 cell line P M W K , supporting our claim that in melanoma, curcumin-induced cell death is not mediated by a p53 apoptotic pathway (Figure 5.4). Furthermore, expression of the apoptotic family members, Bcl-2 and Bax remains unchanged after curcumin treatment over time (Figure 5.4). Since pro-apoptotic Bax is known to be a p53 downstream target, this result also supports our hypothesis that p53 involvement is precluded during this cell death mechanism. 5.2.3 Curcumin-induced Apoptosis Can Inhibit N F - K B and XIAP It is wel l established that curcumin inhibits the pro-survival transcription factor N F - K B in various cell types (Singh and Aggarwal, 1995; Han et al, 1999; Jobin et al, 1999). To confirm this role in our system, we looked at expression of both the p65 and p50 subunits from the N F -K B precursor complex. A s shown in Figure 5.4, after curcumin treatment, both subunits are significantly reduced with time in M M R U and P M W K cells. In addition, we have, for the first time, shown that the expression of the X-l inked inhibitor of apoptosis (XIAP) is dramatically reduced after curcumin treatment. X I A P expression is known to be an NF-KB-dependent target involved in the binding and suppression of caspase protease activation (Stehlik et al, 1998; Hida et al, 2000; Hofer-Warbinek et al, 2000). Futhermore, a recent report demonstrates that Bcl-2 activation can prevent X I A P cleavage of TRAIL-induced apoptosis (Fulda et al, 2002). 96 Figure 5.3 Curcumin-induced apoptosis in melanoma cells is p53-independent. (A) Dose-dependent cell death by curcumin treatment in 4 p 5 3 W T (RPEP, M M R U , M M A N , Sk-mel-5) and 4 p 5 3 M U T (Sk-mel-28, Sk-mel-2, P M W K , M E W O ) melanoma cell lines. Cells were treated with 0, 30, 60, or 100 p M curcumin for 48 h and cell survival was determined by S R B assay. (B) M M R U and P M W K cells were treated with 60 u M curcumin for 0, 12, 24, or 48 h then stained with PI and apoptotic cells counted. Data represent mean + SD from three independent experiments. (C) Microscopic images of Pl-stained M M R U and P M W K cells untreated (left panel) or treated with 60 u M curcumin for 24 h (right panel). Cells were viewed and photographed under 400X magnification using a fluorescence microscope. Arrows indicate apoptotic cells. 97 RPEP MMRU MMAN Sk-rrel-5 Sk-mel-28 Sk-mel-2 PMWK MEWO Cell Lines B 100 N O 80 0 S W © 60 O u '+= 40 O a § . 20 < 0 • MMRU • P M W K 1 1 control 12h 24h 48h Curcumin Treatment Mr jRf / / A W MMRU PMWK 98 Therefore, this correlates well with our data demonstrating that curcumin function does not induce Bcl-2 allowing for rapid degradation of X I A P involving an NF-KB-dependent feedback mechanism. 5.2.4 Curcumin Induces Apoptosis by Activation of Caspases During apoptosis, a series of proteolytic cleavages of various intracellular polypeptides is initiated (Earnshaw et al., 1999). Most of the proteolytic degradation results from the action of a unique family of cysteine-dependent proteases called caspases. We next investigated the role of caspase activation by curcumin in our melanoma system. A s shown in Figure 5.4, there is a clear reduction of the procaspase-3 zymogen. This is in agreement with other studies demonstrating a caspase-3-mediated pathway in rat histiocytoma (Bhaumik et al., 1999; Khar and A l i , 1999). A n increase in poly(ADP-ribose) polymerase (PARP) cleavage over time, as represented by the accumulation of the lower P A R P band in Figure 5.4, confirms caspase-3 activation by curcumin. It is also of interest to note that the caspase-3 activation and P A R P cleavage occurs slightly faster in P M W K than in M M R U cells, which correlates with the faster degradation o f X I A P in P M W K cells, supporting the current model that X I A P inhibits caspase activation. Furthermore, we see a distinct decrease of procaspase-8 expression while the procaspase-9 expression remains unchanged with time. We propose that in our melanoma cell system, curcumin selectively induces apoptosis through a caspase-8-mediated pathway. The contemporary model of caspase-8 function is downstream of the death receptor-mediated pathways involving CD95/Fas /APO- l and TNFcc receptor while caspase-9 acts downstream of the mitochondrial-mediated pathway components such as Apaf-1 and cytochrome c (Earnshaw et al, 1999; Ashikenazi and Dixi t , 1998). We confirm this hypothesis by demonstrating that in the presence of caspase-8-specific inhibitor, Z - I E T D - F M K , and the broad-base caspase inhibitor, Z - V A D - F M K , curcumin-induced apoptosis is dramatically reduced 99 Figure 5.4 The effects of curcumin on pro- and anti-apoptotic gene expression. M M R U (p53W T) and P M W K (p53M U T) cells were treated with 60 u M curcumin for 0, 12, 24 or 48 h. Proteins were extracted and subjected to Western blot analysis. Densitometry values were obtained as described in Materials and Methods, section 2.2.14. MMRU PMWK C 12 24 48h "^HWMW l^iiBiiMp'5 i^iiiiiifiiisi"'*" ^mwKKt^1 1 1.3 0.9 1 tQtagg- a t f ^ t e ^ ^ ^ ^ ' • : *8W^^Pp ' ^WRwwIIPr ^BPBdlff' 1 1.9 1.4 1.2 — • — ' — — 1 0.8 0.7 0.9 •**«|* * ^ J M ^ ^ * • • 1 0.7 0.4 0.2 1 0.6 0.5 0.3 1 0.6 0.1 0.05 » mm — — 1 0.5 0.3 0.1 TUPSBUPP1 'HWBdHPP^  ^ M s p w P 1 * ^ 1 0.6 0.3 0.06 mmm 1 1 0.8 0.9 1 2.5 4.7 9.9 1 0.9 1.8 2.1 ^^^^^^ ^^^^^^ P^tifS^ f^SP^  ^PJ^ PP^^ ^ C 12 24 48h 0.8 0.8 0.7 1.1 0.9 0.9 0.8 1.2 1 0.9 0.6 0.2 0.8 0.7 0.2 0.1 0.1 0.03 0.2 0.2 0.1 1.2 0.4 0.05 0.7 0.9 1 25.8 19.2 16.5 1 1.5 2.7 p53 Bax Bcl -2 p65 p50 XIAP Caspase-3 Caspase-8 Caspase-9 PARP Fas p-actin 100 (Figure 5.5). Compared to the control cells (without curcumin), M M R U cells show a significant increase in survival after 24 h treatment with 60 p M curcumin and 50 p M of specific caspase inhibitor. A smaller protection by the caspase-3 inhibitor, A c - D M Q D - C H O , suggests that other downstream caspases may also be activated by caspase-8. Predictably, cells treated with the caspase-9 inhibitor, Z - L E H D - F M K , had a similar percentage of survival to that of controls (20%), indicating that the caspase-9 inhibitor did not prevent curcumin-induced cell death (Figure 5.5). 5.2.5 Curcumin Induces Fas Expression and Aggregation We next determined that inhibition of membrane-mediated death receptors plays a role in transducing the apoptotic signals. First, we used Western blot analysis to examine whether curcumin induces Fas expression. A s shown in Figure 5.4, Fas expression was induced by at least 2-fold in both M M R U and P M W K cell lines after 60 p M curcumin treatment for 48 h. Since death receptor family members such as T N F R and p a s A P ° - 1 / C D 9 5 duster upon ligand binding and induce a death signal (Ashikenazi and Dixi t , 1998), we next investigated i f curcumin induces Fas receptor clustering. Using a FITC-conjugated anti-Fas antibody, we demonstrate that the Fas-receptor aggregates and most likely initiates the death signals. In the control cells (without curcumin treatment) only weak staining was observed in both M M R U and P M W K (Figure 5.6A, left panels). However, in the curcumin-treated cells, we detected strong fluorescence staining in a punctate pattern indicative of Fas aggregation in these cell lines (Figure 5.6A, right panels). To rule out that the strong fluorescence staining after curcumin treatment is not due to trapping of FITC-conjugated antibodies on the cell surface, we stained M M R U and P M W K cells with an anti-integrin p i FITC-conjugated antibody. Our results indicated that curcumin did not enhance the fluorescent intensity of the cells stained with anti-integrin p i FITC-conjugated cells (data not shown). 101 Figure 5.5 Curcumin-induced apoptosis in melanoma cell lines is reduced after treatment with caspase inhibitors. M M R U (p53 W T ) and P M W K (p53 M U T ) cells were pretreated with 50 p M of the caspase inhibitors (caspase-3 inhibitor, A c - D M Q D - C H O ; caspase-8 inhibitor, Z - I E T D - F M K ; caspase-9 inhibitor, Z - L E H D - F M K ; broad-base caspase inhibitor, Z - V A D - F M K ) for 3 h, and then treated with 60 p M curcumin for 24 h. (Top panel) Photograph of S R B in 96-well plate. (Bottom panel) The cell survival was quantified for S R B absorbance at 405 nm. Experiments were carried out in triplicate. 96-well plate Control Broad III Vlll IX Caspase Inhibitor 102 To confirm that curcumin induces apoptosis through the Fas receptor pathway, we co-treated the P M W K cells with curcumin and anti-Fas antibody used to induce apoptosis. A s shown in Figure 5.6B, the percentage of cell death increases with the amount of Fas antisera. To further confirm that Fas aggregation initiates apoptosis signaling in this system, we pre-incubated cells at 4°C which was shown to block 95% of Fas aggregation (Kulms et al, 1999) and then quantitated the extent of cell survival after curcumin treatment with respect to control cells. In Figures 5.7A and 5.7B, we demonstrate that at different concentrations of curcumin, blockage of death receptor clustering at 4°C significantly inhibited curcumin-induced cell death in M M R U , supporting our hypothesis that curcumin induces apoptosis in melanoma cells by activating the Fas death receptor/caspase-8 pathway. Fas aggregation is known to recruit F A D D protein to the plasma membrane that in turn activates procaspase-8 (Nagata, 1999). To demonstrate that curcumin-induces cell death via a Fas-FADD-caspase-8 pathway, we transfected the M M R U melanoma cells with a dominant negative F A D D expression vector ( F A D D - D N ) to block F a s - F A D D binding. Figure 5.7C shows that F A D D - D N prevented curcumin-induced cell death, further confirming that curcumin-induced cell death is due to the activation of Fas death receptor pathway. To investigate whether curcumin-induced Fas aggregation is due to increased expression of Fas ligand (FasL), we assessed FasL m R N A levels after curcumin treatment. A s shown in Figure 5.8, there is no appreciable change in the expression of the FasL transcript in either M M R U or P M W K cell lines after treatment with 60 p M curcumin for either 3 or 6 h. This means that after curcumin treatment there is no de novo synthesis of FasL. Moreover, we could not detect any FasL protein by Western blot from these cells compared to a Jurkat positive control (data not shown) suggesting that the Fas aggregation we are detecting is independent of its ligand. 103 5.3 D I S C U S S I O N The lack of effective treatments for malignant melanoma requires new therapeutic strategies to be designed. Since many anticancer drugs k i l l tumour cells by inducing apoptosis, this represents a valid mechanism to exploit. The ability of many natural compounds to initiate the same type of cellular response without the toxicity associated with synthetic drugs suggests that they could be used as potentially chemopreventive and chemotherapeutic agents (Manson et al., 2000). Polyphenols as a group of compounds seem to possess many of the desirable qualities for anticancer treatment. One such natural agent is curcumin. 5.3.1 Curcumin Induces P53-Independent Apoptosis in Melanoma We have shown that curcumin induces cell death in melanoma cell lines in a concentration and time-dependent manner (Figure 5.1). A t concentrations greater than 30 p M there is significant cell death up to approximately 100 p M when virtually all cells are dead. We chose to use a specific concentration of 60 p M as this provided an average survival rate of about 10% (Figure 5.3A). We found that maximum cell death was induced by 48 h and beyond this time cell degradation was too extensive to measure by D N A fragmentation (data not shown). Our results indicate that the apoptosis induced by curcumin is p5 3-independent. In eight different melanoma cell lines tested, there was no significant difference in the cell survival rate despite the fact that half o f the cell lines harbor a mutant p53 protein and have been shown to respond differently to anticancer drug treatment (Figure 5.3A) (L i et al, 1998). The PI staining further demonstrates that apoptosis is being activated. The presence of apoptotic bodies occurred in both wild-type and mutant p53 melanoma cell lines (Figure 5.3C) and supports the hypothesis for a p5 3-independence. This is in agreement with other studies demonstrating that p53 expression is reduced or unchanged after curcumin treatment (Jiang etal., 1996; Chen and 104 Figure 5.6 Curcumin induces cell death by activating Fas receptor aggregation. (A) Microscopic images of Fas aggregation. M M R U and P M W K cells untreated (left panels) or treated with 60 p M curcumin for 6 h (right panels) then stained with FITC-conjugated anti-Fas antibody and visualized under fluorescence. Cells were viewed and photographed under 400X magnification using a fluorescent microscope. Arrows indicate Fas-aggregated, punctate staining regions. (B) Curcumin-induced cell death is enhanced by treatment with Fas antibody. P M W K cells were seeded on to 96-well microtitre plates at 1.2xl0 4 /well, grown for 24 h, then treated with the D X 2 anti-Fas antibody at increasing concentrations (1.5, 3, 6 pg/ml) for 24 h. Fol lowing this treatment, cells were further incubated with 60 p M curcumin for 24 h. Ce l l death was determined by S R B assay. 105 106 Figure 5 . 7 Blockage of Fas aggregation by low-temperature incubation on curcumin-induced apoptosis of M M R U cells. Cells were placed at 4°C for 30 min, then treated with curcumin for 1 h at 4°C (see Materials and Methods, section 2.2.11) and incubated with fresh medium for 24 h (A) or treated with curcumin for 24 h (B). Cells incubated at 37°C were used as controls. Ce l l survival was determined by S R B assay. (C) Overexpression of F A D D - D N inhibits curcumin-induced cell death. M M R U cells were transfected with 0.2 pg of either G F P or F A D D - D N for 24 h using Effectene reagent followed by 60 p M curcumin treatment for 24 h. Untreated cells were used as controls. Cel l survival was determined by S R B assay. Data represent mean + SD from three independent experiments. 107 108 Figure 5 .8 Northern blot analysis o f Fas ligand m R N A expression levels after curcumin treatment. M M R U and P M W K cells were treated with curcumin (60 u M ) for 0, 3, or 6 h. Total R N A was extracted from the cells and subjected to Northern blot analysis. 28S r R N A was used as an internal control. M M R U P M W K 0 3 6 0 3 6 h FasL 28S 109 Huang, 1998; Chen and Tan, 1998). The results strongly suggest that in our system, other mechanisms besides p5 3-mediated cell death must be important for curcumin signaling. A comparison of the protein expression levels of critical apoptosis-controlling genes after curcumin exposure provided us with further molecular clues as to its mode o f action. We confirm our hypothesis that curcumin treatment is p53-independent as the p53 expression profile remains unchanged over time (Figure 5.4). The C M - 1 antisera used can recognize both wi ld -type and mutant forms of p53 and so accurately depicts the expression from both cell lines. N o change in p53 expression after curcumin was also observed in either a rat vascular smooth muscle cell line or a human breast tumour cell line (Chen and Huang, 1998; Mehta et al, 1997). Our results demonstrating no change in the expression of Bcl-2 or Bax expression further consolidates the model. Bcl-2 and Bax expression are regulated by p53 both in vitro and in vivo and Bax is a direct target of p53 transcriptional activation (Miyashita and Reed, 1995). Upon exposure to curcumin, a lack of change supports the view that curcumin does not trigger the p53/Bax-mediated apoptosis pathway. A Bcl-2-independent mechanism after curcumin treatment has been observed in other cell types (Jiang et al., 1996; Jee et al., 1998; Mehta et al., 1997). Another p53 target, p 2 l w a f l / c i p l is not increased after curcumin treatment (data not shown) also attesting to a lack of p53 induction. 5.3.2 Curcumin Inhibits the Pro-Survival Pathway Curcumin appears to inhibit N F - K B pro-survival responses. Numerous studies have established an interaction between N F - K B suppression and curcumin activity with varying physiological effects (Singh and Aggarwal, 1995; Han et al, 1999; Jobin et al, 1999). We also demonstrate that both subunits of the N F - K B complex, p65 and p50, are significantly down-regulated after curcumin treatment (Figure 5.4). Accumulating evidence suggests that the apoptotic process is clearly inhibited by activation of N F - K B , which appears to be modulated through induction of 110 anti-apoptotic gene products (Miyashita and Reed, 1995; Wang et al, 1998, 1999). In particular, we show for the first time that the X-l inked inhibitor of apoptosis (XIAP) , which is known to be both a target and regulator of N F - K B activation (Stehlik et al., 1998; Hofer-Warbinek et al., 2000), is dramatically decreased after exposure to curcumin. From a clinical perspective, N F - K B activation in response to chemotherapy is a principal mechanism activating tumour resistance and therefore the inhibition of N F - K B is a new approach to adjuvant cancer treatment (Wang et al, 1999). 5.3.3 Curcumin Activates the Caspase Cascade The relationship between N F - K B as a requirement for the prevention of apoptosis and the large family o f apoptotic activators, the caspases, is becoming increasingly complex (Kawakami et al, 1999; Hida et al, 2000). The current model involves N F - K B activation of IAPs followed by LAP prevention and inhibition of caspase activity. Upon recognizing the first steps of this model in our system, we analyzed the caspase component to this mechanism. We demonstrate a clear time-dependent reduction in procaspase-3 expression (Figure 5.4). Our results agree with the observations by Khar and A l i (1999) showing a curcumin-dependent activation of caspase-3 in the A K - 5 tumour cell line. In contrast, one report found that curcumin-induced cell death in Jurkat cells occurs in a caspase-3-independent fashion, suggesting that caspase-3 activation by curcumin is cell-type specific (Piwocka et al, 1999). We observed a time-dependent increase in P A R P cleavage confirming caspase-3 activity and there was a clear reduction in procaspase-8 with little change in procaspase-9. These results suggest that apoptosis by curcumin proceeds through a membrane or death receptor pathway in melanoma cell lines rather than the intrinsic p53 and mitochondrial-mediated cascade. Treatment of M M R U and P M W K with caspase-8-specific or broad-base caspase inhibitors showed a significant increase in cell survival in the 111 presence of curcumin but no significant change with the caspase-9 inhibitor which further validates the involvement o f caspase-8 but not caspase-9. Stronger inhibition o f apoptosis by the caspase-8 inhibitor versus the caspase-3 inhibitor suggests that other caspase family members such as caspase-7 may also be activated by caspase-8. This rationale has precedence by the fact that high levels of a caspase-8/FLICE (FADD-l ike IL-1 beta-converting enzyme)-inhibitory protein, FLIP , has been demonstrated to contribute to melanoma chemoresistance (Irmler et al., 1997). 5.3.4 Curcumin Induces Membrane-Mediated Apoptosis The role of membrane death receptor-mediated apoptosis is well-established. Ligation of Fas by FasL or a cross-linking antibody results in receptor trimerization followed by binding of the adaptor molecule F A D D to the cytoplasmic domain of the receptor (Nagata, 1999; Kaufmann and Earnshaw, 2000). F A D D in turn recruits and activates procaspase-8 (Nagata, 1999; Siegel et al., 1998). Caspase-8 then cleaves procaspase-3 and -7, and active caspase-3 cleaves procaspase-6 (Ashikenazi and Dixi t , 1998). Similar pathways appear to be activated for other death receptors, e.g., T N F R 1 and death receptors D R 3 , D R 4 , D R 5 , and D R 6 , although the ligands and adaptor molecules are different (Earnshaw et al., 1999). Our results demonstrate for the first time that curcumin induces apoptosis in melanoma cells by activating caspase-8, the downstream target of membrane death receptors, such as Fas and T N F R . Curcumin-induced Fas aggregation is clearly shown using a FITC-conjugated anti-Fas antibody (Figure 5.6A). The fact that prevention of Fas clustering and F a s - F A D D binding by transfection of a F A D D - D N expression vector further supports the model that curcumin activates the death receptor-initiated apoptosis pathway (Figure 5.7). We also clearly demonstrate that curcumin-induced Fas aggregation in these melanoma cells is independent of FasL expression (Figure 5.8). A similar mechanism of Fas ligand-independent, FADD-mediated activation of the Fas death pathway was 112 observed in human colon and leukemic cell lines after cisplatin and etoposide treatment (Micheau et al, 1999; Shao et al., 2001). Furthermore, U V light was also shown to directly stimulate Fas and thereby activate the Fas pathway to induce apoptosis independently of Fas ligand in a keratinocyte cell line (Aragane et al., 1998). These results suggest a common mechanism whereby divergent stimuli can activate membrane-associated events that target the Fas apoptotic pathway in a manner that precludes its natural ligand, FasL. This is consistent with data indicating that melanoma cells lack expression of FasL (Chappell et al., 1999). Taken together, our results show that curcumin induces apoptosis in melanoma cell lines in a manner that is independent of p53 and the Bcl-2 family, but that activates the death receptor Fas-initiated FADD/caspase-8-dependent apoptosis pathway (Figure 5.9). This research has provided a more detailed mechanism of curcumin action in melanoma cells and should further establish its use as a valid chemopreventive and chemotherapeutic agent. 113 Figure 5 . 9 Molecular pathways of curcumin action. Curcumin can (1) induce apoptosis in human M M cell lines in a dose- and time-dependent manner but independent o f p53, (2) inhibit XJ.AP and suppresse N F - K B while activating caspases-3 and -8 but not - 9 . Therefore this supports a model that curcumin activates extrinsic FasL-independent (small ' X ' ) membrane-mediated rather than intrinsic mitochondrial-mediated (large ' X ' ) events. 114 C H A P T E R 6. G E N E R A L C O N C L U S I O N S 6.1 S U M M A R Y This work has been an elaboration of ongoing efforts to sensitize melanoma to chemotherapeutic agents. It has become clear that apart from the hereditary component of melanoma which may only account for about 25% of mutations in large melanoma kindreds (Pollack and Trent, 2000), the majority of sporadic forms of melanoma are directly related to mutations enabling subversion of normal apoptotic mechanisms of anticancer drugs. Therefore, we have focused on pro-apoptotic mechanisms to investigate the modulation of chemosensitivity. Because p53 mutation often correlates with degree of invasion and consequently poorer prognosis in melanoma, we sought to evaluate its biological role in human melanoma. We have established that the mutational status of p53 has a direct effect on the chemosensitivity of melanoma cell lines. Our data suggests a mechanistic role for p53 in the repression of anti-apoptotic Bcl -2 family members and the multidrug transporters, particularly P-gp (Li et al, 2000). To follow-up on the potential interaction between P-gp and p53, we analyzed an in vivo system (see Appendix I). The problem with using a tumour cell line is the uncertainty of other mutations affecting other loci. In our comprehensive review of p53 and P-gp (Bush and L i , 2002), we outlined the necessity for a more thorough examination of multidrug transporters in />53-deficient mice as this represents a clean genetic system. Recently, we have completed work addressing the regulatory role for p53 of Mdrl genes and the functional significance of increased drug transporters in primary skin cells. Our results favour the model that dysfunction or elimination of p53 relieves the suppression of P-gp and this is consistent with our melanoma data (L i etal, 1999) 115 Accumulating data indicates that although p53 is an important factor in chemosensitivity o f melanoma, the p53 mutational status cannot entirely explain the overwhelming resistance o f melanoma cells to chemotherapy. D N A sequencing has indicated that p53 mutations occur in only 20-30% of melanomas, while the majority of patients with melanomas (over 85%) have poor response to chemotherapy. Therefore, other apoptosis genes must also be involved in melanoma chemosensitivity. In other words, dysfunction of either upstream or downstream components in the p53 pathway leading to apoptosis could be inactivated. Two apparent candidates are Chk2 and Apaf-1 (Satyamoorfhy et al., 2000; Soengas et al., 2001). Oddly, in eight different melanoma cell lines, we did not see a significant difference in endogenous Apaf-1 expression levels among the cell lines. This is difficult to resolve in light of the fact that there is a strong connection between families with germline p53 mutations (Li-Fraumeni syndrome) and increased incidence of melanoma (Eng et al., 2001). This would argue in favour of the importance for p53 dysfunction in melanoma. Previous work from collaborators demonstrated an increase in the anti-apoptotic members (Mcl-1 and B C 1 - X L ) o f the Bcl-2 family in advancing melanoma and not in the preneoplastic benign nevi (Tang et al., 1998). We chose to pursue this mechanism further by applying a 'gene therapy' tactic to compensate for the anti-apoptotic behaviour of melanoma. We used several pro-apoptotic genes participating downstream of p53 to overcome any p53-dependent role in the mitochondria-mediated cell death mechanism. However, these pro-apoptotic genes act either immediately upstream of the mitochondria including Bax, and the newly characterized Noxa, and P U M A or the genes were immediately downstream of the mitochondria, Apaf-1 and procaspase-3, involved in the execution of the apoptogenic signals. Ectopic overexpression of these genes can effectively sensitize melanoma cells by themselves and could further enhance the cytotoxic effect of several different drugs. There may be a p5 3-dependence to the activity of these proteins as they are known to be transcriptional targets of p53. Stable cell lines expressing these pro-death factors are also 116 sensitized to drug treatment. We believe this approach is feasible and the data provides a proof-of-principle that these genes are worthwhile long-term targets either through up-regulation by gene-delivery methods or stabilization by small molecule species. In another strategy aimed at delineating the utility of dietary antioxidants in cancer, we studied the effects of curcumin. Curcumin belongs to a large group of antioxidants that are renowned for their therapeutic value in numerous diseases. Other studies suggested that p53 may be involved in mediating the benefits of curcumin in other tumour cell types. Our effort to define the pro-apoptotic mechanism of curcumin in melanoma cells suggested that a p53-independent pathway was activated. In fact, a death receptor-mediated mechanism was found. This is consistent with newer data showing that other stimuli such as anticancer agents and U V R can activate the same pathway (Bush et al., 2001). We feel antioxidant adjunct treatment defines another potential route to exploit for chemosensitization. Thus, we have investigated three basic strategies to decipher possible mechanisms leading to the control of chemoresistance and which could increase the sensitivity of melanoma cells to drug-induced apoptosis. In the literature, it is clear that simultaneous combination treatment of cancers including melanoma is the future in therapy (Bonaccorsi et al., 2001). New therapy regimes for melanoma try to take advantage of multi-modal treatments such as multiple drugs or combined chemo/immunotherapy (Eton et al., 2002). M y work emphasizes this rationale of combinatorial medicine. We advocate a similar design for future studies pursuing this logic and are continuing in such a manner. 6.2 F U T U R E D I R E C T I O N S To advance our understanding o f ectopic expression o f pro-apoptotic genes, we are currently undertaking several directions to improve expression efficiency. We are in the process of 117 establishing adenoviral vectors to transduce the Noxa and PUMA genes into melanoma cells to achieve higher (nearly 100% expression) in the test cells. Following this, our goal is to treat subcutaneous melanoma tumours developed in immunodeficient (SCTD) mice with the Noxa and P U M A adenoviral particles. Ultimately, an in vivo model demonstrating regression of subcutaneous tumours would validate the use of these genes in a preclinical setting as is ongoing for late-phase p53 trials (Merritt et al., 2001). Furthermore, we are continuing with other vector combinations and selecting for stable cell lines with inducible promoters that better regulate gene expression in vitro. We are also developing multicistronic vectors to target multiple pathways. If for instance, a melanoma cell is avoiding apoptosis by suppressing both mitochondrial and death receptor mechanisms then simultaneous overexpression of F A D D , Bax, and procaspase-3 should have severe detrimental effects for the viability of that cell. Alternatively, a recent study confirmed that down-regulating an oncogene and up-regulating a tumour suppressor gene can enhance apoptosis greater than either single gene (Su et al, 2001). Another approach would take advantage of the antioxidant properties of curcumin combined with overexpression of caspase-8 and drug treatment. We envision numerous combinations like these examples. The crucial question of how to target such a potentially lethal combination to melanoma cells still remains. Therefore, we are further refining our work on the melanoma-specific promoter, tyrosinase. Speculation on some new regulatory possibilities include other melanoma-specific genes such as the tyrosinase-related proteins-1 and -2 (TRP-1, -2), the melanoma-associated antigens ( M A G E ) , or the microphthalmia-associated transcription factor (MITF) involved in melanocyte differentiation. 118 REFERENCES Afaq F, Adhami V M , Ahmad N , Mukhtar H . Botanical anti-oxidants for chemoprevention of photocarcinogenesis. Front. Biosci. 2002, 7: 784-92. Albrechtsen N , Dornreiter I, Grosse F, K i m E , Wiesmuller L , Deppert W . Maintenance of genomic integrity by p53: complementary roles for activated and non-activated p53. Oncogene 1999, 18: 7706-17. Ammon H P , Wahl M A . Pharmacology of Curcuma longa. Planta Med. 1991, 57: 1-7; Ananthanarayanan M , Balasubramanian N , Makishima M , Mangelsdorf D J , Suchy FJ . Human bile salt export pump promoter is transactivated by the Farnesoid X Receptor/Bile A c i d Receptor. J. Biol. Chem. 2001, 276: 28857-65. Arafat W O , Gomez-Navarro J, Xiang J, Barnes M N , Mahasreshti P, Alvarez R D , Siegal G P , Badib A O , Buchsbaum D , Curiel D T , Stackhouse M A . A n adenovirus encoding proapoptotic Bax induces apoptosis and enhances the radiation effect in human ovarian cancer. Mol. Ther. 2000, 1: 545-54. Aragane Y , Kulms D , Metze D , Wilkes G , Poppelmann B , Luger T A , Schwarz T. Ultraviolet light induces apoptosis via direct activation of CD95 (Fas/APO-1) independently o f its ligand C D 9 5 L . J. Cell Biol. 1998, 140: 171-82. Ashikenazi A , Dix i t V M . Death receptors: signaling and modulation. Science 1998, 281: 1305-8. Ashkenazi A , Pai R C , Fong S, Leung S, Lawrence D A , Marsters S A , Blackie C , Chang L , McMurtrey A E , Hebert A , DeForge L , Koumenis IL, Lewis D , Harris L , Bussiere J, Koeppen H , Shahrokh Z , Schwall R H . Safety and anti-tumor activity of recombinant soluble Apo2 ligand. J. Clin. Invest. 1999, 104: 155-62. Azuine M A , Bhide S. Chemopreventive effect o f turmeric against stomach and skin tumors induced by chemical carcinogenesis in Swiss mice. Nutr. Cancer 1992, 17: 77-83. Backlund M G , Trasti SL , Backlund D C , Cressman V L , Godfrey V , Kol ler B H . Impact of ionizing radiation and genetic background on mammary tumorigenesis in p5 3-deficient mice. Cancer Res. 2001, 61: 6577-82. Bae I, Smith M L , Sheikh M S , Zhan Q, Scudiero D A , Friend S H , O'Connor P M , Fornace A J Jr. A n abnormality in the p53 pathway following y-irradiation in many wild-type p53 human melanoma lines. Cancer Res. 1996, 56: 840-7. Bakker T R , Reed D , Renno T, Jongeneel C V . Efficient adenoviral transfer o f NF-kappaB inhibitor sensitizes melanoma to tumor necrosis factor-mediated apoptosis. Int. J. Cancer 1999, 80: 320-3. Balajee A S , Bohr V A . Genomic heterogeneity of nucleotide excision repair. Gene 2000, 250: 15-30. 119 Balch C M , Houghton A , Sober A , Soong S-J, eds. Cutaneous Melanoma. 3rd ed. St Louis, M o : Quality Medical Publishing; 1998. Balint E E , Vousden K H . Activation and activities o f the p53 tumour suppressor protein. Br. J. Cancer 2001, 85: 1813-23. Barak Y , Juven T, Haffner R, Oren M . Mdm2 expression is induced by wi ld type p53 activity. EMBOJ. 1993, 12: 461-8. Bargonetti J, Manfredi JJ. Multiple roles of the tumor suppressor p53. Curr. Opin. Oncol. 2002, 14: 86-91. Bassi L , Palitti F, Mosesso P, Natarajan A T . Distribution of camptothecin-induced break points in Chinese hamster cells treated in late S and G2 phases o f the cell cycle. Mutagenesis 1998, 13: 257-61. Bates S, Phillips A C , Clark P A , Stott F, Peters G , Ludwig R L , Vousden K H . p l 4 A R F links the tumour suppressors R B and p53. Nature 1998, 395: 124-5. Bean L J , Stark G R . Phosphorylation of serines 15 and 37 is necessary for efficient accumulation of p53 following irradiation with U V . Oncogene 2001, 20: 1076-84. Beck JF, Brugger D , Brischwein K , L i u C, Bader P, Niethammer D , Gekeler V . Anti-cancer drug-mediated induction of multidrug resistance-associated genes and protein kinase C isozymes in the T-lymphoblastoid cell line C C R F - C E M and in blasts from patients with acute lymphoblastic leukemias. Jpn. J. Cancer Res. 2001, 92: 896-903. Beckman G , Birgander R, Sjalander A , Saha N , Holmberg P A , K ive l a A , Beckman L . Is p53 polymorphism maintained by natural selection? Hum. Hered. 1994, 44: 266-70. Benathan M , Alvero-Jackson H , M o o y A M , Scaletta C, Frenk E . Relationship between melanogenesis, glutathione levels and melphalan toxicity in human melanoma cells. Melanoma Res. 1992,2: 305-14. Beukers R, Berends, W . Isolation and identification o f the irradiation product o f thymine. Biochim. Biophys. Acta 1960, 41: 550-1. Bhaumik S, Anjum R, Rangaraj N . Curcumin mediated apoptosis in A K - 5 tumor cells involves the production of reactive oxygen intermediates. FEBS Lett. 1999, 456: 311-14. Bian J, Sun Y . Transcriptional activation by p53 of the human type IV collagenase (gelatinase A or matrix metalloproteinase 2) promoter. Mol. Cell. Biol. 1997, 17: 6330-8. Bickers D R , Athar M . Novel approaches to chemoprevention of skin cancer. J. Dermatol. 2000, 27: 691-5. Blagosklonny M V . P53: an ubiquitous target of anti-cancer drugs. Int. J. Cancer 2002, 98: 161-6. 120 Boise L H , Gonzalez-Garcia M , Postema C E , Ding L , Lindsten T, Turka L A , Mao X , Nunez G , Thompson C B . bcl-x, a bcl-2-related gene that functions as a dominant regulator o f apoptotic cell death. Cell 1993, 74: 597-608. Bonaccorsi P, Ansel JC, Armstrong C A . Management of high-risk melanoma. Dermatol. Clin. 2001, 19: 727-35. Brambilla E . Negoescu A . Gazzeri S. Lantuejoul S. Moro D . Brambilla C. C o l l JL . Apoptosis-related factors p53, Bcl -2 , and Bax in neuroendocrine lung tumors. Am. J. Pathol. 1996, 149: 1941-52. Budhram-Mahadeo V , Morris PJ , Smith M D , Midgley C A , Boxer L M , Latchman D S . p53 suppresses the activation of the Bcl-2 promoter by the Brn-3a P O U family transcription factor. J. Biol. Chem. 1999, 274: 15237-44. Bush J A , Cheung K J Jr, L i G . Curcumin induces apoptosis in human melanoma cells through a Fas receptor/caspase-8 pathway independent of p53. Exp. Cell Res. 2001, 271: 305-14. Bush J A , L i G . Cancer chemoresistance: the relationship between p53 and multidrug transporters. Int. J. Cancer 2002, 98: 323-30. Byers H R , Etoh T, Doherty JR, Sober A J , M i h m M C Jr. Cel l migration and actin organization in cultured human primary, recurrent cutaneous and metastatic melanoma. Time-lapse and image analysis. Am. J. Pathol. 1991, 139: 423-35. Caltagirone S, Rossi C , Poggi A , Ranelletti F O , Natali P G , Brunetti M , Aie l lo F B , Piantelli M . Flavonoids apigenin and quercetin inhibit melanoma growth and metastatic potential. Int. J. Cancer 2000, 87: 595-600. Chan M M , Huang HI, Fenton M R , Fong D . In vivo inhibition of nitric oxide synthase gene expression by curcumin, a cancer preventive natural product with anti-inflammatory properties. Biochem. Pharmacol. 1998, 55: 1955-62. Chappell D B , Zaks T Z , Rosenberg S A , Restifo N P . Human melanoma cells do not express Fas (Apo- l /CD95) ligand. Cancer Res. 1999, 59: 59-62. Chatterjee D , Wyche J H , Pantazis P. Induction of apoptosis in malignant and camptothecin-resistant human cells. Ann. N.Y. Acad. Sci. 1996, 803: 143-56. Chen H W , Huang H C . Effect of curcumin on cell cycle progression and apoptosis in vascular smooth muscle cells. Br. J. Pharmacol. 1998, 124: 1029-40. Chen Y C , K u o T C , Lin-Shiau S Y , L i n J K . Induction o f HSP70 gene expression by modulation of C a ( + 2 ) ion and cellular p53 protein by curcumin in colorectal carcinoma cells. Molec. Carcinogenesis 1996, 17: 224-34. Chen Y R , Tan T H . Inhibition of the c-Jun terminal kinase ( JNK) signaling pathway by curcumin. Oncogene 1998, 17: 173-8. 121 Cheung K J Jr, Bush J A , Jia W , L i G. Expression of the novel tumor suppressor p33(INGl) is independent of p53. Br. J. Cancer 2000, 83: 1468-72. Childs S, Y e h R L , Hu i D , L ing V . Taxol resistance mediated by transfection of the liver-specific sister gene of P-glycoprotein. Cancer Res. 1998, 58: 4160-7. Chin K V , Ueda K , Pastan I, Gottesman M M . Modulation o f activity o f the promoter o f the human M D R 1 gene by Ras and p53. Science 1992, 255: 459-62. Chinnaiyan A M , O'Rourke K , Tewari M , Dixi t V M . F A D D , a novel death domain-containing protein, interacts with the death domain of Fas and initiates apoptosis. Cell 1995, 81: 505-12. Clahsen P C , van de Velde CJ , Duval C , Pallud C, Mandard A M , Delobelle-Deroide A , van den Broek L , Sahmoud T M , van de Vijver M J . p53 protein accumulation and response to adjuvant chemotherapy in premenopausal women with node-negative early breast cancer. J. Clin. Oncol. 1998, 16:470-9. Cohen G L , Falkson CI. Current treatment options for malignant melanoma. Drugs 1998, 55: 791-9. Colman M S , Afshari C A , Barrett JC. Regulation of p53 stability and activity in response to genotoxic stress. Mutat. Res. 2000, 462: 179-88. Conseil G , Perez-Victoria J M , Jault J M , Gamarro F , Goffeau A , Hofmann J, D i Pietro A . Protein kinase C effectors bind to multidrug A B C transporters and inhibit their activity. Biochemistry 2001,40: 2564-71. Creagh E M , Martin SJ. caspases: cellular demolition experts. Biochem. Soc. Trans. 2001, 29: 696-702. Cristofolini M , B o i S, Girlando S, Zumiani G , Cristofolini P, Dal la Palma P, Doglioni C , Barbarcschi M . p53 protein expression in nevi and melanomas. Arch. Dermatol. 1993, 129: 739-43. Croop J M , Raymond M , Haber D , Devault A , A r c e d R J , Gros P, Housman D E . The three mouse multidrug resistance (mdr) genes are expressed in a tissue-specific manner in normal mouse tissues. Mol. Cell Biol. 1989, 9: 1346-50. Dameron K M , Volpert O V , Tainsky M A , Bouck N . Control of angiogenesis in fibroblasts by p53 regulation of thrombospondin-1. Science 1994, 265: 1582-4. Das A , Chendil D , Dey S, Mohiuddin M , Mohiuddin M , Milbrandt J, Rangnekar V M , Ahmed M M . Ionizing radiation down-regulates p53 protein in primary Egr-1-/- mouse embryonic fibroblast cells causing enhanced resistance to apoptosis. J. Biol. Chem. 2001, 276: 3279-86. de Gruijl FR . Photocarcinogenesis: U V A vs U V B . Methods Enzymol. 2000, 319: 359-66. de Kant E , Heide I, Thiede C, Herrmann R, Rochlitz C F . M D R 1 expression correlated with mutant expression in colorectal cancer metastases. J. Cancer Res. Clin. Oncol. 1996, 122: 671-5. 122 Dean M , Rzhetsky A , Allikmets R. The human ATP-binding cassette ( A B C ) transporter superfamily. Genome Res. 2001, 11: 1156-66. Deb SP, Munoz R M , Brown D R , Subler M A , Deb S. Wild-type human p53 activates the human epidermal growth factor receptor promoter. Oncogene 1994, 9: 1341-9. Demeule M , Labelle M , Regina A , Berthelet F , Beliveau R. Isolation of endothelial cells from brain, lung, and kidney: expression of the multidrug resistance P-glycoprotein isoforms. Biochem. Biophys. Res. Commun. 2001, 281: 827-34. Desai SD, L i u L F , Vazquez-Abad D , D'Arpa P. Ubiquitin-dependent destruction of topoisomerase I is stimulated by the anti-tumor drug camptothecin. J. Biol. Chem. 1997, 272: 24159-64. Devine SE, Hussain A , Davide JP, Melera P W . Fu l l length and alternatively spliced pgpl transcripts in multidrug-resistant Chinese hamster lung cells. J. Biol. Chem. 1991, 266: 4545-55. Dhir R, Buschman E , Gros P. Structural and functional characterization of the mouse multidrug resistance gene family. Bull. Cancer 1990, 77: 1125-9. Diccianni M B , Y u J, Hsiao M , Mukherjee S, Shao L E , Y u A L . Cl inical significance of p53 mutations in relapsed T-cell acute lymphoblastic leukemia. Blood 1994, 84: 3105-12. Donehower L A , Harvey M , Slagle B L , McArthur M J , Montgomery C A Jr, Butel JS, Bradley A . M i c e deficient for p53 are developmentally normal but susceptible to spontaneous tumours. Nature 1992,356:215-21. Doyle L A , Yang W , Abruzzo L V , Krogmann T, Gao Y , Rishi A K , Ross D D . A multidrug resistance transporter from human M C F - 7 breast cancer cells. Proc. Natl. Acad. Sci. U.S.A. 1998, 95: 15665-70. D'Souza S, X i n H , Walter S, Choubey D . The gene encoding p202, an interferon-inducible negative regulator o f the p53 tumor suppressor, is a target o f p53-mediated transcriptional repression. J. Biol. Chem. 2001, 276: 298-305. Duesberg P, Stindl R, Hehlmann R. Origin of multidrug resistance in cells with and without multidrug resistance genes: chromosome reassortments catalyzed by aneuploidy. Proc. Natl. Acad. Sci. U.S.A. 2001, 98: 11283-8. Dumaz N , van Kranen H J , de Vries A , Berg R J , Wester P W , van Krei j l C F , Sarasin A , Daya-Grosjean L , de Gruijl FR . The role of U V - B light in skin carcinogenesis through the analysis of p53 mutations in squamous cell carcinomas of hairless mice. Carcinogenesis 1997, 18: 897-904. Dummer R, Nestle F O , Burg G . Report on the Eighth World Congress on Cancers of the Skin 18-21 July 2001, Zurich, Switzerland. Melanoma Res. 2001, 11: 549-50. Earnshaw W C , Martins L M , Kaufmann S H . Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu. Rev. Biochem. 1999, 68: 383-424. 123 El-Deiry W S , Kern SE, Pietenpol J A , Kinzler K W , Vogelstein B . Definition of a consensus binding site for p53. Nat. Genet. 1992, 1: 45-9. El-Deiry W S , Tokino T, Velculescu V E , Levy D B , Parsons R, Trent J M , L i n D , Mercer W E , Kinzler K W , Vogelstein B . W A F 1 , a potential mediator of p53 tumor suppression. Cell 1993, 75: 817-25. Eliopoulos A G . Kerr D J . Herod J. Hodgkins L . Krajewski S. Reed JC. Young L S . The control of apoptosis and drug resistance in ovarian cancer: influence of p53 and Bcl -2 . Oncogene 1995, 11: 1217-28. Eng C , Hampel H , de la Chapelle A . Genetic testing for cancer predisposition. Annu. Rev. Med. 2001, 52: 371-400. Eton O, Legha SS, Bedikian A Y , Lee JJ, Buzaid A C , Hodges C, Ring SE, Papadopoulos N E , Plager C, East M J , Zhan F, Benjamin R S . Sequential biochemotherapy versus chemotherapy for metastatic melanoma: results from a phase III randomized trial. J. Clin. Oncol. 2002, 20: 2045-52 Faleiro L R , Kobayashi R. Fearnhead H , Lazebnik Y . Multiple species of CPP32 and Mch2 are the major active caspases present in apoptotic cells. EMBO J. 1997, 16: 2271-81. Fan S, el-Deiry W S , Bae I, Freeman J, Jondle D , Bhatia K , Fornace A J Jr, Magrath I, Kohn K W , O'Connor P M . p53 gene mutations are associated with decreased sensitivity of human lymphoma cells to D N A damaging agents. Cancer Res. 1994, 54: 5824-30. Findley H W , Gu L , Yeager A M , Zhou M . Expression and regulation of bcl-2, bcl -xl , and bax correlate with p53 status and sensitivity to apoptosis in childhood acute lymphoblastic leukemia. Blood 1997, 89: 2986-93. Fink D , Aebi S, Howel l SB. The role of D N A mismatch repair in drug resistance. Clin. Cancer Res. 1998,4: 1-6. Ford JC, al-Khodairy F , Fotou E , Sheldrick K S , Griffiths D J , Carr A M . 14-3-3 protein homologs required for the D N A damage checkpoint in fission yeast. Science 1994, 265: 533-5. Foster B A , Coffey H A , Mor in M J , Rastinejad F. Pharmacological rescue o f mutant p53 conformation and function. Science 1999, 286: 2507-10. Friedberg E C . Out o f the shadows and into the light: the emergence o f D N A repair. Trends Biochem. Sci. 1995,20: 381. Friedrich K , Wieder T, V o n Haefen C, Radetzki S, Janicke R, Schulze-Osthoff K , Dorken B , Daniel PT. Overexpression of caspase-3 restores sensitivity for drug-induced apoptosis in breast cancer cell lines with acquired drug resistance. Oncogene 2001, 20: 2749-60. Fulda S, Meyer E , Debatin K M . Inhibition of TRAIL-induced apoptosis by Bcl-2 overexpression. Oncogene 2002, 21: 2283-94. 124 Galati G , Teng S, Moridani M Y , Chan TS, O'Brien PJ . Cancer chemoprevention and apoptosis mechanisms induced by dietary polyphenolics. Drug Metabol. Drug Interact. 2000, 17: 311-49. Galimberti S, Marchetti A , Buttitta F, Carnicelli V , Pellegrini S, Bevilacqua G , Petrini M . Multidrug resistance related genes and p53 expression in human non small cell lung cancer. Anti-cancer Res. 1998, 18: 2973-6. G i l l P K , Gescher A , Gant T W . Regulation of M D R 1 promoter activity in human breast carcinoma cells by protein kinase C isozymes alpha and theta. Eur. J. Biochem. 2001, 268: 4151-7. Goldwasser F, Bae I, Fornace A J Jr, Pommier Y . Differential G A D D 4 5 , p 2 1 C I P l / W A F l , M C L -1 and topoisomerase JJ gene induction and secondary D N A fragmentation after camptothecin-induced D N A damage in two mutant p53 human colon cancer cell lines. Oncol. Res. 1996, 8: 317-23. Grant S W , Kyshtoobayeva A S , Kurosaki T, Jakowatz J, Fruehauf JP. Mutant p53 correlates with reduced expression of thrombospondin-1, increased angiogenesis, and metastatic progression in melanoma. Cancer Detect Prev. 1998, 22: 185-94. Green D R , Reed JC. Mitochondria and apoptosis. Science 1998, 281: 1309-12. Grossman D , K i m PJ , Schechner JS, Alt ier i D C . Inhibition of melanoma tumor growth in vivo by survivin targeting. Proc. Natl. Acad. Sci. U.S.A. 2001, 98: 635-40. Grossman SR, Perez M , Kung A L , Joseph M , Mansur C, X iao Z X , Kumar S, Howley P M , Livingston D M . p300 /MDM2 complexes participate in MDM2-mediated p53 degradation. Mol. Cell 1998, 2: 405-15. Grover R , Wilson G D . Bcl-2 expression in malignant melanoma and its prognostic significance. Eur. J. Surg. Oncol. 1996, 22: 347-9. Haffner R, Oren M . Biochemical preperties and biological effects of p53. Curr. Opin. Genet. Dev. 1995, 5: 84-90. Hajek P, Vi l l an i G , Attardi G. Rate-limiting step preceding cytochrome c release in cells primed for Fas-mediated apoptosis revealed by analysis of cellular mosaicism of respiratory changes. J. Biol. Chem. 2001, 276: 606-15. Han SS, Chung ST, Robertson D A , Ranjan D , Bondada S. Curcumin causes the growth arrest and apoptosis of B cell lymphoma by downregulation of egr-1, c-myc, b c l - X L , NF-kappa B , and p53. Clin. Immunol. 1999, 93: 152-61. Hanahan D , Weinberg R A . The hallmarks of cancer. Cell 2000, 100: 57-70. Harris L C , Remack JS, Houghton PJ , Brent TP. Wild-type p53 suppresses transcription of the human 06-methylguanine-DNA methyltransferase gene. Cancer Res. 1996, 56: 2029-32. 125 Hartmann A , Blaszyk H , Cunnigham JS, McGovern R M , Schroeder JS, Helander SD, Pittelkow M R , Sommer SS, Kovach JS. Overexpression and mutations of p53 in metastatic melanomas. Int. J. Cancer 1996, 67: 313-7. Haupt Y , Maya R, Kazaz A , Oren M . Mdm2 promotes the rapid degradation of p53. Nature 1997, 387: 296-9. Hayes JD, M c M a h o n M . Molecular basis for the contribution of the anti-oxidant responsive element to cancer chemoprevention. Cancer Lett. 2001, 174: 103-13. Hermeking H , Lengauer C, Polyak K , He T C , Zhang L , Thiagalingam S, Kinzler K W , Vogelstein B . 14-3-3 sigma is a p53-regulated inhibitor o f G 2 / M progression. Mol. Cell 1997, 1: 3-11. Hersey P, Zhang X D . How melanoma cells evade trail-induced apoptosis. Nature Rev. Cancer 2001, 1: 142-50. Hida A , Kawakami A , Nakashima T, Yamasaki S, Sakai H , Urayama S, Ida H , Nakamura H , Migi ta K , Kawabe Y , Eguchi K . Nuclear factor-kappaB and caspases co-operatively regulate the activation and apoptosis of human macrophages. Immunology 2000, 99: 553-60. Hijikata M , Kato N , Sato T, Kagami Y , Shimotohno K . Molecular cloning and characterization of a c D N A for a novel phorbol-12-myristate-13-acetate-responsive gene that is highly expressed in an adult T-cell leukemia cell line. J. Virol. 1990, 64: 4632-9. Hirata H , Takahashi A , Kobayashi S, Yonehara S, Sawai H , Okazaki T, Yamamoto K , Sasada M . caspases are activated in a branched protease cascade and control distinct downstream processes in Fas-induced apoptosis. J. Exp. Med. 1998, 187: 587-600. Hofer-Warbinek R, Schmid J A , Stehlik C, Binder B R , Lipp J, de Martin R. Activation of N F -kappa B by X I A P , the X chromosome-linked inhibitor of apoptosis, in endothelial cells involves T A K 1 . J. Biol. Chem. 2000, 275: 22064-8. Hoffman W H , Biade S, Zilfou JT, Chen J, Murphy M . Transcriptional repression of the anti-apoptotic survivin gene by wi ld type p53. J. Biol. Chem. 2002, 277: 3247-57. Hollstein M , Soussi T, Thomas G , von Brevern M C , Bartsch H . P53 gene alterations in human tumors: perspectives for cancer control. Recent Results Cancer Res. 1997, 143: 369-89. Hopper E , Belinsky M G , Zeng H , Tosolini A , Testa JR, Kruh G D . Analysis o f the structure and expression pattern of M R P 7 (ABCC10) , a new member of the M R P subfamily. Cancer Lett. 2001, 162: 181-91. Howe H L , Wingo P A , Thun M J , Ries L A , Rosenberg H M , Feigal E G , Edwards B K . Annual report to the nation on the status of cancer (1973 through 1998), featuring cancers with recent increasing trends. J. Natl. Cancer Inst. 2001, 93: 824-42. 126 Hrycyna C A . Molecular genetic analysis and biochemical characterization of mammalian P-glycoproteins involved in multidrug resistance. Semin. Cell Dev. Biol. 2001, 12: 247-56. Hsu SD, Singh B B , Lewis JB, Borke JL , Dickinson DP , Drake L , Caughman G B , Schuster G S . Chemoprevention of oral cancer by green tea. Gen. Dent. 2002, 50: 140-6. H u Y , Tanzer L R , Cao J, Geringer C D , Moore R E . Use of long R T - P C R to characterize splice variant m R N A s . Biotechniques 1998, 25: 224-9. Huang M T , M a W , Y e n P, X i e JG , Han J, Frenkel K , Grunberger D , Conney A H . Inhibitory effects of topical application of low doses of curcumin on 12-0-tetradecanoylphorbol-13-acetate-induced tumor promotion and oxidized D N A bases in mouse epidermis. Carcinogenesis 1997, 18: 83-8. Huang M T , Newmark H L , Frenkel K . Inhibitory effects o f curcumin on tumorigenesis in mice. J. Cell Biochem. 1997,27: S26-34. Huh W K , Gomez-Navarro J, Arafat W O , Xiang J, Mahasreshti PJ , Alvarez R D , Barnes M N , Curiel D T . Bax-induced apoptosis as a novel gene therapy approach for carcinoma o f the cervix. Gynecol. Oncol. 2001, 83: 370-7. Hwang B J , Ford J M , Hanawalt P C , Chu G . Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair. Proc. Natl. Acad. Sci. U.S.A. 1999, 96: 424-8. Iotsova V , Stehelin D . Down-regulation of fibronectin gene expression by the p53 tumor suppressor protein. Cell Growth Differ. 1996, 7: 629-34. Irmler M , Thome M , Hahne M , Schneider P, Hofmann K , Steiner V , Bodmer JL , Schroter M , Burns K , Mattmann C, Rimoldi D , French L E , Tschopp J. Inhibition of death receptor signals by cellular FLIP . Nature 1997, 388: 190-5. Ishikawa T, K u o M T , Furuta K , Suzuki M . The human multidrug resistance-associated protein ( M R P ) gene family: from biological function to drug molecular design. Clin. Chem. Lab. Med. 2000, 38: 893-7. Iwadate Y , Tagawa M , Fujimoto S, Hirose M , Namba H , Sueyoshi K , Sakiyama S, Yamaura A . Mutation of the p53 gene in human astrocytic tumours correlates with increased resistance to DNA-damaging agents but not to anti-microtubule anti-cancer agents. Br. J. Cancer 1998, 77: 547-51. Jackson R J , Howel l M T , Kaminski A . The novel mechanism o f initiation o f picornavirus R N A translation. Trends Biochem. Sci. 1990, 15: 477-83. Jansen B , Schlagbauer-Wadl H , Brown B D , Bryan R N , van Elsas A , Mul ler M , W o l f f K , Eichler H G , Pehamberger H . bcl-2 anti-sense therapy chemosensitizes human melanoma in SCID mice. Nat. Med. 1998, 4: 232-4. 127 Jee S H , Shen S C , Tseng C R , Chiu H C , Kuo M L . Curcumin induces a p53-dependent apoptosis in human basal cell carcinoma cells. J. Invest. Dermatol. 1998, 11: 656-61. Jenrette J M . Malignant melanoma: the role of radiation therapy revisited. Semin. Oncol. 1996, ' 23: 759-62. Jiang M C , Yang-Yen H F , L i n J K , Y e n JJ. Differential regulation of p53, c-Myc, Bc l -2 and Bax protein expression during apoptosis induced by widely divergent stimuli in human hepatoblastoma cells. Oncogene 1996, 13: 609-16. Jobin C, Bradham C A , Russo M P , Juma B , Narula A S , Brenner D A , Sartor R B . Curcumin blocks cytokine-mediated NF-kappa B activation and proinflammatory gene expression by inhibiting inhibitory factor I-kappa B kinase activity. J. Immunol. 1999, 163: 3474-83. Johnson R A , Ince T A , Scotto K W . Transcriptional repression by p53 through direct binding to a novel D N A element. / . Biol. Chem. 2001, 276: 27716-20. Jones C B , Clements M K , Wasi S, Daoud SS. Sensitivity to camptothecin of human breast carcinoma and normal endothelial cells. Cancer Chemotherapy Pharmacol. 1997, 40: 475-83. Juliano R L , L ing V . A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim. Biophys. Acta 1976, 455: 152-62. Kagawa S, Pearson S A , J i L , X u K , McDonnel l TJ , Swisher S G , Roth J A , Fang B . A binary adenoviral vector system for expressing high levels of the proapoptotic gene bax. Gene Ther. 2000, 7: 75-9. Kamarajan P, Sun N K , Sun C L , Chao C C . Apaf-1 overexpression partially overcomes apoptotic resistance in a cisplatin-selected HeLa cell line. FEBSLett. 2001, 505: 206-12. Kamesaki H . Mechanisms involved in chemotherapy-induced apoptosis and their implications in cancer chemotherapy. Int. J. Hematol. 1998, 68: 29-43. Kanaya T, K y o S, Hamada K , Takakura M , Kitagawa Y , Harada H , Inoue M . Adenoviral expression of p53 represses telomerase activity through down-regulation of human telomerase reverse transcriptase transcription. Clin. Cancer Res. 2000, 6: 1239-47. Kannan K , Kaminski N , Rechavi G , Jakob-Hirsch J, Amariglio N , G ivo l D . D N A microarray analysis of genes involved in p53 mediated apoptosis: activation of Apaf-1. Oncogene 2001, 20: 3449-55. Kanoko M , Ueda M , Nagano T, Ichihashi M . Expression of p53 protein in melanoma progression./. Dermatol. Sci. 1996, 12: 97-103. Kantharidis P, El-Osta A , deSilva M , Wal l D M , H u X F , Slater A , Nadalin G , Parkin JD, Zalcberg JR. Altered methylation of the human A b c b l promoter is associated with acquired multidrug resistance. Clin. Cancer Res. 1997, 3: 2025-32. 128 Kasof G M , Gomes B C . L i v i n , a novel inhibitor of apoptosis protein family member. J. Biol. Chem. 2001,276: 3238-46. Kato K , Ito H , Kamei K , Iwamoto I. Stimulation of the stress-induced expression of stress proteins by curcumin in cultured cells and in rat tissues in vivo. Cell Stress Chaperones 1998, 3: 152-60. Kaufmann S H , Earnshaw W C . Induction of apoptosis by cancer chemotherapy. Exp. Cell Res. 2000, 256: 42-9. Kawakami A , Nakashima T, Sakai H , Urayama S, Yamasaki S, Hida A , Tsuboi M , Nakamura H , Ida H , Migi ta K , Kawabe Y , Eguchi K . Inhibition of caspase cascade by H T L V - I tax through induction of NF-kappaB nuclear translocation. Blood 1999, 94: 3847-54. Kawamori T, Lubet R, Steele V E , Kel lof f G J , Kaskey R B , Rao C V , Reddy B S . Chemopreventive effect of curcumin, a naturally occurring anti-inflammatory agent, during the promotion/progression stages of colon cancer. Cancer Res. 1999, 59: 597-601. Kedersha N L , Rome L H . Isolation and characterization of a novel ribonucleoprotein particle: large structures contain a single species of small R N A . J. Cell Biol. 1986, 103: 699-709. Kerr JF, Wyl l ie A H , Currie A R . Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 1972, 26: 239-57. Keshelava N , Zuo II, Chen P, Waidyaratne S N , Luna M C , Gomer C J , Triche TJ , Reynolds C P . Loss of p53 function confers high-level multidrug resistance in neuroblastoma cell lines. Cancer Res. 2001,61:6185-93. Khar A , A l i A M . Anti-tumor activity of curcumin is mediated through the induction of apoptosis in A K - 5 tumor cells. FEBS Lett. 1999, 445: 165-8. K i m C J , Reintgen D S , Balch C M . The new melanoma staging system. Cancer Control 2002, 9: 9-15. K i m D J , Takasuka N , Nishino H , Tsuda H . Chemoprevention of lung cancer by lycopene. Biofactors 2000, 13: 95-102. K i m K , Lee C K , Sayers, TJ , Muegge K , Durum S K . The trophic action of IL-7 on pro-T cells: inhibitin of apoptosis of p ro -T l , -T2, and -T3 cells correlates with Bcl -2 and Bax levels and is independent of Fas andp53 pathways, d. Immunol. 1998, 160: 5735-41. Klaunig JE , Kamendulis L M . Mechanisms of cancer chemoprevention in hepatic carcinogenesis: modulation of focal lesion growth in mice. Toxicol. Sci. 1999, 52: 101-6. Kleihues P, Schauble B , zur Hausen A , Esteve J, Ohgaki H . Tumors associated with p53 germline mutations: a synopsis of 91 families. Am. J. Pathol. 1997, 150: 1-13. 129 Kobayashi T, Ruan S, Clodi K , Kliche K O , Shiku H , Andreeff M , Zhang W . Overexpression of bax gene sensitizes K562 erythroleukemia cells to apoptosis induced by selective chemotherapeutic agents. Oncogene 1998, 16: 1587-91. K o h H K . Cutaneous melanoma. New Engl. J. Med. 1991, 325: 171-82. Kong A N , Y u R, Hebbar V , Chen C, Owuor E , H u R, Ee R, Mandlekar S. Signal transduction events elicited by cancer prevention compounds. Mutat. Res. 2001, 480-481: 231-41. Korabiowska M , Brinck U , Hoenig JF, Bartkowski SB, Kellner S, Marx D , Schauer A . Significance of P-53 anti-gen in malignant melanomas and naevi of the head and neck area. Anti-cancer Res. 1995, 15: 885-9. Korutla L , Cheung J Y , Mendelsohn J, Kumar R . Inhibition o f ligand-induced activation o f epidermal growth factor receptor tyrosine phosphorylation by curcumin. Carcinogenesis 1995, 16: 1741-5. Kowalski P, Wichert A , Holm PS, Dietel M , Lage H . Selection and characterization of a high-activity ribozyme directed against the anti-neoplastic drug resistance-associated A B C transporter B C R P / M X R / A B C G 2 . Cancer Gene Ther. 2001, 8: 185-92. Kulms D , Poppelmann B , Yarosh D , Luger T A , Kruntmann J, Schwarz T. Nuclear and cell membrane effects contribute independently to the induction of apoptosis in human cells exposed to U V B radiation. Proc. Natl. Acad. Sci. U.S.A. 1999, 96: 7974-9. Kupenvasser C, Hurlbut G D , Kittrell FS , Dickinson E S , Laucirica R, Medina D , Naber SP, Jerry D J . Development of spontaneous mammary tumors in B A L B / c p53 heterozygous mice. A model for Li-Fraumeni syndrome. Am. J. Pathol. 2000, 157: 2151-59. Lage H , Christmann M , Kern_MA, Dietel M , Pick M , . K a i n a B , Schadendorf D . Expression_of DNA'repai r proteins h M S H 2 , h M S H 6 , h M L H l , 06-methylguanine-DNA methyltransferase and N-methylpurine-DNA glycosylase in melanoma cells with acquired drug resistance. Int. J. Cancer 1999, 80: 744-50. r Lamb P, Crawford L . Characterization of the human p53 gene. Mol. Cell. Biol. 1986, 6: 1379-85. Lambert JD , Meyers R O , Timmermann B N , Dorr R T . Tetra-O-methylnordihydroguaiaretic acid inhibits melanoma in vivo. Cancer Lett. 2001, 171: 47-56. Lange C , Walther W , Schwabe H , Stein U . Cloning and initial analysis of the human multidrug resistance-related M V P / L R P gene promoter. Biochem. Biophys. Res. Commun. 2000, 278: 125-33. Lassam L A , From I, Kahn H J . Overexpression o f p53 is a late event in the development of, malignant melanoma. Cancer Res. 1993, 53: 2235-8. Lazebnik Y A , Cole S, Cooke C A , Nelson W G , Earnshaw W C . Nuclear events o f apoptosis in vitro in cell-free mitotic extracts: a model system for analysis of the active phase of apoptosis. J. Cell Biol. 1993, 123: 7-22. 130 Lecureur Vj- Thottassery J V , Sun D , Schuetz E G , Lahti J, Zambetti G P , Schuetz JD. M d r l b facilitates p53-mediated cell death and p53 is required for M d r l b upregulation in vivo. Oncogene 2001,20:303-13. Lee A , DeJong G , Guo J, B u X , Jia W W . Bax expressed from a herpes viral vector enhances the efficacy of N,N'-bis(2-hydroxyethyl)-N-nitrosourea treatment in a rat glioma model. Cancer Gene Ther. 2000, 7: 1113-9. Lee K C , Crowe A J , Barton M C . p53-mediated repression of alpha-fetoprotein gene expression by specific D N A binding. Mol. Cell. Biol. 1999, 19: 1279-88. Levin D B , Wilson K , Valadares de Amor im G , Webber J, Kenny P, Kusser W . Detection of p53 mutations in benign and dysplastic nevi. Cancer Res. 1995, 55: 4278-82. L i G , Bush J A , Ho V C . p53-dependent apoptosis in melanoma cells after treatment with camptothecin. J. Invest. Dermatol. 2000, 114: 514-9. L i G , Ho V C , Trotter M , Horsman D E , Tron V A . p53 mutation in metastatic melanomas and primary melanomas from sun-exposed and sun-protected sites. J. Eur. Acad. Dermatol. & Venereol. 1995, 4: 48-53. L i G , Mitchel l D L , Ho V C , Reed JC, Tron V A . Decreased D N A repair but normal apoptosis in ultraviolet-irradiated skin of p53-transgenic mice. Am. J. Pathol. 1996, 148: 1113-23. L i G , Tang L , Zhou X , Tron V , Ho V . Chemotherapy-induced apoptosis in melanoma cells is p53 dependent. Melanoma Res. 1998, 8: 17-23. L i M , Hurren R, Zastawny R L , Ling V , Buick R N . Regulation and expression of multidrug resistance ( M D R ) transcripts in the intestinal epithelium. Br. J. Cancer 1999, 80: 1123-31. L i X , Marani M , Y u J, Nan B , Roth J A , Kagawa S, Fang B , Denner L , Marcel l i M . Adenovirus-mediated Bax overexpression for the induction of therapeutic apoptosis in prostate cancer. Cancer Res. 2001, 61: 186-91. L i n Y , M a W , Benchimol S. Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. Nat. Genet. 2000, 26: 122-7. Lindahl T, Wood R D . Quality control by D N A repair. Science 1999, 286: 1897-905. L inn S C , Honkoop A H , Hoekman K , van der Valk P, Pinedo H M , Giaccone G . p53 and P-glycoprotein are often co-expressed and are associated with poor prognosis in breast cancer. Br. J. Cancer 1996, 74: 63-8. L i u J, Grogan L , Nau M M , Allegra C J , Chu E , Wright JJ. Physical interaction between p53 and primary response gene Egr-1. Int. J. Oncol. 2001, 18: 863-70. L i u X , Zou H , Slaughter C, Wang X . D F F , a heterodimeric protein that functions downstream of caspase-3 to trigger D N A fragmentation during apoptosis. Cell 1997, 89: 175-84. 131 L i u Y , Kulesz-Martin M . p53 protein at the hub of cellular D N A damage response pathways through sequence-specific and non-sequence-specific D N A binding. Carcinogenesis 2001, 22: 851-60. L i u Z G , Hsu H , Goeddel D V , Kar in M . Dissection of T N F receptor 1 effector functions: J N K activation is not linked to apoptosis while NF-kappaB activation prevents cell death. Cell 1996, 87: 565-76. Lowe SL , Rubinchik S, Honda T, McDonnel l TJ , Dong J Y , Norris JS. Prostate-specific expression of Bax delivered by an adenoviral vector induces apoptosis in L N C a P prostate cancer cells. Gene Ther. 2001, 8: 1363-71. L u Y P , Chang R L , L o u Y R , Huang M T , Newmark H L , Reuhl K R , Conney A H . Effect of curcumin on 12-O-tetradecanoylphorbol-13-acetate- and ultraviolet B light-induced expression of c-Jun and c-Fos in JB6 cells and in mouse epidermis. Carcinogenesis 1994, 15: 2363-70. Lubet R A , Zhang Z , Wiseman R W , Y o u M . Use o f p53 transgenic mice in the development o f cancer models for multiple purposes. Exp. Lung Res. 2000, 26: 581-93. Luu Y , Bush J, Cheung K J Jr, L i G . The p53 Stabilizing Compound CP-31398 Induces Apoptosis by Activating the Intrinsic Bax/Mitochondrial/Caspase-9 Pathway. Exp. Cell. Res. 2002, 276: 214-22. Madden M J , Morrow C S , Nakagawa M , Goldsmith M E , Fairchild C R , Cowan K H . Identification of 5' and 3' sequences involved in the regulation of transcription o f the human abcbl gene in vivo. J. Biol. Chem. 1993, 268: 8290-7. Madden SL. , Galella E A , Zhu J, Bertelsen A H , Beaudry G A . S A G E transcript profiles for p53-dependent growth regulation. Oncogene 1997, 15: 1079-85. M a k i C G , Howley P M . Ubiquitination of p53 and p21 is differentially affected by ionizing and U V radiation. Mol. Cell. Biol. 1997, 17: 355-63. Malafa M P , Fokum F D , Mowlav i A , Abusief M , K i n g M . Vitamin E inhibits melanoma growth in mice. Surgery 2002, 131: 85-91. Manson M M , Gescher A , Hudson E A , Plummer S M , Squires M S , Prigent S A . Blocking and suppressing mechanisms of chemoprevention by dietary constituents. Toxicol. Lett. 2000, 113: 499-505. Marks R. Epidemiology o f melanoma. Clin. Exp. Dermatol. 2000, 25: 459-63. Martinou JC , Green D R . Breaking the mitochondrial barrier. Nat. Rev. Mol. Cell. Biol. 2001, 2: 63-7. Mehta K , Pantazis P, McQueen T, Aggarwal B B . Anti-proliferative effect of curcumin (diferuloylmethane) against human breast tumor cell lines. Anti-cancer Drugs 1997, 8: 470-81. 132 Menon L G , Kuttan R, Kuttan G . Inhibition of lung metastasis in mice induced by B16F10 melanoma cells by polyphenolic compounds. Cancer Lett. 1995, 95: 221-5. Mel lon I, Spivak G , Hanawalt P C . Selective removal of transcription-blocking D N A damage from the transcribed strand of the mammalian D H F R gene. Cell 1987, 51: 241-9. Merritt J A , Roth J A , Logothetis CJ . Clinical evaluation of adenoviral-mediated p53 gene transfer: review of I N G N 201 studies. Semin. Oncol. 2001, 28: 105-14. Micheau O, Solary E , Hammann A , Dimanche-Boitrel M T . Fas ligand-independent, F A D D -mediated activation o f the Fas death pathway by anti-cancer drugs. J. Biol. Chem. 1999, 274: 7987-92. Middleton M R , Lee S M , Arance A , Wood M , Thatcher N , Margison G P . 06-methylguanine formation, repair protein depletion and clinical outcome with a 4 hr schedule of temozolomide in the treatment of advanced melanoma: results of a phase II study. Int. J. Cancer 2000, 88: 469-73. Miyashita T, Krajewski S, Krajewska M , Wang H G , L i n H K , Liebermann D A , Hoffman B , Reed JC. Tumor suppressor p53 is a regulator of bcl-2 and bax gene expression in vitro and in vivo. Oncogene 1994, 9: 1799-805. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional activator of the human bax gene. Cell 1995, 80: 293-9. M o H , Elson C E . Apoptosis and cell-cycle arrest in human and murine tumor cells are initiated by isoprenoids. J. Nutr. 1999, 129: 804-13. Momand J, Zambetti G P , Olson D C , George D , Levine A J . The mdm-2 oncogene product forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992, 69: 1237-45. Moroni M C , Hickman E S , Denchi E L , Caprara G , Co l l i E , Cecconi F, Mul ler H , Hel in K . Apaf-1 is a transcriptional target for E2F and p53. Nat. Cell Biol. 2001, 3: 552-8. Morris G F , Bischoff JR, Mathews M B . Transcriptional activation of the human proliferating-cell nuclear anti-gen promoter by p53. Proc. Natl. Acad. Sci. U.S.A. 1996, 93: 895-9. Muggia F M . Dimery I. Arbuck SG. Camptothecin and its analogs. A n overview of their potential in cancer therapeutics. Ann. N.Y. Acad. Sci. 1996, 803: 213-23. Mullenders L H , Berneburg M . Photoimmunology and nucleotide excision repair: impact of transcription coupled and global genome excision repair. J. Photochem. Photobiol. B 2001, 65: 97-100. Munker M , Munker R, Saxton R E , Koeffler H P . Effect of recombinant monokines, lymphokines, and other agents on clonal proliferation of human lung cancer cell lines. Cancer Res. 1987, 47: 4081-5. 133 Murphy M , A h n J, Walker K K , Hoffman W H , Evans R M , Levine A J , George D L . Transcriptional repression by wild-type p53 utilizes histone deacetylases, mediated by interaction with mSin3a. Genes Dev. 1999, 13: 2490-501. Murphy M , Hinman A , Levine A J . Wild-type p53 negatively regulates the expression of a microtubule-associated protein. Genes Dev. 1996, 10: 2971-80. Muz io M , Chinnaiyan A M , Kischkel F C , O'Rourke K , Shevchenko A , N i J, Scaffidi C , Bretz JD, Zhang M , Gentz R, Mann M , Krammer P H , Peter M E , Dixi t V M . F L I C E , a novel F A D D -homologous ICE/CED-3- l ike protease, is recruited to the CD95 (Fas/APO-1) death—inducing signaling complex. Cell 1996, 85: 817-27. Nagabhushan M , Bhide S V . Curcumin as an inhibitor of cancer. J. Am. Coll. Nutr. 1992, 11: 92-8. Nagane M , Pan G , Weddle JJ, Dixi t V M , Cavenee W K , Huang H J . Increased death receptor 5 expression by chemotherapeutic agents in human gliomas causes synergistic cytotoxicity with tumor necrosis factor-related apoptosis-inducing ligand in vitro and in vivo. Cancer Res. 2000, 60: 847-53. Nagata S, Golstein P. The Fas death factor. Science 1995, 267: 1449-56. Nagata S. Fas ligand-induced apoptosis. Annu. Rev. Genet. 1999, 33: 29-55. Nakano K , Vousden K H . P U M A , a novel proapoptotic gene, is induced by p53. Mol. Cell 2001, 7: 683-94. Neuhouser M L , Kristal A R , Patterson R E , Goodman PJ , Thompson LM. Dietary supplement use in the Prostate Cancer Prevention Trial: implications for prevention trials. Nutr. Cancer 2001, 39: 12-8. Nylander K , Bourdon JC, Bray SE, Gibbs N K , K a y R, Hart I, Ha l l P A . Transcriptional activation of tyrosinase and TRP-1 by p53 links U V irradiation to the protective tanning response. J. Pathol. 2000, 190: 39-46. Oda E , Ohki R, Murasawa H , Nemoto J, Shibue T, Yamashita T, Tokino T, Taniguchi T, Tanaka N . Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 2000, 288: 1053-8. Oehm A , Behrmann I, Falk W , Pawlita M , Maier G , Klas C, Li-Weber M , Richards S, Dhein J, Trauth B C . Purification and molecular cloning of the APO-1 cell surface anti-gen, a member of the tumor necrosis factor/nerve growth factor receptor superfamily. Sequence identity with the Fas anti-gen. J. Biol. Chem. 1992, 267: 10709-15. Olhoffer IH, Bolognia JL . What's new in the treatment of cutaneous melanoma? Semin. Cutan. Med. Surg. 1998, 17: 96-107. 134 Olie R A , Hafner C, Kuttel R, Sigrist B , Willers J, Durnmer R, Ha l l J, Stahel R A , Zangemeister-Wittke U . Bcl -2 and bc l -xL anti-sense oligonucleotides induce apoptosis in melanoma cells of different clinical stages. J. Invest. Dermatol. 2002, 118: 505-12. Oltvai Z N , Mi l l iman C L , Korsmeyer SJ. Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death. Cell 1993, 74: 609-19. Ori A , Zauberman A , Doitsh G , Paran N , Oren M , Shaul Y . p53 binds and represses the H B V enhancer: an adjacent enhancer element can reverse the transcription effect o f p53. EMBO J. 1998, 17: 544-53. Owen-Schaub L B , van Golen K L , H i l l L L , Price JE. Fas and Fas ligand interactions suppress melanoma lung metastasis. J. Exp. Med. 1998, 188: 1717-23. Owen-Schaub L B , Zhang W , Cusack JC, Angelo L S , Santee S M , Fujiwara T, Roth J A , Deisseroth A B , Zhang W W , Kruzel E , Radinsky R. Wild-type human p53 and a temperature-sensitive mutant induce Fas/APO-1 expression. Mol. Cell. Biol. 1995, 15: 3032-40. Pantazis P. The water-insoluble camptothecin analogues: promising drugs for the effective treatment of haematological malignancies. Leukemia Res. 1995, 19: 775-88. Park B J , Brown C K , H u Y , Alexander H R , Horti J, Raje S, Figg W D , Bartlett D L . Augmentation o f melanoma-specific gene expression using a tandem melanocyte-specific enhancer results in increased cytotoxicity of the purine nucleoside phosphorylase gene in melanoma. Hum. Gene Ther. 1999, 10: 889-98. Paunesku T, Mittal S, Protic M , Oryhon J, Korolev S V , Joachimiak A , Woloschak G E . Proliferating cell nuclear anti-gen ( P C N A ) : ringmaster of the genome. Int. J. Radiat. Biol. 2001, 77: 1007-21. Pepper C , Hoy T, Bentley P. Elevated Bcl-2/Bax are a consistent feature of apoptosis resistance in B-ce l l chronic lymphocytic leukaemia and are correlated with in vivo chemoresistance. Leukemia & Lymphoma 1998, 28: 355-61. Perkins C, K i m C N , Fang G , Bhalla K N . Overexpression of Apaf-1 promotes apoptosis of untreated and paclitaxel- or etoposide-treated HL-60 cells. Cancer Res. 1998, 58: 4561-6. Piwocka K , Zablocki K , Wieckowski M R , Skierski J, Feiga I, Szopa J, Drela N , Wojtczak L , Sikora E . A novel apoptosis-like pathway, independent of mitochondria and caspases, induced by curcumin in human lymphoblastoid T (Jurkat) cells. Exp. Cell Res. 1999, 249: 299-307. Pollock P M , Trent J M . The genetics of cutaneous melanoma. Clin. Lab. Med. 2000, 20: 667-90. Price B D , Calderwood S K . Increased sequence-specific p53 -DNA binding activity after D N A damage is attenuated by phorbol esters. Oncogene 1993, 8: 3055-62. Raisova M , Hossini A M , Eberle J, Riebeling C, Wieder T, Sturm I, Daniel PT, Orfanos C E , Geilen C C . The Bax/Bcl-2 ratio determines the susceptibility of human melanoma cells to CD95/Fas-mediated apoptosis. J. Invest. Dermatol. 2001, 117: 333-40. 135 Raymond M , Gros P. Cell-specific activity of cis-acting regulatory elements in the promoter of the mouse multidrug resistance gene m d r l . Mol. Cell. Biol. 1990, 10: 6036-40. Reed R J . Atypical spitz nevus/tumor. Hum. Pathol. 1999, 30: 1523-6. Reimer C L , Borras A M , Kurdistani S K , Garreau JR, Chung M , Aaronson S A , Lee S W . Altered regulation of cyclin G in human breast cancer and its specific localization at replication foci in response to D N A damage in p53+/+ cells. J. Biol. Chem. 1999, 274: 11022-9. Reintgen D , Cruse C W , Atkins M . Cutaneous malignant melanoma. Clin. Dermatol. 2001, 19: 253-61. Reisman D , Greenberg M , Rotter V . Human p53 oncogene contains one promoter upstream o f exon 1 and a second, stronger promoter within intron 1. Proc. Natl. Acad. Sci. U.S.A. 1988, 85: 5146-50. Reynolds JE, Yang T, Qian L , Jenkinson JD, Zhou P, Eastman A , Craig R W . M c l - 1 , a member of the Bcl -2 family, delays apoptosis induced by c-Myc overexpression in Chinese hamster ovary cells. Cancer Res. 1994, 54: 6348-52. Roninson IB, Chin JE, Choi K G , Gros P, Housman D E , Fojo A , Shen D W , Gottesman M M , Pastan I. Isolation of human mdr D N A sequences amplified in multidrug-resistant K B carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 1986, 83: 4538-42. Roperch JP, Alvaro V , Prieur S, Tuynder M , Nemani M , Lethrosne F , Piouffre L , Gendron M C , Israeli D , Dausset J, Oren M , Amson R, Telerman A . Inhibition of presenilin 1 expression is promoted by p53 and p 2 1 W A F - l and results in apoptosis and tumor suppression. Nat. Med. 1998,4: 835-8. Roses D F , Karp N S , Oratz R, Dubin N , Harris M N , Speyer J, Boyd A , Golomb F M , Ransohoff J, Dugan M , Valentine FT , Cooper J, Bystryn JC. Survival with regional and distant metastases from cutaneous malignant melanoma. Surg. Gynecol. Obstet. 1991, 172: 262-8. Rothenberg M L . Topoisomerase I inhibitors: review and update. Ann. Oncol. 1997, 8: 837-55. Rudel T, Bokoch G M . Membrane and morphological changes in apoptotic cells regulated by caspase-mediated activation of P A K 2 . Science 1997, 276: 1571-4. Ruiz-Ruiz C, Lopez-Rivas A . Mitochondria-dependent and -independent mechanisms in T R A I L -induced apoptosis are both regulated by interferon-{gamma} in human breast tumor cells. Biochem J. 2002, (in press). Runger T M , Emmert S, Schadendorf D , Diem C, Epe B , Hellfritsch D . Alterations of D N A repair in melanoma cell lines resistant to cisplatin, fotemustine, or etpposide. J. Invest. Dermatol. 2000,114:34-9. Saleem A , Edwards T K , Rasheed Z , Rubin E H . Mechanisms of resistance to camptothecins. Ann. N.Y. Acad. Sci. 2000, 922: 46-55. 136 Salganik R l . The benefits and hazards of anti-oxidants: controlling apoptosis and other protective mechanisms in cancer patients and the human population. J. Am. Coll. Nutr. 2001, 20: 464S-72S. Sampath J, Sun D , K i d d V J , Grenet J, Gandhi A , Shapiro L H , Wang Q, Zambetti G P , Schuetz JD. Mutant p53 cooperates with E T S and selectively up-regulates human M D R 1 not M R P 1 . J. Biol. Chem. 2001, 276: 39359-67. Sato T, Hanada M , Bodrug S, Irie S, Iwama N , Boise L H , Thompson C B , Golemis E , Fong L , Wang H G , Reed JC. Interactions among members of the Bcl-2 protein family analyzed with a yeast two-hybrid system. Proc. Natl. Acad. Sci. U.S.A. 1994, 91: 9238-42. Satyamoorthy K , Chehab N H , Waterman M J , Lien M C , El-Deiry W S , Herlyn M , Halazonetis T D . Aberrant regulation and function of wild-type p53 in radioresistant melanoma cells. Cell Growth Differ. 2000, 11: 467-74. Schadendorf D , Jurgovsky K , Kohlmus C M , Czarnetzki B M . Glutathione and related enzymes in tumor progression and metastases of human melanoma. J. Invest. Dermatol. 1995, 105: 109-12. Scheffer G L , Schroeijers A B , Izquierdo M A , Wiemer E A , Scheper R J . Lung resistance related protein/major vault protein and vaults in multidrug-resistant cancer. Curr. Opin. Oncol. 2000, 12: 550-6. Scheffer G L , Wijngaard P L , Flens M J , Izquierdo M A , Slovak M L , Pinedo H M , Meijer C J , Clevers H C , Scheper R J . The drug resistance-related protein L R P is the human major vault protein. Nat. Med. 1995, 1: 578-82. Schiengold M , Schwantes L , Schwartsmann G , Chies J A , Nardi N B . Multidrug resistance gene expression during the murine ontogeny. Mech. Ageing Dev. 2001, 122: 255-70. Schinkel A H , Mayer U , Wagenaar E , M o l C A , van Deemter L , Smit JJ, van der Valk M A , Voordouw A C , Spits H , van Tellingen O, Zijlmans J M , Fibbe W E , Borst P. Normal viability and altered pharmacokinetics in mice lacking mdrl-type (drug-transporting) P-glycoproteins. Proc. Natl. Acad. Sci. U.S.A. 1997, 94: 4028-33. Scott G A , Haake A R . Keratinocytes regulate melanocyte number in human fetal and neonatal skin equivalents. J. Invest. Dermatol. 1991, 97: 776-81. Sekiguchi I, Suzuki M , Ohwada M , Sato I. Apoptosis as a criterion for sensitivity of uterine cervical adenocarcinoma to cisplatin. Oncology Reports 1998, 5: 1077-8. Setlow R B , Carrier W L . Pyrimidine dimers in ultraviolet-irradiated D N A ' s . / . Mol. Biol. 1966, 17:237-54. Shao R G , Cao C X , Nieves-Neira W, Dimahche-Boitrel M T , Solary E , Pommier Y . Activation of the Fas pathway independently of Fas ligand during apoptosis induced by camptothecin in p53 mutant human colon carcinoma cells. Oncogene 2001, 20: 1852-59. Shimizu S, Narita M , Tsujimoto Y . Bcl-2 family proteins regulate the release of apoptogenic cytochrome c by the mitochondrial channel V D A C . Nature 1999, 399: 483-7. 137 Shin M S , Park W S , K i m S Y , K i m H S , Kang SJ, Song K Y , Park J Y , Dong S M , P i J H , Oh R R , Lee J Y , Y o o N J , Lee S H . Alterations of Fas (Apo- l /CD95) gene in cutaneous malignant melanoma. Am. J. Pathol. 1999, 154: 1785-91. Shin T H , Paterson A J , Kudlow JE. p53 stimulates transcription from the human transforming growth factor alpha promoter: a potential growth-stimulatory role for p53. Mol. Cell. Biol. 1995, 15: 4694-701. Shinomiya N . New concepts in radiation-induced apoptosis: 'premitotic apoptosis' and 'postmitotic apoptosis'. J. Cell Mol. Med. 2001, 5: 240-53. Shinoura N , Sakurai S, Shibasaki F, Asai A , Kir ino T, Hamada H . Co-transduction of Apaf-1 and caspase-9 highly enhances p53-mediated apoptosis in gliomas. Br. J. Cancer 2002, 86: 587-95. Siegel R M , Martin D A , Zheng L , N g S Y , Bertin J, Cohen J, Lenardo M J . Death-effector filaments: novel cytoplasmic structures that recruit caspases and trigger apoptosis. J. Cell Biol. 1998, 141: 1243-53. Sigal A , Rotter V . Oncogenic mutations of the p53 tumor suppressor: the demons of the guardian of the genome. Cancer Res. 2000, 60: 6788-93. Singh S, Aggarwal B B . Activation of transcription factor NF-kappa B is suppressed by curcumin (diferuloylmethane). J. Biol. Chem. 1995, 270: 24995-5000. Skehan P, Storeng R, Scudiero D , Monks A , M c M a h o n J, Vist ica D , Warren JT, Bokesch H , Kenney S, Boyd M R . New colorimetric cytotoxicity assay for anti-cancer-drug screening. J. Natl. Cancer. Inst. 1990, 82: 1107-12. Slichenmyer W J , Nelson W G , Slebos R J , Kastan M B . Loss of a p53-associated G l checkpoint does not decrease cell survival following D N A damage. Cancer Res. 1993, 53: 4164-8. Slominski A , Wortsman J, Carlson A J , Matsuoka L Y , Balch C M , M i h m M C . Malignant melanoma. Arch. Pathol. Lab. Med. 2001, 125: 1295-306. Smit JJ, M o l C A , van Deemter L , Wagenaar E , Schinkel A H , Borst P. Characterization of the promoter region of the human M D R 3 P-glycoprotein gene. Biochim. Biophys. Acta 1995, 1261: 44-56. Smith M L , Chen IT, Zhan Q, Bae I, Chen C Y , Gilmer T M , Kastan M B , O'Connor P M , Fornace A J Jr. Interaction of the p53-regulated protein Gadd45 with proliferating cell nuclear anti-gen. Science 1994, 266: 1376-80. Soengas M S , Capodieci P, Polsky D , Mora J, Esteller M , Opitz-Araya X , McCombie R, Herman J G , Gerald W L , Lazebnik Y A , Cordon-Cardo C, Lowe SW. Inactivation of the apoptosis effector Apaf-1 in malignant melanoma. Nature 2001, 409: 207-11. Soussi T, Beroud C. Assessing TP53 status in human tumours to evaluate clinical outcome. Nature Rev. Cancer 2001, 1: 233-40. 138 Sparrow L E , Soong R, Dawkins H J , Iacopetta B J , Heenan PJ. p53 gene mutation and expression in naevi and melaomas. Melanoma Res. 1995, 5: 93-100. Srivastava R K . T R A J L / A p o - 2 L : mechanisms and clinical applications in cancer. Neoplasia 2001,3: 535-46. Stehlik C, de Martin R, Kumabashiri I, Schmid J A , Binder B R , Lipp J. Nuclear factor (NF)-kappaB-regulated X-chromosome-linked iap gene expression protects endothelial cells from tumor necrosis factor alpha-induced apoptosis. J. Exp. Med. 1998, 188: 211-6. Stein U , Walther W , Wunderlich V . Point mutations in the abcbl promoter of human osteosarcomas are associated with in vitro responsiveness to multidrug resistance relevant drugs. Eur. J. Cancer 1994, 30A: 1541-5. Stewart C F . Zamboni W C . Crom W R . Gajjar A . Heideman R L . Furman W L . Meyer W H . Houghton PJ . Pratt C B . Topoisomerase I interactive drugs in children with cancer. Invest. New Drugs 1996, 14: 37-47. Stockel B , Kon ig J, Nies A T , Cu i Y , Brom M , Keppler D . Characterization of the 5'-flanking region of the human multidrug resistance protein 2 (MRP2) gene and its regulation in comparison withthe multidrug resistance protein 3 (MRP3) gene. Eur. J. Biochem. 2000, 267: 1347-58. Stommel J M , Marchenko N D , Jimenez GS , M o l l U M , Hope TJ , Wahl G M . A leucine-rich nuclear export signal in the p53 tetramerization domain: regulation of subcellular localization andp53 activity by N E S masking. EMBO J. 1999, 18: 1660-72. Storey A , Thomas M , Kali ta A , Harwood C, Gardiol D , Mantovani F, Breuer J, Leigh J M , Matlashewski G , Banks L . Role of a p53 polymorphism in the development of human papillomavirus-associated cancer. Nature 1998, 393: 229-34. Strauss B E , Haas M . The region 3' to the major transcriptional start site of the A b c b l downstream promoter mediates activation by a subset of mutant P53 proteins. Biochem. Biophys. Res. Commun. 1995, 217: 333-40. Stretch JR, Gatter K C , Ralfkiaer E , Lane D P , Harris A L . Expression of mutant p53 in melanoma. Cancer Res. 1991, 51: 5976-9. Su Z , Lebedeva IV, Gopalkrishnan R V , Goldstein NI , Stein C A , Reed JC , Dent P, Fisher P B . A combinatorial approach for selectively inducing programmed cell death in human pancreatic cancer cells. Proc. Natl. Acad. Sci. U.S.A. 2001, 98: 10332-7. Subbaramaiah K , Altorki N , Chung W J , Mestre JR, Sampat A , Dannenberg A J . Inhibition of cyclooxygenase-2 gene expression byp53. J. Biol. Chem. 1999, 274: 10,911-5. Szabo G , Hirobe T, Flynn E A , Garcia R l . The biology of the melanocyte. Prog. Clin. Biol. Res. 1988,256:463-74. 139 Tai Y T , Strobel T, Kufe D , Cannistra S A . In vivo cytotoxicity of ovarian cancer cells through tumor-selective expression of the B A X gene. Cancer Res. 1999, 59: 2121-6. Takada T, Suzuki H , Sugiyama Y . Characterization of 5'-flanking region of human M R P 3 . Biochem. Biophys. Res. Commun. 2000, 270: 728-32. Tan B , Piwnica-Worms D , Ratner L . Multidrug resistance transporters and modulation. Curr. Opin. Oncol. 2000, 12: 450-8. Tanaka T, Uchiumi T, Hinoshita E , Inokuchi A , Toh S, Wada M , Takano H , Kohno K , Kuwano M . The human multidrug resistance protein 2 gene: functional characterization of the 5'-flanking region and expression in hepatic cells. Hepatology 1999, 30: 1507-12. Tang L , Tron V A , Reed JC, M a h K J , Krajewska M , L i G , Zhou X , Ho V C , Trotter M J . Expression of apoptosis regulators in cutaneous malignant melanoma. Clin. Cancer Res. 1998, 4: 1865-71. Thomas W D , Zhang X D , Franco A V , Nguyen T, Hersey P. TNF-related apoptosis-inducing ligand-induced apoptosis of melanoma is associated with changes in mitochondrial membrane potential and perinuclear clustering of mitochondria. J. Immunol. 2000, 165: 5612-20. Thottassery J V , Sun D , Zambetti G P , Troutman A , Sukhatme V P , Schuetz E G , Schuetz JD. S p l and egr-1 have opposing effects on the regulation of the rat Pgp2/Mdrlb gene. J. Biol. Chem. 1999, 274: 3199-206. Thottassery J V , Zambetti G P , Arimori K , Schuetz E G , Schuetz JD. p53-dependent regulation o f M D R 1 gene expression causes selective resistance to chemotherapeutic agents. Proc. Natl. Acad. Sci. U.S.A. 1997, 94: 11037-42. Timmer T, de Vries E G , de Jong S. Fas receptor-mediated apoptosis: a clinical application? J. Pathol. 2002, 196: 125-34. Tron V A , Trotter M J , Tang L , Krajewska M , Reed JC, Ho V C , L i G . p53-regulated apoptosis is differentiation-dependent in U V B irradiated mouse keratinocytes. Am. J. Pathol. 1998, 153: 579-85. Tsatmali M , Ancans J, Thody A J . Melanocyte function and its control by melanocortin peptides. J. Histochem. Cytochem. 2002, 50: 125-33. Tsuruta Y , Mandai M , Konishi I, Kuroda H , Kusakari T, Yura Y , Hamid A A , Tamura I, Kar iya M , Fujii S. Combination effect of adenovirus-mediated pro-apoptotic bax gene transfer with cisplatin or paclitaxel treatment in ovarian cancer cell lines. Eur. J. Cancer 2001, 37: 531-41. Turton N J , Judah D J , Ri ley J, Davies R, Lipson D , Styles J A , Smith A G , Gant T W . Gene expression and amplification in breast carcinoma cells with intrinsic and acquired doxorubicin resistance. Oncogene 2001, 20: 1300-6. 140 Ueba T, Nosaka T, Takahashi J A , Shibata F, Florkiewicz R Z , Vogelstein B , Oda Y , Kikuchi H , Hatanaka M . Transcriptional regulation of basic fibroblast growth factor gene by p53 in human glioblastoma and hepatocellular carcinoma cells. Proc. Natl. Acad. Sci. U.S.A. 1994, 91: 9009-13. Ueda K , Pastan I, Gottesman M M . Isolation and sequence of the promoter region of the human multidrug-resistance (P-glycoprotein) gene. J. Biol. Chem. 1987, 262: 17432-6. Vancoill ie G , Lambert J, Nayaert J M . Melanocyte biology and its implications for the clinician. Eur. J. Dermatol. 1999, 9: 241-51. Vucic D , Stennicke H R , Pisabarro M T , Salvesen GS , Dixi t V M . M L - I A P , a novel inhibitor of apoptosis that is preferentially expressed in human melanomas. Curr. Biol. 2000, 10: 1359-66. Waga S, Harmon GJ , Beach D , Stillman B . The p21 inhibitor of cyclin-dependent kinases controls D N A replication byinteraction with P C N A . Nature 1994, 369: 574-8. Wal l M E , Wani M C . Camptothecin and taxol: discovery to clinic—thirteenth Bruce F. Cain Memorial Award Lecture. Cancer Res. 1995, 55: 753-60. Wang Q, Beck W T . Transcriptional suppression of multidrug resistance-associated protein ( M R P ) gene expression by wild-type p53. Cancer Res. 1998, 58: 5762-9. Wang C Y , Mayo M W , Komeluk R G , Goeddel D V , Baldwin A S Jr. NF-kappaB anti-apoptosis: induction o f T R A F 1 and T R A F 2 and c - I A P l and C-IAP2 to suppress caspase-8 activation. Science 1998, 281: 1680-3. Wang C Y , Cusack JC Jr, L i u R, Baldwin A S Jr. Control of inducible chemoresistance: enhanced anti-tumor therapy through increased apoptosis by inhibition of NF-kappaB. Nat. Med. 1999, 5: 412-7. Wang L , W u Q, Qiu P, M i r z a A , McGui rk M , Kirschmeier P, Greene JR, Wang Y , Pickett C B , L i u S. Analyses of p53 target genes in the human genome by bioinformatics and microarray approaches. J. Biol. Chem. 2001, 276: 43604-10. Wargovich M J , Jimenez A , M c K e e K , Steele V E , Velasco M , Woods J, Price R, Gray K , Kel lof f GJ . Efficacy of potential chemopreventive agents on rat colon aberrant crypt formation and progression. Carcinogenesis 2000, 21: 1149-55. Wattel E , Preudhomme C, Hecquet B , Vanrumbeke M , Quesnel B , Dervite I, More l P, Fenaux P. p53 mutations are associated with resistance to chemotherapy and short survival in hematologic malignancies. Blood 1994, 84: 3148-57. Weiss J, Heine M , Arden K C , Korner B , Pi lch H , Herbst R A , Jung E G . Mutation and expression ofTP53 in malignant melanomas. Recent Results Cancer Res. 1995, 139: 137-54. Whiteman D C , Whiteman C A , Green A C . Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies. Cancer Causes Control 2001, 12: 69-82. 141 W o l l E , Bedikian A , Legha SS. Uveal melanoma: natural history and treatment options for metastatic disease. Melanoma Res. 1999, 9: 575-81. Wong J, L i P X , Klamut H J . A novel p53 transcriptional repressor element (p53 T R E ) and the asymmetrical contribution of two p53 binding sites modulate the response of the PTGF-beta promoter to p53. Biol. Chem. 2002, (in press). Woo M , Hakem R, Soengas M S , Duncan G S , Shahinian A , Kag i D , Hakem A , McCurrach M , Khoo W , Kaufman S A , Senaldi G , Howard T, Lowe SW, Mak T W . Essential contribution of caspase 3/CPP32 to apoptosis and its associated nuclear changes. Genes Dev. 1998, 12: 806-19. W u G S , Burns T F , McDonald E R 3rd, Jiang W , Meng R, Krantz ID, Kao G , Gan D D , Zhou J Y , Muschel R, Hamilton SR, Spinner N B , Markowitz S, W u G , el-Deiry W S . K I L L E R / D R 5 is a D N A damage-inducible p53-regulated death receptor gene. Nat. Genet. 1997, 17: 141-3. W u X W , Bayle J H , Olson D , Levine A J . The p53 mdm-2 autoregulatory feedback loop. Genes Dev\993, 7: 1126-32. Wyl l ie A H . Apoptosis and carcinogenesis. Eur. J. Cell Biol. 1997, 73: 189-97. Xiong Y , Hannon G J , Zhang H , Casso D , Kobayashi R, Beach D . p21 is a universal inhibitor of cyclin kinases. Nature 1993, 366: 701-4. X u H , Raafat el-Gewely M . P53-responsive genes and the potential for cancer diagnostics and therapeutics development. Biotechnol. Annu. Rev. 2001, 7: 131-64. Yamamoto M , Takahashi H , Saitoh K , Horikoshi T, Takahashi M . Expression of the p53 protein in malignant melanoma as a prognostic indicator. Arch. Dermatol. Res. 1995, 287: 147-51. Yang C S , Landau J M , Huang M T , Newmark H L . Inhibition of carcinogenesis by dietary polyphenolic compounds. Annu. Rev. Nutr. 2001, 21: 381-406. Y u J, Zhang L , Hwang P M , Kinzler K W , Vogelstein B . P U M A induces the rapid apoptosis of colorectal cancer cells. Mol. Cell 2001, 7: 673-82. Y u R, Hebbar V , K i m D W , Mandlekar S, Pezzuto J M , Kong A N . Resveratrol inhibits phorbol ester and UV-induced activator protein 1 activation by interfering with mitogen-activated protein kinase pathways. Mol. Pharmacol. 2001, 60: 217-24. Zamzami N , Marchetti P, Castedo M , Zanin C, Vayssiere JL , Petit P X , Kroemer G . Reduction in mitochondrial potential constitutes an early irreversible step of programmed lymphocyte death in vivo. J. Exp. Med. 1995, 181: 1661-72. Zastawny R L , Salvino R, Chen J, Benchimol S, L ing V . The core promoter region of the P-glycoprotein gene is sufficient to confer differential responsiveness to wild-type and mutant p53. Oncogene 1993, 8: 1529-35. Zeng Q, Tan Y H , Hong W . A single plasmid vector (pSTAR) mediating efficient tetracycline-induced gene expression. Anal. Biochem. 1998, 259: 187-94. 142 Zhang K , Mack P, Wong K P . Glutathione-related mechanisms in cellular resistance to anti-cancer drugs. Int. J. Oncol. 1998, 12: 871-82. Zhang W , Geiman D E , Shields J M , Dang D T , Mahatan C S , Kaestner K H , Biggs JR, Kraft A S , Yang V W . The gut-enriched Kruppel-like factor (Kruppel-like factor 4) mediates the transactivating effect of p53 on the p 2 1 W A F l / C i p l promoter. Biol. Chem. 2000, 275: 18391-8. Zhang X D , Franco A , Myers K , Gray C, Nguyen T, Hersey P. Relation of TNF-related apoptosis-inducing ligand (TRAIL) receptor and FLICE-inhibitory protein expression to T R A I L -induced apoptosis o f melanoma. Cancer Res. 1999, 59: 2747-53. Zhou G , Kuo M T . NF-kappaB-mediated induction of M d r l b expression by insulin in rat hepatoma cells. J. Biol. Chem. 1997, 272: 15174-83. Zhou G , Kuo M T . Wild-type p53-mediated induction of rat M d r l b expression by the anti-cancer drug daunorubicin. Biol. Chem. 1998, 273: 15387-94. Zhu Q, Center M S . Cloning and sequence analysis of the promoter region of the M R P gene of H L 6 0 cells isolated for resistance to adriamycin. Cancer Res. 1994, 54: 4488-92. Zimmermann K C , Bonzon C, Green D R . The machinery of programmed cell death. Pharmacol. Ther. 2001,92: 57-70. Zornig M , Hueber A , Baum W , Evan G . Apoptosis regulators and their role in tumorigenesis. Biochim. Biophys. Acta. 2001, 1551: F l - 3 7 . Zou H , Henzel W J , L i u X , Lutschg A , Wang X . Apaf-1, a human protein homologous to C. elegans C E D - 4 , participates in cytochrome c-dependent activation o f caspase-3. Cell 1997, 90: 405-13. 143 APPENDIX I. RELATIONSHIP BETWEEN P53 AND P-GLYCOPROTEIN A.1 R A T I O N A L E A N D H Y P O T H E S I S During cancer progression, tumour cells often establish resistance to a broad spectrum of drug substrates frequently mediated by members o f the A B C transporter superfamily. These are large transmembrane proteins localized to apical surfaces of cells that actively efflux xenobiotics out o f the cell interior through hydrolysis o f A T P (Dean et al., 2001). A common mechanism for the multidrug resistance ( M D R ) phenotype of cancer is due to overexpression of the Mdrl ( H U G O designation, ABCB1) gene product, P-gp. In humans, P-gp is encoded by two loci , Mdrl and Mdr2/3 (ABCB4) but in rodents encoded by a small three-member gene family, Mdrla, Mdrlb, and Mdrl (Hrycyna, 2001; Hsu et al., 1989); however, to date, a role in multidrug resistance has not been attributed to either Mdr2/3 for humans or the Mdr2 gene product for rodents (Smit, 1993). Numerous studies over the past decade have confirmed transcriptional control of the Mdrl promoter by p53 (reviewed in Bush and L i , 2002). The traditional model proposed that normal p53 could transrepress Mdrl transcription through mechanisms acting independently of a p53 consensus binding sequence while mutants of p53 would activate transcription of both human and rodent Mdrl promoters (Chin et al., 1992; Zastawny et al., 1993) (Figure A . l ) . This has been demonstrated in multiple mammalian cell lines using a variety of promoter constructs, reporter genes and p53 versions in many different genetic backgrounds (i.e., wild-type, mutant, or nullizygous for p53 and drug-sensitive or resistant for P-gp). Unfortunately, few studies have established a correlation with endogenous P-gp protein levels so functional relevance is questionable (Thottassery et al., 1997). More recently, new data has sought to refine the model by further validating that mutant p53 may interact with other transcription factors such as ETS to 144 Figure A .1 The classic hypothetical model for p53 regulation of the Mdrl promoter. Normal p53 may interact directly with a sequence in the Mdrl promoter or indirectly with other transcription factors to repress P-gp expression. Mutation o f p53 may relieve this suppression or facilitate transcriptional activation and by default an increase in P-gp occurs. 145 up-regulate Mdrl transcription (Sampath et al, 2001). A unique orientation of the conventional p53 DNA-bind ing sequence was identified in the human Mdrl promoter to serve as the repressive element (Johnson et al., 2001). Another study asserts that p53 potentially interacts with histone deacetylases via an adaptor protein, mSin3a, and the T A T A binding protein to mediate transcriptional repression of other p53-repressed genes (Murphy et al., 1999). Finally, a novel transcriptional repressor element has been recently confirmed in the Placental Transforming Growth Factor Beta (PTGF-beta) promoter, an important downstream mediator of D N A damage signaling and a transcriptional target of p53 (Wong et al., 2002). From chapter 3, we found that P-gp was down-regulated in a melanoma cell line with normal p53 after anticancer drug treatment but was unchanged in an isogenic cell line with mutant p53. We have additional results that support a role for increased P-gp expression both at the transcriptional and translational level in melanoma cell lines compared to melanocytes (Figure A.2) . Furthermore, this difference is amplified in those cell lines with mutant p53 compared to the wild-type p53 cell lines and our group has shown previously that this reflects the p53 status (Li et al, 1998 and Figure A.2) . Nevertheless, dissecting the relationship between p53 and P-gp is not unequivocal in tumour cell lines. Therefore, we sought to further investigate the connection of these pivotal pathways by utilizing a 'clean' genetic system. Despite the multiple levels at which p53 seems to control P-gp-induced drug resistance, few studies have addressed this relationship in vivo. The complexity' is further amplified, particularly in rodents, considering that there are two P-gp isoforms, M d r l a and M d r l b , with extensive homology of over 90% (Dhir et al., 1990; Lecureur et al., 2001) and that they are co-expressed in some tissues yet also show tissue specificities (Croop et al., 1989). Furthermore, variations in the temporal distribution of the isoforms have been determined (Schiengold et al., 2001) including a developmental conversion from M d r l a / M d r l b to predominantly M d r l a within intestinal epithelium (L i et al, 1999). Although alternative-splicing has not been adequately confirmed 146 Figure A . 2 Increased expression o f P-gp in p53 mutant melanoma cell lines. Melanoma cell lines were assessed for their Mdrl m R N A expression and P-gp protein expression compared to melanocytes. M M A N and M M R U are p 5 3 W T while M e W o and SK-mel-110 are p 5 3 M U T (denoted by asterisk). p85, the small regulatory subunit o f phosphatidylinositol 3'-kinase was used as a high molecular weight control for loading. Densitometry was performed as outlined in Materials and Methods, section 2.2.14. for the mouse MdrlalMdrlb genes, it is implicated for human Mdrl (Hu et al., 1998) and possibly hamster pgpl (Devine et al., 1991). Clearly, a complex regulatory mechanism exists. We asked the question, does the dysregulation of p53 increase endogenous P-gp expression in vivo? Remarkably, however, the endogenous tissue expression of the P-gp isoforms has not been established in vivo within the context of a p53-n\x\\ background for multiple tissues. Therefore, we have utilized the /?53-knockout mouse model as a means to gain further insight into the expressional regulation o f P-gp by p53. Our hypothesis is that gene products o f the mouse A B C B 1 homologues ( M d r l a and M d r l b ) may be up-regulated cells from a p53'A background compared to cells from ap53+/+ background. A . 2 R E S U L T S A.2.1 Dysfunction of P53 Increases Expression of P-gp We isolated primary skin cells from 4-week old mice (dermal fibroblasts or mFbs) to show that both Mdrla and Mdrlb m R N A levels are dramatically increased in the p53~/~ cells compared with p53+/+ cells (Figure A . 3 A ) . More significantly, we show that this is recapitulated at the protein level (Figure A .3B) . In an attempt to demonstrate that we can get reversion of the molecular phenotype, we transfected a wild-type mouse p53 expression plasmid into early passage p53'A fibroblasts, and then assessed P-gp protein levels after 48 h. We found a clear reduction in the amount of P-gp expression in the transfected cell line versus the control p53'A fibroblasts transfected with a G F P expression vector to evaluate our transfection efficiency (Figure A . 3 C ) . In a reciprocal experiment, we were able to overexpress a mutant mouse p53 into early passage p53+/+ fibroblasts and show an increase of endogenous P-gp expression compared to the control cells perhaps by interfering with wild-type p53 function (Figure A . 3 C ) . 148 Figure A . 3 Increased Mdrla/b and P-gp in p 5 3 " cells. (A) R T - P C R of Mdrla and Mdrlb isoforms in dermal fibroblasts from p 5 3 + + and p 5 3 ' A mice and NIH3T3 cells. (B) Total protein was isolated from fibroblasts and probed with antisera to P-gp that recognizes both M d r l a and M d r l b isoforms or P-actin. (C) Early passage murine fibroblasts were transiently transfected with either a normal mouse p 5 3 plasmid (pPW) or an N-termimal deletion mutant mouse p 5 3 plasmid (pMR53) to determine the P-gp expression. Mdrla Mdrlb GAPDH B p53 + / + pSZ-1-r r i P-gp Act in 1 11 149 Using a reporter construct containing a 5'-flanking sequence of the Mdrlb promoter ( -1165 to +84) linked to a C A T reporter gene (WT-Mdrlb::CAT) or another version in which the p53 consensus binding sequence has been mutated (M\JT-Mdrlb::CAT) (Figure A . 4 A ) , we expressed these plasmids in p53+/+ and p53'A fibroblasts (Raymond and Gros, 1990). After standardizing to a p C A T control plasmid containing a generic promoter, surprisingly, there was a modest reduction in C A T expression in p53+/+ fibroblasts transfected with the WT-Mdrlb reporter while in contrast, there was a dramatic increase in C A T expression in p53'A cells (Figure A.4D) . This is contrary to other results showing that human p53 activates a rat Mdrlb promoter (Lecureur et al., 2001). Interestingly, we see a comparable increase over the control in both cell lines with the MUT-Mdrlb reporter. This may suggest some p53-independent basal C A T activation from other cz's-acting regulatory elements not mutated on this construct (Raymond and Gros, 1990; Zhou and Kuo, 1997). A.2.2 The Funct ional Significance of Elevated P-gp The results from section A.2.1 support a generalized model that the absence or dysfunction of p53 can promote increased P-gp expression. We further confirm this hypothesis with functional assays determining survival after either doxorubicin ( D O X ) or vincristine (VPN) treatment. Since D O X and V P N are known substrates of P-gp activity and commonly used to show changes in the resistance mechanism associated with P-gp overexpression, we treated both p53+/+ and p53'A fibroblasts of equal passage with increasing doses of these drugs for 48 h then quantitated the relative survival. There is an obvious difference in the susceptibility of the different cell genotypes. The p53'A fibroblasts are dramatically more resistant to D O X and V I N than the p53+/+ control cells (Figure A . 5 A and B) . The strong resistance in the p53~'~ fibroblasts could be substantially reversed in the presence of the P-gp inhibitor, verapamil ( V E R ) (Figure A .5C) . Furthermore, several studies have demonstrated an increase in P K C isozymes in drug-resistant 150 Figure A.4 Activation of the Mdrlb promoter in the absence of p53. (A) Schematic of the Mdrlb-CAT reporter plasmids. (B) Early passage mFbs were transfected with either a C A T control plasmid, WT-Mdrlb::CAT, or MUT-Mdrlb-CAT. C A T expression was detected by a C A T E L I S A (see Materials & Methods, section 2.2.15). These results represent the mean of duplicate experiments. -197 p53 site -178 -1165 — \ \ — G A A C A c G T a a AGACAAGTCT FcIF • T — A - CAT \NT-Mdr1b M\JT-Mdr1b B 500 400 c .2 300 o 3 T3 = 200 < ^0Q^ • CAT control • MUT-Mdr1b B WT-Mdr1b p53+/+ p53-/-151 Figure A . 5 Comparison of drug sensitivity between p53+/+ and p53v~ mFbs. Ce l l survival was assessed by S R B assay in early passage mFbs grown in 24-well plates and treated with increasing concentrations of (A) doxorubicin ( D O X ) or (B) vincristine (VIN) for 48 h. (C) The p53'A fibroblasts were incubated with or without 5 p M verapamil ( V E R ) for 1 h, then D O X was added and the cells were cultured for 48 h followed by quantitation for cell survival. 600 1200 DOX Dose (nM) 1800 152 cells lines (Beck et al, 2001; Conseil et al, 2001; G i l l et al, 2001). We validate these observations by Western blot analysis exhibiting higher levels of PKCct , 6, and C, but not P K C e in the p53" /_ fibroblasts compared to the p 5 3 + / + controls (Figure A.6) . Although correlative, our findings confirm data by G i l l et al (2001) that increased levels of PKCct and 0 occur in connection with P-gp expression in cells that have acquired multidrug resistance. A.2.3 P-gp Expression Is Increased in Specific Tissues The organ-specific differences for the murine P-gp isoforms may be due to the nature of the polypeptide structures and thus they may have different yet overlapping substrate specificity (Schinkel et al, 1997). Alternatively, their respective promoters may contain specific response elements that provide for tissue-specific expression. Although the tissue distribution of the Mdrla and Mdrlb isoforms has been established, we found that there was a clear increase in the P-gp levels in certain organs of the /?53-knockout samples over the wild-type tissues. Specifically, we detected higher expression in kidney, spleen, and testis of p53'A mice. This was evaluated at both the message and protein levels (Figures A . 7 A and B) . In contrast, we found no difference for P-gp expression in brain, heart, liver, lung, muscle or thymus, although variable levels were evident among the different organs. Our R T - P C R results demonstrate that Mdrla, but not Mdrlb, predominates in brain tissue supporting the findings by Demeule et al, (2001); however, overamplification after additional cycles (40 cycles) produced a detectable band that was more intense in the p53'A tissue (Figure A.7C) . Finally, we confirm our Western blot and R T - P C R results with in situ immunohistochemical staining in selected tissues. Comparable P-gp expression is seen in mouse heart from p53+/+ and p53'A mice. However, in kidney samples there is considerably stronger P-gp staining of the proximal tubules in the knockout tissue than thep53 + / + tissue consistent with the results from Western analysis (Figure A .7D) . 153 Figure A . 6 Correlative increase in expression levels of P K C isozymes. The P K C family of serine/threonine kinases is involved in signal transduction for a number of different cell processes. Some P K C isozymes are increased in classic multidrug resistant cell lines where P-gp is overexpressed. Total protein was isolated from p53+/+, p53~!~ or NIH3T3 fibroblasts, run on 10% S D S - P A G E gels for Western blotting as above, and probed with antisera to PKCcc, 9, s, or p-actin as a control for protein loading. Densitometry was performed as outlined in Materials and Methods, section 2.2.14. These represent the average results of duplicate Western blots. 154 A.3 D I S C U S S I O N A.3.1 The Va l id i ty of the In Vivo Observations Versus the In Vitro System We feel that establishing the tissue-specific differences for the P-gp multidrug transporter in the context of the />53-knockout mouse could have insightful ramifications on the study of tumour treatment and in vivo carcinogenesis. Most functional studies ascertaining the regulatory relationship between p53 and increases in drug transporters are based on promoter/reporter control of MDR-related genes by p53 mutants in long-established cell lines. Very few studies have determined the actual transporter protein expression and fewer utilized primary cells as we have done. Therefore, in those p53-mx\\ tissues with higher basal levels of MdrlalMdrlb transcripts and P-glycoprotein, a repressional rather than activational role for p53 is implicated which has particular significance for the Mdrlb isoform. This is confounded by recent data from Schuetz and coworkers (Lecureur et al., 2001) as well as Kuo and colleagues (Zhou and Kuo , 1998). However, we find several shortcomings with the previous reports. The in vitro activation of the rat or mouse Mdrlb promoter was performed using human normal and mutant p53 c D N A s frequently in human (SAOS-2, p53-null and Rb-inactivated) or rat (H-4-IJ-E) tumour cell lines with very minimal promoter constructs. We would caution that this may be a very synthetic system. The core domain of mouse and human p53 are known to differ by 15% at base residues. A s well , there are mutational differences between the homologues, for example, UV-induced mutations do not occur at exactly the same positions, which may have functional manifestations on the protein (Dumaz et al, 1997). Although it seems reasonable to qualify the activation or repression of P-gp in terms of the complex regulatory mechanisms a number o f questions remain. It was shown that the transcription factor S p l activates rat Mdrlb while Egr-1 represses the promoter (Thottassery et al., 1999) suggesting that p53 and Egr-1 have opposing function. This contradicts other experiments showing that wild-type but not mutant p53 physically 155 Figure A .7 Organ-specific expression of P-gp protein and Mdrla or Mdrlb m R N A . (A) 12-week old p53+/+ (+) and p53'A (-) male mice were sacrificed and the organs removed for protein extraction in lysis buffer. A n ' * ' denotes higher P-gp expression in the p53'A tissues. Densitometry values were obtained as described in Materials and Methods, section 2.2.14. (B and C) Mouse organs were homogenized with a microhomogenizer and R T - P C R performed as described in Materials and Methods, section 2.2.13, except for the additional P C R cycles from the magnified region. (D) Immunohistochemistry was performed on 6-micron sections of paraffin-embedded blocks o f biopsies from 12-week old male mice. Samples were blocked for 20 min in non-specific rabbit serum and immunolabelled using the polyclonal anti-P-gp rabbit antibody or P-actin. 156 Spleen* + -Lung Testis* + - + -Brain Liver + - + -Heart Thymus + - + -Muscle Kidney* + - + -mm | P-gp • m l I s 1 3.2 1 1.2 1 4.1 1 1.1 1 0.9 1 1.3 1 1.2 1 1 1 3.5 Control + P-9P Control + P-9P H E A R T * • • •« • • • • 4 ~» $ * • * -* ft* 9 p53 p53^ K I D N E Y 157 interacts with Egr-1 (Liu, 2001) or that in response to ionizing radiation in mouse embryonic fibroblasts, Egr-1-induced apoptosis requires functional p53 (Das et al., 2001). Moreover, p53 was shown to negatively regulate the human Mrpl promoter while S p l could activate it in a competitive mechanism (Wang and Beck, 1998). A.3.2 The Role of P53 in Transrepressing Other Apoptosis Genes The discovery that p53 activates Mdrlb in a sequence-specific DNA-bind ing manner apparently does not correspond with high throughput genomic screens using microarray (Kannan et al., 2001) or serial analysis of gene expression (SAGE)(Madden et al, 1997) techniques to identify />J5-responsive targets for which to our knowledge, Mdrlb was not identified. Other genes with authentic p53-binding sites can be repressed such as the anti-apoptotic gene, Survivin (Hoffman et al., 2002) and the interferon-inducible, p202 gene (D'Souza et al., 2001). Finally, in mouse liver, Lecureur et al. (2001) show by RNase protection assay that Mdrlb could have higher basal expression in the p53'A versus the p53+/+ liver tissue. We detected very high levels (after 30 P C R cycles) o f both the Mdrla and Mdrlb transcripts by R T - P C R in liver and therefore were not able to quantitatively determine a difference (data not shown). A.3.3 The Absence of P53 Increases P-gp Isoforms in Some Tissues The body of literature overwhelmingly substantiates that loss of p53 whether as an early or late event in tumourigenesis contributes to cancer severity and directly to chemoresistance through a combination of properties including inhibition of apoptotic mechanisms and increases in endogenous drug transporters. We show that the absence of p53 directly causes an up-regulation of a mouse Mdrlb promoter in p53'A fibroblasts and that several different tissues show increased expression of Mdrla and Mdrlb isoforms. The fact that P-gp appears to be expressed at different levels in some tissues of p53+/+ and p53'A origin and similar levels in others under non-158 inducible conditions is intriguing. Our results imply that basal expression of P-gp isoforms is repressed either directly or indirectly by p53. This has particular ramifications on tumourigenesis and potential treatment modalities. Since p53 is mutated in over half of all human cancers (Hollstein et al., 1997), and has a clear involvement in maintaining genomic stability, the absence o f normal p53 may by default increase endogenous P-gp isoform expression and thus create a multidrug-resistant cell. Recent data from Hehlmann and coworkers (Duesberg et al., 2001) supports a model that aneuploidy mechanisms may contribute to drug resistance in cancers and tumour cell lines. While formally, the p53~A mouse fibroblasts we used may be resistant as a result of amplication in the Mdrl locus, we find this less likely because they are early passage cells, and thus, we suspect the increased resistance to be dependent on p53 loss. A.3.4 The Importance of In Vivo Model Systems for Human Cancers The genetic background on which mutations and biochemical pathways are evaluated may largely determine the incidence of tumours. Therefore, knowing that p53 status is a factor in a wide variety of cancers (Backlund et al., 2001; Keshelava et al., 2001), genetically engineered mice with p53 perturbations are an invaluable tool for cancer research (Lubet et ah, 2000). More precisely, we believe our observations have particular importance for understanding chemoresistance susceptibility for individuals with predisposing p53 germline mutations such as Li-Fraumeni syndrome (Kleihues et ah, 1997; Kuperwasser et al., 2000). Because P-gp is expressed in the same tissues in humans and rodents, direct correlations can be made with respect to innate organ drug resistance. Although mice do not succumb to traditional skin cancers as humans suffer, the potential to understand the biology of skin cells lacking p53 is a step towards that understanding. With the ever-expanding list o f A B C transporters involved in the M D R phenomenon, the />J3-deficient mouse is a useful model system for determining organ-specific chemotherapy regimes to better target a particular cancer type. Moreover, it can be used 159 to estimate the potential inherent resistance for certain substrates based on the predominance P-gp or in fact other multidrug transporter isoforms. 160 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090502/manifest

Comment

Related Items