Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Novel therapeutic targeting of apoptosis and survival pathways in melanoma Karst, Alison Marie 2008

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2008_fall_karst_alison.pdf [ 22.03MB ]
Metadata
JSON: 24-1.0066678.json
JSON-LD: 24-1.0066678-ld.json
RDF/XML (Pretty): 24-1.0066678-rdf.xml
RDF/JSON: 24-1.0066678-rdf.json
Turtle: 24-1.0066678-turtle.txt
N-Triples: 24-1.0066678-rdf-ntriples.txt
Original Record: 24-1.0066678-source.json
Full Text
24-1.0066678-fulltext.txt
Citation
24-1.0066678.ris

Full Text

 NOVEL THERAPEUTIC TARGETING OF APOPTOSIS AND SURVIVAL PATHWAYS IN MELANOMA  by  Alison Marie Karst  B.Sc., Saint Mary’s University, 1998   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate Studies (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  October 2008  ©Alison Marie Karst, 2008   ii ABSTRACT Cutaneous malignant melanoma is an aggressive form of skin cancer, characterized by strong chemoresistance and poor patient prognosis.  The molecular mechanisms underlying its resistance to chemotherapy remain unclear but are speculated to involve dysregulation of apoptosis and reinforcement of survival signaling.  In this work, we show that aberrant expression of two key proteins, PUMA and p-Akt, is associated with melanoma tumor progression and poor patient survival.  We report that PUMA expression is reduced in melanoma tumors compared to dysplastic nevi, while p-Akt expression is elevated in melanoma tissue compared to dysplastic nevi. We propose a two-pronged therapeutic strategy of (1) boosting PUMA expression and (2) inhibiting Akt phosphorylation.  We demonstrate that exogenous overexpression of PUMA, via adenoviral-mediated gene expression (ad-PUMA), forces melanoma cells to undergo rapid mitochondrial-mediated apoptosis in vitro. We also report that a small molecule Akt inhibitor, API-2, greatly inhibits melanoma cell growth in vitro.  Using a SCID mouse melanoma xenograft model, we show that combination treatment of ad-PUMA and API-2 dramatically suppresses tumor growth in an additive manner, leading to over 80% growth inhibition compared to controls. We also investigate the role of NF-κB overexpression in melanoma. Our lab previously reported that expression of the p50 subunit of NF-κB, in particular, correlates with melanoma progression and poor patient survival. Here, we use cDNA microarray analysis to show that p50 overexpression upregulates IL-6 in melanoma cells. We further demonstrate that p50-mediated IL-6 expression stimulates the growth of endothelial cells in vitro and promotes angiogenesis in vivo. This work supports the hypothesis that melanoma cells exploit multiple  iii mechanisms to sustain a survival advantage, including: 1) suppression of apoptosis (via PUMA down-regulation), 2) increased activation of survival pathways (via increased p- Akt), and 3) upregulation of pro-angiogenic factors (via p50-mediated IL-6 induction). This work suggests that the specific targeting of one or more key mediators of these processes may be an effective therapeutic strategy for treating malignant melanoma.  iv TABLE OF CONTENTS Abstract……………………………………………………………..……………….……... Table of Contents…………………………………………...…………..………………….. List of Tables…………………………………………...…………………………………… List of Figures……………………………………………………………….……………... Acknowledgements…………………………………………...…………………………..... Dedication………………………………………………………………………………….. Co-authorship Statement……………...…………………………………………………..... ii iv vii viii x xi xii 1 1. General Introduction………………..……………...…………………………………... 1.1  Melanoma biology………………………………………………………………... 1.1.1 Role of cutaneous melanocytes…………………………………………... 1.1.2 Melanocyte transformation …………………………..………………….. 1.2 Melanoma epidemiology and risk factors………………………………………... 1.3 Melanoma diagnosis, prognosis, and treatment………..………………………… 1.4 Apoptosis dysregulation in melanoma……………..…..…….....................……... 1.4.1 Regulation of apoptosis by the Bcl-2 family……………………….......... 1.4.2 Bcl-2 family proteins in cancer…………………………………………... 1.4.3 The role of PUMA in cancer.…….…..………………………..….…........ 1.5 Akt activation in cancer……………………………..…......……….............…..... 1.5.1 PI3K/Akt signaling pathway …………………………………………….. 1.5.2 De-regulated Akt activity in cancer………………………..…………….. 1.6 NF-κB pathway activation in cancer ……………..……...………...…....………. 1.6.1 NF-κB signaling pathway …………………………………...……........... 1.6.2 De-regulated NF-κB activity in cancer…………………..……................. 1.6.3 Interleukin-6 in cancer……………………………..………..……............ 1.7 Hypothesis and objectives ……………..……...………...…....………………….. 1.8    References………………………………………………………………………... 1 1 1 3 4 6 8 8 13 14 17 17 19 21 21 24 25 26 28 2.  PUMA Expression Is Significantly Reduced in Human Cutaneous Melanomas……… 2.1  Introduction…………………………………………………………..……….….. 2.2 Materials and methods………………...…………………………………...….….  2.2.1 Melanoma tissue microarray………………………………….…….......... 2.2.2 Immunohistochemical TMA analysis…...………………………….......... 2.2.3 Cell culture………………………………………...………………........... 2.2.4 Adenoviral constructs……………………………...…………………...... 2.2.5 Adenoviral transduction…………………………...………………..…..... 2.2.6 Reverse transcription polymerase chain reaction (RT-PCR)………..….... 2.2.7 Flow cytometric analysis of DNA fragmentation....…………………....... 2.3  Results and discussion…………..…………………………………..……….….. 41 41 43 43 43 44 44 44 45 45 46  v 2.4 References………………………………………………………………………... 56 3.  Role of PUMA and Phosphorylated Akt in Melanoma Cell Growth, Apoptosis, and Patient Survival................................................................................................................ 3.1  Introduction…………………………………………………………...……….…. 3.2 Materials and methods………………...…………………………………...……..  3.2.1 Melanoma tissue microarray…………………………………….……...... 3.2.2 Immunohistochemical TMA analysis…...………………………….…..... 3.2.3 Cell culture………………………………………...……….…………...... 3.2.4 Adenoviral constructs……………………………...…….……………..... 3.2.5 Hoechst staining…………………………………...…….……………...... 3.2.6 Subcellular fractionation……………………….……..………………...... 3.2.7 Clonogenic assay………………………………...........………………..... 3.2.8 Ad-PUMA treatment in vivo……………..……...........………………...... 3.2.9 API-2 treatment in vivo………………..………….......………………...... 3.2.10 Western blot……………………………………….......………………..... 3.2.11 Sulforhodamine B (SRB) assay…………..……….......………………..... 3.2.12 Ad-PUMA and API-2 combination treatment in vivo……………...…..... 3.3 Results and discussion…………………………………….…………………..… 3.3.1 Weak PUMA expression and strong Akt phosphorylation cooperatively  reduce melanoma patient survival…………………………….………..... 3.3.2 Ad-PUMA kills melanoma cells via rapid induction of mitochondrial- mediated apoptosis …......…………………………………….………...... 3.3.3 Ad-PUMA inhibits growth of melanoma tumor xenografts…….…...…... 3.3.4 Ad-PUMA and API-2 cooperatively inhibit melanoma cell growth…...... 3.3.5 Ad-PUMA and API-2 cooperatively inhibit growth of melanoma tumor xenografts……………………………………………………………....... 3.4 References………………………………………………………………………...  58 58 60 60 60 60 60 60 61 62 62 62 62 63 63 64  64  66 71 74  79 82 4.  Nuclear Factor Kappa B Subunit p50 Promotes Melanoma Angiogenesis by Upregulating Interleukin-6 expression………………...………………………………. 4.1  Introduction………………………………………………………….…..……….. 4.2 Materials and methods………………...…………………………….………...….  4.2.1 Cell culture………………...……………………………….…………...... 4.2.2 Transfection………………………...…...………………….…………..... 4.2.3 Lentiviral transduction……...……………………...……….…………..... 4.2.4 Western blot……………………………………......……….…………..... 4.2.5 Enzyme-linked immunosorbent assay (ELISA)…...…….……………..... 4.2.6 Sulforhodamine B (SRB) cell growth assay………………..…………..... 4.2.7 cDNA microarray analysis and real-time qPCR......……….…………...... 4.2.8 Conditioned medium………………………............………….………...... 4.2.9 Chromatin immunoprecipitation (ChIP) assay.........…………….……..... 4.2.10 In vivo angiogenesis assay and immunofluorescent staining.........……..... 4.3 Results …………………………………………………………………….…..…. 4.3.1 NF-κB p50 does not affect cell growth rate…….……………….……...... 4.3.2 NF-κB p50 upregulates IL-6 gene expression……..……….………….....  85 85 86 86 87 87 87 88 88 88 89 90 90 91 91 91  vi  4.3.3 Activating transcription factor 3 (ATF3) inhibits NF-κB p50-mediated IL-6 upregulation ……………………………………………..…………. 4.3.4 Silencing NF-κB p105/p50 expression inhibits IL-6 expression……........ 4.3.5 NF-κB p50 expression by MMRU cells stimulates endothelial cell growth………………………………………………………….………… 4.3.6    NF-κB p50 expression by MMRU cells promotes angiogenesis………… 4.4 Discussion…...…………………………………………………….…………...… 4.5 References…...…………………………………………………….…………...… 5.  Concluding Remarks………..…………………………….……………….……............ 5.1  Targeting apoptosis dysregulation in melanoma………………....…………….... 5.1.1 Anti-apoptotic versus pro-apoptotic gene therapy: the case for ad-PUMA 5.1.2 PUMA gene therapy in pre-clinical models of melanoma………..…...…. 5.1.3 Adenoviral-based gene therapy: is it clinically possible?........................... 5.2 Targeting the PI3K/Akt pathway in melanoma……………………...……..……. 5.2.1 Inhibiting constitutive Akt activation…………………...……………...... 5.2.2 Pharmacological Akt inhibitors………...………………...…………..….. 5.3 Dual target therapy of melanoma…………………..……………...…………...… 5.4 Targeting angiogenesis in melanoma………………………...…………...……… 5.4.1 Anti-angiogenic therapeutics……...……………………...…………….... 5.4.2 Targeting NF-κB proteins to inhibit angiogenesis ………………………. 5.5 Summary…………………………………………………………………………. 5.6 References………………………………………………………………………... Appendix    Animal Care Committee approval certificates for in vivo experiments…….....   99 103  108 108 111  117  121 121 121 123 126 131 131 132 134 137 137 139 140 142  149  vii LIST OF TABLES Table 2.1 PUMA expression and clinicopathological characteristics…………….…….. Table 4.1 Primer pairs used for qPCR……………………………………………….…. Table 4.2 179 genes differentially expressed in the NF-κB p50-overexpressing clone compared to the vector clone………………………………………...… 49 89  94 1  viii LIST OF FIGURES Figure 1.1 Progression of melanocyte transformation…….…………………………….. Figure 1.2 Bcl-2 protein family………….………………………………………………. Figure 1.3 Induction of apoptosis by BH3-only proteins………………………………... Figure 1.4 PI3K/Akt signal transduction pathway.……………………............................ Figure 1.5 NF-κB signal transduction pathways…………………………..……………. Figure 2.1 PUMA expression in human dysplastic nevi, primary melanoma, and metastatic melanoma tumor tissues………………………………………….. Figure 2.2  Kaplan–Meier survival analyses of melanoma patients……..……………..… Figure 2.3 Adenoviral-mediated overexpression of PUMA in melanoma cell lines…..... Figure 2.4 Ad-PUMA-induced apoptosis of melanoma cells………………….………... Figure 3.1 The combination of weak PUMA and strong p-Akt expression in human melanoma tumors correlates with poor 5-year patient survival……………… Figure 3.2 Ad-PUMA induces the formation of apoptotic bodies in human melanoma cells……………………………………………………….………….………. Figure 3.3 Ad-PUMA rapidly induces mitochondria-mediated apoptosis in human melanoma cells……………………………………………………………….. Figure 3.4 Ad-PUMA severely inhibits colony formation by human melanoma cells….. Figure 3.5 Ad-PUMA inhibits MMRU tumor growth in vivo……………….………….. Figure 3.6 Ad-PUMA inhibits MMAN tumor growth in vivo…………….…………….. Figure 3.7 API-2 inhibits p-Akt expression by human melanoma cells in vitro……..….. Figure 3.8 API-2 inhibits the survival of human melanoma cells in vitro………….…… Figure 3.9 API-2 enhances ad-PUMA-mediated growth inhibition in vitro…….…….… Figure 3.10 API-2 enhances ad-PUMA-mediated growth inhibition in vivo…..………..... Figure 4.1 Effect of NF-κB p50 overexpression on melanoma cell growth…….............. 2 9 11 18 23  47 50 53 54  65  67  69 70 72 73 75 77 78 80 92 1  ix Figure 4.2 cDNA microarray analysis of the NF-κB p50 and vector stable clones……... Figure 4.3 IL-6 transcription is highly upregulated in NF-κB p50 stable clones……….. Figure 4.4 NF-κB p50 induces IL-6 mRNA transcription………………………………. Figure 4.5 NF-κB p50 induces IL-6 protein expression………………………………… Figure 4.6 NF-κB p50 protein binds to the IL-6 gene promoter consensus sequence…... Figure 4.7 ATF3 expression inhibits p50-mediated upregulation of IL-6 transcripts…… Figure 4.8 ATF3 expression inhibits p50-mediated upregulation of IL-6 protein expression……………………………………………………………………. Figure 4.9 Characterization of the sh-p50 stable clone………………………………….. Figure 4.10 Knockdown of p50 expression reduced IL-6 expression…………………….. Figure 4.11 NF-κB p50 expression promotes endothelial cell growth in vitro…………… Figure 4.12 NF-κB p50 expression induces VEGF upregulation………………………… Figure 4.13 NF-κB p50 expression promotes angiogenesis in vivo……………………..... 93 98 100 101 102 104 105 106 107 109 110 112 1   x ACKNOWLEDGEMENTS I would like to thank Nhu Le for his assistance with statistical analyses, David Huntsman and Nikita Makretsov for their help with tissue microarray construction, Andrew Coldman for providing patient survival data, and Anne Haegert for her assistance with the cDNA microarray analysis.  I would also like to thank Magdalena Martinka for sharing her expertise in melanoma pathology with our lab and her assistance with scoring tissue microarray analyses.  I would also like to acknowledge all the members of the Li Lab (past and present) for their contributions, but especially Derek Dai and Ronald Wong.  Derek was a co- author on two of my papers and we worked closely together for many months.  I thank Derek for his patience and professionalism.  Ronald is another superb colleague; his thoughtful discussions of experimental design and his supportive nature helped me to complete this work.  Thank you Ronald for never being too busy to listen.  Lastly, I would like to thank my supervisor, Dr. Gang Li.  Your guidance over the last six years has taught me to set my goals high and to persevere.  You will always have my admiration and respect.  xi DEDICATION This work is dedicated to my parents, Gerald and Audrene Karst.  Dad, thank you for teaching me a strong work ethic.  Mom, thank you for raising me to believe I could become whatever I dreamed.  Your unwavering support has enabled me to succeed. Words cannot express my gratitude… I only hope I can make you both proud.  xii CO-AUTHORSHIP STATEMENT Chapters 2 and 3 were co-authored by Derek Dai.  Derek contributed greatly to the tissue microarray analyses, adenoviral amplification, and animal work included in this thesis. Chapter 4 was co-authored by Kai Gao.  Kai conducted the cDNA microarray analysis (Figures 4.2 and 4.3) that provided the rationale for most of the ensuing experiments.     1 1.   GENERAL INTRODUCTION 1.1 MELANOMA BIOLOGY Melanoma arises from the uncontrolled proliferation of pigment-producing cells called melanocytes.  The vast majority of human melanocytes reside in the skin and therefore cutaneous melanoma is the predominant form of the disease in humans.  A small number of melanomas (4-5%) arise from melanocytes of the eye, anogenital regions, mucosal surfaces, nail beds, and leptomeninges (Hussein, 2008).  Cutaneous melanocytes originate from the embryonic neural crest.  Melanocyte precursor cells, called melanoblasts, migrate through the dermal layer of the skin during fetal development, eventually finding their niche at the dermal/epidermal border (Takeda et al., 2007). 1.1.1 Role of cutaneous melanocytes Human skin is composed of two main layers - the dermis and epidermis - beneath which lies a layer of adipose and connective tissue (Figure 1.1A).  At the dermal/epidermal junction resides a layer of basal cells composed of keratinocytes and melanocytes, occurring at a ratio of approximately 10:1 (Burgeson and Christiano, 1997).  The basal keratinocytes give rise to multiple layers of overlying epidermal keratinocytes, which form the major structural component of the epidermis.  Keratinocyte layers are constantly being pushed upwards, during which time they differentiate, becoming enucleated, flattened, and deposited with keratin protein.  Terminally differentiated keratinocytes form a dead surface layer of cells (the stratum corneum) that protects our skin from the entry of foreign matter, infectious agents, and water loss (Eckert and Rorke, 1989).  2 Figure 1.1 Progression of melanocyte transformation. (A) Melanocytes in normal skin are distributed evenly along the dermal/epidermal junction. (B) Benign proliferations of melanocytes (nevi) commonly occur.  Atypical (dysplastic) nevi show a degree of dysplasia but are still considered non/pre-malignant. (C) The primary stage of melanoma, called the radial growth phase, is characterized by disorderly growth of epidermal melanocytes, including pagetoid spread (upward migration). (D) The vertical growth phase of melanoma involves dermal invasion and leads directly to melanoma metastasis. Adapted from Gray-Schopfer et al. (2007); reproduced with permission from Nature Publishing Group.    3  A major function of melanocytes is to protect keratinocytes from cellular damage by supplying them with a photo-protective pigment called melanin (Lin and Fisher, 2007).  Melanin absorbs ultraviolet radiation from the sun, thus minimizing free-radical generation and DNA damage (Agar and Young, 2005).  Melanocytes in the basal cell layer synthesize melanin and deliver it to nearby keratinocytes via dendritic processes extending from the melanocyte cell body.  Each melanocyte serves approximately 40 suprabasal keratinocytes.  Phenotypic differences in skin color are determined not by the number of cutaneous melanocytes, but by the amount and type of melanin produced, as well as its distribution pattern (Yamaguchi et al., 2007). 1.1.2 Melanocyte transformation It is quite common to find benign proliferations of melanocytes, termed nevi, or commonly called "moles" (Figure 1.1B).  Nevi may be either congenital or acquired during childhood or adolescence (Schaffer, 2007).  They appear as nest-like melanocytic cell clusters in the lower epidermis or dermis.  Benign nevi have well-defined borders and do not interfere with surrounding cutaneous structures such as hair follicles, nerves, or blood vessels.  Some nevi are classified as "atypical" or "dysplastic", referring to the presence of abnormal clinical or histologic features, respectively.  These nevi exhibit a certain degree of architectural and cytological atypia, but are not considered to be malignant.  Many dermatologists believe that the dysplastic nevus represents an intermediate stage of evolution from benign nevus to malignant melanoma (Hussein, 2005).  However, this hypothesis remains controversial, as many dysplastic nevi never develop into melanomas.  4  A disorderly proliferation of melanocytes or nevus cells suggests malignant behavior and may signal the transformation of melanocytes to melanoma (Figure 1.1C). Typically, melanoma cells spread first through the epidermis, which is described as the radial growth phase (RGP).  However, melanoma is a highly aggressive and invasive malignancy.  As such, it will usually progress rapidly to a vertical growth phase (VGP) and invade the underlying dermis (Figure 1.1D).  Downward vertical growth facilitates contact with vascular and lymphatic vessels, thus providing a direct route for melanoma metastasis (Gray-Schopfer et al., 2007)  1.2  MELANOMA EPIDEMIOLOGY AND RISK FACTORS In the United States, melanoma incidence rates have been rising steadily for decades. During the 1970's and 80's, melanoma incidence increased by about 5% per year in both genders (Espey et al., 2007).  Recently, this trend has slowed somewhat; for the period of 1995-2004, melanoma incidence increased by about 3% per year in both genders (Espey et al., 2007).  Melanoma is currently the sixth and seventh most common cancer among American men and women, respectively (Jemal et al., 2008).  In Canada, incidence rates are lower but continue to increase.  From 1995 to 2004, melanoma incidence increased by 1.8% per year in Canadian men and 1.0% per year in Canadian women (Canadian Cancer Society/National Cancer Institute of Canada: Canadian Cancer Statistics 2008) International rates of melanoma incidence vary according to a combination of ethnic background and geographic location; they are highest in Caucasian populations and increase with proximity to the equator (Lens and Dawes, 2004).  The highest melanoma incidence rates in the world occur in Queensland, Australia, where there are 56 new cases  5 per year per 100,000 men and 43 new cases per year per 100,000 women (Leiter and Garbe, 2008).  As with nearly all cancers, the risk factors for melanoma are both genetic and environmental in nature.  On the genetic side, individuals having fair skin, freckles, and who sunburn easily are at higher risk for melanoma (Evans et al., 1988).  Risk also increases with the total number of melanocytic nevi present on the body (Holly et al., 1987), as approximately one quarter of melanomas are histologically associated with a nevus (Bevona et al., 2003).  The likelihood of developing melanoma is further heightened by the presence of very large nevi (> 5 mm diameter) (Naldi et al., 2000) or a high number of dysplastic nevi (Titus-Ernstoff et al., 2005).  Nevi patterns are a heritable trait and therefore cases of melanoma frequently cluster in families (Goldstein et al., 1993).  In fact, a familial syndrome called "dysplastic nevus syndrome" or "atypical mole syndrome" has long been recognized as a major risk factor for melanoma (Kraemer and Greene, 1985).  Familial melanoma accounts for 5-12% of all cases (Goldstein and Tucker, 2001) and is most commonly associated with a germ line mutation in the cyclin- dependent kinase inhibitor 2A (CDKN2A) gene, which codes for the tumor suppressor proteins p16INK4A and p14ARF (Bishop et al., 2002).  The primary environmental risk factor for melanoma is exposure to ultraviolet radiation (UVR), usually as sunlight (Jemal et al., 2001).  While non-melanoma skin cancers are associated with chronic sun damage and cumulative UVR exposure, melanomas appear to be linked to intense, intermittent doses of UVR (Jhappan et al., 2003).  Consistent with this hypothesis is the observation that the torso and legs are the most common sites of melanoma in men and women, respectively; sites that are protected  6 from the sun for most of the year, but may be subjected to high-level UVR exposure during the summer months (Jhappan et al., 2003).  Moreover, a history of childhood sunburns has been shown to be an important risk factor for melanoma (Whiteman et al., 2001).  Individuals with risk factors for melanoma must take care to limit UVR exposure, avoid sunburn, and most importantly, perform routine skin self-examinations.  1.3 MELANOMA DIAGNOSIS, PROGNOSIS AND TREATMENT The diagnostic clinical features of melanoma are represented by the acronym ABCDE: asymmetry, border irregularity, color variation, diameter (> 6 mm), and evolution (change over time) (Abbasi et al., 2004).  It is based on these characteristics that patients are referred to a dermatologist for assessment.  Dermatologists use high-resolution optical devices called dermatoscopes to examine suspicious lesions more closely and determine whether they may be malignant (Argenziano et al., 2003).  In the case of uncertainty, the lesion must be biopsied and evaluated by an experienced histopathologist.  Pathological diagnosis of melanocytic lesions is very complex and takes into consideration a whole range of architectural and cytological features.  Upon diagnosis, the most important prognostic factor for a melanoma patient is the depth of invasion or "thickness" of their tumor.  Thickness is quantified according to the Breslow method - i.e. by measuring the vertical distance (in millimeters) from the granular layer of the epidermis to the deepest cell of the melanoma lesion (Breslow, 1970).  Other important prognostic factors include ulceration (lack of intact epidermis), satellite tumor foci (micrometastasis), and regional lymph node involvement (macrometastasis).  Each of these factors are used to stage melanoma patients as either  7 stage I, II, III, or IV (Balch et al., 2001).  In stage I, tumor tissue can usually be completely surgically resected with excellent prognosis (10-year survival rate of 80- 100%).  However, prognosis rapidly worsens with advancing disease stage.  Patients in stages II, III, and IV have 10-year survival rates of about 50-65%, 25-50%, and 0-15%, respectively (Balch et al., 2001; Jhappan et al., 2003).  Unfortunately, there are currently no effective therapies for advanced melanoma. The only cytotoxic drug ever approved by the U.S. Food and Drug Administration (FDA) for the treatment of metastatic melanoma is the alkylating agent dacarbazine (DTIC). DTIC was first introduced in the 1970's and produces partial responses in only 13-20% of patients, with a median duration of 5-6 months (Chapman et al., 1999).  Complete responses are seen in less than 5% of patients (Chapman et al., 1999).  Countless attempts have been made over the years to combine DTIC with other chemo- or immuno- therapeutic agents in order to improve response rates.  However, no consistent survival advantage has ever emerged from such studies (Eggermont and Kirkwood, 2004).  In the United States, high dose interferon alpha is often given as an adjuvant treatment for advanced stage melanoma.  However, the treatment is highly toxic and its use is controversial because it has not shown consistent overall survival benefits in clinical trials (Kalaaji, 2007).  Sadly, over the past three decades, virtually no progress has been made in the treatment of metastatic melanoma.  This dismal fact illustrates the dire need for the development of novel therapeutic strategies for this disease.    8 1.4 APOPTOSIS DYSREGULATION IN MELANOMA The notorious resistance of melanoma to conventional cancer treatments likely stems from an underlying resistance to apoptosis.  Apoptosis is the mechanism of programmed cell death by which the body rids itself of damaged, genetically defective, or superfluous cells (Willis and Adams, 2005).  Apoptosis plays a critical role in tumor suppression by preventing the accumulation of cells with tumorigenic potential.  Melanoma tumors have been shown to exhibit lower rates of spontaneous apoptosis than other solid tumor types (Gilchrest et al., 1999).  It is widely believed that melanocytic cells acquire a resistance to apoptosis during the progression from normal melanocyte to dysplastic nevus cell to melanoma cell (Hussein et al., 2003).  In accordance with this notion, melanocytic nevus cells show a greater resistance to apoptosis than do melanocytes, when grown in collagen gels (Alanko et al., 1999).  Furthermore, most melanoma cells lines are highly resistant to drug-induced apoptosis (Li et al., 1998). 1.4.1 Regulation of apoptosis by the Bcl-2 family Chemotherapy and radiation treatments both kill cells by inducing DNA damage, which activates a type of cell death know as "intrinsic" or "mitochondrial-mediated" apoptosis (Fisher, 1994).  Mitochondrial-mediated apoptosis is regulated by the B-cell leukemia/lymphoma 2 (Bcl-2) family of proteins (Figure 1.2).  This family is comprised of both pro- and anti-apoptotic members, all of which contain one or more Bcl-2 homology (BH) domains.  Bcl-2 family proteins can be divided into three functional groups: 1) anti-apoptotic Bcl-2-like members, 2) pro-apoptotic Bax-like members, and 3) pro-apoptotic BH3-only members (Willis and Adams, 2005).  The cell controls apoptosis Figure 1.2 Bcl-2 protein family. Bcl-2 proteins are divided into three functional groups: 1) Bcl-2-like, 2) Bax-like, and 3) BH3-only proteins. Anti-apoptotic Bcl-2-like proteins functionally oppose Bax-like and BH3-only proteins, which are both pro- apoptotic. Most Bcl-2-like proteins contain four Bcl-2 homology (BH) domains; BH1, BH2, BH3, and BH4. Bax-like proteins contain two or three BH domains. BH3-only proteins contain only a BH3 domain. Most Bcl-2 family members also express a trans- membrane (TM) domain that facilitates mitochondrial membrane localization.    TMBH2BH1BH3 BH3 BH3 TM BH2BH1BH3 TM Noxa Bim Bmf Bik Hrk  Puma Bid Bad Bax Bak Bok Bcl-2 Bcl-XL Bcl-w Mcl-1* A1 BH4 BH3-only proteins: Bax-like proteins: Bcl-2-like proteins: Anti- apoptotic Pro- apoptotic *Mcl-1 does not contain a BH4 domain.  9  10 by carefully balancing the expression/activity levels of anti- and pro-apoptotic Bcl-2 family proteins.  BH3-only proteins function as the “damage sensors” of the cell.  They are activated in response to a diverse range of stress stimuli including cytokine withdrawal, loss of adhesion to the extracellular matrix, DNA damage, and oncogene activation (Shibue and Taniguchi, 2006).  BH3-only proteins are activated via either transcriptional upregulation (PUMA, Noxa, Bim, Hrk) or posttranslational modification (Bid, Bim, Bmf, Bad, Bik) (Cory and Adams, 2002).  The particular BH3-only protein(s) activated in a given situation depends on the nature of the cytotoxic stimulus and the tissue type involved.  Once activated, BH3-only proteins bind to their Bcl-2-like and Bax-like relatives (Shibue and Taniguchi, 2006) (Figure 1.3).  The binding of BH3-only proteins to Bcl-2-like proteins is thought to neutralize their pro-survival function and “prime” the cell for apoptosis (Willis and Adams, 2005).  This antagonizing interaction is mediated by the BH3 domain, an amphipathic α-helix, which inserts into a hydrophobic groove formed by the BH1, BH2 and BH3 domains on the surface of Bcl-2-like protein (Liu et al., 2003).  Changes induced by BH3-only proteins are also thought to activate pro- apoptotic Bax and Bak proteins (Cheng et al., 2001; Zong et al., 2001).  Bax or Bak activation is required for apoptosis to proceed; activated Bax/Bak form oligomers that permeabilize the outer membrane of the mitochondria (Cheng et al., 2001; Zong et al., 2001).  Disruption of this membrane results in the leakage of pro-apoptotic factors [e.g. cytochrome c, second mitochondria-derived activator of caspases (Smac), apoptosis inducing factor (AIF)] into the cytosol, subsequent apoptosome formation, activation of caspases, and inevitably, cell death (Green and Kroemer, 2004).  11 Figure 1.3 Induction of apoptosis by BH3-only proteins. BH3-only proteins are activated in response to DNA damage or other cellular stress via transcriptional upregulation (Puma, Noxa, Bim, Hrk) or posttranslational modification (Bid, Bim, Bmf, Bad, Bik). Activated BH3-only proteins interact with Bcl-2-like proteins at the mitochondrial membrane, neutralizing their pro-survival function and “priming” the cell for apoptosis. Activated BH3-only proteins are also thought to activate pro-apoptotic Bax and Bak proteins. Activated Bax/Bak form homo-oligomers that lead to permeabilization of the mitochondrial outer membrane, release of cytochrome c, caspase activation, and, ultimately, apoptotic cell death.   12 A systematic study of BH3-only protein binding affinities revealed that most BH3-only proteins bind selectively, not promiscuously, to Bcl-2-like anti-apoptotic proteins (Chen et al., 2005).  Although PUMA, Bim, and activated Bid (tBid) can bind to and neutralize all five Bcl-2-like proteins (Bcl-2, Bcl-XL, Bcl-w, Mcl-1, and A1), other BH3-only members can not.  Specifically, Bad and Bmf bind only to Bcl-2, Bcl-XL, and Bcl-w; Bik and Hrk bind only to Bcl-XL, Bcl-w, and A1; and Noxa binds only to A1 and Mcl-1 (Willis and Adams, 2005).  Therefore, some BH3-only proteins, like PUMA, appear to play broader, more dominant roles in the apoptotic program, whereas others act with more narrow situational specificity.  As previously mentioned, BH3-only proteins require Bax/Bak activation in order to induce apoptosis (Cheng et al., 2001; Zong et al., 2001).  However, the exact mechanism underlying this activation has not been fully elucidated.  Two models have been postulated to account for the role of BH3-only proteins in Bax/Bak activation.  The first, proposed by Letai et al., asserts that some BH3-only proteins can directly activate Bax/Bak.  This model classifies each BH3-only protein as either an “activator” or “sensitizer” (Letai et al., 2002).  Activators (e.g. tBid and Bim) interact directly with Bax/Bak, inducing their oligomerization.  However, they are prevented from doing so under normal conditions because they are sequestered in the cytosol by anti-apoptotic Bcl-2-like proteins.  Sensitizers (e.g. Bad and Bik) are activated following cytotoxic stress, whereupon they bind to Bcl-2-like proteins and cause them to release bound activators (Letai et al., 2002).  An alternative model, proposed by Willis et al., proposes that BH3-only proteins activate Bax/Bak indirectly, by displacing them from anti- apoptotic Bcl-2-like proteins.  This model asserts that under normal conditions, Mcl-1  13 and Bcl-XL sequester an active form of Bak in the cytosol (Willis et al., 2005). Following cytotoxic stimuli, activated BH3-only proteins must displace Bak from both Mcl-1 and Bcl-XL in order to achieve Bak activation.  An analogous mechanism, involving an active form of Bax, supposedly occurs at the mitochondrial membrane (Willis et al., 2005).  Currently, there is experimental evidence to support both models (Willis and Adams, 2005).  Therefore, further study will be required to resolve the issue. 1.4.2 Bcl-2 family proteins in cancer Bcl-2-like anti-apoptotic proteins have long been implicated in tumorigenesis.  The most well known example is the deregulation of Bcl-2 expression in human follicular lymphoma.  This disease is characterized by a t(14;18) chromosomal translocation that places the Bcl-2 gene under control of an immunoglobulin heavy chain promoter, resulting in constitutive expression of Bcl-2 protein.  Bcl-2 and Bcl-XL overexpression have also been linked to other malignancies, including acute promyelocytic leukemia, breast cancer, and pancreatic β-cell cancer (Coultas and Strasser, 2003).  Meanwhile, the absence of pro-apoptotic Bax-like proteins is associated with tumor progression.  For example, loss or mutation of Bax may contribute to colon cancer (Rampino et al., 1997), mammary tumors (Shibata et al., 1999), and brain tumors (Yin et al., 1997).  However, it is only in recent years that researchers have begun to examine the roles of pro-apoptotic BH3-only proteins in tumorigenesis.  Because BH3-only proteins are critical effectors of apoptosis following cellular damage, their proper regulation is a crucial component of tumor suppression.  Absence or loss of BH3-only protein expression may promote an apoptotic resistance phenotype.  In the context of cancer treatment, this presents a major therapeutic obstacle.  14 1.4.3 The role of PUMA in cancer PUMA (p53 upregulated modulator of apoptosis) was discovered by three independent groups, two of which were searching for p53-inducible target genes (Nakano and Vousden, 2001; Yu et al., 2001), and one of which was screening for Bcl-2-binding partners (Han et al., 2001).  The PUMA gene contains two consensus p53 binding sites within its promoter region, indicating that it is a direct transcriptional target of p53 (Han et al., 2001; Nakano and Vousden, 2001; Yu et al., 2001).  PUMA is not known to undergo posttranslational modifications, but upon translation, localizes to mitochondria where it antagonizes anti-apoptotic members of the Bcl-2 protein (Strasser, 2005). PUMA expression is strongly and rapidly upregulated in response DNA damaging agents, such as adriamycin, 5-FU, actinomycin D, etoposide, and ionizing radiation, in a p53-dependent manner (Han et al., 2001; Nakano and Vousden, 2001; Yu et al., 2001). However, PUMA expression can also be induced by p53-independent apoptotic stimuli including glucocorticoid treatment, serum starvation, hypoxia, and cytokine withdrawal (Han et al., 2001; Nakano and Vousden, 2001; Yu et al., 2001; You et al., 2006).  Less is known about p53-independent mechanisms of PUMA induction. PUMA-/- mice do not suffer developmental abnormalities, but thymocytes and myeloid progenitors from these animals are protected against apoptosis induced by anticancer drugs or γ-irradiation.  Thymocytes from PUMA-/- mice are also resistant to cytokine withdrawal, glucocorticoid treatment, staurosporine, and the phorbol ester PMA (Jeffers et al., 2003; Villunger et al., 2003).  Of all known BH3-only proteins, PUMA is the most potent inducer of apoptosis (Chen et al., 2005).  Exogenous PUMA expression causes rapid and complete cell death of many different malignant cell types, including  15 colorectal cancer, lung cancer, osteosarcoma, glioma, head and neck cancer, and melanoma cells (Hoque et al., 2003; Ito et al., 2005; Karst et al., 2005; Nakano and Vousden, 2001; Yu et al., 2001).  Meanwhile, PUMA-/- cells show a marked resistance to apoptosis.  There is abundant experimental evidence supporting the notion that PUMA is a critical mediator of p53-dependent apoptosis.  For example, mouse embryonic fibroblasts (MEFs) transduced with the adenoviral oncogene E1A are normally rendered sensitive to p53-mediated apoptosis (Lowe et al., 1993), but PUMA-/- MEFs are strongly resistant to DNA damaging agents under the same conditions (Villunger et al., 2003). Colorectal cancer cells and lung cancer cells lacking wild-type p53 are unable to upregulate PUMA following DNA damage (Nakano and Vousden, 2001; Yu et al., 2001).  Similarly, colon carcinoma cells expressing the human papillomavirus (HPV) E6 protein, which rapidly degrades p53 (Kessis et al., 1993), cannot express PUMA following DNA damage (Nakano and Vousden, 2001). Loss of PUMA has not been shown to cause cancer.  However, inhibition of PUMA expression does promote certain malignant phenotypes.  For example, Hemann et al. reported that shRNA-mediated knockdown of PUMA potently induces transformation of MEFs co-expressing the oncogenes E1A and ras (Hemann et al., 2004).  It is well established that p53 mutations cooperate with E1A and ras to promote oncogenic transformation (Debbas and White, 1993; Hemann et al., 2004; Lowe et al., 1994).  Thus, PUMA loss mimics p53 mutation in this context.  In the same study by Hemann et al., suppression of PUMA accelerated Eµ-myc-induced lymphomagenesis (Hemann et al., 2004).  Eµ-myc transgenic mice express the c-myc oncogene from an Ig heavy chain enhancer and develop B cell lymphomas at an early age (Adams et al., 1985).  16 Furthermore, hematopoietic stem cells (HSCs) from these Eµ-myc transgenic mice will give rise to lymphomas if they are transferred into healthy recipient mice.  It has been shown that suppression or deletion of p53 in this Eµ-myc transgenic model results in accelerated lymphomagenesis with a more aggressive phenotype (Hemann et al., 2003; Schmitt et al., 1999).  To determine whether deletion of PUMA would have a similar effect, shRNA against PUMA was used to stably suppress its expression in Eµ-myc HSCs.  Upon transferring these cells to normal recipients, they observed dramatically accelerated lymphomagenesis with reduced latency in all recipient animals.  Thus, PUMA loss mimics p53 loss and cooperates with myc in lymphomagenesis.  Overall, these results indicate that PUMA loss can recapitulate the effects of p53 loss/mutation and is a bona fide tumor suppressor. Considering PUMA’s essential role in apoptosis and its “p53-like” tumor suppressing effects, it seems intuitive that PUMA loss may contribute to human cancer. Chromosomal locus 19q13.3, containing the PUMA gene, is known to be frequently lost in human gliomas (Yong et al., 1995), neuroblastomas (Mora et al., 2001), and certain B- cell lymphomas (Shimazaki et al., 2000).  Deletion of chromosomal arm 19q has also been reported in head and neck squamous cell carcinomas (HNSCCs) and lung cancers (Johns et al., 1996; Miura et al., 1990; Mutirangura et al., 1997; Nawroz et al., 1994; Sanchez-Cespedes et al., 2001; Wong et al., 2002).  Hoque et al. reported loss of heterozygosity (LOH) at 19q in 56% of HNSCCs and 27% of primary lung cancer samples (N=30) (Hoque et al., 2003). Mutational analyses of PUMA in cancer have so far found no evidence of gene mutation (Ahn et al., 2008; Hoque et al., 2003; Yoo et al., 2007).  However, a recent  17 study of Burkitt lymphomas suggests that PUMA may be silenced in tumor cells (Garrison et al., 2008).  Garrison et al. reported that approximately 40% of primary human Burkitt lymphomas fail to express detectable levels of PUMA, and this observation was associated with DNA methylation in many cases.  Furthermore, using a mouse model, they presented evidence that Eµ-myc lymphomas naturally select against PUMA protein expression.  1.5 AKT ACTIVATION IN CANCER 1.5.1 PI3K/Akt signaling pathway The serine/threonine kinase Akt [also called protein kinase B (PKB)] is a major downstream target of the enzyme phosphatidylinositol 3'-kinase (PI3K).  PI3K is recruited to and activated by stimulated cell surface growth factor receptors in response to ligand binding (Burgering and Coffer, 1995; Franke et al., 1995).  Active PI3K then catalyzes the production of a second messenger molecule, phosphatidylinositol-3,4,5- trisphosphate (PIP3) (Figure 1.4).  PIP3 recruits cytosolic Akt to the plasma membrane via its pleckstrin homology (PH) domain and causes a change in its conformation, allowing it to be phosphorylated by phosphoinositide-dependent kinase 1 (PDK1) and other kinases (Nicholson and Anderson, 2002).  Akt is activated by phosphorylation of both its Ser473 and Thr308 residues (Harlan et al., 1995).  Activated Akt acts on a variety of downstream targets to promote cell survival, growth, and proliferation.  Akt promotes cell survival by suppressing several mediators of apoptosis.  For example, it phosphorylates and inhibits the pro-apoptotic proteins Bad Figure 1.4 PI3K/Akt signal transduction pathway.  Stimulated growth factor receptors recruit adaptor proteins (Grb, Sos), which in turn leads to Ras-mediated PI3K activation. Active PI3K catalyzes the conversion of phosphatidylinositol-4,5-bisphosphate (PIP2) to phosphatidylinositol-3,4,5-trisphosphate (PIP3).  PIP3 recruits cytosolic Akt to the plasma membrane, where it is phosphorylated and activated by PDK1 and other kinases. Activated Akt acts on a variety of downstream targets to promote cell survival (via inhibition of apoptosis), cell growth (via enhanced protein translation) and cell proliferation.  See text for a detailed explanation.   S6K1 mTOR Raptor PI(4,5)P2 PI(3,4,5)P3 Ras Grb2 Sos PI3K PTEN AKT PDK1 Hdm2 4E-BP1p53 Bad Bim  mTORC1 FasL Apoptosis Protein translation p27 FOXO Proliferation  18  19 (Datta et al., 1997) and caspase-9 (Fujita et al., 1999).  Akt also phosphorylates and inhibits forkhead (FOXO) transcription factors, thus blocking FOXO-mediated transcription of the pro-apoptotic genes Bim and Fas Ligand (FasL) and transcription of the cyclin-dependent kinase inhibitor p27Kip1 (Reagan-Shaw and Ahmad, 2007).  Another substrate of Akt is the human double minute (Hdm2) oncogene.  Akt phosphorylation of Hdm2 promotes its translocation to the nucleus, where it triggers p53 protein degradation and inhibits activation of p53-dependent apoptotsis (Mayo and Donner, 2001).  Activated Akt promotes cell growth by activating mTOR complex 1 (mTORC1), which enhances translation and protein synthesis through its downstream targets S6K1 and eukaryotic initiation factor 4E binding protein 1 (4E-BP1) (Chiang and Abraham, 2007).  Akt stimulates cell proliferation by phosphorylating p27Kip1, thus impairing its import into the nucleus where it induces cell cycle arrest (Liang et al., 2002).  Activated Akt also enhances glycogen synthesis by phoshorylating and inactivating glycogen synthase kinase 3 (GSK-3) (van Weeren et al., 1998). 1.5.2 De-regulated Akt activity in cancer Inappropriate activation of Akt has been observed in many types of cancer and is most often attributed to: 1) activating mutations of PIK3CA, 2) PIK3CA and/or Akt gene amplification, or 3) allelic loss and/or inactivating mutations of PTEN (Osaki et al., 2004).  PIK3CA codes for the catalytic subunit (P110α) of PI3K; activating mutations of this gene increase PI3K activity, resulting in elevated PIP3 levels and, in turn, increased activation of Akt.  A high frequency of PIK3CA mutations have been found in cancers of the breast, lung, and colon, stomach, and brain (Samuels et al., 2004).  Gene amplifications of either PIK3CA and/or Akt have been reported in ovarian (Shayesteh et  20 al., 1999; Bellacosa et al., 1995), cervical (Ma et al., 2000), skin (Byun et al., 2003; Woenckhaus et al., 2002), gastric (Byun et al., 2003; Staal, 1987), pancreatic (Cheng et al., 1996), and brain (Knobbe and Reifenberger, 2003) tumors.  Generation of PIP3 by PI3K is negatively regulated by phosphatase and tensin homolog (PTEN).  PTEN dephosphorylates PIP3 and thus attenuates PI3K signaling.  Allelic loss and/or mutations of PTEN have been documented in breast (Garcia et al., 1999), bladder (Wang et al., 2000), prostate (Ittmann, 1998), endometrium (An et al., 2004), and brain cancers (Rasheed et al., 1999), often correlated with advanced or high grade disease.  Silencing of PTEN, due to promoter hypermethylation, has been reported in colorectal cancer (Goel et al., 2004).  Stahl et al. have demonstrated the importance of PTEN loss in melanoma by transferring an intact chromosome 10 into PTEN-null cells in order to express the protein at normal physiological levels (Stahl et al., 2003).  They found that PTEN expression inhibited tumor growth in mice, unless PTEN was subsequently deleted through loss of heterozygosity (LOH).  PTEN loss by this mechanism induced Akt activation and suppressed apoptosis in melanoma tumors, leading to enhanced growth.  Immunohistochemical detection of activated Akt (p-AktSer473) is frequently used to assess the level of active Akt in human tumor samples.  Strong p-AktSer473 immunoreactivity has been described in numerous cancer types and is correlated with poor prognosis in glioma (Ermoian et al., 2002), head and neck cancer (Lee et al., 2001), pancreatic ductal carcinoma (Yamamoto et al., 2004), gastric cancer (Nam et al., 2003), and breast cancer (Pérez-Tenorio et al., 2002), and melanoma (Dai et al., 2005).  Studies of Akt in melanoma indicate that Akt is an important player in tumor progression.  Akt is constitutively activated in 43% to 67% of melanomas (Robertson,  21 2005) and p-Akt levels are significantly higher in malignant melanoma tissues compared to normal and dysplastic nevi (Dai et al., 2005; Dhawan and Richmond, 2002).  Our lab has previously shown that p-Akt expression correlates directly with melanoma invasion and progression, and Akt is an independent prognostic determinant of 5-year patient survival, with strong p-Akt expression predicting for worse prognosis (Dai et al., 2005).  1.6 NF-κB ACTIVATION IN CANCER 1.6.1 NF-κB signaling pathway The mammalian nuclear factor kappa B (NF-κB) family of transcription factors is comprised of five members: p105/p50 (NF-κB1), p100/p52 (NF-κB2), RelA (p65), RelB, and c-Rel.  All members contains a highly conserved N-terminal Rel homology domain (RHD) that facilitates their dimerization and binding to DNA (Gilmore, 1990).  NF-κB proteins dimerize in different combinations to form transcriptionally active complexes that target specific gene promoters (Ghosh and Karin, 2002).  Multiple subunit pairings are possible, including both homo- and heterodimers, thus giving rise to the complexity with which NF-κB regulates gene transcription.  NF-κB is a key mediator of inflammatory and immune responses.  It is activated in response to pro-inflammatory cytokines [e.g. tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), epidermal growth factor (EGF)], T- and B-cell mitogens, bacteria and lipopolysaccharides (LPS), viruses, viral proteins, double-stranded RNA, and other forms of cellular stress (Perkins and Gilmore, 2006).  In normal cells, NF-κB activation is tightly controlled; activation is transient and subject to negative feedback.  22  NF-κB proteins may be subdivided into “NF-κB” subunits (p105/p50 and p100/p52) and “Rel” subunits (RelA, RelB, and c-Rel).  NF-κB subunits p50 and p52 are derived from precursor proteins p105 and p100, respectively, via proteolytic cleavage. NF-κB and Rel subunits partner in different combinations to form various transcriptional complexes.  The most commonly observed NF-κB transcriptional complex is the p50/p65 (or p50/RelA) heterodimer.  Rel subunits bear a C-terminal transactivation domain not present in NF-κB subunits.  Therefore, dimers not containing a Rel subunit (i.e. p50 and p52 homodimers) are thought to require a transcriptional co-activator in order to initiate gene transcription (Gilmore, 2006).  Under basal conditions, Rel-containing NF-κB dimers are sequestered in the cytosol by inhibitor of NF-κB (IκB) proteins, thus preventing their entry into the nucleus. IκB-bound dimers may be activated via one of two pathways - the canonical or non- canonical pathway (Figure 1.5).  In both pathways, IκB kinase (IKK) complexes are induced by external stimuli to phosphorylate IκB proteins, inducing their rapid degradation and the liberation of NF-κB dimers (Scheidereit, 2006).  These active dimers then enter the nucleus where they bind to target gene promoters and initiate transcription. The canonical and non-canonical pathways differ in that they employ different types of IKK complexes and generate different forms of active NF-κB.  A third pathway leading to NF-κB-mediated transcriptional activity, independent of IκB regulation, has also been described.  In this pathway p50 or p52 homodimers freely enter the nucleus but must combine with a transcriptional activator, such as Bcl-3, to activate gene transcription (Gilmore, 2006).  The regulatory mechanisms of this latter pathway remain poorly understood. Figure 1.5 NF-κB signal transduction pathways. In the canonical and non-canonical pathways, NF-κB dimers are maintained in the cytoplasm by interaction with IκB molecules.  Stimulation of cell surface receptors recruits adaptor proteins to the receptor which, in turn, recruit an IKK complex.  The IKK complex phosphorylates NF-κB- sequestering proteins [IκB (canonical) or p100 (non-canonical)], thus liberating NF-κB dimers and allowing them to enter the nucleus and initiate transcription.  The canonical and non-canonical pathways differ in that they employ different types of IKK complexes and generate different forms of active NF-κB.  In pathway 3, p50 homodimers enter the nucleus, where they become transcriptional activators by virtue of interaction with co- activators (e.g. Bcl-3).  How pathway 3 is regulated is not understood.  Adapted from Gilmore TD. (2006); reproduced with permission from Nature Publishing Group.    23  24 1.6.2 De-regulated NF-κB activity in cancer Because NF-κB regulates so many different genes, it can act as either a tumor promoter or a tumor suppressor, depending on the cell type and stimulus involved (Perkins and Gilmore, 2006).  NF-κB's oncogenic effects derive from its roles in promoting cell growth and survival.  For example, NF-κB can bind to and activate the cyclin D1 promoter, thus driving cell cycle progression (Joyce et al., 2001).  NF-κB can also directly inhibit apoptosis by inducing the transcription of anti-apoptotic genes Bcl-2 (Catz and Johnson, 2001) and Bcl-XL (Chen et al., 2000).  In addition, NF-κB has been implicated in tumor invasion and angiogenesis because it can upregulate the transcription of matrix metalloproteinases (MMPs) (Bond et al., 1998) as well as pro-angiogenic genes like interleukin-8 (IL-8) (Huang et al., 2000a) and vascular endothelial growth factor (VEGF) (Levine et al., 2003).  Constitutive activation of NF-κB complexes has been observed in cancers of the breast (Sovak et al., 1997), prostate (Sweeney et al., 2004), colon (Kojima et al., 2004), pancreas (Wang et al., 1999), and ovary (Huang et al., 2000c), among others.  In melanoma, p50 and p65 proteins are reportedly overexpressed in the nuclei of dysplastic nevi and melanoma cells compared to those of normal nevi and healthy melanocytes, respectively (McNulty et al., 2004; McNulty et al., 2001).  Elevated nuclear p50/p65 levels in melanoma have been attributed to increased IκB expression and/or activity (Shattuck-Brandt and Richmond, 1997; Yang and Richmond, 2001) or strong expression of NF-κB-inducing kinase (NIK) (Dhawan and Richmond., 2002).  Constitutive NF-κB activation has been shown to promote tumorigenicity, angiogenesis, and metastasis of  25 melanoma cells in nude mice, mediated by IL-8 expression (Huang et al., 2000a; Huang et al., 2000b). 1.6.3 Interleukin-6 in cancer Interleukin-6 (IL-6) is a pleiotropic cytokine most notably involved in immune responses and inflammation and, as such, is a transcriptional target of NF-κB (Sehgal et al., 1995). Deregulation of IL-6 impacts multiple physiological processes and is therefore associated with a variety of diseases including arthritis, lupus, Crohn's disease, inflammatory bowel disease, and cardiovascular disease, among others (Hong et al., 2007).  IL-6 also appears to play a role in cancer.  For example, elevated IL-6 serum levels correlate with poor prognosis in multiple myeloma (Ludwig et al., 1991), lymphoma (Seymour et al., 1995), ovarian cancer (Plante et al., 1994), prostate cancer (Twillie et al., 1995), and renal cell carcinoma (Blay et al., 1992).  In melanoma, the role of IL-6 appears to depend upon the stage of progression.  The growth of early-stage "metastatically incompetent" melanoma cell lines is inhibited by IL-6 and they undergo G0/G1 cell cycle arrest upon IL-6 transfection (Lu and Kerbel, 1993; Sun et al., 1992).  Paradoxically, metastatic melanoma cell lines are resistant to IL-6-mediated growth inhibition and may even secrete high levels of IL-6 (Hong et al., 2007; Komyod et al., 2007).  Furthermore, it has been shown that IL-6 enhances the metastatic potential of advanced-stage melanoma cells (De Galdeano et al., 1998).  These studies suggest that IL-6 may switch from a paracrine growth inhibitor to an autocrine growth stimulator over the course of melanoma tumor progression (Lu and Kerbel, 1993).    26 1.7 HYPOTHESIS AND OBJECTIVES In the first part of this study, we sought to determine whether expression of the BH3-only protein PUMA is relevant in melanoma progression.  As previously mentioned, overexpression of anti-apoptotic Bcl-2 family proteins has been implicated in tumorigenesis (Coultas and Strasser, 2003).  High levels of Bcl-2, Bcl-XL, and Mcl-1 expression are frequently observed in human melanoma tissues and are thought to contribute to chemotherapeutic resistance by hampering apoptosis (Tang et al., 1998; Selzer et al., 1998).  We hypothesized that the dysregulation of pro-apoptotic proteins may be of equal or greater importance in maintaining the strong apoptosis "barrier" in melanoma cells.  We chose to focus on the novel BH3-only protein PUMA because of its potent cell-killing abilities and "p53-like" tumor suppressing effects.  Our two main objectives were: 1) to examine whether PUMA expression levels affect melanoma tumor progression, and 2) to determine whether manipulation of PUMA expression can be exploited to induce apoptosis of melanoma cells.  In the second part of this work, we considered the idea that melanoma tumors rely upon multiple pathways to sustain a survival advantage.  Our lab has previously shown that strong Akt activation correlates with melanoma progression and survival (Dai et al., 2005).  We therefore hypothesized that melanoma cells simultaneously suppress apoptosis and promote survival through PUMA down-regulation and Akt hyperactivation, respectively.  To test this hypothesis, we defined the following objectives: 1) to identify patterns of PUMA and p-Akt expression in melanoma tissues and determine whether they cooperatively influence patient survival rates, and 2) to  27 determine whether combined therapeutic targeting of both pathways can enhance melanoma tumor suppression, possibly in a synergistic manner.  Lastly, we sought to examine the role of NF-κB p50 in melanoma development. Our lab previously reported that expression of the p50 subunit in particular correlated with tumor progression and is associated with poor 5-year survival of melanoma patients (Gao et al., 2006).  Based on these observations, we hypothesized that p50 must be regulating some mediator of cell growth, survival, or angiogenesis.  In the present study we sought to determine how p50 expression confers a growth or survival advantage to melanoma cells.  Our objectives were: 1) to identify target genes activated/upregulated by p50 overexpression, and 2) to determine how p50-induced changes in cell behavior can drive melanoma tumor progression.   28 1.8 REFERENCES Abbasi NR, Shaw HM, Rigel DS, Friedman RJ, McCarthy WH, Osman I, Kopf AW, Polsky D (2004). Early diagnosis of cutaneous melanoma: revisiting the ABCD criteria. JAMA 292: 2771-6.  Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, Palmiter RD, Brinster RL (1985). The c-myc oncogene driven by immunoglobulin enhancers induces lymphoid malignancy in transgenic mice. Nature 318: 533-8.  Agar N, Young AR (2005). Melanogenesis: a photoprotective response to DNA damage? Mutat Res 571: 121-32.  Ahn CH, Jeong EG, Kim SS, Lee JW, Lee SH, Kim SH, Kim MS, Yoo NJ, Lee SH (2008). Expressional and mutational analysis of pro-apoptotic Bcl-2 member PUMA in hepatocellular carcinomas. Dig Dis Sci 53: 1395-9.  Alanko T, Rosenberg M, Saksela O (1999). FGF expression allows nevus cells to survive in three-dimensional collagen gel under conditions that induce apoptosis in normal human melanocytes. J Invest Dermatol 113: 111-6.  An HJ, Logani S, Isacson C, Ellenson LH (2004). Molecular characterization of uterine clear cell carcinoma. Mod Pathol 17: 530-7.  Argenziano G, Soyer HP, Chimenti S, Talamini R, Corona R, Sera F, Binder M, Cerroni L, De Rosa G, Ferrara G, Hofmann-Wellenhof R, Landthaler M, Menzies SW, Pehamberger H, Piccolo D, Rabinovitz HS, Schiffner R, Staibano S, Stolz W, Bartenjev I, Blum A, Braun R, Cabo H, Carli P, De Giorgi V, Fleming MG, Grichnik JM, Grin CM, Halpern AC, Johr R, Katz B, Kenet RO, Kittler H, Kreusch J, Malvehy J, Mazzocchetti G, Oliviero M, Ozdemir F, Peris K, Perotti R, Perusquia A, Pizzichetta MA, Puig S, Rao B, Rubegni P, Saida T, Scalvenzi M, Seidenari S, Stanganelli I, Tanaka M, Westerhoff K, Wolf IH, Braun-Falco O, Kerl H, Nishikawa T, Wolff K, Kopf AW (2003). Dermoscopy of pigmented skin lesions: results of a consensus meeting via the Internet. J Am Acad Dermatol 48: 679-93.  Balch CM, Buzaid AC, Soong SJ, Atkins MB, Cascinelli N, Coit DG, Fleming ID, Gershenwald JE, Houghton A Jr, Kirkwood JM, McMasters KM, Mihm MF, Morton DL, Reintgen DS, Ross MI, Sober A, Thompson JA, Thompson JF (2001). Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 19: 3635-48.  Bellacosa A, de Feo D, Godwin AK, Bell DW, Cheng JQ, Altomare DA, Wan M, Dubeau L, Scambia G, Masciullo V, Ferrandina G, Benedetti Panici P, Mancuso S, Neri G, Testa JR (1995). Molecular alterations of the AKT2 oncogene in ovarian and breast carcinomas. Int J Cancer 64: 280-5.   29 Bevona C, Goggins W, Quinn T, Fullerton J, Tsao H (2003). Cutaneous melanomas associated with nevi. Arch Dermatol 139: 1620-4.  Bishop DT, Demenais F, Goldstein AM, Bergman W, Bishop JN, Bressac-de Paillerets B, Chompret A, Ghiorzo P, Gruis N, Hansson J, Harland M, Hayward N, Holland EA, Mann GJ, Mantelli M, Nancarrow D, Platz A, Tucker MA; Melanoma Genetics Consortium (2002). Geographical variation in the penetrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 94: 894-903.  Blay JS, Negrier S, Combaret V, Attali S, Goillot E, Merrouche Y, Mercatello A, Ravault A, Tourani JM, Moskovtchenko JF, Philip T, Farvot M (1992). Serum level of interleukin 6 as a prognosis factor in metastatic renal cell carcinoma. Cancer Res 52: 3317-22.  Bond M, Fabunmi RP, Baker AH, Newby AC (1998). Synergistic upregulation of metalloproteinase-9 by growth factors and inflammatory cytokines: an absolute requirement for transcription factor NF-kappa B. FEBS Lett 435: 29-34.  Breslow A (1970). Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 172: 902-8.  Burgering BM, Coffer PJ (1995). Protein kinase B (c-Akt) in phosphatidylinositol-3-OH kinase signal transduction. Nature 376: 599-602.  Burgeson RE, Christiano AM (1997). The dermal-epidermal junction. Curr Opin Cell Biol 9: 651-8.  Byun DS, Cho K, Ryu BK, Lee MG, Park JI, Chae KS, Kim HJ, Chi SG (2003). Frequent monoallelic deletion of PTEN and its reciprocal associatioin with PIK3CA amplification in gastric carcinoma. Int J Cancer 104: 318-27.  Canadian Cancer Society/National Cancer Institute of Canada: Canadian Cancer Statistics 2008, Toronto, Canada, 2008.  Catz SD, Johnson JL (2001). Transcriptional regulation of bcl-2 by nuclear factor kappa B and its significance in prostate cancer. Oncogene 20: 7342-51.  Chapman PB, Einhorn LH, Meyers ML, Saxman S, Destro AN, Panageas KS, Begg CB, Agarwala SS, Schuchter LM, Ernstoff MS, Houghton AN, Kirkwood JM (1999). Phase III multicenter randomized trial of the Dartmouth regimen versus dacarbazine in patients with metastatic melanoma. J Clin Oncol 17: 2745-51.  Chen C, Edelstein LC, Gélinas C (2000). The Rel/NF-kappaB family directly activates expression of the apoptosis inhibitor Bcl-x(L). Mol Cell Biol 20: 2687-95.   30 Chen L, Willis SN, Wei A, Smith BJ, Fletcher JI, Hinds MG, Colman PM, Day CL, Adams JM, Huang DC (2005). Differential targeting of prosurvival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function. Mol Cell 17: 393-403.  Cheng EH, Wei MC, Weiler S, Flavell RA, Mak TW, Lindsten T, Korsmeyer SJ (2001). BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK- mediated mitochondrial apoptosis. Mol Cell 8: 705-11.  Cheng JQ, Ruggeri B, Klein WM, Sonoda G, Altomare DA, Watson DK, Testa JR (1996). Amplification of AKT2 in human pancreatic cells and inhibition of AKT2 expression and tumorigenicity by antisense RNA. Proc Natl Acad Sci U S A 93: 3636-41.  Chiang GG, Abraham RT (2007). Targeting the mTOR signaling network in cancer. Trends Mol Med 13: 433-42.  Cory S, Adams JM. (2002). The Bcl2 family: regulators of the cellular life-or-death switch. Nat Rev Cancer 2: 647-56.  Coultas L, Strasser A (2003). The role of the Bcl-2 protein family in cancer. Semin Cancer Biol 13: 115-23.  Dai DL, Martinka M, Li G (2005). Prognostic significance of activated Akt expression in melanoma: a clinicopathologic study of 292 cases. J Clin Oncol 23: 1473-82.  Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y, Greenberg ME (1997). Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91: 231-41.  De Galdeano G, Boyano D, Smith-Zubiaga I, Alvarez A, Cantón I, Cañavate L (1998). Involvement of interleukin-6 in the biology and metastatic activity of B16F10 melanoma cells. Eur Cytokine Netw 9: 187-92.  Debbas M, White E (1993). Wild-type p53 mediates apoptosis by E1A, which is inhibited by E1B. Genes Dev 7: 546-54.  Dhawan P, Richmond A (2002). A novel NF-kappa B-inducing kinase-MAPK signaling pathway up-regulates NF-kappa B activity in melanoma cells. J Biol Chem 277: 7920-8.  Eckert RL, Rorke EA (1989). Molecular biology of keratinocyte differentiation. Environ Health Perspect 80: 109-16.  Eggermont AM, Kirkwood JM (2004). Re-evaluating the role of dacarbazine in metastatic melanoma: what have we learned in 30 years? Eur J Cancer 40: 1825-36.   31 Ermoian RP, Furniss CS, Lamborn KR, Basila D, Berger MS, Gottschalk AR, Nicholas MK, Stokoe D, Haas-Kogan DA (2002). Dysregulation of PTEN and protein kinase B is associated with glioma histology and patient survival. Clin Cancer Res 8: 1100-6.  Espey DK, Wu XC, Swan J, Wiggins C, Jim MA, Ward E, Wingo PA, Howe HL, Ries LA, Miller BA, Jemal A, Ahmed F, Cobb N, Kaur JS, Edwards BK (2007). Annual report to the nation on the status of cancer, 1975-2004, featuring cancer in American Indians and Alaska Natives. Cancer 110: 2119-52.  Evans RD, Kopf AW, Lew RA, Rigel DS, Bart RS, Friedman RJ, Rivers JK (1988). Risk factors for the development of malignant melanoma--I: Review of case-control studies. J Dermatol Surg Oncol 14: 393-408.  Fisher DE (1994). Apoptosis in cancer therapy: crossing the threshold. Cell 78: 539-42.  Franke TF, Yang SI, Chan TO, Datta K, Kazlauskas A, Morrison DK, Kaplan DR, Tsichlis PN (1995). The protein kinase encoded by the Akt proto-oncogene is a target of the PDGF-activated phosphatidylinositol 3-kinase. Cell 81: 727-36.  Fujita E, Jinbo A, Matuzaki H, Konishi H, Kikkawa U, Momoi T (1999). Akt phosphorylation site found in human caspase-9 is absent in mouse caspase-9. Biochem Biophys Res Commun 264: 550-5.  Gao K, Dai DL, Martinka M, Li G (2006). Prognostic significance of nuclear factor- kappaB p105/p50 in human melanoma and its role in cell migration. Cancer Res 66: 8382-8.  Garcia JM, Silva JM, Dominguez G, Gonzalez R, Navarro A, Carretero L, Provencio M, España P, Bonilla F (1999). Allelic loss of the PTEN region (10q23) in breast carcinomas of poor pathophenotype. Breast Cancer Res Treat 57: 237-43.  Garrison SP, Jeffers JR, Yang C, Nilsson JA, Hall MA, Rehg JE, Yue W, Yu J, Zhang L, Onciu M, Sample JT, Cleveland JL, Zambetti GP (2008). Selection Against PUMA Gene Expression in Myc-Driven B Cell Lymphomagenesis. Mol Cell Biol 28: 5391-402.  Ghosh S, Karin M (2002). Missing pieces in the NF-kappaB puzzle. Cell 109 Suppl: S81-96.  Gilchrest BA, Eller MS, Geller AC, Yaar M (1999). The pathogenesis of melanoma induced by ultraviolet radiation. N Engl J Med 340: 1341-8.  Gilmore TD (1990). NF-kappa B, KBF1, dorsal, and related matters. Cell 62: 841-3.  Gilmore TD (2006). Introduction to NF-kappaB: players, pathways, perspectives. Oncogene 25: 6680-4.   32 Goel A, Arnold CN, Niedzwiecki D, Carethers JM, Dowell JM, Wasserman L, Compton C, Mayer RJ, Bertagnolli MM, Boland CR (2004). Frequent inactivation of PTEN by promoter hypermethylation in microsatellite instability-high sporadic colorectal cancers. Cancer Res 64: 3014-21.  Goldstein AM, Tucker MA (2001). Genetic epidemiology of cutaneous melanoma: a global perspective. Arch Dermatol 137: 1493-6.  Goldstein AM, Tucker MA, Crutcher WA, Hartge P, Sagebiel RW (1993). The inheritance pattern of dysplastic naevi in families of dysplastic naevus patients. Melanoma Res 3: 15-22.  Gray-Schopfer V, Wellbrock C, Marais R (2007). Melanoma biology and new targeted therapy. Nature 445: 851-7.  Green DR, Kroemer G (2004). The pathophysiology of mitochondrial cell death. Science 305: 626-9.  Han J, Flemington C, Houghton AB, Gu Z, Zambetti GP, Lutz RJ, Zhu L, Chittenden T (2001). Expression of bbc3, a pro-apoptotic BH3-only gene, is regulated by diverse cell death and survival signals. Proc Natl Acad Sci U S A 98: 11318-23.  Harlan JE, Yoon HS, Hajduk PJ, Fesik SW (1995). Structural characterization of the interaction between a pleckstrin homology domain and phosphatidylinositol 4,5- bisphosphate. Biochemistry 34: 9859-64.  Hemann MT, Fridman JS, Zilfou JT, Hernando E, Paddison PJ, Cordon-Cardo C, Hannon GJ, Lowe SW (2003). An epi-allelic series of p53 hypomorphs created by stable RNAi produces distinct tumor phenotypes in vivo. Nat Genet 33: 396-400.  Hemann MT, Zilfou JT, Zhao Z, Burgess DJ, Hannon GJ, Lowe SW (2004). Suppression of tumorigenesis by the p53 target PUMA. Proc Natl Acad Sci U S A 101: 9333-8.  Holly EA, Kelly JW, Shpall SN, Chiu SH (1987). Number of melanocytic nevi as a major risk factor for malignant melanoma. J Am Acad Dermatol 17: 459-68.  Hong DS, Angelo LS, Kurzrock R (2007). Interleukin-6 and its receptor in cancer: implications for Translational Therapeutics. Cancer 110: 1911-28.  Hoque MO, Begum S, Sommer M, Lee T, Trink B, Ratovitski E, Sidransky D (2003). PUMA in head and neck cancer. Cancer Lett 199: 75-81.  Huang S, DeGuzman A, Bucana CD, Fidler IJ (2000a). Level of interleukin-8 expression by metastatic human melanoma cells directly correlates with constitutive NF-kappaB activity. Cytokines Cell Mol Ther 6: 9-17.   33 Huang S, DeGuzman A, Bucana CD, Fidler IJ (2000b). Nuclear factor-kappaB activity correlates with growth, angiogenesis, and metastasis of human melanoma cells in nude mice. Clin Cancer Res 6: 2573-81.  Huang S, Robinson JB, Deguzman A, Bucana CD, Fidler IJ (2000c). Blockade of nuclear factor-kappaB signaling inhibits angiogenesis and tumorigenicity of human ovarian cancer cells by suppressing expression of vascular endothelial growth factor and interleukin 8. Cancer Res 60: 5334-9.  Hussein MR (2005). Melanocytic dysplastic naevi occupy the middle ground between benign melanocytic naevi and cutaneous malignant melanomas: emerging clues. J Clin Pathol 58: 453-6.  Hussein MR (2008). Extracutaneous malignant melanomas. Cancer Invest 26: 516-34.  Hussein MR, Haemel AK, Wood GS. (2003). Apoptosis and melanoma: molecular mechanisms. J Pathol 199: 275-88.  Ito H, Kanzawa T, Miyoshi T, Hirohata S, Kyo S, Iwamaru A, Aoki H, Kondo Y, Kondo S (2005). Therapeutic efficacy of PUMA for malignant glioma cells regardless of p53 status. Hum Gene Ther 16: 685-98.  Ittmann MM (1998). Chromosome 10 alterations in prostate adenocarcinoma (review). Oncol Rep 5: 1329-35.  Jeffers JR, Parganas E, Lee Y, Yang C, Wang J, Brennan J, MacLean KH, Han J, Chittenden T, Ihle JN, McKinnon PJ, Cleveland JL, Zambetti GP (2003). Puma is an essential mediator of p53-dependent and -independent apoptotic pathways. Cancer Cell 4: 321-8.  Jemal A, Devesa SS, Hartge P, Tucker MA (2001). Recent trends in cutaneous melanoma incidence among whites in the United States. J Natl Cancer Inst 93: 678-83.  Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ. (2008). Cancer statistics, 2008. CA Cancer J Clin 58: 71-96.  Jhappan C, Noonan FP, Merlino G (2003). Ultraviolet radiation and cutaneous malignant melanoma. Oncogene 22: 3099-112.  Johns MM 3rd, Westra WH, Califano JA, Eisele D, Koch WM, Sidransky D (1996). Allelotype of salivary gland tumors. Cancer Res 56: 1151-4.  Joyce D, Albanese C, Steer J, Fu M, Bouzahzah B, Pestell RG (2001). NF-kappaB and cell-cycle regulation: the cyclin connection. Cytokine Growth Factor Rev 12: 73-90.   34 Kalaaji AN (2007). Cytokine therapy in advanced melanoma. J Drugs Dermatol 6: 374- 8.  Karst AM, Dai DL, Martinka M, Li G (2005). PUMA expression is significantly reduced in human cutaneous melanomas. Oncogene 24: 1111-6.  Kessis TD, Slebos RJ, Nelson WG, Kastan MB, Plunkett BS, Han SM, Lorincz AT, Hedrick L, Cho KR (1993). Human papillomavirus 16 E6 expression disrupts the p53- mediated cellular response to DNA damage. Proc Natl Acad Sci U S A 90: 3988-92.  Knobbe CB, Reifenberger G (2003). Genetic alterations and aberrant expression of genes related to the phosphatidyl-inositol-3'-kinase/protein kinase B (Akt) signal transduction pathway in glioblastomas. Brain Pathol 13: 507-18.  Kojima M, Morisaki T, Sasaki N, Nakano K, Mibu R, Tanaka M, Katano M (2004). Increased nuclear factor-kB activation in human colorectal carcinoma and its correlation with tumor progression. Anticancer Res 24: 675-81.  Komyod W, Böhm M, Metze D, Heinrich PC, Behrmann I (2007). Constitutive suppressor of cytokine signaling 3 expression confers a growth advantage to a human melanoma cell line. Mol Cancer Res 5: 271-81.  Kraemer KH, Greene MH (1985). Dysplastic nevus syndrome. Familial and sporadic precursors of cutaneous melanoma. Dermatol Clin 3: 225-37.  Lee JI, Soria JC, Hassan KA, El-Naggar AK, Tang X, Liu DD, Hong WK, Mao L (2001). Loss of PTEN expression as a prognostic marker for tongue cancer. Arch Otolaryngol Head Neck Surg 127: 1441-5.  Leiter U, Garbe C (2008). Epidemiology of melanoma and nonmelanoma skin cancer-- the role of sunlight. Adv Exp Med Biol 624: 89-103.  Lens MB, Dawes M (2004). Global perspectives of contemporary epidemiological trends of cutaneous malignant melanoma. Br J Dermatol 150: 179-85.  Letai A, Bassik MC, Walensky LD, Sorcinelli MD, Weiler S, Korsmeyer SJ (2002). Distinct BH3 domains either sensitize or activate mitochondrial apoptosis, serving as prototype cancer therapeutics. Cancer Cell 2: 183-92.  Levine L, Lucci JA 3rd, Pazdrak B, Cheng JZ, Guo YS, Townsend CM Jr, Hellmich MR (2003). Bombesin stimulates nuclear factor kappa B activation and expression of proangiogenic factors in prostate cancer cells. Cancer Res 63: 3495-502.  Li G, Tang L, Zhou X, Tron V, Ho V (1998). Chemotherapy-induced apoptosis in melanoma cells is p53 dependent. Melanoma Res 8: 17-23.   35 Liang J, Zubovitz J, Petrocelli T, Kotchetkov R, Connor MK, Han K, Lee JH, Ciarallo S, Catzavelos C, Beniston R, Franssen E, Slingerland JM (2002). PKB/Akt phosphorylates p27, impairs nuclear import of p27 and opposes p27-mediated G1 arrest. Nat Med 8: 1153-60.  Lin JY, Fisher DE (2007). Melanocyte biology and skin pigmentation. Nature 445: 843- 50.  Liu FT, Newland AC, Jia L (2003). Bax conformational change is a crucial step for PUMA-mediated apoptosis in human leukemia. Biochem Biophys Res Commun 310: 956- 62.  Lowe SW, Jacks T, Housman DE, Ruley HE (1994). Abrogation of oncogene-associated apoptosis allows transformation of p53-deficient cells. Proc Natl Acad Sci U S A 91: 2026-30.  Lowe SW, Ruley HE, Jacks T, Housman DE (1993). p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell 74: 957-67.  Lu C, Kerbel RS (1993). Interleukin-6 undergoes transition from paracrine growth inhibitor to autocrine stimulator during human melanoma progression. J Cell Biol 120: 1281-8.  Ludwig H, Nachbaur DM, Fritz E, Krainer M, Huber H (1991). Interleukin-6 is a prognostic factor in multiple myeloma. Blood 77: 2794-5.  Ma YY, Wei SJ, Lin YC, Lung JC, Chang TC, Whang-Peng J, Liu JM, Yang DM, Yang WK, Shen CY (2000). PIK3CA as an oncogene in cervical cancer. Oncogene 19: 2739- 44.  Mayo LD, Donner DB (2001). A phosphatidylinositol 3-kinase/Akt pathway promotes translocation of Mdm2 from the cytoplasm to the nucleus. Proc Am Soc Clin Oncol 98: 11598-603.  McNulty SE, del Rosario R, Cen D, Meyskens FL Jr, Yang S (2004). Comparative expression of NFkappaB proteins in melanocytes of normal skin vs. benign intradermal naevus and human metastatic melanoma biopsies. Pigment Cell Res 17: 173-80.  McNulty SE, Tohidian NB, Meyskens FL Jr (2001). RelA, p50 and inhibitor of kappa B alpha are elevated in human metastatic melanoma cells and respond aberrantly to ultraviolet light B. Pigment Cell Res 14: 456-65.  Miura I, Siegfried JM, Resau J, Keller SM, Zhou JY, Testa JR (1990). Chromosome alterations in 21 non-small cell lung carcinomas. Genes Chromosomes Cancer 2: 328-38.   36 Mora J, Cheung NK, Chen L, Qin J, Gerald W (2001). Loss of heterozygosity at 19q13.3 is associated with locally aggressive neuroblastoma. Clin Cancer Res 7: 1358-61.  Mutirangura A, Tanunyutthawongese C, Pornthanakasem W, Kerekhanjanarong V, Sriuranpong V, Yenrudi S, Supiyaphun P, Voravud N (1997). Genomic alterations in nasopharyngeal carcinoma: loss of heterozygosity and Epstein-Barr virus infection. Br J Cancer 76: 770-6.  Nakano K, Vousden KH (2001). PUMA, a novel proapoptotic gene, is induced by p53. Mol Cell 7: 683-94.  Naldi L, Lorenzo Imberti G, Parazzini F, Gallus S, La Vecchia C (2000). Pigmentary traits, modalities of sun reaction, history of sunburns, and melanocytic nevi as risk factors for cutaneous malignant melanoma in the Italian population: results of a collaborative case-control study. Cancer 88: 703-10.  Nam SY, Lee HS, Jung GA, Choi J, Cho SJ, Kim MK, Kim WH, Lee BL (2003). Akt/PKB activation in gastric carcinomas correlates with clinicopathologic variables and prognosis. APMIS 111: 1105-13.  Nawroz H, van der Riet P, Hruban RH, Koch W, Ruppert JM, Sidransky D (1994). Allelotype of head and neck squamous cell carcinoma. Cancer Res 54: 1152-5.  Nicholson KM, Anderson NG (2002). The protein kinase B/Akt signalling pathway in human malignancy. Cell Signal 14: 381-95.  Osaki M, Oshimura M, Ito H (2004). PI3K-Akt pathway: its functions and alterations in human cancer. Apoptosis 9: 667-76.  Pérez-Tenorio G, Stål O; Southeast Sweden Breast Cancer Group (2002). Activation of AKT/PKB in breast cancer predicts a worse outcome among endocrine treated patients. Br J Cancer 86: 540-5.  Perkins ND, Gilmore TD (2006). Good cop, bad cop: the different faces of NF-kappaB. Cell Death Differ 13: 759-72.  Plante M, Rubin SC, Wong GY, Federici MG, Finstad CL, Gastl GA (1994). Interleukin- 6 level in serum and ascites as a prognostic factor in patients with epithelial ovarian cancer. Cancer 73: 1882-8.  Rampino N, Yamamoto H, Ionov Y, Li Y, Sawai H, Reed JC, Perucho M (1997). Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science 275: 967-9.  Rasheed BK, Wiltshire RN, Bigner SH, Bigner DD (1999). Molecular pathogenesis of malignant gliomas. Curr Opin Oncol 11: 162-7.  37  Reagan-Shaw S, Ahmad N (2007). The role of Forkhead-box Class O (FoxO) transcription factors in cancer: a target for the management of cancer. Toxicol Appl Pharmacol 224: 360-8.  Robertson GP (2005). Functional and therapeutic significance of Akt deregulation in malignant melanoma. Cancer Metastasis Rev 24: 273-85.  Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, Yan H, Gazdar A, Powell SM, Riggins GJ, Willson JK, Markowitz S, Kinzler KW, Vogelstein B, Velculescu VE (2004). High frequency of mutations of the PIK3CA gene in human cancers. Science 304: 554.  Sanchez-Cespedes M, Ahrendt SA, Piantadosi S, Rosell R, Monzo M, Wu L, Westra WH, Yang SC, Jen J, Sidransky D (2001). Chromosomal alterations in lung adenocarcinoma from smokers and nonsmokers. Cancer Res 61: 1309-13.  Schaffer JV (2007). Pigmented lesions in children: when to worry. Curr Opin Pediatr 19: 430-40.  Scheidereit C (2006). IkappaB kinase complexes: gateways to NF-kappaB activation and transcription. Oncogene 25: 6685-705.  Schmitt CA, McCurrach ME, de Stanchina E, Wallace-Brodeur RR, Lowe SW (1999). INK4a/ARF mutations accelerate lymphomagenesis and promote chemoresistance by disabling p53. Genes Dev 13: 2670-7.  Sehgal PB, Wang L, Rayanade R, Pan H, Margulies L (1995). Interleukin-6-type cytokines. Ann N Y Acad Sci 762: 1-14.  Selzer E, Schlagbauer-Wadl H, Okamoto I, Pehamberger H, Pötter R, Jansen B (1998). Expression of Bcl-2 family members in human melanocytes, in melanoma metastases and in melanoma cell lines. Melanoma Res 8: 197-203.  Seymour JF, Talpaz M, Cabanillas F, Wetzler M, Kurzrock R (1995). Serum interleukin- 6 levels correlate with prognosis in diffuse large-cell lymphoma. J Clin Oncol 13: 575- 82.  Shattuck-Brandt RL, Richmond A (1997). Enhanced Degradation of I-kappaB alpha Contributes to Endogenous Activation of NF-kB in Hs294T Melanoma Cells. Cancer Res 57: 3032-9.  Shayesteh L, Lu Y, Kuo WL, Baldocchi R, Godfrey T, Collins C, Pinkel D, Powell B, Mills GB, Gray JW (1999). PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet 21: 99-102.   38 Shibata MA, Liu ML, Knudson MC, Shibata E, Yoshidome K, Bandey T, Korsmeyer SJ, Green JE (1999). Haploid loss of bax leads to accelerated mammary tumor development in C3(1)/SV40-TAg transgenic mice: reduction in protective apoptotic response at the preneoplastic stage. EMBO J 18: 2692-701.  Shibue T, Taniguchi T (2006). BH3-only proteins: integrated control point of apoptosis. Int J Cancer 119: 2036-43.  Shimazaki C, Inaba T, Nakagawa M (2000). B-cell lymphoma-associated hemophagocytic syndrome. Leuk Lymphoma 38: 121-30.  Sovak MA, Bellas RE, Kim DW, Zanieski GJ, Rogers AE, Traish AM, Sonenshein GE (1997). Aberrant nuclear factor-kappaB/Rel expression and the pathogenesis of breast cancer. J Clin Invest 100: 2952-60.  Staal SP (1987). Molecular cloning of the akt oncogene and its human homologues AKT1 and AKT2: amplification of AKT1 in a primary human gastric adenocarcinoma. Proc Natl Acad Sci U S A 84: 5034-7.  Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP (2003). Loss of PTEN promotes tumor development in malignant melanoma. Cancer Res 63: 2881-90.  Strasser A (2005). The role of BH3-only proteins in the immune system. Nat Rev Immunol 5: 189-200.  Sun WH, Kreisle RA, Phillips AW, Ershler WB (1992). In vivo and in vitro characteristics of interleukin 6-transfected B16 melanoma cells. Cancer Res 52: 5412-5.  Sweeney C, Li L, Shanmugam R, Bhat-Nakshatri P, Jayaprakasan V, Baldridge LA, Gardner T, Smith M, Nakshatri H, Cheng L (2004). Nuclear factor-kappaB is constitutively activated in prostate cancer in vitro and is overexpressed in prostatic intraepithelial neoplasia and adenocarcinoma of the prostate. Clin Cancer Res 10: 5501- 7.  Tang L, Tron VA, Reed JC, Mah KJ, Krajewska M, Li G, Zhou X, Ho VC, Trotter MJ (1998). Expression of apoptosis regulators in cutaneous malignant melanoma. Clin Cancer Res 4: 1865-71.  Takeda K, Takahashi NH, Shibahara S (2007). Neuroendocrine functions of melanocytes: beyond the skin-deep melanin maker. Tohoku J Exp Med 211: 201-21.  Titus-Ernstoff L, Perry AE, Spencer SK, Gibson JJ, Cole BF, Ernstoff MS (2005). Pigmentary characteristics and moles in relation to melanoma risk. Int J Cancer 116: 144-9.   39 Twillie DA, Eisenberger MA, Carducci MA, Hseih WS, Kim WY, Simons JW (1995). Interleukin-6: a candidate mediator of human prostate cancer morbidity. Urology 45: 542-29.  van Weeren PC, de Bruyn KM, de Vries-Smits AM, van Lint J, Burgering BM (1998). Essential role for protein kinase B (PKB) in insulin-induced glycogen synthase kinase 3 inactivation. Characterization of dominant-negative mutant of PKB. J Biol Chem 273: 13150-6.  Villunger A, Michalak EM, Coultas L, Müllauer F, Böck G, Ausserlechner MJ, Adams JM, Strasser A (2003). p53- and drug-induced apoptotic responses mediated by BH3-only proteins puma and noxa. Science 302: 1036-8.  Wang DS, Rieger-Christ K, Latini JM, Moinzadeh A, Stoffel J, Pezza JA, Saini K, Libertino JA, Summerhayes IC (2000). Molecular analysis of PTEN and MXI1 in primary bladder carcinoma. Int J Cancer 88: 620-5.  Wang W, Abbruzzese JL, Evans DB, Larry L, Cleary KR, Chiao PJ (1999). The nuclear factor-kappa B RelA transcription factor is constitutively activated in human pancreatic adenocarcinoma cells. Clin Cancer Res 5: 119-27.  Whiteman DC, Whiteman CA, Green AC (2001). Childhood sun exposure as a risk factor for melanoma: a systematic review of epidemiologic studies. Cancer Causes Control 12: 69-82.  Willis SN, Adams JM (2005). Life in the balance: how BH3-only proteins induce apoptosis. Curr Opin Cell Biol 17: 617-25.  Willis SN, Chen L, Dewson G, Wei A, Naik E, Fletcher JI, Adams JM, Huang DC (2005). Proapoptotic Bak is sequestered by Mcl-1 and Bcl-xL, but not Bcl-2, until displaced by BH3-only proteins. Genes Dev 19: 1294-305.  Woenckhaus J, Steger K, Werner E, Fenic I, Gamerdinger U, Dreyer T, Stahl U (2002). Genomic gain of PIK3CA and increased expression of p110alpha are associated with progression of dysplasia into invasive squamous cell carcinoma. J Pathol 198: 335-42.  Wong MP, Lam WK, Wang E, Chiu SW, Lam CL, Chung LP (2002). Primary adenocarcinomas of the lung in nonsmokers show a distinct pattern of allelic imbalance. Cancer Res 62: 4464-8.  Yamaguchi Y, Brenner, M., Hearing VJ (2007). The regulation of skin pigmentation. J Biol Chem 282: 27557-61.  Yamamoto S, Tomita Y, Hoshida Y, Morooka T, Nagano H, Dono K, Umeshita K, Sakon M, Ishikawa O, Ohigashi H, Nakamori S, Monden M, Aozasa K (2004). Prognostic  40 significance of activated Akt expression in pancreatic ductal adenocarcinoma. Clin Cancer Res 10: 2846-50.  Yang J, Richmond A (2001). Constitutive IkappaB kinase activity correlates with nuclear factor-kappaB activation in human melanoma cells. Cancer Res 61: 4901-9.  Yin C, Knudson CM, Korsmeyer SJ, Van Dyke T (1997). Bax suppresses tumorigenesis and stimulates apoptosis in vivo. Nature 385: 637-40.  Yong WH, Ueki K, Chou D, Reeves SA, von Deimling A, Gusella JF, Mohrenweiser HW, Buckler AJ, Louis DN (1995). Cloning of a highly conserved human protein serine- threonine phosphatase gene from the glioma candidate region on chromosome 19q13.3. Genomics 29: 533-6.  Yoo NJ, Lee JW, Jeong EG, Lee SH (2007). Immunohistochemical analysis of pro- apoptotic PUMA protein and mutational analysis of PUMA gene in gastric carcinomas. Dig Liver Dis 39: 222-7.  You H, Pellegrini M, Tsuchihara K, Yamamoto K, Hacker G, Erlacher M, Villunger A, Mak TW (2006). FOXO3a-dependent regulation of Puma in response to cytokine/growth factor withdrawal. J Exp Med 203: 1657-63.  Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B (2001). PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell 7: 673-82.  Zong WX, Lindsten T, Ross AJ, MacGregor GR, Thompson CB (2001). BH3-only proteins that bind pro-survival Bcl-2 family members fail to induce apoptosis in the absence of Bax and Bak. Genes Dev 15: 1481-6.     41 2.   PUMA EXPRESSION IS SIGNIFICANTLY REDUCED IN HUMAN CUTANEOUS MELANOMAS1 2.1 INTRODUCTION Malignant melanoma is an aggressive and lethal form of skin cancer, the incidence of which is rising rapidly among Caucasians populations (Jemal et al., 2008).  In the United States, lifetime risk of melanoma has reached an all-time high, with 62,480 new cases expected to be diagnosed in 2008 (Jemal et al., 2008).  Although melanoma is curable through early diagnosis and surgical excision (Balch et al., 2001), malignant lesions that go unnoticed are highly invasive and capable of rapid metastasis to other organs (Houghton and Polsky, 2002).  Consequently, patients diagnosed at late stages of tumor progression are faced with very poor prognosis; average survival is only 6–10 months for those with metastatic melanoma (Balch et al., 2001).  A major obstacle in treating metastatic melanoma is its stubborn resistance to conventional chemotherapies.  Large randomized trials of anticancer drugs, including nitrosoureas, taxanes, vinca alkaloids, and platinum compounds, have failed to produce significant responses in melanoma tumors (Soengas and Lowe, 2003).  Even dacarbazine (DTIC), the only drug approved by the FDA specifically for metastatic melanoma, brings about a complete response in less than 5% of cases (Serrone et al., 2000).  The molecular basis for melanoma's drug resistance is poorly understood.  Several mechanisms responsible for chemoresistance in other tumor types have been studied in melanoma but do not appear to be major factors (Grossman and Altieri, 2001).  An alternative explanation for melanoma's drug resistance lies in the dysregulation of apoptosis.  Most  1 A version of this chapter has been published.  Karst AM, Dai DL, Martinka M, Li G. PUMA expression is significantly reduced in human cutaneous melanomas.  Oncogene 2005, 24:1111-6.  42 anticancer drugs exert their cytotoxic effects by inducing apoptosis (Fisher, 1994); therefore, treatment success depends upon functional cell death machinery.  Defects in one or more components of the apoptotic pathway may result in resistance to drug- induced apoptosis.  Many molecular changes have the potential to cause apoptotic dysregulation, including activation of anti-apoptotic factors, inactivation of pro-apoptotic effectors, and/or reinforcement of survival signals (Soengas and Lowe, 2003).  Overexpression of apoptosis inhibitors such as Survivin and FLIP (FLICE-inhibitory protein) have been reported in malignant melanoma (Grossman et al., 1999; Irmler et al., 1997), whereas the apoptosis effector Apaf-1 has been found to be inactivated, presumably by methylation (Soengas et al., 2001).  Unlike many other human cancers, melanomas rarely harbor p53 mutations (Chin et al., 1998; Serrone and Hersey, 1999).  Therefore, other components of the p53 pathway, either upstream or downstream of p53, are likely defective in melanoma.  One such candidate is the pro-apoptotic gene PUMA (p53 upregulated modulator of apoptosis).  PUMA codes for a BH3-only mitochondrial protein belonging to the Bcl-2 family of apoptotic regulators.  PUMA was discovered in 2001 by three independent groups, two of which were searching for p53-inducible target genes (Nakano nad Vousden, 2001; Yu et al., 2001), and one of which was screening for Bcl-2-binding partners (Han et al., 2001).  PUMA expression causes rapid p53-dependent apoptosis and growth inhibition (Yu et al., 2001), induces cytochrome c release, and activates procaspases 3 and 9 (Nakano and Vousden, 2001), suggesting a prominent role in the intrinsic apoptotic pathway.  43 2.2 MATERIALS AND METHODS 2.2.1 Melanoma tissue microarray A tissue microarray (TMA) was constructed using paraffin-embedded human biopsies obtained from the 1990 to 1998 archives of the Department of Pathology at Vancouver General Hospital (Vancouver, British Columbia, Canada).  A total of 237 biopsies, including 118 primary melanomas, 53 metastatic melanomas, and 66 dysplastic nevi, were used for TMA construction.  Each biopsy was stained with hematoxylin and eosin and examined by a pathologist to identify the most suitable area for coring.  The selected area was deemed to be representative of the tumor biopsy as a whole.  The paraffin blocks containing the biopsies were then cored (0.6 mm diameter) in duplicate and transferred to a recipient block.  The completed recipient block was sectioned (4 µm) with a microtome and applied to adhesive-coated slides. 2.2.2 Immunohistochemical TMA analysis Paraffin was removed from slides by washing with xylene.  Tissues were then re- hydrated through a series of washes with 100%, 90%, and 80% ethanol, followed by PBS.  Antigen retrieval was performed by heating samples in 10 mM sodium citrate (pH 6.0).  Endogenous peroxidase activity was quenched with 0.04% hydrogen peroxide. Tissues were blocked with universal blocking serum (DAKO Diagnostics, Mississauga, ON, Canada), incubated with primary PUMA antibody (Imgenex, San Diego, CA, USA), followed by biotinylated secondary antibody (DAKO), and streptavidin–HRP (DAKO). Samples were developed using DAB substrate (Vector Laboratories, Burlington, ON, Canada) and counterstained with hematoxylin.  A negative control was performed by omitting primary antibody during the incubation step.  44 2.2.3 Cell culture Metastatic melanoma cell lines MMRU and MMAN were kind gifts from Dr. H.R. Byers (Boston University School of Medicine, Boston, MA).  These cell lines were established from fresh surgical specimens of melanoma removed from the cervical (MMRU) and inguinal (MMAN) lymph nodes of two melanoma patients (Byers et al., 1991). Malignant melanoma cell lines Sk-mel-110 and MeWo were kind gifts from Dr. A.P. Albino (Memorial Sloan-Kettering Cancer Center, New York, NY).  The p53 status of these cell lines has been previously determined by DNA sequencing; MMRU and MMAN contain wild-type p53 (Li et al., 1995), while Sk-mel-110 and MeWo contain mutant p53 (Bae et al., 1996).  The human embryonic kidney cell line HEK293, used for adenovirus amplification, was a kind gift from Dr. S. Dedhar (BC Cancer Research Centre, Vancouver, BC).  All cells were maintained in Dulbecco’s Modified Eagle Medium (Invitrogen, Burlington, ON, Canada) supplemented with 10% fetal bovine serum (HyClone, Logan, UT) in a humidified atmosphere of 5% CO2 at 37°C. 2.2.4 Adenoviral constructs A replication-deficient (E1 and E3 deleted) recombinant adenovirus containing human PUMAα cDNA under control of the cytomegalovirus (CMV) promoter was constructed using the AdEasy vector system from Qbiogene (Irvine, CA).  A control virus expressing green fluorescent protein (GFP) was constructed using the same method. 2.2.5 Adenoviral transduction Adenoviral constructs were amplified by multiple rounds of infection of 293 cells.  The adenoviruses were harvested by freezing and thawing the cells three times to induce cell lysis.  Crude lysates were cleared by centrifugation and titred using the TCID50 (tissue  45 culture infectious dose 50) assay.  Briefly, 293 cells were seeded into 96-well plates at a density of 5000 cells/well and serial dilutions of the adenovirus were added to each row of the plate.  After 5 days, cells were examined for evidence of cytopathic effect (CPE) and the number of positive wells was counted.  The TCID50 value and corresponding viral titre were calculated by the Reed-Muench method (Reed and Muench, 1938). 2.2.6 Reverse transcription polymerase chain reaction (RT-PCR) Total cellular RNA was extracted using Trizol (Invitrogen) and first strand cDNA was prepared from total cellular RNA using Superscript II RT (Invitrogen).  PCR primer pairs were designed using the online Primer 3 program (Rozen and Skaletsky, 2000).  PCR reactions were performed as follows: 1) initial denaturing at 94°C for 3 min, 2) 25-30 cycles of 94°C for 1 min (denaturing), 55°C for 1 min (annealing), and 72°C for 1 min (extension), and 3) final extension at 72°C for 10 min.  The PCR primer pairs used were: PUMA forward, 5’-CGACCTCAACGCACAGTA-3’; PUMA reverse, 5’- CCTAATTGGGCTCCATCTCG-3’; GAPDH forward, 5’- CTCATGACCACAGTCCATGCCATC-3’; GADPH reverse, 5’- CTGCTTCACCACCTTCTTGATGTC-3’. 2.2.7 Flow cytometric analysis of DNA fragmentation Cells transduced with ad-PUMA or ad-GFP for 72 h were stained with 50 µg/ml propidium iodide (Sigma-Aldrich Canada Ltd, Oakville, ON, Canada) for 30 min.  DNA fragmentation was detected using a Coulter Epics XL-MCL flow cytometer (Beckman Coulter, Mississauga, ON, Canada).  The percentage of cells exhibiting sub-G1 DNA was determined Expo32 ADC v1.2 acquisition and analysis software (Applied Cytometry Plano, TX).  46 2.3 RESULTS AND DISCUSSION We sought to determine whether PUMA has any relevance in the development and/or progression of malignant melanoma.  To this end, we compared PUMA expression levels in human melanoma biopsies at various stages of disease progression, using tissue microarray and immunohistochemical staining.  Although some cores were lost during the staining process, 107 primary melanomas, 51 metastatic melanomas, and 64 dysplastic nevi could be evaluated for PUMA staining. PUMA staining was evaluated by two blinded independent observers (including one dermatopathologist) and a consensus score was reached for each biopsy core. Staining was mainly cytoplasmic; intensity was scored as weak (1+), moderate (2+), strong (3+), or very strong (4+) (Figure 2.1A-D).  The staining pattern observed within individual biopsy cores was fairly homogenous; in 95% of nevi cores and 85% of tumor cores, all of the nevi/tumor cells exhibited uniform staining intensity.  In the remaining 5% of nevi cores and 15% of tumor cores, where some degree of heterogeneity was observed, a staining score was assigned according to which intensity level was most prevalent throughout the core.  Comparison of duplicate cores (from the same biopsy) revealed identical staining in 91% of nevi duplicates and 94% of tumor duplicates.  In cases of inconsistent staining between duplicates, the core with the higher staining score was used for analyses.  Differential staining of replicate cores likely reflects the heterogeneous and highly metastatic nature of malignant melanoma.  As melanoma quickly progresses, associated changes in gene expression may occur non-uniformly throughout the tumor.  47 Figure 2.1 PUMA expression in human dysplastic nevi, primary melanoma, and metastatic melanoma tumor tissues.  (A-D), Different levels of PUMA staining observed in the tissue microarray: (A) weak PUMA expression (1+) in a metastatic melanoma; (B) moderate PUMA expression (2+) in a primary melanoma; (C) strong PUMA expression (3+) in a dysplastic nevus; and (D) very strong PUMA expression (4+) in a dysplastic nevus. Magnification, × 400.  (E) PUMA expression is significantly weaker in primary melanoma tissue compared to dysplastic nevi (P<0.0001,  χ2 test). Similarly, PUMA expression is significantly weaker in metastatic melanoma tissue compared to primary melanoma tissue (P=0.001, Fisher’s exact test).   48 Although the level of PUMA varied in both nevi and melanoma tissues, PUMA expression was significantly weaker in primary melanoma than in dysplastic nevi (P<0.0001, χ2 test) (Figure 2.1E).  PUMA expression was further reduced in metastatic melanomas, compared to primary tumors (P=0.001, Fisher's exact test).  These results demonstrate for the first time that PUMA expression  is  lower  in  malignant  melanoma compared to dysplastic nevi.   This is significant because it identifies a possible point of dysfunction along the intrinsic cell death pathway that may contribute to malignant transformation.  To determine the implications of PUMA reduction in melanoma tissue, we looked for correlations between PUMA expression in tumors and clinicopathological parameters (Table 2.1).  Contingency table analyses revealed no associations between PUMA expression and gender, age, tumor thickness, ulceration, histological tumor subtype, or anatomical site.  To examine the effect of PUMA expression on survival, Kaplan–Meier survival curves were plotted for patients whose primary tumors exhibited weak (1+) PUMA staining versus those whose primary melanoma exhibited moderate or strong (2+, 3+) staining.  PUMA expression in primary melanoma tissue correlated with both overall and disease-specific 5-year survival (P=0.01 and P<0.01, respectively, log-rank test) (Figure 2.2A and B).  The correlation was even more significant when primary and metastatic melanoma cases were combined for the analysis; both overall and disease-specific 5-year survival rates were significantly worse for patients with weak (1+) PUMA expression in tumor tissue compared to those with moderate or strong expression (2+, 3+) (P<0.005 and P<0.001, respectively, log-rank test) (Figure 2.2C and D).  To determine whether  49  Table 2.1 PUMA expression and clinicopathological characteristics.               Intensity of PUMA staining           Total    P-value*         1+       2+       3+  Primary melanoma Gender  Male    22 (34%)  39 (61%)  3 (5%)  64  Female   19 (44%)  22 (51%)  2 (5%)  43  >0.05 Age (years)  21–38   8 (50%)  7 (44%)  1 (6%)  16  39–57    14 (36%)  24 (61.5%)  1 (2.5%)  39  58–75    9 (39%)  19 (57%)  1 (4%)  29  76–94    10 (43%)  11 (48%)  2 (9%)  23  >0.05 Tumor thickness (mm)  ≤ 0.75    8 (27.5%)  19 (65.5%)  2 (7%)  29  0.76–1.5   11 (39%)  17 (61%)  0 (0%)  28  1.51–3.0   9 (36%)  13 (52%)  3 (12%)  25  > 3.0    13 (52%)  12 (48%)  0 (0%)  25  >0.05 Ulceration  Positive   9 (45%)  10 (50%)  1 (5%)  20  Negative   24 (39%)  36 (58%)  2 (3%)  62  Unspecified   8 (32%)  15 (60%)  2 (8%)  25  >0.05 Tumor subtype  Lentigo maligna  8 (42%)  10 (53%)  1 (5%)  19  Nodular   7 (44%)  9 (56%)  0 (0%)  16  Superficial spreading 15 (31%)  31 (63%)  3 (6%)  49  Unspecified   11 (48%)  11 (48%)  1 (4%)  23  >0.05 Site  Head/neck   11 (55%)  7 (35%)  2 (10%)  20  Trunk    12 (27%)  32 (73%)  0 (0%)  44  Upper extremities  7 (37%)  10 (53%)  2 (10%)  19  Lower extremities  11 (46%)  12 (50%)  1 (4%)  24 >0.05  Metastatic melanoma Gender  Male    22 (63%)  12 (34%)  1 (3%)  35  Female   12 (75%)  4 (25%)  0 (0%)  16  >0.05 Age (years)  21–38   7 (100%)  0 (0%)  0 (0%)  7  39–5   7 13 (68%)  6 (32%)  0 (0%)  19  58–75    8 (53%)  6 (40%)  1 (7%)  15  76–94    6 (60%)  4 (40%)  0 (0%)  10 >0.05  *χ2 or Fisher’s exact test for weak (1+) versus moderate–strong (2+, 3+) PUMA expression.  50 Figure 2.2  Kaplan–Meier survival analyses of melanoma patients. (A) Overall survival of primary melanoma patients. (B) Disease-specific survival of primary melanoma patients. (C) Overall survival of all melanoma patients (including primary and metastatic cases). (D) Disease-specific survival of all melanoma patients.       51 PUMA expression could independently predict melanoma survival outcome, Cox multivariate regression analysis was employed to adjust for all clinicopathological factors listed in Table 2.1.  Multivariate analysis revealed that the overall 5-year mortality risk for melanoma patients with weak (1+) PUMA expression was 2.33 times higher than for patients with moderate or strong (2+, 3+) PUMA expression  (R2=2.33,  P=0.05).   In the case of disease-specific survival, the 5-year mortality risk was 2.89 times higher for patients with weak (1+) PUMA expression over patients with moderate or strong (2+, 3+) expression (R2=2.89, P=0.05).  Our results clearly indicate that low PUMA expression in either primary or metastatic melanoma tissue is associated with poor prognosis, suggesting that PUMA reduction may serve as a molecular marker for aggressive disease.  PUMA is a potent pro-apoptotic effector that is rapidly induced in cells following DNA damage (Han et al., 2001) and is required for p53-induced apoptosis (Yu et al., 2003).  Induction of p53-mediated apoptosis is the principal mechanism by which most chemotherapeutic drugs kill tumor cells (Fisher, 1994).  Consequently, downregulation of PUMA may severely hamper a cell's ability to respond to chemotherapy.  It has been shown that PUMA is necessary for the apoptosis induced by the DNA-damaging drug adriamycin in colorectal cancer cells (Yu et al., 2003) and it is possible that other chemotherapeutic drugs also require PUMA.  Reduced PUMA expression might explain why melanomas typically exhibit low apoptotic indices (Glinsky et al., 1997; Staunton and Gaffney, 1995) and highly chemoresistant phenotypes, despite their low frequency of p53 mutation (Chin et al., 1998; Serrone and Hersey, 1999).  If low PUMA expression renders melanoma resistant to apoptosis, then exogenous expression of PUMA should reverse such resistance.  To test this hypothesis,  52 we transduced four melanoma cell lines with an adenoviral construct containing human PUMA cDNA (Figure 2.3).  We found that both wild-type p53 and mutant p53 melanoma cells underwent a significant amount of dose-dependent apoptosis following infection with ad-PUMA compared to infection with the ad-GFP control virus (Figure 2.4).  This result suggests that adenoviral delivery of PUMA may be a valuable gene therapy tool for sensitizing melanoma tumors to apoptosis.  In our study, patients with weak PUMA expression had significantly poorer prognoses.  As treatment histories were not available, it is not clear whether chemoresistance was responsible for these patients' deaths.  However, the results of this study lead us to speculate that reduced PUMA expression may contribute to chemoresistance and thus decreased survival of melanoma patients.  Of course, it is unlikely that aberrant expression of a single gene leads to melanoma development and associated drug resistance.  Dysregulation of several apoptotic regulators, in combination with PUMA loss, is probably responsible for melanoma chemoresistance.  For example, the apoptotic inhibitors Mcl-1, Bcl-XL, XIAP, Livin, and Survivin are elevated in some melanoma cell lines, while pro-apoptotic Bax protein is abnormally low, compared to normal melanocytes (Bowen et al., 2003).  We propose a model in which PUMA downregulation in melanoma creates an apoptotic-resistant phenotype that responds poorly to chemotherapeutic drug treatment. Over the course of such treatment, cells exhibiting low PUMA expression enjoy a selective advantage, enabling them to survive and increase in proportion despite being subjected to pro-apoptotic stimuli.  If treatment fails and metastasis occurs, the metastatic tissue is more likely to have reduced PUMA expression than was the primary tumor.  53 Figure 2.3 Adenoviral-mediated overexpression of PUMA in melanoma cell lines. Cells were transduced with ad-PUMA (+) or ad-GFP control (-) for 72 h.  PUMA mRNA transcripts were detected by RT–PCR.  GADPH mRNA transcripts were measured as a loading control.   54 Figure 2.4 Ad-PUMA-induced apoptosis of melanoma cells.  Both wild-type p53 (A, MMRU; B, MMAN) and mutant p53 (C, Sk-mel-110; D, MeWo) melanoma cell lines underwent significant DNA fragmentation following ad-PUMA infection compared to infection with an ad-GFP control.  Top panels are representative images of control or ad- PUMA-infected cells (2.25 × 106 pfu/ml).  Magnification, × 400.  Bottom panels show the percentage of apoptotic (sub-G1) cells, as determined by flow cytometric analysis.   55  This would explain our finding that PUMA reduction is more prevalent in metastatic than in primary melanoma tumors.  Delivery of PUMA to melanoma tissue via adenoviral- based gene therapy may kill tumor cells and/or sensitize them to chemotherapy.  In conclusion, our study has demonstrated that PUMA expression is dramatically reduced in melanoma tumor tissue compared to dysplastic nevi, and is further reduced in metastatic melanomas compared to primary melanomas.  We have shown a strong correlation between 5-year survival and PUMA expression in melanoma tissue, suggesting that reduced PUMA expression may serve as a molecular marker for aggressive disease.  Finally, we have demonstrated that forced PUMA expression in melanoma cells causes significant apoptotic cell death.   56 2.4 REFERENCES Bae I, Smith ML, Sheikh MS, Zhan Q, Scudiero DA, Friend SH, O'Connor PM, Fornace AJ Jr (1996). An abnormality in the p53 pathway following gamma-irradiation in many wild-type p53 human melanoma lines. Cancer Res 56: 840-7.  Balch CM, Buzaid AC, Soong SJ, Atkins MB, Cascinelli N, Coit DG, Fleming ID, Gershenwald JE, Houghton A Jr, Kirkwood JM, McMasters KM, Mihm MF, Morton DL, Reintgen DS, Ross MI, Sober A, Thompson JA, Thompson JF (2001). Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 19: 3635-48.  Bowen AR, Hanks AN, Allen SM, Alexander A, Diedrich MJ, Grossman D (2003). Apoptosis regulators and responses in human melanocytic and keratinocytic cells. J Invest Dermatol 120: 48-55.  Byers HR, Etoh T, Doherty JR, Sober AJ, Mihm MC Jr (1991). Cell migration and actin organization in cultured human primary, recurrent cutaneous and metastatic melanoma. Time-lapse and image analysis. Am J Pathol 139: 423-35.  Chin L, Merlino G, DePinho RA (1998). Malignant melanoma: modern black plague and genetic black box. Genes Dev 12: 3467-81.  Fisher DE (1994). Apoptosis in cancer therapy: crossing the threshold. Cell 78: 539-42.  Glinsky GV, Glinsky VV, Ivanova AB, Hueser CJ (1997). Apoptosis and metastasis: increased apoptosis resistance of metastatic cancer cells is associated with the profound deficiency of apoptosis execution mechanisms. Cancer Lett 115: 185-93.  Grossman D, Altieri DC (2001). Drug resistance in melanoma: mechanisms, apoptosis, and new potential therapeutic targets. Cancer Metastasis Rev 20: 3-11.  Grossman D, McNiff JM, Li F, Altieri DC (1999). Expression and targeting of the apoptosis inhibitor, survivin, in human melanoma. J Invest Dermatol 113: 1076-81.  Han J, Flemington C, Houghton AB, Gu Z, Zambetti GP, Lutz RJ, Zhu L, Chittenden T (2001). Expression of bbc3, a pro-apoptotic BH3-only gene, is regulated by diverse cell death and survival signals. Proc Natl Acad Sci U S A 98: 11318-23.  Houghton AN, Polsky D (2002). Focus on melanoma. Cancer Cell 2: 275-8.  Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V, Bodmer JL, Schröter M, Burns K, Mattmann C, Rimoldi D, French LE, Tschopp J (1997). Inhibition of death receptor signals by cellular FLIP. Nature 388: 190-5.   57 Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun MJ (2008). Cancer statistics, 2008. CA Cancer J Clin 58: 71-96.  Li G, Ho VC, Trotter M, Horsman DE, Tron VA (1995). p53 mutation in metastatic melanomas and primary melanomasfrom sun-exposed skin and sun-protected sites. J Eur Acad Dermatol Venereol 4: 48-53.  Nakano K, Vousden KH (2001). PUMA, a novel proapoptotic gene, is induced by p53. Mol Cell 7: 683-94.  Reed LJ, and Muench H (1938). A simple method of estimating fifty percent endpoints. Am J Hyg 27: 493-7.  Rozen S, Skaletsky, HJ (2000). Primer3 on the WWW for general users and for biologist programmers. Humana Press: Totowa, NJ.  Serrone L, Hersey P (1999). The chemoresistance of human malignant melanoma: an update. Melanoma Res 9: 51-8.  Serrone L, Zeuli M, Sega FM, Cognetti F (2000). Dacarbazine-based chemotherapy for metastatic melanoma: thirty-year experience overview. J Exp Clin Cancer Res 19: 21-34.  Soengas MS, Capodieci P, Polsky D, Mora J, Esteller M, Opitz-Araya X, McCombie R, Herman JG, Gerald WL, Lazebnik YA, Cordón-Cardó C, Lowe SW (2001). Inactivation of the apoptosis effector Apaf-1 in malignant melanoma. Nature 409: 207-11.  Soengas MS, Lowe SW (2003). Apoptosis and melanoma chemoresistance. Oncogene 22: 3138-51.  Staunton MJ, Gaffney EF (1995). Tumor type is a determinant of susceptibility to apoptosis. Am J Clin Pathol 103: 300-7.  Yu J, Wang Z, Kinzler KW, Vogelstein B, Zhang L (2003). PUMA mediates the apoptotic response to p53 in colorectal cancer cells. Proc Natl Acad Sci U S A 100: 1931- 6.  Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B (2001). PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell 7: 673-82.     58 3. ROLE OF PUMA AND PHOSPHORYLATED AKT IN MELANOMA CELL GROWTH, APOPTOSIS, AND PATIENT SURVIVAL2 3.1 INTRODUCTION Melanoma is a lethal and highly invasive form of skin cancer arising from the abnormal proliferation of epidermal melanocytes (Houghton and Polsky, 2002).  Treatment of the disease is made difficult by its strong resistance to conventional cancer treatments, such as radiation and chemotherapy (Soengas and Lowe, 2003).  The molecular mechanisms underlying this resistance have not been clearly elucidated.  However, it is speculated that dysregulation of apoptosis may be responsible for the chemoresistant phenotype of melanoma.  Apoptotic dysregulation may manifest itself in many forms: through inactivation of pro-apoptotic effectors, overexpression of anti-apoptotic proteins, or hyperactivation of survival signaling pathways (Soengas and Lowe, 2003). We have previously reported that expression of the pro-apoptotic protein PUMA is significantly reduced in human cutaneous melanomas (Karst et al., 2005).  PUMA is an essential mediator of cell death and plays a key functional role in the process of p53- mediated apoptosis (Jeffers et al., 2003; Villunger et al., 2003).  PUMA is a BH3-only protein belonging to the Bcl-2 family of proteins that regulate apoptosis (Nakano and Vousden, 2001; Yu et al., 2001).  PUMA is a direct transcriptional target of p53, and its expression is up-regulated most notably in response to DNA damage (Nakano and Vousden, 2001; Yu et al., 2001).  Like all the BH3-only proteins, PUMA binds to and neutralizes anti-apoptotic members of the Bcl-2 family to promote apoptosis (Chen et al.,  2 A version of this chapter has been published.  Karst AM, Dai DL, Cheng JQ, Li G. Role of p53 up- regulated modulator of apoptosis and phosphorylated Akt in melanoma cell growth, apoptosis, and patient survival. Cancer Res 2006, 66:9221-6.  59 2005).  Although most BH3-only proteins interact with only a subset of anti-apoptotic Bcl-2 proteins, PUMA targets all of them, making PUMA a particularly potent effector of apoptosis (Chipuk et al., 2005).  In addition to neutralizing anti-apoptotic proteins, BH3- only proteins must also activate either Bax or Bak protein to induce cell death (Cheng et al., 2001; Zong et al., 2001).  Whether PUMA accomplishes this task directly or indirectly remains unresolved (Willis and Adams, 2005).  Aside from inactivation of pro- apoptotic effectors, hyperactivation of survival and/or proliferation signaling pathways is thought to contribute to the aggressive nature of melanoma.  Notably, the serine/threonine kinase Akt has been found to be constitutively activated in 43% to 67% of melanomas (Robertson, 2005).  Under normal conditions, Akt is activated indirectly via growth factor stimulation of cell surface receptors.  Stimulated receptors recruit and activate PI3K, resulting in production of a second messenger molecule, PIP3 (Nicholson and Anderson, 2002).  PIP3 in turn recruits Akt to the plasma membrane, where it is activated and, in turn,  promotes the transcription of various cell survival and proliferation genes (Karin and Lin, 2002).  We and others have reported that phosphorylated Akt (p-Akt) plays an important role in melanoma progression and invasion (Dai et al., 2005; Dhawan et al., 2002; Robertson, 2005; Stahl et al., 2003; Stahl et al., 2004). In this study we sought to investigate the roles of PUMA and p-Akt expression in human cutaneous melanoma progression and patient survival and to determine whether these two molecules may serve as therapeutic targets for malignant melanoma.  60 3.2 MATERIALS AND METHODS 3.2.1 Melanoma tissue microarray Tissue microarray construction is described in chapter 2.2.1. 3.2.2 Immunohistochemical TMA analysis TMA samples were analyzed using the immunohistochemical methods described in chapter 2.2.2, using PUMA (Imgenex) and p-Akt-Ser473 (Cell Signaling Technology, Davers, MA) primary antibodies.  In the present analysis, 146 biopsies (99 primary tumors and 47 metastatic tumors) could be evaluated for both PUMA and p-Akt-Ser473 staining.  PUMA staining was mostly weak to moderate, with a few cores staining strongly to very strongly.  Conversely, p-Akt staining was mostly moderate to strong, with a few cores staining weakly.  Thus, to obtain categories of sufficient size for the present statistical analysis, both PUMA and p-Akt staining intensities were categorized simply as "weak" or "strong". 3.2.3 Cell Culture Cell culture conditions for melanoma cells are described in chapter 2.2.3. 3.2.4 Adenoviral constructs Construction of ad-PUMA and ad-GFP adenoviruses is described in chapter 2.2.4. 3.2.5 Hoechst staining Cells were seeded onto cover slips in 6-well plates and infected with ad-PUMA or ad- GFP for 48 hours.  Samples were then stained with the nuclear dye Hoechst 33258 (Sigma-Aldrich).  Nuclei were visualized by immunofluorescent microscopy, and the number of apoptotic cells in each sample was counted.  61 3.2.6 Subcellular fractionation Cells were infected with ad-PUMA or ad-GFP for 0, 4, 8, or 12 hours.  Cells were harvested and resuspended in mitochondrial isolation buffer [20 mM HEPES-KOH (pH 7.5), 10 mM KCl, 1.5 mM MgCl2, 1 mM sodium EDTA, 1 mM sodium EGTA, 1 mM DTT, 250 mM sucrose] containing protease inhibitors (Roche Diagnostics, Laval, QC, Canada).  Cells were lysed by passage through a 26-gauge needle and then centrifuged at 2,000 × g for 10 minutes to remove unbroken cells, plasma membrane fragments, and nuclei.  The resulting supernatant was centrifuged at 10,000 × g for 20 minutes to obtain a mitochondrial pellet.  The pellet was washed with mitochondrial isolation buffer and solubilized in 30 µl of TNC buffer [10 mM Tris-acetate (pH 8.0), 0.5% NP40, 5 mM CaCl2] containing protease inhibitors.  The cytosolic fraction was obtained by centrifugation of the supernatant at 50,000 × g for 1 hour.  Protein concentrations were quantified by Bradford assay using Bio-Rad protein assay dye reagent (Bio-Rad, Mississauga, ON, Canada).  Proteins were separated on 10% SDS-polyacrylamide gels and blotted onto polyvinylidene difluoride (PVDF) membranes (Bio-Rad).  Membranes were incubated with primary antibodies for: PUMA (Imgenex), Bax (Upstate Cell Signaling Solutions, Lake Placid, NY), Smac (Imgenex), cytochrome c (Cell Signaling Technology), caspase-9 (Cell Signaling Technology), caspase-3 (Cell Signaling Technology), β-actin (Sigma-Aldrich), and Hsp-60 (Stressgen Biotechnologies, Victoria, BC).     62 3.2.7  Clonogenic assay Cells infected with ad-PUMA or ad-GFP were seeded in 60 mm tissue culture plates at a density of 3,000 cells per plate.  After 14 days, cells were fixed with 3.7% formalin and stained with 2% crystal violet, and the number of colonies on each plate was counted. 3.2.8 Ad-PUMA treatment in vivo Tumors were induced by injecting 1.5 × 107 melanoma cells subcutaneously into the right flanks of 12-week-old male severe combined immunodeficient (SCID) mice.  When tumors were palpable, mice were randomized into two treatment groups.  Tumors were injected intratumorally with 1 × 108 plaque-forming units (pfu) of either ad-PUMA or ad- GFP in a volume of 100 µl in a single pass using 30-gauge needles.  Adenovirus treatments were given every 3 days until the end of the experiment.  Tumor dimensions were measured using calipers.  Tumor volumes were calculated according to the following formula: V = L × W2 × π/6 (Tomayko and Reynolds, 1989).  Experiments were stopped when the tumor burdens of ad-GFP control groups became excessive. 3.2.9 API-2 treatment in vitro The API-2 (Akt/PKB signaling inhibitor-2) compound was supplied by Dr. J.Q. Cheng (H. Lee Moffitt Cancer Center, Tampa, FL), dissolved in DMSO.  For in vitro assays, cells were treated at 50% confluency with complete medium containing 5 to 50 µM of API-2. 3.2.10 Western blot Total cellular protein extracts were prepared using triple detergent buffer (20 mM Tris-Cl [pH 8.0], 150 mM NaCl, 0.1% sodium dodecyl sulfate, 1% Nonidet P-40, 0.5% sodium deoxycholate) containing protease and phosphatase inhibitors (Roche Diagnostics, Laval,  63 QC, Canada).  Protein samples were quantified, separated on SDS-PAGE gels, and electroblotted onto membranes as described in chapter 3.2.6.  Membranes were incubated with primary antibodies for: p-Akt-Ser473 (Cell Signaling Technology) and β-actin (Sigma-Aldrich). 3.2.11 Sulforhodamine B (SRB) assay Cell survival was determined by the sulforhodamine B (SRB) assay.  SRB is a histological dye that binds to protein basic amino acid residues in cells fixed with trichloroacetic acid (TCA).  It provides a sensitive index of cellular protein content (that is linear over a cell density range of at least 2 orders of magnitude) and can be used to accurately measure cytotoxicity following drug treatment (Skehan et al., 1990).  Cells treated in 96-well plates were fixed at the appropriate time points with 10% trichloroacetic acid (TCA), stained with 0.4% SRB in 1% acetic acid, then de-stained with 1% acetic acid.  Cell density was quantified by dissolving bound dye in 10 mM Tris (pH 10.5) followed by colorimetric determination at 550 nm. 3.2.12 Ad-PUMA and API-2 combination treatment in vivo Cells (1 × 107) were injected subcutaneously into the right flanks of 6-week-old male SCID mice.  When tumors were palpable, mice were randomized into four treatment groups.  Animals were treated with 1 × 108 pfu of ad-PUMA or ad-GFP virus, delivered by intratumoral injection once every 3 days.  API-2 was administered via intraperitoneal injection at a dose of 1 mg/kg/day (in 15% DMSO).  Tumor dimensions were measured using calipers.  Experiments were stopped when the tumor burden of the double negative control group (ad-GFP + DMSO) became excessive.   64 3.3 RESULTS AND DISCUSSION 3.3.1 Weak PUMA expression and strong Akt phosphorylation cooperatively reduce melanoma patient survival We have previously shown, using tissue microarray analysis, that weak PUMA expression in melanoma tumors correlates with poor 5-year survival of melanoma patients (Karst et al., 2005).  In a subsequent study, we reported that p-Akt expression in human melanoma is strongly associated with invasion and progression and that p-Akt levels inversely correlate with melanoma patient survival (Dai et al., 2005).  Based on these results, we were interested in determining whether PUMA and p-Akt have a cooperative effect on melanoma survival rates.  Because both aforementioned studies were carried out on the same tissue microarray, we were able to combine the two data sets for our present analysis.  To assess the combined effect of weak PUMA expression and strong p-Akt expression on disease-specific survival, Kaplan-Meier survival curves were plotted for patients whose primary tumors exhibited:  1) weak p-Akt and strong PUMA, 2) strong p-Akt or weak PUMA (but not both), and 3) strong p-Akt and weak PUMA expression.  We found that patients with both weak PUMA expression and strong p-Akt expression in their primary tumor tissue had significantly worse 5-year survival than patients having either strong p-Akt or weak ad-PUMA alone (P=0.0069, log-rank test) (Figure 3.1A).  In a combined analysis of primary and metastatic tumors, the results were even more significant; patients exhibiting both weak PUMA and strong p-Akt expression in tumor tissue again had reduced 5-year survival (P=0.0004, log-rank test) (Figure 3.1B). Strikingly, there were no deaths (within 5 years) among patients exhibiting both strong PUMA and weak p-Akt expression in primary tumor tissue (Figure 3.1A).  Our data  65 Figure 3.1 The combination of weak PUMA and strong p-Akt expression in human melanoma tumors correlates with poor 5-year patient survival.  Kaplan-Meier survival curves were plotted for (A) primary melanoma patients (n=99) and (B) all melanoma patients (n=146, including 99 primary melanomas and 47 metastatic melanomas), based on the expression levels of PUMA and p-Akt in tumor tissue.  Patients with both weak PUMA and strong p-Akt expression had significantly worse 5-year disease-specific survival than patients with either weak PUMA or strong p-Akt expression in primary melanomas (P=0.0069) and in primary and metastatic melanomas combined (P=0.0004, log-rank test).   66 suggest that aberrant expression of PUMA and p-Akt may cooperatively contribute to melanoma tumor progression, resulting in worse patient prognosis.  Based on these results, we hypothesized that a two-pronged strategy, one that simultaneously addresses weak PUMA expression and hyperphosphorylation of Akt, may prove effective in treating melanoma and improving patient prognosis.  To induce PUMA expression in melanoma tissue, we used an adenoviral-based gene therapy approach. 3.3.2 Ad-PUMA kills melanoma cells via rapid induction of mitochondrial- mediated apoptosis The ability of PUMA to induce cell death has been well documented in various human cancer cell lines, including colorectal (Nakano and Vousden, 2001; Yu et al., 2003; Yu et al., 2001), lung (Nakano and Vousden, 2001; Yu et al., 2001), head and neck (Hoque et al., 2003), osteosarcoma (Nakano and Vousden, 2001), and glioma (Ito et al., 2005).  We have previously shown that adenoviral-mediated expression of PUMA causes massive death of melanoma cells within 72 hours of infection (Figure 2.4).  Here, we show that ad-PUMA induces morphologic changes associated with apoptosis, notably chromatin fragmentation and the formation of apoptotic bodies.  In both wild-type p53 (MMRU and MMAN) and mutant p53 (Sk-mel-110 and MeWo) melanoma cell lines, infection with ad-PUMA produced a large proportion (25-45%) of apoptotic cells (Figure 3.2).  In three cell lines (MMRU, MMAN, and MeWo), 37% to 45% of cells exhibited apoptotic bodies after 48 hours of ad-PUMA infection.  The number was slightly lower (25%) for the Sk- mel-110 cell line because it is the most susceptible to adenoviral infection (data not shown).  Consequently, apoptosis was more efficiently induced in Sk-mel-110 and many apoptotic cells had detached before in situ Hoechst staining was conducted.  67 Figure 3.2 Ad-PUMA induces the formation of apoptotic bodies in human melanoma cells.  (A) Hoechst staining of MMRU cells infected with ad-PUMA or ad-GFP for 48 h. (B) The percentage of cells exhibiting apoptotic bodies after 48 h of ad-PUMA infection in two wild-type p53 (MMRU, MMAN) and two mutant p53 (Sk-mel-110, MeWo) melanoma cell lines.    68  PUMA reportedly induces apoptosis through the intrinsic, mitochondrial- mediated pathway (Nakano and Vousden, 2001; Yu et al., 2001).  Activation of this pathway is characterized by translocation of cytosolic Bax to the mitochondrial membrane, release of Smac/DIABLO and cytochrome c from mitochondria, subsequent activation of caspase-3 and caspase-9, and, inevitably, cell death.  To confirm that our ad- PUMA construct activates this pathway in melanoma cells, we examined the cytosolic and mitochondrial protein expression of ad-PUMA-transduced MMRU cells.  Cells infected with ad-PUMA for 0, 4, 8, or 12 hours were subjected to subcellular fractionation to separate cytosolic and mitochondrial proteins.  Western blot analysis of the protein extracts showed that PUMA was strongly expressed in mitochondria at 8 hours post- infection (Figure 3.3), demonstrating the ability of ad-PUMA to rapidly induce gene expression.  PUMA expression was accompanied by translocation of cytosolic Bax to the mitochondria, release of Smac and cytochrome c from mitochondria into the cytosol, and cleavage of procaspase-3 and procaspase-9 (Figure 3.3).  These data confirm that ad- PUMA induces melanoma cell death via mitochondrial-mediated apoptosis.  To determine whether ad-PUMA suppresses the long-term survival of melanoma cells, clonogenic assays were performed.  We found that infection with ad-PUMA significantly impaired the abilities of both wild-type p53 (MMRU and MMAN) and mutant p53 (Sk-mel-110 and MeWo) melanoma cells to form colonies over a 14-day period (Figure 3.4).  Ad-PUMA infection inhibited colony formation by 80% to 90% compared to the ad-GFP control.  69 Figure 3.3 Ad-PUMA rapidly induces mitochondrial-mediated apoptosis in human melanoma cells.  Western blot analysis of MMRU cells infected with ad-PUMA or ad- GFP for 0-12 h.  Samples were separated into cytosolic and mitochondrial fractions to detect changes in protein localization.    70 Figure 3.4 Ad-PUMA severely inhibits colony formation by human melanoma cells. (A) Clonogenic assay of MMRU cells transduced with ad-PUMA or ad-GFP.  Cells were infected for 4 h, then re-seeded at low density.  (B) The percentage of colonies formed in two wild-type p53 (MMRU, MMAN) and two mutant p53 (Sk-mel-110, MeWo) melanoma cell lines.   71 3.3.3 Ad-PUMA inhibits growth of melanoma tumor xenografts In light of the potent cell killing ability of ad-PUMA in vitro, we hypothesized that ad- PUMA would be highly effective at inhibiting melanoma tumor growth in vivo. Melanoma tumors are notoriously resistant to both chemotherapy and radiotherapy and thus are difficult to treat.  Ad-PUMA gene therapy may be an effective alternative to conventional treatments that rely on cytotoxic DNA-damaging agents to induce apoptosis. To test the efficacy of ad-PUMA in vivo, we used a SCID mouse model in which tumors were induced by subcutaneous injection of MMRU (n=6) or MMAN (n=10) human melanoma cells.  Once tumors were palpable, mice were randomly assigned to either ad- PUMA or ad-GFP treatment groups.  Virus (1 × 108 pfu) was administered by intratumoral injection every 3 days.  As shown in Figure 3.5, melanoma tumors injected with ad-PUMA grew more slowly than those injected with ad-GFP control virus.  After 60 days of treatment, MMRU tumors treated with ad-PUMA were 60% smaller than controls (P<0.01) (Figure 3.5).  MMAN tumor growth was also inhibited; ad-PUMA- treated tumors were 40% smaller than controls after 52 days of treatment (P<0.05) (Figure 3.6).  The 20% difference in treatment efficacy between the cell lines likely reflects a difference in susceptibility to adenoviral infection.  Our results indicate that ad-PUMA indeed inhibits melanoma tumor growth in vivo.  However, the growth-inhibitory effect observed in vivo was not as strong as that seen in colony formation assays in vitro.  We speculate that two factors may contribute to this apparent discrepancy: 1) insufficient distribution of virus throughout tumors and 2) leakage of virus from tumor tissue.  In our study, tumors were injected with adenovirus in a volume of 100 µl in a single pass using 30-gauge needles as described previously  72 Figure 3.5 Ad-PUMA inhibits MMRU tumor growth in vivo. (A) MMRU tumors excised from SCID mice after treatment with ad-PUMA or ad-GFP for 60 days. (B) Average volume of MMRU tumors (n=6) treated with ad-PUMA or ad-GFP, beginning 11 days after tumor induction.     73 Figure 3.6 Ad-PUMA inhibits MMAN tumor growth in vivo. (A) MMAN tumors excised from SCID mice after treatment with ad-PUMA or ad-GFP for 52 days. (B) Average volume of MMAN tumors (n=10) treated with ad-PUMA or ad-GFP, beginning 13 days after tumor induction.   74 (Wolkersdörfer et al., 2004).  Although a single intratumoral injection may be adequate for the delivery of replication-competent adenoviruses, which can "spread" from cell to cell, this administration method may not have sufficiently distributed our replication- deficient ad-PUMA construct throughout the tumor, especially as tumor volumes increased.  Additionally, we suspect that some viruses may have "leaked" from tumors, reducing transduction efficiency.  To circumvent these technical problems, we propose that a conditionally replicating adenoviral vector under the control of a melanocytic tissue-specific promoter be used to improve the efficiency of PUMA gene delivery to melanoma tumor tissues. 3.3.4 Ad-PUMA and API-2 cooperatively inhibit melanoma cell growth We next hypothesized that the inhibitory effects of ad-PUMA on melanoma cell growth could be enhanced by eliminating prominent cell survival and/or proliferation signals from melanoma cells.  Based on our finding that p-Akt is associated with melanoma invasion, progression, and poor prognosis (Dai et al., 2005), we sought to determine whether combining p-Akt inhibition with ad-PUMA treatment would be of therapeutic value.  Western blot analyses of the malignant melanoma cell lines MMRU and MMAN revealed that both cell lines strongly express activated Akt phosphorylated on Ser473 (Figure 3.7).  To eliminate this survival/proliferation signal, we treated cells with the small-molecule Akt pathway inhibitor API-2.  This compound is reported to specifically inhibit Akt phosphorylation and kinase activity without altering the activity of related kinases, such as PI3K, PDK1, protein kinase C, serum- and glucocorticoid-inducible kinase, protein kinase A, signal transducers and activators of transcription 3, extracellular signal-regulated kinase 1/2, or c-Jun NH2-terminal kinase (Yang et al., 2004).  Treatment  75 Figure 3.7 API-2 inhibits p-Akt expression by human melanoma cells in vitro.  Western blot analysis of p-Akt expression in two human melanoma cell lines, (A) MMRU and (B) MMAN, treated with API-2 for 24 h.     76 of melanoma cells with API-2 reduced their Akt phosphorylation levels (Figure 3.7). Furthermore, API-2 inhibited MMRU and MMAN cell survival in a dose- and time- dependent manner (Figure 3.8).  At the lowest dose tested (5 µM), 48 hours of API-2 treatment inhibited MMRU and MMAN cell survival by 35% and 45%, respectively, compared with treatment with vehicle alone.  To determine whether API-2 and ad-PUMA could act synergistically to inhibit melanoma cell survival, MMRU cells were infected with ad-PUMA or ad-GFP and treated  with  20  µM  API-2  for  48 hours.   The cells  were  analyzed  by  SRB assay  to determine the percentage of surviving cells.  We found that API-2 treatment enhanced ad- PUMA-mediated cell death in an additive manner compared to treatment with the DMSO vehicle (Figure 3.9).  Overall, these data are consistent with the hypothesis that weak PUMA expression combined with strong p-Akt expression promotes melanoma cell survival.  It is not unexpected that combining ad-PUMA and API-2 treatments had an additive, but not synergistic, effect on cell survival because PUMA and Akt function in two distinct signaling pathways and are not known to interact with each other.  To determine whether API-2 could augment ad-PUMA-induced apoptosis of melanoma cells, we analyzed cells by flow cytometry following combination treatment with ad-PUMA and API-2.  Although other groups have reported that Akt inhibition induces apoptosis in various mammalian cell lines (Cheng et al., 1997; Datta et al., 1999; Sun et al., 2001; Yang et al., 2004), API-2 did not seem to induce a significant degree of apoptosis in the melanoma cell lines we tested (data not shown).  This suggests that API-2 may function primarily to inhibit growth rather than promote apoptosis in human melanoma cells.  77 Figure 3.8 API-2 inhibits the survival of human melanoma cells in vitro. SRB cell survival assays of (A) MMRU cells and (B) MMAN cells treated with 5-50 µM of API-2 for 24 or 48 h.    78 Figure 3.9 API-2 enhances ad-PUMA-mediated growth inhibition in vitro.  SRB cell survival assays of (A) MMRU and (B) MMAN cells treated both ad-PUMA (MOI 2) and API-2 (20 µM) for 48 h.    79 3.3.5 Ad-PUMA and API-2 cooperatively inhibit growth of melanoma tumor xenografts To determine whether API-2 could enhance ad-PUMA-mediated inhibition of melanoma tumor growth, we used a SCID mouse tumor xenograft model.  Animals bearing MMRU tumors (N=20) were randomized into four groups and treated with either (a) ad-PUMA and API-2, (b) ad-PUMA and DMSO (vehicle), (c) ad-GFP (control virus) and API-2, or (d) ad-GFP and DMSO as described in chapter 3.2.12.  As expected, ad-PUMA significantly inhibited MMRU tumor growth (Figure 3.10).  After 54 days of treatment (18 viral injections), tumors treated with ad-PUMA were 57% smaller than those treated with ad-GFP (P<0.005, t-test).  API-2 also dramatically slowed MMRU tumor growth rates; after 54 days of treatment, tumors dosed with API-2 were 46% smaller than those treated with DMSO vehicle (P=0.005, t-test).  When API-2 and ad-PUMA treatments were combined, however, a marked enhancement of growth inhibition was observed; 54 days of the combination treatment (ad-PUMA and API-2) yielded tumors that were 81% smaller than the negative control group (ad-GFP and DMSO; P<0.001, t-test).  Our data indicate that inhibiting p-Akt expression clearly augments the inhibitory effect of ad- PUMA on melanoma growth in vivo.  This suggests that combining a p-Akt inhibitor drug with ad-PUMA gene therapy may be of significant therapeutic value in treating malignant melanoma.  In summary, we have presented the first evidence that PUMA and Akt may together play an important role in melanoma progression and patient survival.  Our data perhaps explains why melanoma is both highly invasive and resistant to chemotherapy; an inability to express PUMA could severely impair p53-dependent apoptosis, while at the  80 Figure 3.10 API-2 enhances ad-PUMA-mediated growth inhibition in vivo.  (A) MMRU tumors excised from SCID mice 55 days after commencement of combination treatment or treatment with the indicated controls.  (B) Average volume of MMRU tumors (N=20) treated for 55 days with various combinations ad-PUMA, ad-GFP (control virus), API-2, and DMSO (vehicle).  81 same time hyperphosphorylation of Akt may promote aggressive tumor growth.  Our novel finding that combined dysregulation of PUMA and p-Akt expression results in poorer prognosis than aberrant expression of either protein alone provides a strong rationale for dual therapeutic targeting of PUMA and Akt in melanoma.  In further support of this principle, we have shown that p-Akt inhibition enhances the inhibitory effect of ad-PUMA on melanoma cell survival in vitro and melanoma tumor growth in vivo.  Thus, by reversing the "weak PUMA and strong p-Akt" expression pattern in melanoma tissue, we can effectively induce cell death and suppress proliferation.  These results, taken together, form the basis for a novel and potentially more effective therapeutic strategy in the treatment of malignant melanoma.   82 3.4 REFERENCES Chen L, Willis SN, Wei A, Smith BJ, Fletcher JI, Hinds MG, Colman PM, Day CL, Adams JM, Huang DC (2005). Differential targeting of prosurvival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function. Mol Cell 17: 393-403.  Cheng EH, Wei MC, Weiler S, Flavell RA, Mak TW, Lindsten T, Korsmeyer SJ (2001). BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK- mediated mitochondrial apoptosis. Mol Cell 8: 705-11.  Cheng JQ, Altomare DA, Klein MA, Lee WC, Kruh GD, Lissy NA, Testa JR (1997). Transforming activity and mitosis-related expression of the AKT2 oncogene: evidence suggesting a link between cell cycle regulation and oncogenesis. Oncogene 14: 2793-801.  Chipuk JE, Bouchier-Hayes L, Kuwana T, Newmeyer DD, Green DR (2005). PUMA couples the nuclear and cytoplasmic proapoptotic function of p53. Science 309: 1732-5.  Dai DL, Martinka M, Li G (2005). Prognostic significance of activated Akt expression in melanoma: a clinicopathologic study of 292 cases. J Clin Oncol 23: 1473-82.  Datta SR, Brunet A, Greenberg ME (1999). Cellular survival: a play in three Akts. Genes Dev 13: 2905-27.  Dhawan P, Singh AB, Ellis DL, Richmond A (2002). Constitutive activation of Akt/protein kinase B in melanoma leads to up-regulation of nuclear factor-kappaB and tumor progression. Cancer Res 62: 7335-42.  Hoque MO, Begum S, Sommer M, Lee T, Trink B, Ratovitski E, Sidransky D (2003). PUMA in head and neck cancer. Cancer Lett 199: 75-81.  Houghton AN, Polsky D (2002). Focus on melanoma. Cancer Cell 2: 275-8.  Ito H, Kanzawa T, Miyoshi T, Hirohata S, Kyo S, Iwamaru A, Aoki H, Kondo Y, Kondo S (2005). Therapeutic efficacy of PUMA for malignant glioma cells regardless of p53 status. Hum Gene Ther 16: 685-98.  Jeffers JR, Parganas E, Lee Y, Yang C, Wang J, Brennan J, MacLean KH, Han J, Chittenden T, Ihle JN, McKinnon PJ, Cleveland JL, Zambetti GP (2003). Puma is an essential mediator of p53-dependent and -independent apoptotic pathways. Cancer Cell 4: 321-8.  Karin M, Lin A (2002). NF-kappaB at the crossroads of life and death. Nat Immunol 3: 221-7.  Karst AM, Dai DL, Martinka M, Li G. (2005). PUMA expression is significantly reduced in human cutaneous melanomas. Oncogene 24: 1111-6.  83  Nakano K, Vousden KH (2001). PUMA, a novel proapoptotic gene, is induced by p53. Mol Cell 7: 683-94.  Nicholson KM, Anderson NG (2002). The protein kinase B/Akt signalling pathway in human malignancy. Cell Signal 14: 381-95.  Robertson GP (2005). Functional and therapeutic significance of Akt deregulation in malignant melanoma. Cancer Metastasis Rev 24: 273-85.  Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica D, Warren JT, Bokesch H, Kenney S, Boyd MR (1990). New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 82: 1107-12.  Soengas MS, Lowe SW (2003). Apoptosis and melanoma chemoresistance. Oncogene 22: 3138-51.  Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP (2003). Loss of PTEN promotes tumor development in malignant melanoma. Cancer Res 63: 2881-90.  Stahl JM, Sharma A, Cheung M, Zimmerman M, Cheng JQ, Bosenberg MW, Kester M, Sandirasegarane L, Robertson GP (2004). Deregulated Akt3 activity promotes development of malignant melanoma. Cancer Res 64: 7002-10.  Sun M, Wang G, Paciga JE, Feldman RI, Yuan ZQ, Ma XL, Shelley SA, Jove R, Tsichlis PN, Nicosia SV, Cheng JQ (2001). AKT1/PKBalpha kinase is frequently elevated in human cancers and its constitutive activation is required for oncogenic transformation in NIH3T3 cells. Am J Pathol 159: 431-7.  Tomayko MM, Reynolds CP (1989). Determination of subcutaneous tumor size in athymic (nude) mice. Cancer Chemother Pharmacol 24: 148-54.  Villunger A, Michalak EM, Coultas L, Müllauer F, Böck G, Ausserlechner MJ, Adams JM, Strasser A (2003). p53- and drug-induced apoptotic responses mediated by BH3-only proteins puma and noxa. Science 302: 1036-8.  Willis SN, Adams JM (2005). Life in the balance: how BH3-only proteins induce apoptosis. Curr Opin Cell Biol 17: 617-25.  Wolkersdörfer GW, Morris JC, Ehninger G, Ramsey WJ (2004). Trans-complementing adenoviral vectors for oncolytic therapy of malignant melanoma. J Gene Med 6: 652-62.  Yang L, Dan HC, Sun M, Liu Q, Sun XM, Feldman RI, Hamilton AD, Polokoff M, Nicosia SV, Herlyn M, Sebti SM, Cheng JQ (2004). Akt/protein kinase B signaling inhibitor-2, a selective small molecule inhibitor of Akt signaling with antitumor activity in cancer cells overexpressing Akt. Cancer Res 64: 4394-9.  84  Yu J, Wang Z, Kinzler KW, Vogelstein B, Zhang L (2003). PUMA mediates the apoptotic response to p53 in colorectal cancer cells. Proc Natl Acad Sci U S A 100: 1931- 6.  Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B (2001). PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell 7: 673-82.  Zong WX, Lindsten T, Ross AJ, MacGregor GR, Thompson CB (2001). BH3-only proteins that bind pro-survival Bcl-2 family members fail to induce apoptosis in the absence of Bax and Bak. Genes Dev 15: 1481-6.     85 4.   NUCLEAR FACTOR KAPPA B SUBUNIT P50 PROMOTES MELANOMA ANGIOGENESIS BY UPREGULATING INTERLEUKIN-6 EXPRESSION3 4.1 INTRODUCTION The NF-κB family of transcription factors regulates expression of multiple genes involved in the immune and stress responses, inflammation, cell proliferation, and cell survival.  Constitutive activation of NF-κB is associated with diseases of chronic inflammation and immune dysfunction, as well as cancer (Kumar et al., 2004; Yamamoto et al., 2001).  There are five NF-κB proteins [p105/p50 (NF-κB1), p100/p52 (NF-κB2), RelA (p65), RelB, and c-Rel], which dimerize to form the various hetero- and homo-dimeric NF-κB transcriptional complexes (Gilmore, 1990).  Different NF-κB dimers are activated in different situations, allowing NF-κB to regulate target gene transcription with great degree of contextual specificity (Ghosh and Karin, 2002).  The vast majority of studies of NF-κB in cancer have focused solely on the p50/p65 (or p50/RelA) heterodimer, which is considered to be the "classical" form of NF- κB.  However, an increasing number of reports indicate that nuclear p50 protein expression may be elevated in cancer tissue independently of p65.  Overexpression of the p50 subunit in particular and evidence of p50 homodimer activity has been observed in cervical carcinoma (Prusty et al., 2005), squamous cell carcinomas (Budunova et al., 1999; Mishra et al., 2006),  nasopharyngeal  carcinoma (Thornburg et al., 2003),  non- small cell lung carcinoma (Mukhopadhyay et al., 1995), and lymphoma (Kurland et al.,  3 A version of this chapter has been accepted for publication.  Karst AM, Gao K, Nelson C, Li G. Nuclear factor kappa b subunit p50 promotes melanoma angiogenesis by upregulating interleukin-6 expression. Revised manuscript submitted to Int J Cancer on July11, 2008; 24 pages.  86 2001; Mathas et al., 2005).  Our lab has previously reported that p50 expression correlates strongly with melanoma progression in a panel of human melanoma biopsies and that nuclear p50 expression is an independent predictor of 5-year survival in melanoma patients (Gao et al., 2006).  In the present report we show that p50 overexpression in melanoma cells strongly induces interleukin-6 (IL-6) upregulation.  IL-6 is a pleiotropic cytokine most notably involved in immune responses and inflammation, and often upregulated in cancer (Sehgal et al., 1995).  In melanoma, IL-6 is has been shown to inhibit early-stage growth but enhance metastatic potential during late-stage growth (Komyod et al., 2007; Lu and Kerbel, 1993; Sun et al., 1992).  Here, we demonstrate that upregulation of IL-6, mediated by p50 overexpression in melanoma cells, enhances the growth of endothelial cells in vitro and promotes melanoma angiogenesis in vivo.  4.2 MATERIALS AND METHODS 4.2.1 Cell culture Cell culture conditions for melanoma cells are described in chapter 2.2.3.  Human umbilical vein endothelial cells (HUVECs), a kind gift from Dr. A. Karsan (BC Cancer Research Centre, Vancouver, BC), were cultured in Kaighn’s Modified Ham’s F-12K medium (Mediatech, Manassas, VA) supplemented with endothelial cell growth supplement (BD Biosciences, Mississauga, ON, Canada) and 10% fetal bovine serum. HUVECs were maintained in a humidified atmosphere of 5% CO2 and 37°C.    87 4.2.2 Transfection Expression plasmids encoding human NF-κB p50 and ATF3 were kindly provided by Dr. R.P. Ricciardi (University of Pennsylvania, Philadelphia, PA) and Dr. D.F. Steiner (University of Chicago, Chicago, IL), respectively.  Transient and stable transfections were performed using Effectene reagent (Qiagen, Mississauga, ON, Canada).  NF-κB p50 stably-transfected MMRU cells were selected by adding 800 µg/ml of G418 (Sigma- Aldrich) to the culture medium for 2-3 weeks.  Single clones were then transferred to individual plates and maintained in 400 µg/ml of G418.  Clones were tested by western blot to verify NF-κB p50 overexpression. 4.2.3 Lentiviral transduction Mission TRC shRNA lentiviral transduction particles expressing short hairpin RNA (shRNA) targeting NF-κB p105 and lentiviral negative control particles were purchased from Sigma-Aldrich.  Transduction of MMRU cells was performed according to the manufacturer's instructions and stably-transduced cells were selected by adding 1 µg/ml of puromycin (Sigma-Aldrich) to the culture medium for 1 week.  Single clones were then transferred to individual plates and tested by western blot to verify NF-κB p50 knockdown. 4.2.4 Western blot Western blot analysis was performed as described in chapter 3.2.10 using the following primary antibodies: polyclonal rabbit anti-human p105/p50 (Santa Cruz Biotechnology, Santa Cruz, CA) and monoclonal mouse anti-actin (Sigma-Aldrich).  Signals were detected using an Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE).  88 4.2.5 Enzyme-linked immunosorbent assay (ELISA) To measure IL-6 protein levels in conditioned media, a human IL-6 ELISA kit from eBioscience (San Diego, CA) was used according to the manufacturer’s instructions. 4.2.6 Sulforhodamine B (SRB) cell growth assay To compare cell growth rates, cells were seeded in 96-well plates.  At each time point, cell density was quantified by SRB assay, as described in chapter 3.2.11.  The initial time point (0 h) for the cell growth rate assay was measured by fixing cells immediately after they had attached to the tissue culture plate, 8 h after seeding.  Subsequent time points were measured by fixing cells 24, 48, and 72 h later.  Relative rates of cell growth were calculated as a ratio of the cell density at each time point over the cell density at 0 h. 4.2.7 cDNA microarray analysis and real-time qPCR Total cellular RNA was extracted using Trizol (Invitrogen) and evaluated with a 2100 Bioanalyzer (Agilent Technologies, Mississauga, ON, Canada).  A microarray of 21,000 70-mer human oligo probes representing 21,000 genes (Operon Biotechnologies, Huntsville, AL) printed in duplicate onto aminopropylsilane-coated slides was supplied by the Microarray Facility of the Prostate Centre at Vancouver General Hospital.  The arrays were hybridized with three DNA dendrimer probes generated from 10 µg of RNA from triplicate samples of the p50 or vector clones.  Microarrays were scanned with GenePix Autoloader 4200AL (Molecular Devices, Sunnyvale, CA).  Signal quality and quantity were assessed using Imagene 6.0.1 software (BioDiscovery, El Segundo, CA). The raw data was analyzed using GeneSpring 7.1 software (Agilent Technologies).  Gene expression profiles were compared by t-test and hierarchical clustering with Pearson correlation.  Real-time qPCR reactions were performed in triplicate with SYBR Green  89 PCR Master Mix and using a 7900HT Real-time qPCR System thermal cycler (Applied Biosystems, Foster City, CA).  The primers for real-time qPCR are listed in Table 4.1. Table 4.1 Primer pairs used for qPCR. Target gene  Primer sequence IL-6     Forward 5’-AGTCCAGCCTGAGGGCTCTT-3’  Reverse 5’-GCCCAGTGGACAGGTTTCTG-3’ VIPR1   Forward 5’-TCCAAGTCTCAGTGGCTTCATCT-3’  Reverse 5’-GGAGGGCAGCTCTTGATTCC-3’ SLC14A1   Forward 5’-CTGTTCACGGCCTATCTTGGA-3’  Reverse 5’-GGAACAATAGCGTGGCCAAA-3’ Megsin Forward 5’-GCCAAAGTGGAGCGAGTTGA-3’  Reverse 5’-CGTTCTTGATTTTGCCATGTGT-3’ IL-1α Forward 5’-CATGAAGGCTGCATGGATCA-3’  Reverse 5’-TGGTTGCTACTACCACCATGCT-3’ MYBPC1 Forward 5’-TGCCAGAGAATCCTGTTTATCAATAA-3’  Reverse 5’-TTACGAAGAGCTCAGTGGAACATT-3’ TRIM9 Forward 5’-GATGGCAACGGTGGTCAATT-3’  Reverse 5’-TGACCCGAGCGTTGTATGTG-3’ NGAL Forward 5’-CAGGAGAACTTCATCCGCTTCT-3’  Reverse 5’-TGTGCACTCAGCCGTCGATA-3’ S100B    Forward 5’-GGCGATGGAGACCCTCATC-3’  Reverse 5’-CGTCTGCAGCAGCTCTTTCA-3’ SULF1 Forward 5’-AATGCTGCCCATCCACATG-3’  Reverse 5’-CATGTTATACAGCCTCTCCACAGAA-3’ Actin Forward 5’-GCTCTTTTCCAGCCTTCCTT-3’  Reverse 5’-CGGATGTCAACGTCACACTT-3’  4.2.8 Conditioned medium Cells were cultured in 6-well tissue culture plates with 1.5 ml of fresh, complete medium for 24 h.  Conditioned medium was then collected and either added directly to HUVECs (for HUVEC growth assays) or stored at -80°C for subsequent protein quantification by ELISA.  Unconditioned medium containing recombinant IL-6 protein (eBioscience) was used as a positive control for IL-6-mediated stimulation of HUVEC growth.  90 4.2.9 Chromatin immunoprecipitation (ChIP) assay Formaldehyde-fixed cells were immunoprecipitated with human anti-p50 antibody (Santa Cruz Biotechnology) and the associated genomic DNA was analyzed by PCR using two sets of primers, each designed to amplify a different region of the IL-6 promoter consensus sequence (300 – 424 bp). Primer pair 1 (forward 5’- TGAGCCTCAGACATCTCCAG-3’, reverse 5’-AGACATGCCAAAGTGCTGAG-3’) generated a 300 bp product.  Primer pair 2 (forward 5’-TCATCGGAAAATCCCACATT- 3’, reverse 5’-AAGACATGCCAAAGTGCTGA-3’) generated a 242 bp product. 4.2.10 In vivo angiogenesis assay and immunofluorescent staining NF-κB p50, sh-control, or sh-p50 cells (1 × 106), supported by Matrigel (BD Biosciences), were implanted subcutaneously into the flanks of 6-week-old male nude mice.  The animals were sacrificed 10 days later whereupon the matrigel plugs were excised, photographed, and immediately embedded in Tissue-Tek O.C.T. compound (Sakura Finetek USA, Torrance, CA).  Embedded tissues were frozen at -80°C for 1 h, then sectioned (5 µm thickness) using a CM1850 cryostat (Leica Microsystems, Richmond Hill, ON, Canada) and applied to glass slides.  To visualize blood vessels, tissue sections were fixed with 4% paraformaldehyde and immunofluorescent staining for CD31 expression was performed using anti-mouse CD31 antibody (BD Biosciences) and Alexa Fluor 488 secondary antibody (Invitrogen).  Sections were counterstained with Hoechst 33258 (Sigma-Aldrich).  Fluorescent signals were visualized using a DM400 microscope (Leica Microsystems) equipped with a Retiga 1300i CCD digital camera and Openlab 4.0.2 software (Improvision, Waltham, MA).  CD-31 staining shows the extent  91 of host angiogenesis induced by p50, sh-control, and sh-p50 implanted cells.  Hoechst nuclear staining indicates the overall cell density of each matrigel plug.  4.3 RESULTS 4.3.1 NF-κB p50 does not affect cell growth rate To assess the effects of p50 overexpression on melanoma cells, the metastatic melanoma cell line MMRU was transfected with an expression plasmid encoding human NF-κB p50 or an empty vector control.  A stably-transfected clone was established by G418 selection and western blot was used to verify overexpression of p50 protein (Figure 4.1A).  A sulforhodamine B (SRB) assay was used to compare growth rates of the p50 and vector clones, and no significant differences were detected (Figure 4.1B).  This result indicates that NF-κB p50 expression levels do not affect the growth rate of MMRU cells in vitro. 4.3.2 NF-κB p50 upregulates IL-6 gene expression To determine the global effects of p50 overexpression and to identify potential downstream targets of p50, we compared the gene expression profiles of p50 and vector clones by cDNA microarray analysis.  Of 21,000 genes screened, 179 were found to have significantly altered expression (≥ 2-fold difference) in the p50 clone compared to the vector clone (Figure 4.2 and Table 4.2).  The 10 genes with most altered expression were: IL-6, VIPR1, SLC14A1, Megsin, IL-1α, MYBPC1, TRIM9, NGAL, S100B, and SULF1. The most highly upregulated gene was IL-6, which exhibited a 15-fold increase in transcription following p50 overexpression (Figure 4.3).  To validate the cDNA microarray results, MMRU cells were transfected with p50 or vector and analyzed by real-time qPCR.  A similar trend of altered gene expression was observed in MMRU cells  92 Figure 4.1 Effect of NF-κB p50 overexpression on melanoma cell growth. (A) Western blot analysis of p50 protein expression in MMRU cells stably transfected with an expression plasmid encoding human NF-κB p50 or an empty vector control. (B) Comparison of the cell growth rates of p50 and vector clones, determined by SRB growth cell growth assay.    93 Figure 4.2 cDNA microarray analysis of the NF-κB p50 and vector stable clones. Cluster analysis indicated that 179 genes (of 21,000 screened) were differentially expressed between the p50 and vector clones.     94 Table 4.2 179 genes differentially expressed in the NF-κB p50-overexpressing clone compared to the vector clone. Probe ID Gene name P value Fold change H200010620 Interleukin-6  (IL-6) 0.0006 15.4 H200020274 Vasoactive intestinal polypeptide receptor 1 (VIPR1) 0.0013 10.2 H200007987 Solute carrier family 14, member 1 (SLC14A1) 0.0144 8.57 H200013276 Megsin 0.0271 7.60 H200000451 Interleukin-1 alpha (IL-1α) 0.0328 6.74 H200014828 Protein MICAL-2 0.0054 5.66 H200016225 APOBEC3G  0.0189 5.59 H200016159 Transcription factor 19  0.0367 5.56 H200011805 Normal mucosa of esophagus specific gene 1 protein 0.0369 5.32 H200012604  0.0017 5.31 H200004964  0.0056 5.17 H200019450  0.0060 5.05 H200012985 Cardiotrophin-like cytokine 0.0283 4.75 H200016773 Integrin alpha-2 precursor 0.0055 4.62 H200008209 Neuron navigator 3 0.0182 4.48 H200010451  0.0067 3.86 H200004451 S100 calcium-binding protein A2  0.0130 3.71 H200001031 Synaptopodin 0.0031 3.57 H200002766 Probable ubiquitin carboxyl-terminal hydrolase CYLD 0.0124 3.46 H200008438 Unc-112 related protein 1  0.0015 3.43 H200013648 Smoothelin 0.0002 3.36 H200001630 Protein-glutamine gamma-glutamyltransferase  0.0055 3.32 H200007979 Ephrin type-A receptor 2 precursor 0.0148 3.31 H200001892 CD166 antigen precursor   0.0030 3.29 H200011012 Tumor necrosis factor ligand superfamily member 7 0.0310 3.21 H200002394  0.0459 3.20 H200020254 Intermediate filament protein syncoilin  0.0041 3.11 H200013413 Low-density lipoprotein receptor-related protein 3  0.0013 3.01 H200008365 Cyclin-dependent kinase inhibitor 1 (p21)  0.0004 2.99 H200004941 Leupaxin.  0.0105 2.97 H200002230 Lymphocyte specific adapter protein Lnk  0.0042 2.95 H200003337 Keratin, type II cytoskeletal 7  0.0117 2.95 H200013959 Fibroblast growth factor-20  0.0214 2.92 H200009616  0.0080 2.76 H200008363 Urokinase plasminogen activator surface receptor  0.0368 2.74 H200002718 Amino acid transporter system A1 0.0210 2.72 H200000253 Natural killer cells protein 4 precursor 0.0054 2.70 H200011441  0.0439 2.70 H200009647 Tissue factor pathway inhibitor 2 precursor (TFPI- 2) 0.0216 2.64 H200006072 Ribonucleoside-diphosphate reductase M2 chain  0.0071 2.63 H200007562 Numb protein homolog 0.0060 2.62 H200000264  0.0258 2.61 H200008377  0.0377 2.59 H200020138  0.0396 2.59 H200000265 Vitamin K-dependent protein Z precursor 0.0133 2.57  95 Probe ID Gene name P value Fold change H200005768  0.0075 2.56 H200008789 Mitochondrial ribosomal protein S24 0.0056 2.53 H200002641 Protein-tyrosine sulfotransferase 1  0.0004 2.49 H200020015  0.0134 2.49 H200011394  0.0328 2.48 H200007011 Interstitial collagenase precursor  0.0088 2.47 H200007725 Atrial natriuretic peptide receptor A precursor 0.0332 2.45 H200013798 Mitogen-activated protein kinase kinase kinase 5  0.0254 2.43 H200001880 Tumor necrosis factor receptor superfamily member Fn14 0.0027 2.41 H200016556 Integrin alpha-3 precursor  0.0047 2.39 H200005863 Vascular endothelial growth factor A precursor 0.0021 2.37 H200004003 G1/S-Specific cyclin E2 0.0405 2.36 H200001697 CYR61 protein precursor   0.0071 2.34 H200006989 G1/S-specific cyclin D1  0.0011 2.34 H200009948 Fat-specific protein FSP27 homolog 0.0385 2.34 H200009591  0.0452 2.32 H200015006 Related RAS viral (r-ras) oncogene homolog 2 0.0122 2.28 H200009934 S100 calcium-binding protein A16  0.0054 2.27 H200007438 Intimal thickness-related receptor 0.0044 2.26 H200011636 Nef-associated factor 1  0.0062 2.25 H200006805 DNA polymerase gamma subunit 1 0.0479 2.24 H200009351  0.0193 2.23 H200002747  0.0009 2.22 H200016224 NKG2D ligand 2 precursor 0.0131 2.22 H200006754 Keratin, type I cytoskeletal 15  0.0344 2.22 H200003873 Cell division cycle 7-related protein kinase  0.0082 2.21 H200009668  0.0140 2.20 H200020163 Putative HIF-prolyl hydroxylase PH-4  0.0224 2.16 H200003655 Transmembrane 4 superfamily member 7 (Novel antigen 2) 0.0097 2.15 H200014627 Smad ubiquitination regulatory factor 2  0.0112 2.15 H200007997 Poliovirus receptor precursor  0.0014 2.14 H200013928 Pleckstrin homology-like domain family A member 2 0.0011 2.14 H200007065 Cytochrome b5 0.0021 2.10 H200001239  0.0204 2.10 H200004832  0.0226 2.08 H200005879 Bone morphogenetic protein 2 precursor (BMP-2) (BMP-2A). 0.0329 2.08 H200006069 Aldose reductase  0.0019 2.08 H200011606 Collagen alpha 1(VI) chain precursor 0.0036 2.08 H200007050 Rho GDP-dissociation inhibitor 2  0.0320 2.07 H200005065 Vacuolar protein sorting 13A (Chorein) 0.0424 2.04 H200011786 SET binding factor 1 0.0393 2.04 H200017722 Carboxypeptidase A6 precursor  0.0138 2.03 H200018928  0.0147 2.03 H200004256 Cell division cycle associated protein 4  0.0400 2.01 H200006120 Connective tissue growth factor precursor  0.0268 2.00 H200011490 Latrophilin 1 precursor 0.0009 0.50 H200003924  0.0115 0.50 H200008568 POU domain, class 2, transcription factor 1 0.0143 0.50 H200015843  0.0029 0.50 H200005932 Aldehyde dehydrogenase family 7 member A1  0.0052 0.49  96 Probe ID Gene name P value Fold change H200017492 Calcium/calmodulin-dependent protein kinase ID 0.0360 0.49 H200006909 Semaphorin 3B precursor 0.0163 0.49 H200013933 Lipopolysaccharide-binding protein precursor (LBP) 0.0156 0.49 H200004493  0.0160 0.48 H200001987 Chondroitin beta1,4 N- acetylgalactosaminyltransferase 0.0027 0.48 H200020576 Mas-related G-protein coupled receptor member X3 0.0163 0.48 H200018683  0.0202 0.48 H200001527  0.0237 0.47 H200004014  0.0207 0.47 H200011538 Sodium- and chloride-dependent transporter XTRP3 0.0402 0.47 H200014523 Protein phosphatase 1 0.0258 0.47 H200010402 Pyruvate carboxylase, mitochondrial precursor  0.0250 0.46 H200004468 Integrin alpha-4 precursor 0.0243 0.46 H200006870 Polycystin 2  0.0065 0.46 H200018060 Prostacyclin synthase  0.0311 0.46 H200001150 Arrestin domain containing 4  0.0126 0.46 H200019523 Ghrelin precursor  0.0449 0.46 H200014839 Receptor protein-tyrosine kinase erbB-3 precursor 0.0006 0.46 H200006348 Sodium/potassium-transporting ATPase beta-3 chain 0.0001 0.46 H200004179  0.0370 0.46 H200013135 Urocortin precursor 0.0065 0.45 H200000972 Formin binding protein 2 (srGAP2) 0.0418 0.45 H200013730 Fucose-1-phosphate guanylyltransferase  0.0315 0.45 H200001712  0.0010 0.44 H200015190 Laminin gamma-1 chain precursor (Laminin B2 chain) 0.0069 0.44 H200000021 Zinc-alpha-2-glycoprotein precursor  0.0320 0.44 H200003440  0.0187 0.44 H200005214 Insulin-induced protein 1. 0.0304 0.43 H200003487 GPI deacylase 0.0152 0.43 H200016671  0.0087 0.42 H200008291  0.0434 0.42 H200000621 Angiopoietin-1 precursor (ANG-1).  0.0036 0.41 H200020855  0.0240 0.41 H200006658 Lipoma HMGIC fusion partner-like 2  0.0154 0.39 H200007274  0.0227 0.39 H200006744 RNA-binding protein with multiple splicing  0.0244 0.39 H200004873 Tumor necrosis factor receptor superfamily member 19 precursor 0.0347 0.39 H200008549 Glutamate receptor, ionotropic kainate 1 precursor  0.0074 0.39 H200017417 Matrix Gla-protein precursor 0.0306 0.39 H200006258 Decorin precursor  0.0200 0.39 H200020667  0.0388 0.38 H200006882 Cbp/p300-interacting transactivator 2 0.0325 0.37 H200014435  0.0474 0.37 H200001979 Autophagy protein 5-like (APG5-like)  0.0448 0.36 H200005990 MARCKS-related protein  0.0145 0.35 H200009209    0.0143 0.35 H200014019 B-cell lymphoma 6 protein  0.0436 0.35  97 Probe ID Gene name P value Fold change H200003394  0.0076 0.35 H200006603 OX-2 membrane glycoprotein precursor  0.0026 0.34 H200008450 Scavenger receptor class B member 1  0.0188 0.33 H200016567 Fc receptor homolog expressed in B cells 0.0246 0.33 H200018140 Sulfate transporter  0.0192 0.33 H200006831 S100 calcium-binding protein A4  0.0004 0.33 H200006854 Versican core protein precursor  0.0074 0.32 H200016273  0.0096 0.31 H200006597 Myocyte-specific enhancer factor 2C.  0.0028 0.31 H200002118  0.0022 0.31 H200017874 Oxysterol binding protein-related protein 10 0.0395 0.31 H200010223 Transcription factor SOX-5 0.0156 0.30 H200007312 Syndecan-3 (SYND3). 0.0021 0.28 H200001575  0.0423 0.28 H200009536 S-100 protein, alpha chain  0.0146 0.28 H200009248 SLIT and NTRK-like protein 6  0.0201 0.27 H200001272 Breast carcinoma amplified sequence 3  0.0006 0.27 H200006318  0.0282 0.25 H200019015 T-box transcription factor TBX2  0.0013 0.25 H200001295 Guanine nucleotide-binding protein gamma-7 subunit. 0.0195 0.25 H200010192 Endothelin B receptor precursor  0.0128 0.25 H200012768 Alpha-methylacyl-CoA racemase  0.0122 0.25 H200003992  0.0046 0.24 H200006197 NDRG1 protein  0.0095 0.24 H200003683 Fatty acid-binding protein, brain (B-FABP)  0.0031 0.22 H200000350 Colipase precursor 0.0216 0.22 H200006566 Receptor-type protein-tyrosine phosphatase zeta 0.0067 0.21 H200013940 Chitinase 3-like protein 2 precursor  0.0247 0.20 H200005066 Collagen alpha 3(IX) chain precursor.  0.0040 0.19 H200005298 Scavenger receptor class F member 2 precursor  0.0256 0.19 H200018207 Aldo-keto reductase family 1 member C1  0.0224 0.17 H200007868 Myosin-binding protein C, slow-type (MYBPC1) 0.0413 0.17 H200005999 Tripartite motif-containing protein 9 (TRIM9) 0.0232 0.16 H200014979 Neutrophil gelatinase-associated lipocalin (NGAL) 0.0317 0.12 H200007028 S100 calcium binding protein B (S100B) 0.0015 0.11 H200015420  0.0027 0.11 H200005722 Extracellular sulfatase 1 (SULF1) 0.0226 0.11   98 Figure 4.3 IL-6 transcription is highly upregulated in NF-κB p50 stable clones.  The top ten genes up/down-regulated in the p50 stable clone, compared to the vector clone, are shown.  99 following transient p50-overexpression (Figure 4.4).  IL-6 was again the most highly transcribed gene of those tested, showing a 16-fold induction.  To verify that enhanced IL-6 transcriptional activity actually led to elevated IL-6 protein translation, we tested conditioned medium from the p50 and vector clones by ELISA.  Indeed, conditioned medium from the p50 clone contained almost 5 times more IL-6 protein than that from the vector clone (Figure 4.5A).  To ascertain whether p50-mediated IL-6 upregulation was specific to the MMRU cell line, we transiently expressed p50 in another metastatic melanoma cell line, MMAN, then analyzed the conditioned medium for IL-6 protein expression.  In MMAN cells, p50 expression induced a 52% increase in IL-6 levels (Figure 4.5B).  To determine whether p50 could be directly regulating IL-6 gene transcription, we tested the ability of p50 protein to bind the IL-6 gene promoter in vivo using a chromatin immunoprecipitation (ChIP) assay.  Formaldehyde-fixed MMRU cells were immunoprecipitated with p50 antibody and the associated genomic DNA was analyzed by PCR with two separate primer pairs, each spanning a different region of the IL-6 promoter (Figure 4.6).  Binding of p50 to the consensus sequence of the IL-6 promoter was clearly detected with both primer pairs, but not with control primers designed to amplify a region upstream of the IL-6 promoter.  This data indicates that p50 is able to bind directly to the IL-6 promoter sequence in vivo and may play a role in activating IL-6 gene transcription in melanoma cells. 4.3.3 Activating transcription factor 3 (ATF3) inhibits NF-κB p50-mediated IL-6 upregulation The transcription factor ATF3 has recently been shown to cooperate with the NF-κB subunit c-Rel to negatively regulate IL-6 gene transcription (Gilchrist et al., 2006).  We  100 Figure 4.4 NF-κB p50 induces IL-6 transcription. Changes in mRNA transcript levels of MMRU cells upon p50 overexpression.  Cells were transiently transfected with p50 or vector for 48 h, then compared by real-time qPCR analysis.  101 Figure 4.5 NF-κB p50 induces IL-6 protein expression.  IL-6 protein levels in conditioned medium from (A) p50 and vector clones, and (B) MMAN cells transiently transfected with p50 or vector control.  Protein levels were measured by ELISA.  102 Figure 4.6 NF-κB p50 protein binds to the IL-6 gene promoter consensus sequence.  In vivo binding of endogenous p50 protein to the IL-6 promoter in MMRU cells was detected by ChIP assay.  The two sets of IL-6 primers amplify different regions of the promoter consensus sequence.  The control primers amplify a region upstream of the IL-6 promoter.  103 therefore tested whether ATF3 expression could interfere with p50-mediated IL-6 induction.  MMRU cells were co-transfected with expression plasmids encoding NF-κB p50 and ATF3 or an empty vector control.  The resulting effect on IL-6 expression was determined by RT-PCR and ELISA.  While p50 efficiently induced IL-6 upregulation at both the transcriptional (Figure 4.4) and translational (Figure 4.5) levels, co-expression of ATF3 significantly inhibited this IL-6 induction (Figures 4.7 and 4.8).  This result suggests that ATF3 may be a negative regulator of NF-κB p50/IL-6 signaling. 4.3.4 Silencing NF-κB p105/p50 expression inhibits IL-6 expression To assess the effects of silencing endogenous NF-κB p50 expression in melanoma, MMRU cells were transduced with a shRNA lentivirus targeting p105 (the precursor protein to p50), or a negative control lentivirus, and stable clones were generated by puromycin selection.  Western blot analysis was used to verify knockdown of both p105 and p50 protein expression in the selected clone (Figure 4.9A).  To determine whether loss of p50 expression had an impact on cell growth, we compared the growth rates of the sh-p50 and sh-control clones by SRB cell growth assay.  Again, p50 expression had no significant effect on MMRU cell growth rates (Figure 4.9B).  Since p50 overexpression upregulates IL-6 expression, we expected that knocking down endogenous p105/p50 expression would have the opposite effect on IL-6 levels.  Indeed, IL-6 transcription was found to be downregulated in sh-p50 clones compared to sh-control clones (Figure 4.10A).  Likewise, conditioned medium from the sh-p50 clone contained only 38% of the IL-6 protein detected in conditioned medium from the sh-control clone (Figure 4.10B). These data provide strong evidence supporting our hypothesis that NF-κB p50 regulates IL-6 gene expression in melanoma.  104 Figure 4.7 ATF3 expression inhibits p50-mediated upregulation of IL-6 transcripts. Transient co-expression of ATF3 and p50 abrogated the increase in IL-6 transcripts seen in cells co-expressing p50 and vector control.  Transcript levels were determined by determined by RT-PCR.  105 Figure 4.8 ATF3 expression inhibits p50-mediated upregulation of IL-6 protein expression.  Transient co-expression of ATF3 and p50 resulted in lower levels of IL-6 protein in conditioned medium, compared to cells co-expressing p50 and vector control. Protein levels were determined by ELISA.   106 Figure 4.9 Characterization of the sh-p50 stable clone. (A) Western blot analysis of p50 protein expression in MMRU cells stably transduced with a shRNA lentivirus targeting p105 (the precursor protein to p50) or a negative control lentivirus. (B) Comparison of cell growth rates of the sh-p50 and sh-control clones, determined by SRB cell growth assay.  107 Figure 4.10 Knockdown of p50 expression reduced IL-6 expression.  (A) IL-6 mRNA levels in the sh-p50 and sh-control clones, determined by RT-PCR. (B) IL-6 protein levels in conditioned medium from the sh-p50 and sh-control clones, determined by ELISA.  108 4.3.5 NF-κB p50 expression by MMRU cells stimulates endothelial cell growth To determine whether p50-overexpressing MMRU cells could promote the growth of endothelial cells, we used conditioned medium from the p50 and vector clones to stimulate HUVECs and measured their relative growth rates using SRB cell growth assays.  HUVECs exposed to conditioned medium from the p50 clone grew 40% faster than those treated with conditioned medium from the vector clone and this effect was abrogated by addition of anti-IL-6 antibody (Figure 4.11A).  Unconditioned medium containing recombinant IL-6 protein, used as a positive control, also stimulated HUVEC growth.  Meanwhile, conditioned medium from the sh-p50 clone was unable to promote HUVEC growth compared to that from the sh-control clone (Figure 4.11B).  In fact, silencing the p50 expression of MMRU cells resulted in a 45% reduction in their ability to stimulate endothelial cell proliferation.  Since IL-6 reportedly upregulates vascular endothelial growth factor (VEGF) to induce angiogenesis (Cohen et al., 1996), we used RT-qPCR to assay for differences in VEGF transcription levels between the p50 clone, sh-p50 clone, and vector controls.  Cells stably overexpressing p50 exhibited a 1.5-fold increase in VEGF transcripts compared to the vector control (Figure 4.12).  Conversely, stable knockdown of p50 expression suppressed VEGF transcription by 1.4-fold.  This result indicates that NF-κB p50 alone may be able to regulate VEGF expression. 4.3.6 NF-κB p50 expression in MMRU cells promotes angiogenesis We next investigated whether p50 expression in MMRU cells could promote the formation of new blood vessels in vivo.  To this end, suspensions of sh-p50, sh-control, and p50 clone cells, supported by matrigel basement membrane matrix, were implanted subcutaneously into the flanks of 6-week-old nude mice.  Ten days following  109 Figure 4.11 NF-κB p50 expression promotes endothelial cell growth in vitro. (A) Conditioned medium from the p50 clone efficiently stimulates HUVEC growth, determined by SRB cell growth assay. The effect is blocked by addition of an antibody against IL-6. (B) Conditioned medium from the sh-p50 clone inhibited HUVEC growth compared to conditioned medium from the sh-control clone, determined by SRB cell growth assay.  110 Figure 4.12 NF-κB p50 expression induces VEGF upregulation.  VEGF mRNA transcript levels in the p50 and sh-p50 clones, compared to vector and sh-control clones, respectively; determined by RT-qPCR.    111 implantation, the matrigel plugs were excised, cryopreserved, and sectioned on slides. The sectioned tissues were analyzed by immunofluorescence for expression of the endothelial cell marker CD31 in order to quantify the extent of murine blood vessel formation induced by the melanoma implants.  Superficial examination of the excised matrigel plugs revealed varying levels of vascularization (Figure 4.13), with very few blood vessels observed in the sh-p50 plugs and more extensive vascularization apparent in the sh-control and p50 plugs.  CD31 staining revealed that the p50 plugs contained denser, more extensive micro vessel networks compared to control plugs (Figure 4.13B). The sh-p50 plugs, on the other hand, exhibited a scanty pattern of endothelial structures, indicative of less extensive angiogenesis supporting these implants.  Our results clearly indicate that NF-κB p50 expression plays a role in melanoma tumor angiogenesis.  4.4 DISCUSSION We have previously reported that NF-κB p50 protein expression strongly correlates with melanoma tumor progression, as determined by tissue microarray analysis of human melanoma biopsies (Gao et al., 2006).  We also reported that strong nuclear p50 expression is an independent prognostic factor predicting for worse 5-year survival of melanoma patients with thick (>2.0 mm) lesions (Gao et al., 2006).  In the present study we sought to characterize the effects of p50 expression in melanoma so as to understand how p50 may confer a survival advantage to melanoma tumor cells. We began by stably overexpressing p50 in a melanoma cell line (MMRU) to see whether p50 enhanced the rate of melanoma cell growth, but found no such effect. We next investigated whether p50 overexpression induced a change in the gene expression  112 Figure 4.13 NF-κB p50 expression promotes angiogenesis in vivo. (A) Matrigel was mixed with p50, sh-control, or sh-p50 cells and implanted subcutaneously in nude mice. Photographs of matrigel plugs excised from mice after 10 days of growth in vivo. (B) Immunofluorescent staining of matrigel plugs for expression of the endothelial cell marker CD-31 (left panels) shows the extent of host angiogenesis induced by implanted p50, sh-control, and sh-p50 clone cells.  Hoechst nuclear staining (right panels) indicates the overall cell density of each matrigel plug.  113 profile of MMRU stable clones.  Using cDNA microarray analysis, we discovered that IL-6 transcription was highly upregulated in the p50 clone compared to the vector clone, and this finding was confirmed in separate experiments using RT-qPCR.  ELISA analysis of conditioned medium from the p50 and vector clones indicated that IL-6 protein production was correspondingly upregulated.  We next tested another metastatic melanoma cell line (MMAN) and found p50 also induced IL-6 expression in these cells. This finding is in agreement with a report by Molnár et al in which IL-6 was secreted by highly metastatic melanoma cell lines (Molnár et al., 2000).  In that case, however, IL-6 production was attributed to the effects of endogenous histamine.  Our data is also consistent with reports of high serum IL-6 levels in patients with metastatic melanoma (Mouawad et al., 1996; Soubrane et al., 2005).  Serum IL-6 levels have been shown to correlate with high tumor burden and poor clinical outcome in melanoma patients (Mouawad et al., 1996) and found to be predictive of both melanoma recurrence (Mouawad et al., 2002) and overall survival (Soubrane et al., 2005; Tas et al., 2005). Although IL-6 secretion by melanoma cells is not a new phenomenon, the observation that IL-6 transcription is driven by expression of the NF-κB subunit p50 is a novel and unprecedented finding. It was somewhat surprising that no classical NF-κB target genes, other than IL-6, were upregulated following p50 overexpression.  However, this may be explained by the fact that most NF-κB “target genes” are in fact transcriptional targets of the p50/p65 dimer or other NF-κB heterodimers.  Overexpression of the p50 subunit alone may not support the activation of such heterodimers, which are regulated by IKK and IκB proteins, and thus require very different mechanisms of activation.  114 In further support of our hypothesis that p50 regulates IL-6 transcription, we showed that p50 protein can physically interact with the IL-6 gene promoter in vivo, using a ChIP assay.  While our data clearly indicated p50 binding, it remains to be determined whether a transcriptional co-activator is required for p50 to induce IL-6 transcription in melanoma cells.  This is most likely, since the p50 subunit lacks a transactivation domain (Gilmore, 2006).  Potential co-activators include Bcl-3, p300, and CREB binding protein (CBP).  The oncoprotein Bcl-3 is a well-known transcriptional co-activator of p50 homodimers (Bours et al., 1993; Fujita et al., 1993; Watanabe et al., 1997) and has been shown to associate with p50 homodimers in several types of cancer cells (Cogswell et al., 2000; Mathas et al., 2005; Mishra et al., 2006; Thornburg et al., 2003).  The closely related co-activating proteins CBP and p300 can also form DNA-binding complexes with p50 homodimers in vivo (Cao et al., 2006; Deng and Wu, 2003). In this report we also identified ATF3 as a potential negative regulator of p50- mediated IL-6 expression by demonstrating that co-expression of ATF3 and p50 abrogated p50-mediated IL-6 upregulation.  ATF3 belongs to the ATF/CREB family of transcription factors and is a negative regulator of pro-inflammatory cytokine expression, usually induced in response to cellular stress (Hai et al., 1999).  A recent study by Gilchrist et al indicated that ATF3 forms transcriptional complexes with NF-κB proteins in Toll-like receptor (TLR)-activated macrophages, whereupon ATF3 negatively regulates the transcription of important cytokines such as interleukin IL-6 and IL-12b (Gilchrist et al., 2006). Complementary to our cDNA microarray results, we next showed that shRNA- mediated knockdown of p50 expression inhibits IL-6 transcription and protein expression  115 levels, thus providing more evidence to support our hypothesis of p50-mediated IL-6 transcription. Finally, we sought to determine the functional significance of p50-mediated IL-6 upregulation in melanoma.  IL-6 is primarily regarded as an inflammatory cytokine (Kishimoto et al., 1995) , but can also promote angiogenesis by virtue of its ability to upregulate VEGF (Cohen et al., 1996; Huang et al., 2004; Nilsson et al., 2005).  We therefore speculated that NF-κB p50-induced IL-6 secretion from melanoma cells may promote tumor angiogenesis, thus driving melanoma progression.  NF-κB has previously been implicated as a mediator of angiogenesis under hypoxic conditions, such as those present in the tumor microenvironment (Koong et al., 1994).  For example, NF-κB reportedly stimulates angiogenesis in ovarian cancer cells by inducing expression of IL-8 and VEGF (Huang et al., 2000b).  Also, NF-κB inhibition (by a dominant-negative mutant) was shown to decrease tumor micro vessel density and IL-8 expression in a metastatic melanoma xenograft model (Huang et al., 2000a).  However, no study to date has specifically implicated the NF-κB p50 protein as a mediator of angiogenesis. Likewise, it has never been demonstrated that the p50 subunit promotes angiogenesis via IL-6 upregulation.  We have shown here for the first time that p50 overexpression alone can induce VEGF transcription and promote the growth of endothelial cells in vitro. Moreover, we have demonstrated that p50 overexpression by melanoma cells enhances the formation of supportive vasculature in vivo, while p50 knockdown inhibits micro vessel formation.  Our work highlights a novel mechanism of melanoma angiogenesis, which is of critical importance since neovascularization is one of the most important steps in the multistage progression of this aggressive malignancy.  116 The data presented in our study provide a logical explanation for why p50 expression is strongly expressed in advanced primary and metastatic melanoma tumors despite having no positive effects on melanoma cell growth rates.  The ability of melanoma tissue to express high levels of IL-6 not only provides autocrine and paracrine growth stimulation (Lázár-Molnár et al., 2000), but may also induce angiogenesis, thus conferring an important survival advantage and metastatic potential to tumor cells.  The results of our study indicate that p50 expression strongly promotes melanoma angiogenesis and that it mediates this effect through upregulation of the pro-angiogenic cytokine IL-6.  An NF-κB p50/IL-6 signaling axis, leading to angiogenesis, has never been described in melanoma.  Based on the present study, we propose that targeting the NF-κB p50/IL-6 pathway may be an effective anti-angiogenic therapeutic strategy for melanoma.   117 4.5 REFERENCES Bours V, Franzoso G, Azarenko V, Park S, Kanno T, Brown K, Siebenlist U (1993). The oncoprotein Bcl-3 directly transactivates through kappa B motifs via association with DNA-binding p50B homodimers. Cell 72: 729-39.  Budunova IV, Perez P, Vaden VR, Spiegelman VS, Slaga TJ, Jorcano JL (1999). Increased expression of p50-NF-kappaB and constitutive activation of NF-kappaB transcription factors during mouse skin carcinogenesis. Oncogene 9: 7423-31.  Cao S, Zhang X, Edwards JP, Mosser DM (2006). NF-kappaB1 (p50) homodimers differentially regulate pro- and anti-inflammatory cytokines in macrophages. J Biol Chem 281: 26041-50.  Cogswell PC, Guttridge DC, Funkhouser WK, Baldwin AS Jr (2000). Selective activation of NF-kappa B subunits in human breast cancer: potential roles for NF-kappa B2/p52 and for Bcl-3. Oncogene 19: 1123-31.  Cohen T, Nahari D, Cerem LW, Neufeld G, Levi BZ (1996). Interleukin 6 induces the expression of vascular endothelial growth factor. J Biol Chem 271: 736-41.  Deng WG, Wu KK (2003). Regulation of inducible nitric oxide synthase expression by p300 and p50 acetylation. J Immunol 171: 6581-8.  Fujita T, Nolan GP, Liou HC, Scott ML, Baltimore D (1993). The candidate proto- oncogene bcl-3 encodes a transcriptional coactivator that activates through NF-kappa B p50 homodimers. Genes Dev 7: 1354-63.  Gao K, Dai DL, Martinka M, Li G (2006). Prognostic significance of nuclear factor- kappaB p105/p50 in human melanoma and its role in cell migration. Cancer Res 66: 8382-8.  Ghosh S, Karin M (2002). Missing pieces in the NF-kappaB puzzle. Cell 109 Suppl: S81-96.  Gilchrist M, Thorsson V, Li B, Rust AG, Korb M, Kennedy K, Hai T, Bolouri H, Aderem A (2006). Systems biology approaches identify ATF3 as a negative regulator of Toll-like receptor 4. Nature 441: 173-8.  Gilmore TD (1990). NF-kappa B, KBF1, dorsal, and related matters. Cell 62: 841-3.  Gilmore TD (2006). Introduction to NF-kappaB: players, pathways, perspectives. Oncogene 25: 6680-4.  Hai T, Wolfgang CD, Marsee DK, Allen AE, Sivaprasad U (1999). ATF3 and stress responses. Gene Expr 7: 321-35.  118  Huang S, DeGuzman A, Bucana CD, Fidler IJ (2000a). Nuclear factor-kappaB activity correlates with growth, angiogenesis, and metastasis of human melanoma cells in nude mice. Clin Cancer Res 6: 2573-81.  Huang S, Robinson JB, Deguzman A, Bucana CD, Fidler IJ (2000b). Blockade of nuclear factor-kappaB signaling inhibits angiogenesis and tumorigenicity of human ovarian cancer cells by suppressing expression of vascular endothelial growth factor and interleukin 8. Cancer Res 60: 5334-9.  Huang SP, Wu MS, Shun CT, Wang HP, Lin MT, Kuo ML, Lin JT (2004). Interleukin-6 increases vascular endothelial growth factor and angiogenesis in gastric carcinoma. J Biomed Sci 11: 517-27.  Kishimoto T, Akira S, Narazaki M, Taga T (1995). Interleukin-6 family of cytokines and gp130. Blood 86: 1243-54.  Komyod W, Böhm M, Metze D, Heinrich PC, Behrmann I (2007). Constitutive suppressor of cytokine signaling 3 expression confers a growth advantage to a human melanoma cell line. Mol Cancer Res 5: 271-81.  Koong AC, Chen EY, Giaccia AJ (1994). Hypoxia causes the activation of nuclear factor kappa B through the phosphorylation of I kappa B alpha on tyrosine residues. Cancer Res 54: 1425-30.  Kumar A, Takada Y, Boriek AM, Aggarwal BB (2004). Nuclear factor-kappaB: its role in health and disease. J Mol Med 82: 434-48.  Kurland JF, Kodym R, Story MD, Spurgers KB, McDonnell TJ, Meyn RE (2001). NF- kappaB1 (p50) homodimers contribute to transcription of the bcl-2 oncogene. J Biol Chem 276: 45380-6.  Lázár-Molnár E, Hegyesi H, Tóth S, Falus A (2000). Autocrine and paracrine regulation by cytokines and growth factors in melanoma. Cytokine 12: 547-54.  Lu C, Kerbel RS (1993). Interleukin-6 undergoes transition from paracrine growth inhibitor to autocrine stimulator during human melanoma progression. J Cell Biol 120: 1281-8.  Mathas S, Jöhrens K, Joos S, Lietz A, Hummel F, Janz M, Jundt F, Anagnostopoulos I, Bommert K, Lichter P, Stein H, Scheidereit C, Dörken B (2005). Elevated NF-kappaB p50 complex formation and Bcl-3 expression in classical Hodgkin, anaplastic large-cell, and other peripheral T-cell lymphomas. Blood 106: 4287-93.   119 Mishra A, Bharti AC, Varghese P, Saluja D, Das BC (2006). Differential expression and activation of NF-kappaB family proteins during oral carcinogenesis: Role of high risk human papillomavirus infection. Int J Cancer 119: 2840-50.  Molnár EL, Hegyesi H, Tóth S, Darvas Z, László V, Szalai C, Falus A (2000). Biosynthesis of interleukin-6, an autocrine growth factor for melanoma, is regulated by melanoma-derived histamine. Semin Cancer Biol 10: 25-8.  Mouawad R, Benhammouda A, Rixe O, Antoine EC, Borel C, Weil M, Khayat D, Soubrane C (1996). Endogenous interleukin 6 levels in patients with metastatic malignant melanoma: correlation with tumour burden. Clin Cancer Res 2: 1405–9.  Mouawad R, Rixe O, Meric JB, Khayat D, Soubrane C (2002). Serum interleukin-6 concentrations as predictive factor of time to progression in metastatic malignant melanoma patients treated by biochemotherapy: a retrospective study. Cytokines Cell Mol Ther 7: 151–6.  Mukhopadhyay T, Roth JA, Maxwell SA (1995). Altered expression of the p50 subunit of the NF-kappa B transcription factor complex in non-small cell lung carcinoma. Oncogene 11: 999-1003.  Nilsson MB, Langley RR, Fidler IJ (2005). Interleukin-6, secreted by human ovarian carcinoma cells, is a potent proangiogenic cytokine. Cancer Res 65: 10794-800.  Prusty BK, Husain SA, Das BC (2005). Constitutive activation of nuclear factor-kB: preferential homodimerization of p50 subunits in cervical carcinoma. Front Biosci 10: 1510-9.  Sehgal PB, Wang L, Rayanade R, Pan H, Margulies L (1995). Interleukin-6-type cytokines. Ann N Y Acad Sci 762: 1-14.  Soubrane C, Rixe O, Meric JB, Khayat D, Mouawad R (2005). Pretreatment serum interleukin-6 concentration as a prognostic factor of overall survival in metastatic malignant melanoma patients treated with biochemotherapy: a retrospective study. Melanoma Res 15: 199-204.  Sun WH, Kreisle RA, Phillips AW, Ershler WB (1992). In vivo and in vitro characteristics of interleukin 6-transfected B16 melanoma cells. Cancer Res 52: 5412-5.  Tas F, Oguz H, Argon A, Duranyildiz D, Camlica H, Yasasever V, Topuz E (2005). The value of serum levels of IL-6, TNF-alpha, and erythropoietin in metastatic malignant melanoma: serum IL-6 level is a valuable prognostic factor at least as serum LDH in advanced melanoma. Med Oncol 22: 241-6.   120 Thornburg NJ, Pathmanathan R, Raab-Traub N (2003). Activation of nuclear factor- kappaB p50 homodimer/Bcl-3 complexes in nasopharyngeal carcinoma. Cancer Res 63: 8293-301.  Watanabe N, Iwamura T, Shinoda T, Fujita T (1997). Regulation of NFKB1 proteins by the candidate oncoprotein BCL-3: generation of NF-kappaB homodimers from the cytoplasmic pool of p50-p105 and nuclear translocation. EMBO J 16: 3609-20.  Yamamoto Y, Gaynor RB (2001). Role of the NF-kappaB pathway in the pathogenesis of human disease states. Curr Mol Med 1: 287-96.     121 5.   CONCLUDING REMARKS In this study, three proteins have been identified as therapeutic targets for the treatment of malignant melanoma: PUMA, p-Akt, and NF-κB p50.  We have demonstrated that these proteins function as mediators of apoptosis, cell survival, and angiogenesis, respectively. The current state of therapeutics targeting each of these pathways, in the context of human cancer, is discussed below.  5.1 TARGETING APOPTOSIS DYSREGULATION IN MELANOMA 5.1.1 Anti-apoptotic versus pro-apoptotic gene therapy: the case for ad-PUMA This study has shown that expression of the pro-apoptotic gene PUMA is significantly reduced in melanoma and that low PUMA levels adversely affect the 5-year survival rates of melanoma patients.  Insufficient expression of pro-apoptotic genes, like PUMA, may prevent apoptosis induction in melanoma tumor tissue, thus promoting chemoresistance and disease progression.  To overcome this problem we have proposed that PUMA protein be overexpressed via adenoviral gene therapy.  While enhancement of pro-apoptotic gene expression is a relatively new anti-cancer strategy, suppression of anti-apoptotic gene expression has been explored for several years.  The most well-known example of this is Bcl-2 antisense treatment.  Expression of Bcl-2 in melanoma has been examined in numerous studies, with somewhat conflicting results; about one third of the data indicates that Bcl-2 is upregulated during melanoma progression, while another third suggests is it downregulated (Bush and Li, 2003).  The remaining one third of studies report that Bcl-2 is highly expressed in all melanocytic tissues, whether benign or malignant.  Despite these inconsistencies, Bcl-2 antisense  122 treatment showed therapeutic potential in a preclinical study of human melanoma xenografts, where it sensitized tumors to DTIC (Jansen et al., 1998).  Based on such results, a Bcl-2 antisense compound called oblimersen was developed and evaluated in clinical trials.  In a randomized phase III trial of metastatic melanoma patients, combination treatment of oblimersen and DTIC nearly doubled the objective response rate (13% versus 7%), but did not significantly improve overall survival (Bedikian et al., 2006).  The failure of Bcl-2 antisense therapy may be attributed to the fact that several anti- apoptotic proteins, in addition to Bcl-2, are frequently overexpressed in melanoma. Therefore, targeting Bcl-2 expression alone is probably not sufficient to overcome the barrier to apoptosis that exists in these tumors.  For example, the Bcl-2 family members Bcl-XL and Mcl-1 are reportedly upregulated in primary and metastatic melanomas, compared to benign nevi (Tang et al., 1998; Selzer et al., 1998).  Moreover, levels of Bcl- XL and Mcl-1 protein expression appear to increase with tumor progression (Tang et al., 1998).  These findings have been confirmed by a number of subsequent studies (Bush and Li, 2003).  As with Bcl-2, antisense oligonucleotides targeting Mcl-1 have been tested preclinically.  In a melanoma xenograft model, Mcl-1 antisense treatment inhibited tumor growth and sensitized tumors to DTIC treatment (Thallinger et al., 2003).  However, Mcl-1 antisense therapy for melanoma has yet to be tested clinically.  Aside from anti-apoptotic Bcl-2 proteins, members of the inhibitor of apoptosis (IAP) family are frequently upregulated in melanoma.  Most notably, X-linked inhibitor of apoptosis (XIAP) and Survivin are highly expressed in melanoma tissues (Grossman et  123 al., 1999; Kluger et al., 2007).  These proteins are caspase inhibitors that block the execution phase of apoptosis (Vucic and Fairbrother, 2007).  Targeting Survivin expression (via inducible expression of mutant Survivin) has been shown to inhibit the growth of melanoma xenografts (Grossman et al., 2001).  Like Bcl-2 and Mcl-1, both XIAP and Survivin antisense oligonucleotides have shown efficacy in preclinical studies of other tumor types and are currently in phase I trials for patients with refractory malignancies (Vucic and Fairbrother, 2007).  Despite all of these efforts in developing antisense therapies to target anti- apoptotic gene expression, a glaring problem remains.  Even if the apoptosis "barrier" is successfully removed through antisense treatment, there is no guarantee that pro- apoptotic genes will be expressed at sufficient levels to induce cell death.  When key apoptosis effectors are severely down-regulated in malignant tissues, altering anti- apoptotic gene expression may not resolve the problem of apoptosis resistance.  For this reason, we have suggested that overexpression of a potently pro-apoptotic gene, like PUMA, is a superior therapeutic strategy.  PUMA not only binds to and neutralizes all anti-apoptotic members of the Bcl-2 protein family, but is also thought to contribute to the activation of Bax and/or Bak proteins, which is required for the induction of mitochondrial-mediated apoptosis. 5.1.2 PUMA gene therapy in pre-clinical models of melanoma The ad-PUMA construct used in this study dramatically inhibited melanoma growth, both in vitro and in vivo, via rapid activation of the intrinsic apoptosis pathway.  Based on these observations, we proposed that ad-PUMA be exploited as a gene therapy tool for the treatment of malignant melanoma.  The in vivo ad-PUMA data presented here is in  124 agreement with a previous study by Ito et al., in which ad-PUMA was tested as a gene therapy for malignant glioma (Ito et al., 2005).  The adenoviral vector used in their study was similar to ours, except that PUMA expression was driven by the human telomerase reverse transcriptase (hTERT) promoter in their system, whereas ours was under control of the cytomegalovirus (CMV) promoter.  Selection of the hTERT promoter was based on observations that telomerase is expressed in most gliomas but not in normal brain tissue (Le et al., 1998).  Akin to our study, Ito et al. demonstrated that ad-PUMA efficiently induced mitochondrial-mediated apoptosis of glioma cells in vitro regardless of p53 mutational status.  They too used a xenograft mouse model to test the efficacy of ad-PUMA in vivo.  Human glioma tumors were induced subcutaneously in nude mice and allowed to grow to a volume of 30-50 mm3 before treatment was initiated.  Despite the fact that a tissue-specific hTERT promoter was incorporated into their adenoviral construct, they did not administer the virus systemically.  Instead, tumors were dosed five days in a row via intra-tumoral injection of ad-PUMA, and tumor volumes were monitored for 20 days.  They reported that ad-PUMA inhibited the growth of glioma xenografts by more than 2-fold; considerably more than in our study.  The difference in in vivo efficacy observed between our two studies may be due to various factors: 1) differential dosing regimens, 2) differences in the growth rates of glioma xenografts compared to melanoma xenografts, 3) variations in the susceptibility of glioma and melanoma tumor cells to adenoviral infection, and 4) the use of different gene promoters.  Shortly before our in vivo ad-PUMA results were published (Karst et al., 2006), Yu et al. reported findings similar to ours, using a human lung cancer model (Yu et al., 2006).  In their study, ad-PUMA (under control of the CMV promoter) inhibited the  125 growth of human lung tumor xenografts by about 80% compared to a control virus expressing a mutant form of PUMA, lacking the BH3 domain.  They also tested the ability of ad-PUMA to sensitize lung cancer cell lines to a variety of chemotherapeutic drugs, including paclitaxel, cisplatin, 5-fluorouracil, etoposide, and doxorubicin.  Low doses of ad-PUMA lowered the IC50 values of these drugs by 3-fold to 10-fold in vitro. However, they did not test the chemosensitizing effects of ad-PUMA in vivo.  A recent report by Chen et al. indicates that ad-PUMA can sensitize drug-resistant choriocarcinoma tumors to chemotherapy (Chen et al., 2007).  In their study, they selected a sub-population of drug-resistant cells from a human choriocarcinoma cell line by treating the cells with a gradually increasing concentration of etoposide over a 6 month period.  The drug-resistant cell line exhibited upregulation of the multidrug resistance protein p-glycoprotein (P-gp) and was highly resistant to etoposide (VP-16), fluorouracil (5-FU), and methotrexate (MTX) treatments.  However, treating the cells with ad-PUMA resulted in 8.66- to 13.06-fold decreases in their IC50 values.  In a human choriocarcinoma xenograft model, intratumoral injection of ad-PUMA enhanced the therapeutic response to etoposide by over 60%, compared to only 6% for xenografts established from the parental, non-drug-resistant cell line.  The results of our study, as well as those reported by other groups, clearly indicate that ad-PUMA has potential as a gene therapy tool, either on its own or as a sensitizing agent in combination with chemotherapy.  It would be particularly interesting to see whether ad-PUMA is able to sensitize melanoma tumors to DTIC, the only standard chemotherapeutic drug used to treat melanoma.  If so, combination treatment of ad- PUMA and DTIC might improve the therapeutic outcomes of melanoma patients.  126 5.1.3 Adenoviral-based gene therapy: is it clinically possible? Although viral-mediated gene transfer is widely employed in preclinical studies, viral- based cancer treatments have been slow to develop in the clinic.  There are several reasons for this.  Firstly, administration of viral vectors induces an immune response in patients that may clear viral particles from the body before they have time to infect malignant tissue (Dobbelstein, 2003).  Secondly, there are safety concerns regarding the use of viruses that make clinical trials difficult.  For example, if replication competent viruses are used, patients may require isolation from the general public to ensure that they do not transmit the therapeutic virus to others (Liu et al., 2004).  Such factors make clinical trials expensive and cumbersome.  Nevertheless, viral-based anti-cancer therapies have been developed and evaluated in clinical trials.  A very well known example is ONYX-015, an oncolytic virus that has been tested in patients with head and neck squamous cell carcinoma (HNSCC). ONYX-015 is an adenovirus that selectively replicates in (and thus kills) p53-deficient cells because it has been engineered to lack the adenoviral E1B gene (Bischoff et al., 1996).  E1B is normally required for adenoviral replication in host cells containing functional p53 (O’Shea et al., 2005).  Since many cancer cells are p53-deficient (Hofseth et al., 2004), ONYX-015 can be used to selectively target malignant tissues.  ONYX-015 was evaluated in phase I and II clinical trials with promising results and is now in phase III trials.  Unfortunately, p53 is rarely mutated in primary melanoma tumors (Papp et al., 1996; Castresana et al., 1993), so ONYX-015 would not a suitable treatment for malignant melanoma.  127  Rather than oncolysis, we have proposed a gene transfer approach with our ad- PUMA construct.  Anti-cancer gene transfer strategies typically employ a replication- deficient retrovirus or adenovirus virus carrying a "suicide" gene - that is, a gene whose expression is toxic or whose gene function is pro-apoptotic (Spencer, 2000).  Despite the enormous potential of gene transfer for cancer treatment, there are two major challenges associated with this approach: 1) targeting gene transfer only to malignant cells while avoiding healthy tissues, and 2) achieving high efficiencies of gene transfer.  The first challenge may be overcome through the use of tissue specific promoters that restrict gene expression to certain cell types.  For example, the tyrosinase gene is expressed only in melanocytic cells.  Tyrosinase is a key enzyme in the melanin biosynthetic pathway (Schallreuter et al., 2007).  Alternatively, the microphthalmia-associated transcription factor (MITF) gene promoter may be used.  This transcription factor is integral to melanocyte development and promotes melanocyte survival by upregulating Bcl-2 expression (McGill et al., 2002).  The promoters of these genes may be incorporated into adenoviral vectors to direct suicide gene expression to melanocytic tissue.  The second challenge is more complicated and will require engineering of improved adenoviral vectors for more efficient gene-transfer.  The most widely used suicide gene in clinical trials to date has been the herpes simplex virus–thymidine kinase (HSV-tk) gene.  Once HSV-tk is introduced into tumor cells, patients are given the drug ganciclovir, which is an acyclic nucleoside analogue. Ganciclovir is phosphorylated by HSV-tk inside cells and becomes incorporated into DNA as ganciclovir-triphosphate, leading to termination of DNA elongation during S- phase (Reid et al., 1988).  This approach initially showed promise when tested in  128 preclinical xenograft models of murine and human melanoma.  For example, in a study by Vile and Hart, B16 murine melanoma cells transfected with a plasmid encoding HSV- tk under control of the tyrosinase promoter were used to establish tumors in vivo, then the animals were treated with Ganciclovir.  Tumor growth was inhibited by nearly 70% in the Ganciclovir-treated animals compared to controls (Vile and Hart, 1993).  In a later study by Bonnekoh et al., an adenovirus carrying the HSV-tk gene (ad-tk) was injected directly into human melanoma xenografts, followed by intraperitoneal ganciclovir treatment.  Ad- tk plus ganciclovir treatment resulted in 50% inhibition of tumor growth compared to controls (Bonnekoh et al., 1996).  Based on such results, HSV-tk gene therapy of melanoma was advanced to clinical trials.  In a phase I/II dose-escalation study utilizing retroviral-mediated delivery of HSV-tk plus ganciclovir, treatment was well tolerated by patients but was not efficacious (Klatzmann et al., 1998).  In vivo transduction efficiency was extremely low and all patients showed tumor progression upon long-term follow-up.  A disadvantage of the latter study is that it employed a retroviral rather than adenoviral vector.  The retroviral life cycle requires viral integration into the host cell genome, whereas adenovirus DNA functions in the extrachromosomal portion of a cell’s nucleus.  It has been reported that retroviral gene transfer is often undermined by host cell "escape mechanisms" such as chromosomal deletion, postinsertional recombination, and gene silencing (Frank et al., 2004).  Therefore, adenoviruses may be superior vectors for the delivery of suicide genes to melanoma.  In addition, adenoviruses can carry a larger DNA load than retroviruses and adenoviruses exhibit higher transduction efficiencies in human cells (St George, 2003).  Furthermore, adenoviruses, unlike retroviruses, can infect and express their genes in non-dividing cells.  The main disadvantage of  129 adenoviruses is the possibility that they may provoke a severe immune response.  Such a response is rare, but has occurred.  In 1999, an 18-year old patient tragically died following systemic administration of a adenoviral-based gene therapy for ornithine transcarbamylase (OTC) deficiency (Raper et al., 2003).  The patient suffered a systemic inflammatory response culminating in multiple organ system failure 98 h post-treatment. Obviously, extensive study of the human immune response to adenoviral vectors is required in order to minimize risk and advance adenoviral gene transfer technology in the future.  Only a handful of adenoviral-mediated gene-transfer strategies have so far undergone phase I testing in melanoma patients.  These studies involved the transfer of p53, interleukin-2 (IL-2), and melanoma differentiation-associated gene-7 (mda-7).  Dummer et al. conducted a phase I proof of concept study in which they tested the ability of an adenoviral vector carrying p53 to induce transgene expression in metastatic melanoma tumor tissue (Dummer et al., 2000).  They reported that p53 transgene mRNA was detected in 80% of melanoma tumors (4 of 5 patients), sampled 48 h after intratumoral injection of the virus.  Side effects of adenoviral treatment were mild, with no clinically significant toxicity was observed.  Stewart et al. conducted a study of IL-2 gene transfer in 23 patients with metastatic melanoma or breast cancer (Stewart et al., 1999).  IL-2 is an immunotherapeutic agent that, when administered in a high-doses, has produced some durable remissions in metastatic melanoma patients (Perez et al., 1991).  However, systemic delivery of the protein is limited by toxicity side-effects and short protein half- life.  For this reason, Stewart et al. hypothesized that gene therapy would be a better way  130 to provide IL-2 to melanoma tissue.  In their study, vector-derived IL-2 mRNA was detected in 80% of biopsies, sampled 7 days after intratumoral injection.  Additionally, increased IL-2 protein expression could be detected in tumor tissue 48 h post-treatement. Although the study was not powered to determine efficacy, local tumor regression at the site of injection was noted in 24% of patients.  Again, adenoviral treatments were well tolerated; inflammation at the site of injection was the most serious side effect.  This study also tested for signs of liver toxicity, which has been reported on some animal models following adenovirus gene delivery (Toloza et al., 1996).  However, no abnormal liver function was detected.  Perhaps the most promising phase I trial is that involving INGN 241, an adenoviral vector expressing mda-7, otherwise known as interleukin-24 (IL-24).  Mda-7 was first identified as a gene downregulated during the terminal differentiation of melanoma cells (Jiang et al., 1995) and was soon found to suppress the growth of various types of cancer cells, but not normal cells (Jiang et al., 1996).   It has since been shown that, similar to PUMA, mda-7 induces apoptosis of cancer cells though mitochondrial pathways involving Bcl-2 family members (Lebedeva et al., 2003).  On this basis, it was considered a good candidate for adenoviral-mediated pro-apoptotic gene therapy of cancer.  Tong et al. treated a group of 22 advanced cancer patients, including 7 melanoma patients, with escalating doses of INGN 241, administered by repeated intratumoral injection (Tong et al., 2005).  Successful gene transfer was demonstrated in 100% of patients, by DNA- and RT-PCR analysis of tumor tissue.  Mda-7 protein expression and induction of apoptosis were also observed in all tumors.  The adenoviral treatments were well tolerated, with T-cell activation and increased cytokine production being the main  131 side effects.  The results of this study are very encouraging, especially considering the parallels to our proposed pro-apoptotic ad-PUMA strategy.  It will be exciting to see how ad-mda-7 fares in phase II/III trials of melanoma patients.  While the phase I studies described here cannot determine the anti-tumor efficacy of adenoviral mediated gene transfer, they certainly demonstrate the safety and feasibility of adenoviral vectors for gene delivery in humans and are a promising first step towards the development of adenoviral-based gene-transfer therapies for the treatment of melanoma.  5.2 TARGETING THE PI3K/AKT PATHWAY IN MELANOMA 5.2.1 Inhibiting constitutive Akt activation The second major focus of this study is the inhibition of constitutive Akt activation in melanoma.  We have shown here that activated Akt (p-Akt) is a major prognostic indicator of melanoma patient survival.  Similar links between p-Akt and prognosis have been demonstrated in other cancer types, including prostate (Kreisberg et al., 2004), breast (Pérez-Tenorio et al., 2002), lung (Tsurutani et al., 2006), pancreas (Yamamoto et al., 2004), brain (Ermoian et al., 2002), stomach (Nam et al., 2003), liver (Nakanishi et al., 2005), endometrium (Terakawa et al., 2003), and blood (Min et al., 2004).  There have been relatively few studies directly investigating the role of Akt in melanoma.  One of the few groups to study Akt in melanoma has reported that the Akt3 isoform, in particular, is responsible for melanoma development in nonfamilial melanomas (Stahl et al., 2004).  They showed that Akt3 expression increases with melanoma tumor progression due to a combination of gene amplification and decreased PTEN function  132 (through PTEN gene loss or haploinsufficiency).  Furthermore, they used siRNA to demonstrate that Akt3 is the primary isoform contributing to the high p-Akt levels of melanoma cells.  The validity of Akt as a therapeutic target in cancer has been verified by pre- clinical studies of other tumor types.  For example, Jetzt et al. inhibited Akt activity in a breast cancer xenograft model via adenoviral-mediated delivery of a dominant negative kinase-dead Akt mutant (ad-Akt-DN) (Jetzt et al., 2003).  Intratumoral expression of ad- Akt-DN inhibited tumor growth by nearly 90% and selectively induced of apoptosis in tumor cells expressing activated Akt, while having little effect on normal or tumor cells lacking activated Akt.  In a subsequent study by Stahl et al., Akt expression or activation were targeted in melanoma cells using siRNA against Akt or by overexpressing PTEN, respectively (Stahl et al., 2004).  The transfected cells were then used to establish melanoma xenografts in nude mice.  In both cases, tumor growth rates were inhibited by over 50% and were accompanied by a dramatic increase in tumor cell apoptosis, compared to control tumors.  Yang et al. reported that the small molecule Akt inhibitor API-2 can suppress the growth of renal cell carcinomas (Yang et al., 2004).  Using a xenograft model, they showed that administration API-2 selectively inhibited the growth of renal cell carcinoma tumors containing high levels of activated Akt by 80-90%, compared to controls.  These studies support our finding that Akt is a viable therapeutic target for the suppression of malignant growth in vivo. 5.2.2 Pharmacological Akt inhibitors To deplete the p-Akt signal in melanoma, we used the small-molecule Akt inhibitor API- 2.  However, there are many other inhibitors available.  In fact, there has been a huge  133 effort on the part of both academia and industry to develop effective pharmacological inhibitors of Akt.  Currently, there are three major classes of Akt inhibitors: 1) lipid- based inhibitors, 2) ATP-competitive inhibitors, and 3) small molecule inhibitors.  Lipid-based inhibitors are by far the most well-developed class of Akt inhibitors. They interfere with the translocation of Akt protein to the plasma membrane, thus preventing Akt activation (LoPiccolo et al., 2007).  The two main types of lipid-based inhibitors are alkylphospholipids (ALPs) and phosphatidylinositol ether lipid analogs (PIAs).  ALPs work by inserting into the plasma membrane, where they accumulate in lipid rafts and interfere with translocation of Akt to the plasma membrane (van Blitterswijk et al., 1987; van der Luit et al., 2002).  PIAs, on the other hand, were designed as structural analogs of PI(3,4)P2 and PI(3,4)P3.  They bind to the PH domain of Akt and thus prevent its recruitment to the plasma membrane (Castillo et al., 2004).  Both ALPs and PIAs have been shown to inhibit cancer cell growth in association with Akt inhibition in preclinical studies (Kondapaka et al., 2003; Gills et al., 2006).  ALPs have also undergone clinical testing, including one phase II study of patients with metastatic melanoma.  Ernst et al. treated 18 metastatic melanoma patients with the ALP perifosine as a single agent (Ernst et al., 2005).  Unfortunately, there were no objective responses; 21% of evaluable patients achieved stable disease, while the remainder exhibited progression.  However, the patients of this study were not tested beforehand to see whether or not their tumor tissue exhibited elevated Akt activity.  Therefore, they may not have been good candidates for Akt inhibitor treatment in the first place.  Although lipid-based inhibitors are well developed, they are not ideal inhibitors of Akt due to their lack of specificity (Lindsley et al., 2007).  ALP integration into the plasma membrane has  134 the potential to disrupt a whole range of membrane-dependent signaling events, while PIAs could inhibit a variety of PH-domain containing proteins other than Akt.  ATP-competitive inhibitors are synthetic compounds are designed to compete for the ATP binding site and therefore inhibit Akt kinase activity (Lindsley et al., 2007). There are a host of such compounds (NL-71-101, doxazosin, 2-pyrimidyl-5- amidothiophene, A-443654, A-423795, etc.) that have shown promise in pre-clinical studies (reviewed by Lindsley et al., 2007), but none have been tested clinically to date.  Small molecule Akt inhibitors include API-2, 9-methoxy-2-methylellipticinium acetate (API-59CJ-OMe), KP372, and others.  Most of these compounds were discovered in high-throughput screens of chemical libraries, looking for compounds that inhibited the growth of cancer cells with elevated Akt activity, but not the growth of normal cells. The mechanisms by which these compounds inhibit Akt activity are generally not known. In the case of API-2, for example, it has been shown that the drug does not inhibit the kinase activities of upstream regulators PI3K or  PDK1 in vitro (Yang et al., 2004).  Nor does it affect the activities of structurally similar kinases PKA, PKCα, or SGK.  Exactly how API-2 specifically targets Akt activity remains to be determined.  5.3 DUAL TARGET THERAPY OF MELANOMA In addition to identifying PUMA and Akt as therapeutic targets for melanoma, this study has presented data in support of dual target therapy of melanoma.  We identified a striking pattern of protein expression in which melanoma patients with both weak PUMA and strong p-Akt expression in their tumor tissue had extremely poor 5-year survival compared to patients with the reverse pattern of protein expression.  This result indicates  135 not only that melanoma tumors frequently exhibit dysregulation of more than one intracellular signaling pathway (here, the p53-mediated apoptosis and PI3K survival pathways), but that combined dysregulation of more than one pathway can dramatically impact prognosis.  This is an important finding because it may directly impact the effectiveness of single-target melanoma therapies.  For example, if gene therapy is used to overexpress a pro-apoptotic gene but Akt remains constitutively activated, the p-Akt survival signal may hamper therapeutic response.  To make matters worse, strong p-Akt signaling may select for a small population of treatment-resistant cells and lead to post-treatment tumor regression.  To minimize such a possibility, it would be prudent to target more than one signaling pathway simultaneously.  We tested this strategy by treating melanoma cells with a combination of ad- PUMA and p-Akt inhibitor, API-2.  While API-2 treatment and ad-PUMA alone exerted growth inhibitory effects, a combination of the two enhanced this effect in an additive manner.  Specifically, combination treatment enhanced growth inhibition by 20-40% over the single-target treatment groups.  Our results clearly indicate that dual target therapy is effective in stalling melanoma tumor progression and that targeting both PUMA and p- Akt may be a promising therapeutic strategy.  Although we are the first to suggest dual targeting of PUMA and p-Akt, the need for dual target cancer therapy has been recognized by many in the cancer research community and is now being tested in preclinical studies.  This approach is especially important for apoptosis-resistant malignancies like melanoma, against which all monotherapies have so far failed.  Several studies provide strong rationale for the  136 targeting of at least two different signaling pathways in melanoma.  A recent study by Meier et al. showed that dual targeting of the MAPK and AKT signaling pathways may be an effective therapeutic strategy for melanoma (Meier et al., 2007).  In their study, a panel of pharmacological inhibitors (BAY 43-9006, PD98059, U0126, wortmannin, LY294002), were used to inhibit the MAPK and AKT pathways in various melanoma cell lines.  While the single-agent treatments generally elicited a poor response, combinations targeting both MAPK and AKT significantly inhibited growth and enhanced apoptosis.  Importantly, they confirmed these results using a regenerated human skin model that mimics the microenvironment of human melanoma.  Another recent study, by Lasithiotakis et al., made a case for simultaneous MAPK and AKT pathway inhibition by combining the RAF inhibitor sorafenib with the mTOR inhibitor rapamycin (Lasithiotakis et al., 2008).  In their study, combination treatment enhanced growth inhibition of 6 metastatic cell lines by 13.0-27.8% over monotherapy and increased cell death rates by about two-fold.  Moreover, sorafenib and rapamycin combination treatment completely suppressed invasive melanoma growth in organotypic culture (i.e. a human dermal reconstruct).  Western blot analysis revealed that co- treatment of melanoma cells with sorafenib and rapamycin completely abolished the expression of anti-apoptotic proteins Bcl-2 and Mcl-1, but not Bcl-XL.  In the dual target studies described above, both of the intended targets are pro- survival/proliferation genes.  The rationale behind this type of approach is the common observation that melanoma tumors exhibit activation of multiple signaling pathways. Therefore, disabling one survival pathway is not sufficient to halt tumor cell growth. However, by disabling two, or even three, of the most highly activated survival pathways,  137 the odds of effectively inhibiting tumor growth significantly improve.  Our study takes a slightly different approach, in that we seek to suppress a major cell survival pathway (the PI3K/Akt pathway) while at the same time activating a potent pro-apoptotic effector (PUMA).  This strategy has the advantage of not only stalling tumor growth, but also providing a powerful stimulus to drive tumor cells towards rapid death.  The main challenge for dual target therapy development is to identify which signaling pathways are best targeted in combination to achieve the most dramatic results with the smallest degree of toxicity.  It is likely that the ideal combination of targets will vary from patient to patient, depending on which particular pathways are most highly activated in each case.  Therefore, it will be necessary to tailor treatments to individual patients in accordance with the results of biopsy testing for pathway activation.  Although this "personalized" approach would certainly introduce a new level of complexity to diagnosis and treatment of melanoma patients, it has the potential to significantly improve disease outcome and represents an exciting new frontier for the treatment of advanced melanoma.  5.4 TARGETING ANGIOGENESIS IN MELANOMA 5.4.1 Anti-angiogenic therapeutics Finally, this work identifies the p50 subunit of NF-κB as a novel pro-angiogenic factor in melanoma.  Tumor angiogenesis (i.e. the formation and development of blood vessels) is a crucial step for both the development of primary tumors and for metastatic outgrowth (Hanahan and Weinberg, 2000).  The initiating step of angiogenesis is the activation of endothelial cells by specific growth factors such as basic fibroblast growth factor (bFGF),  138 vascular endothelial cell growth factor (VEGF), or placental growth factor (PLGF) (Hillen and Griffioen, 2007).  In addition to growth factors, tumor cells may produce a number of other pro-angiogenic factors to enhance their vascularization.  For example, production of IL-8 enhances the migration endothelial cells towards tumor tissue (Li et al., 2003).  Secretion of platelet-derived growth factor (PDGF) acts as a chemoattractant for pericytes, a type of connective tissue cell present in capillary walls (Betsholtz et al., 2005).  Transforming growth factor-β (TGF-β) is secreted to stabilize newly formed vessels and suppresses the immune system (Elliott and Blobe, 2005).  In addition, production of epidermal growth factor (EGF) helps to promote VEGF upregulation (Petit et al., 1997).  Most anti-angiogenic therapies work by blocking pro-angiogenic factors and/or their receptors, with the most common target being the VEGF/VEGFR signaling axis. The first FDA-approved anti-angiogenic drug was a monoclonal antibody against VEGF, called bevacizumab.  Bevacizumab has not yet shown efficacy as a monotherapy, but it has improved the overall survival rates of cancer patients receiving chemotherapy.  In a phase III trial of metastatic colorectal cancer patients, bevacizumab significantly improved overall survival (15.6 versus 20.3 months) when combined with a fluorouracil- based chemotherapeutic regimen (Hurwitz et al., 2004).  Similarly, in a phase III study of advanced nonsmall cell lung cancer (NSCLC), bevacizumab improved  overall patient survival (10.2 versus 12.5 months) in combination with carboplatin/paclitaxel (Sandler et al., 2006).  However, bevacizumab has failed to produce an overall survival benefit in advanced breast cancer patients receiveing paclitaxel or capecitabine treatment (Miller et  139 al., 2007; Miller et al., 2005).  A phase II trial of bevacizumab for the treatment of melanoma is currently underway (Facchetti et al., 2007).  A second major therapeutic agent designed to target angiogenesis is the small molecule tyrosine kinase inhibitor vatalanib, which targets the receptors of VEGF, PDGF, and c-KIT (Scott et al., 2007).  The advantage of vatalanib over bevacizumab is that it is administered orally rather than intravenously.  However, the phase III trial results for vatalanib presented thus far are less favourable than those for bevacizumab (Los et al., 2007). 5.4.2 Targeting NF-κB proteins to inhibit angigogenesis Because NF-κB regulates so many genes and is associated with such a wide variety of diseases, numerous efforts have been made to develop NF-κB inhibitors over the years. In fact, a recent review estimates that there are now 785 inhibitors of NF-κB, including natural products, chemicals, metals, metabolites, synthetic compounds, peptides, and proteins (Gilmore, 2006).  In the realm of cancer, many pre-clinical studies have demonstrated that targeting NF-κB signaling inhibits malignant cell growth.  However, in practical terms, NF-κB is not considered an ideal therapeutic target because its suppression gives rise to immunodeficiency and produces highly toxic side effects (Pacifico and Leonardi, 2006).  Here, we have demonstrated that the p50 subunit of NF-κB mediates angiogenesis in melanoma and that silencing p50 expression inhibits vascularization in vivo over the short-term (i.e. one-week).  These findings suggest that NF-κB p50 may be a major factor in melanoma angigogenesis and thus, a potential anti-angiogenic therapeutic target. However, we have not determined whether p50 knockdown inhibits melanoma tumor  140 growth over the long term (i.e. 1-2 months).  Nor have we identified the nature of the NF- κB transcriptional complex involved.  Based on the present study and previously published data from our lab (Gao et al., 2006), we hypothesize that p50 may be acting as a homodimer to activate IL-6 gene transcription in melanoma.  To test this hypothesis, it will be necessary to investigate p50 dimer formation and DNA binding patterns using an electrophoretic mobility shift assay (EMSA) and to verify that p50 homodimer binding to DNA is required for IL-6 upregultion.  To date, there have been no studies focusing on p50 homodimer formation in melanoma.  If p50 is indeed acting as a homodimer, signaling through an alternate NF-κB pathway and requiring a transcriptional co-activator (as discussed in chapter 4), it may be possible to inhibit pro-angiogenic p50 signaling without inducing the debilitating effects associated with suppression of classical NF-κB signaling.  For instance, if we could identify the specific transcriptional co-activators required for p50-mediated upregulation of IL-6, perhaps we could target those co-activators instead of p50 to inhibit melanoma angiogenesis.  Therefore, it is important to further study this phenomenon and to elucidate its underlying mechanisms.  5.5 SUMMARY Overall, this study has defined important roles for PUMA, p-Akt and NF-κB p50 in melanoma tumor growth, progression, and angigoenesis.  Through tissue microarray analysis, we found that PUMA and p-Akt were significantly reduced and overexpressed, respectively, in melanoma tumors and we identified highly correlative relationships between PUMA expression, p-Akt expression,  141 and 5-year melanoma patient survival.  We demonstrated that exogenous adenoviral- mediated PUMA expression induces melanoma cell death via rapid activation of the intrinsic apoptotsis pathway, and that ad-PUMA gene therapy can dramatically inhibit melanoma tumor growth in vivo.  We also showed that inhibiting Akt activation with a small molecule inhibitor (API-2) enhances ad-PUMA-mediated inhibition of melanoma cell growth.  Furthermore, we demonstrated that the combination of ad-PUMA and API-2 treatments severely suppresses tumorigenic growth in vivo, highlighting the potential for dual-target therapy of melanoma.  Finally, using cDNA micrarray analysis, we showed that the p50 subunit of NFκB upregulates the pro-angiogenic cytokine IL-6 and plays a distinct role in promoting melanoma tumor angiogenesis.  These findings are all important contributions to the understanding of melanoma pathogenesis and may prove useful in the development of novel, more effective treatments for this deadly disease.   142 5.6 REFERENCES Bedikian AY, Millward M, Pehamberger H, Conry R, Gore M, Trefzer U, Pavlick AC, DeConti R, Hersh EM, Hersey P, Kirkwood JM, Haluska FG; Oblimersen Melanoma Study Group (2006). Bcl-2 antisense (oblimersen sodium) plus dacarbazine in patients with advanced melanoma: the Oblimersen Melanoma Study Group. J Clin Oncol 24: 4738-45.  Betsholtz C, Lindblom P, Gerhardt H (2005). Role of pericytes in vascular morphogenesis. EXS 94: 115-25.  Bischoff JR, Kirn DH, Williams A, Heise C, Horn S, Muna M, Ng L, Nye JA, Sampson- Johannes A, Fattaey A, McCormick F (1996). An adenovirus mutant that replicates selectively in p53-deficient human tumor cells. Science 274: 373-6.  Bonnekoh B, Greenhalgh DA, Bundman DS, Kosai K, Chen SH, Finegold MJ, Krieg T, Woo SL, Roop R (1996). Adenoviral-mediated herpes simplex virus-thymidine kinase gene transfer in vivo for treatment of experimental human melanoma. J Invest Dermatol 106: 1163-8.  Bush JA, Li G (2003). The role of Bcl-2 family members in the progression of cutaneous melanoma. Clin Exp Metastasis 20: 531-9.  Castillo SS, Brognard J, Petukhov PA, Zhang C, Tsurutani J, Granville CA, Li M, Jung M, West KA, Gills JG, Kozikowski AP, Dennis PA (2004). Preferential inhibition of Akt and killing of Akt-dependent cancer cells by rationally designed phosphatidylinositol ether lipid analogues. Cancer Res 64: 2782-92.  Castresana JS, Rubio MP, Vázquez JJ, Idoate M, Sober AJ, Seizinger BR, Barnhill RL (1993). Lack of allelic deletion and point mutation as mechanisms of p53 activation in human malignant melanoma. Int J Cancer 55: 562-5.  Chen Y, Qian H, Wang H, Zhang X, Fu M, Liang X, Ma Y, Zhan Q, Lin C, Xiang Y (2007). Ad-PUMA sensitizes drug-resistant choriocarcinoma cells to chemotherapeutic agents. Gynecol Oncol 107: 505-12.  Dobbelstein M (2003). Viruses in therapy—royal road or dead end? Virus Res 92: 219- 21.  Dummer R, Bergh J, Karlsson Y, Horowitz JA, Mulder NH, Huinink DTB , Burg G, Hofbauer G, Osanto S (2000). Biological activity and safety of adenoviral vector- expressed wild-type p53 after intratumoral injection in melanoma and breast cancer patients with p53-overexpressing tumors. Cancer Gene Ther 7: 1069-76.  Elliott RL, Blobe GC (2005). Role of transforming growth factor Beta in human cancer. J Clin Oncol 23: 2078-93.  143  Ermoian RP, Furniss CS, Lamborn KR, Basila D, Berger MS, Gottschalk AR, Nicholas MK, Stokoe D, Haas-Kogan DA (2002). Dysregulation of PTEN and protein kinase B is associated with glioma histology and patient survival. Clin Cancer Res 8: 1100-6.  Ernst DS, Eisenhauer E, Wainman N, Davis M, Lohmann R, Baetz T, Belanger K, Smylie M (2005). Phase II study of perifosine in previously untreated patients with metastatic melanoma. Invest New Drugs 23: 569-75.  Facchetti F, Monzani E, La Porta CA (2007). New perspectives in the treatment of melanoma: anti-angiogenic and anti-lymphangiogenic strategies. Recent Patents Anticancer Drug Discov 2: 73-8.  Frank O, Rudolph C, Heberlein C, von Neuhoff N, Schröck E, Schambach A, Schlegelberger B, Fehse B, Ostertag W, Stocking C, Baum C (2004). Tumor cells escape suicide gene therapy by genetic and epigenetic instability. Blood 104: 3543-9.  Gao K, Dai DL, Martinka M, Li G (2006). Prognostic significance of nuclear factor- kappaB p105/p50 in human melanoma and its role in cell migration. Cancer Res 66: 8382-8.  Gills JJ, Holbeck S, Hollingshead M, Hewitt SM, Kozikowski AP, Dennis PA (2006). Spectrum of activity and molecular correlates of response to phosphatidylinositol ether lipid analogues, novel lipid-based inhibitors of Akt. Mol Cancer Ther 5: 713-22.  Gilmore TD, Herscovitch M (2006). Inhibitors of NF-kappaB signaling: 785 and counting. Oncogene 25: 6887-99.  Grossman D, Kim PJ, Schechner JS, Altieri DC (2001). Inhibition of melanoma tumor growth in vivo by survivin targeting. Proc Natl Acad Sci USA 98: 635–40.  Grossman D, McNiff JM, Li F, Altieri DC (1999). Expression and targeting of the apoptosis inhibitor, survivin, in human melanoma. J Invest Dermatol 113: 1076-81.  Hanahan D, Weinberg RA (2000). The hallmarks of cancer. Cell 100: 57-70.  Hillen F, Griffioen AW (2007). Tumour vascularization: sprouting angiogenesis and beyond. Cancer Metastasis Rev 26: 489-502.  Hofseth LJ, Hussain SP, Harris CC (2004). p53: 25 years after its discovery. Trends Pharmacol Sci 25: 177-81.  Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, Fyfe G, Rogers B, Ross R, Kabbinavar F (2004). Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350: 2335-42.  144  Ito H, Kanzawa T, Miyoshi T, Hirohata S, Kyo S, Iwamaru A, Aoki H, Kondo Y, Kondo S (2005). Therapeutic efficacy of PUMA for malignant glioma cells regardless of p53 status. Hum Gene Ther 16: 685-98.  Jansen B, Schlagbauer-Wadl H, Brown BD, Bryan RN, van Elsas A, Müller M, Wolff K, Eichler HG, Pehamberger H (1998). Bcl-2 antisense therapy chemosensitizes human melanoma in SCID mice. Nat Med 4: 232-4.  Jetzt A, Howe JA, Horn MT, Maxwell E, Yin Z, Johnson D, Kumar CC (2003). Adenoviral-mediated expression of a kinase-dead mutant of Akt induces apoptosis selectively in tumor cells and suppresses tumor growth in mice. Cancer Res 63: 6697- 706.  Jiang H, Lin JJ, Su ZZ, Goldstein NI, Fisher PB (1995). Subtraction hybridization identifies a novel melanoma differentiation associated gene, mda-7, modulated during human melanoma differentiation, growth and progression. Oncogene 11: 2477-86.  Jiang H, Su ZZ, Lin JJ, Goldstein NI, Young CS, Fisher PB (1996). The melanoma differentiation associated gene mda-7 suppresses cancer cell growth. Proc Natl Acad Sci U S A 93: 9160-5.  Karst AM, Dai DL, Cheng JQ, Li G (2006). Role of p53 up-regulated modulator of apoptosis and phosphorylated Akt in melanoma cell growth, apoptosis, and patient survival. Cancer Res 66: 9221-6.  Klatzmann D, Chérin P, Bensimon G, Boyer O, Coutellier A, Charlotte F, Boccaccio C, Salzmann JL, Herson S (1998). A phase I/II dose-escalation study of herpes simplex virus type 1 thymidine kinase "suicide" gene therapy for metastatic melanoma. Study Group on Gene Therapy of Metastatic Melanoma. Hum Gene Ther 9: 2585-94.  Kluger HM, McCarthy MM, Alvero AB, Sznol M, Ariyan S, Camp RL, Rimm DL, Mor G (2007). The X-linked inhibitor of apoptosis protein (XIAP) is up-regulated in metastatic melanoma, and XIAP cleavage by Phenoxodiol is associated with Carboplatin sensitization. J Transl Med 5: 6.  Kondapaka SB, Singh SS, Dasmahapatra GP, Sausville EA, Roy KK (2003). Perifosine, a novel alkylphospholipid, inhibits protein kinase B activation. Mol Cancer Ther 2: 1093- 103.  Kreisberg JI, Malik SN, Prihoda TJ, Bedolla RG, Troyer DA, Kreisberg S, Ghosh PM (2004). Phosphorylation of Akt (Ser473) is an excellent predictor of poor clinical outcome in prostate cancer. Cancer Res 64: 5232-6.  Lasithiotakis KG, Sinnberg TW, Schittek B, Flaherty KT, Kulms D, Maczey E, Garbe C, Meier FE (2008). Combined inhibition of MAPK and mTOR signaling inhibits growth,  145 induces cell death, and abrogates invasive growth of melanoma cells. J Invest Dermatol 128: 2013-23.  Le S, Zhu JJ, Anthony DC, Greider CW, Black PM (1998). Telomerase activity in human gliomas. Neurosurgery 42: 1120-4.  Lebedeva IV, Sarkar D, Su ZZ, Kitada S, Dent P, Stein CA, Reed JC, Fisher PB (2003). Bcl-2 and Bcl-x(L) differentially protect human prostate cancer cells from induction of apoptosis by melanoma differentiation associated gene-7, mda-7/IL-24. Oncogene 22: 8758-73.  Li A, Dubey S, Varney ML, Dave BJ, Singh RK (2003). IL-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J Immunol 170: 3369–76.  Lindsley CW, Barnett SF, Yaroschak M, Bilodeau MT, Layton ME (2007). Recent progress in the development of ATP-competitive and allosteric Akt kinase inhibitors. Curr Top Med Chem 7: 1349-63.  Liu M, Acres B, Balloul JM, Bizouarne N, Paul S, Slos P, Squiban P (2004). Gene-based vaccines and immunotherapeutics. Proc Natl Acad Sci U S A 101: 14567-71.  LoPiccolo J, Granville CA, Gills JJ, Dennis PA (2007). Targeting Akt in cancer therapy. Anticancer Drugs 18: 861-74.  Los M, Roodhart JM, Voest EE (2007). Target practice: lessons from phase III trials with bevacizumab and vatalanib in the treatment of advanced colorectal cancer. Oncologist 12: 443-50.  McGill GG, Horstmann M, Widlund HR, Du J, Motyckova G, Nishimura EK, Lin YL, Ramaswamy S, Avery W, Ding HF, Jordan SA, Jackson IJ, Korsmeyer SJ, Golub TR, Fisher DE (2002). Bcl2 regulation by the melanocyte master regulator Mitf modulates lineage survival and melanoma cell viability. Cell 109: 707-18.  Meier F, Busch S, Lasithiotakis K, Kulms D, Garbe C, Maczey E, Herlyn M, Schittek B (2007). Combined targeting of MAPK and AKT signalling pathways is a promising strategy for melanoma treatment. Br J Dermatol 156: 1204-13.  Miller K, Wang M, Gralow J, Dickler M, Cobleigh M, Perez EA, Shenkier T, Cella D, Davidson NE (2007). Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 357: 2666-76.  Miller KD, Chap LI, Holmes FA, Cobleigh MA, Marcom PK, Fehrenbacher L, Dickler M, Overmoyer BA, Reimann JD, Sing AP, Langmuir V, Rugo HS (2005). Randomized phase III trial of capecitabine compared with bevacizumab plus capecitabine in patients with previously treated metastatic breast cancer. J Clin Oncol 23: 792-9.  146  Min YH, Cheong JW, Kim JY, Eom JI, Lee ST, Hahn JS, Ko YW, Lee MH (2004). Cytoplasmic mislocalization of p27Kip1 protein is associated with constitutive phosphorylation of Akt or protein kinase B and poor prognosis in acute myelogenous leukemia. Cancer Res 64: 5225-31.  Nakanishi K, Sakamoto M, Yamasaki S, Todo S, Hirohashi S (2005). Akt phosphorylation is a risk factor for early disease recurrence and poor prognosis in hepatocellular carcinoma. Cancer 103: 307-12.  Nam SY, Lee HS, Jung GA, Choi J, Cho SJ, Kim MK, Kim WH, Lee BL (2003). Akt/PKB activation in gastric carcinomas correlates with clinicopathologic variables and prognosis. APMIS 111: 1105-13.  O’Shea CC, Soria C, Bagus B, McCormick F (2005). Heat shock phenocopies E1B-55K late functions and selectively sensitizes refractory tumor cells to ONYX-015 oncolytic viral therapy. Cancer Cell 8: 61-74.  Pacifico F, Leonardi A (2006). NF-kappaB in solid tumors. Biochem Pharmacol 72: 1142-52.  Papp T, Jafari M, Schiffmann D (1996). Lack of p53 mutations and loss of heterozygosity in non-cultured human melanocytic lesions. J Cancer Res Clin Oncol 122: 541-8.  Pérez-Tenorio G, Stål O; Southeast Sweden Breast Cancer Group (2002). Activation of AKT/PKB in breast cancer predicts a worse outcome among endocrine treated patients. Br J Cancer 86: 540-5.  Perez EA, Scudder SA, Meyers FA, Tanaka MS, Paradise C, Gandara DR (1991). Weekly 24-hour continuous infusion interleukin-2 for metastatic melanoma and renal cell carcinoma: a phase I study. J Immunother 10: 57-62.  Petit AM, Rak J, Hung MC, Rockwell P, Goldstein N, Fendly B, Kerbel RS (1997). eutralizing antibodies against epidermal growth factor and ErbB-2/neu receptor tyrosine kinases down-regulate vascular endothelial growth factor production by tumor cells in vitro and in vivo: angiogenic implications for signal transduction therapy of solid tumors. Am J Pathol 151: 1523-30.  Raper SE, Chirmule N, Lee FS, Wivel NA, Bagg A, Gao GP, Wilson JM, Batshaw ML (2003). Fatal systemic inflammatory response syndrome in a ornithine transcarbamylase deficient patient following adenoviral gene transfer. Mol Genet Metab 80: 148-58.  Reid R, Mar EC, Huang ES, Topal MD (1988). Insertion and extension of acyclic, dideoxy, and ara nucleotides by herpesviridae, human alpha and human beta polymerases. A unique inhibition mechanism for 9-(1,3-dihydroxy-2- propoxymethyl)guanine triphosphate. J Biol Chem 263: 3898-904.  147  Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, Dowlati A, Lilenbaum R, Johnson DH (2006). Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 355: 2542-50.  Schallreuter KU, Kothari S, Chavan B, Spencer JD (2007). Regulation of melanogenesis- -controversies and new concepts. Exp Dermatol 17: 395-404.  Scott EN, Meinhardt G, Jacques C, Laurent D, Thomas AL (2007). Vatalanib: the clinical development of a tyrosine kinase inhibitor of angiogenesis in solid tumours. Expert Opin Investig Drugs 16: 367-79.  Selzer E, Schlagbauer-Wadl H, Okamoto I, Pehamberger H, Pötter R, Jansen B (1998). Expression of Bcl-2 family members in human melanocytes, in melanoma metastases and in melanoma cell lines. Melanoma Res 8: 197-203.  Spencer DM (2000). Developments in suicide genes for preclinical and clinical applications. Curr Opin Mol Ther 2: 433-40.  St George JA (2003). Gene therapy progress and prospects: adenoviral vectors. Gene Ther 10: 1135-41.  Stahl JM, Sharma A, Cheung M, Zimmerman M, Cheng JQ, Bosenberg MW, Kester M, Sandirasegarane L, Robertson GP (2004). Deregulated Akt3 activity promotes development of malignant melanoma. Cancer Res 64: 7002-10.  Stewart AK, Lassam NJ, Quirt IC, Bailey DJ, Rotstein LE, Krajden M, Dessureault S, Gallinger S, Cappe D, Wan Y, Addison CL, Moen RC, Gauldie J, Graham FL (1999). Adenovector-mediated gene delivery of interleukin-2 in metastatic breast cancer and melanoma: results of a phase 1 clinical trial. Gene Ther 6: 350-63.  Tang L, Tron VA, Reed JC, Mah KJ, Krajewska M, Li G, Zhou X, Ho VC, Trotter MJ (1998). Expression of apoptosis regulators in cutaneous malignant melanoma. Clin Cancer Res 4: 1865-71.  Terakawa N, Kanamori Y, Yoshida S (2003). Loss of PTEN expression followed by Akt phosphorylation is a poor prognostic factor for patients with endometrial cancer. Endocr Relat Cancer 10: 203-8.  Thallinger CWM, Wacheck V, Maierhofer H, Günsberg P, Polterauer P, Pehamberger H, Monia BP, Selzer E, Wolff K, Jansen B (2003). Mcl-1 antisense therapy chemosensitizes human melanoma in a SCID mouse xenotransplantation model. J Invest Dermatol 120: 1081-6.   148 Toloza EM, Hunt K, Swisher S, McBride W, Lau R, Pang S, Rhoades K, Drake T, Belldegrun A, Glaspy J, Economou JS (1996). In vivo cancer gene therapy with a recombinant interleukin-2 adenovirus vector. Cancer Gene Ther 3: 11-7.  Tong AW, Nemunaitis J, Su D, Zhang Y, Cunningham C, Senzer N, Netto G, Rich D, Mhashilkar A, Parker K, Coffee K, Ramesh R, Ekmekcioglu S, Grimm EA, van Wart Hood J, Merritt J, Chada S (2005). Intratumoral injection of INGN 241, a nonreplicating adenovector expressing the melanoma-differentiation associated gene-7 (mda-7/IL24): biologic outcome in advanced cancer patients. Mol Ther 11: 160-72.  Tsurutani J, Fukuoka J, Tsurutani H, Shih JH, Hewitt SM, Travis WD, Jen J, Dennis PA (2006). Evaluation of two phosphorylation sites improves the prognostic significance of Akt activation in non-small-cell lung cancer tumors. J Clin Oncol 24: 306-14.  van Blitterswijk WJ, Hilkmann H, Storme GA (1987). Accumulation of an alkyl lysophospholipid in tumor cell membranes affects membrane fluidity and tumor cell invasion. Lipids 22: 820-3.  van der Luit AH, Budde M, Ruurs P, Verheij M, van Blitterswijk WJ (2002). Alkyl- lysophospholipid accumulates in lipid rafts and induces apoptosis via raft-dependent endocytosis and inhibition of phosphatidylcholine synthesis. J Biol Chem 277: 39541-7.  Vile RG, Hart IR (1993). Use of tissue-specific expression of the herpes simplex virus thymidine kinase gene to inhibit growth of established murine melanomas following direct intratumoral injection of DNA. Cancer Res 53: 3860-4.  Vucic D, Fairbrother WJ (2007). The inhibitor of apoptosis proteins as therapeutic targets in cancer. Clin Cancer Res 13: 5995-6000.  Yamamoto S, Tomita Y, Hoshida Y, Morooka T, Nagano H, Dono K, Umeshita K, Sakon M, Ishikawa O, Ohigashi H, Nakamori S, Monden M, Aozasa K (2004). Prognostic significance of activated Akt expression in pancreatic ductal adenocarcinoma. Clin Cancer Res 10: 2846-50.  Yang L, Dan HC, Sun M, Liu Q, Sun XM, Feldman RI, Hamilton AD, Polokoff M, Nicosia SV, Herlyn M, Sebti SM, Cheng JQ (2004). Akt/protein kinase B signaling inhibitor-2, a selective small molecule inhibitor of Akt signaling with antitumor activity in cancer cells overexpressing Akt. Cancer Res 64: 4394-9.  Yu J, Yue W, Wu B, Zhang L (2006). PUMA sensitizes lung cancer cells to chemotherapeutic agents and irradiation. Clin Cancer Res 12: 2928-36.    APPENDIX  149   150

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0066678/manifest

Comment

Related Items