ios press the molecular pathology of cutaneous melanoma...in melanomas, id1 expression is limited to...

Cancer Biomarkers 9 (2011) 267–286 267DOI 10.3233/CBM-2011-0164IOS Press

The molecular pathology of cutaneousmelanoma

Thomas Bogenriedera and Meenhard Herlynb,∗aBoehringer Ingelheim RCV, Dr. Boehringer Gasse 5-11, 1121 Vienna, AustriabThe Wistar Institute, 3601 Spruce Street, Philadelphia, PA, USA

Abstract. Cutaneous melanoma is a highly aggressive cancer with still limited, but increasingly efficacious, standard treatmentoptions. Recent preclinical and clinical findings support the notion that cutaneous melanoma is not one malignant disorderbut rather a family of distinct molecular diseases. Incorporation of genetic signatures into the conventional histopathologicalclassification of melanoma already has great implications for the management of cutaneous melanoma. Herein, we review ourrapidly growing understanding of the molecular biology of cutaneous melanoma, including the pathogenic roles of the mitogen-associated protein kinase (MAPK) pathway, the phosphatidylinositol 3 kinase [PI3K]/phosphatase and tensin homologue deletedon chromosome 10 [PTEN]/Akt/mammalian target of rapamycin [mTOR])PTEN (phosphatase and tensin homolog) pathway,MET (hepatocyte growth factor), Notch signaling, and other key molecules regulating cell cycle progression and apoptosis. Themutation Val600Glu in the BRAF oncogene (designated BRAF(V600E)) has been associated with clinical benefit from agents thatinhibit BRAF(V600E) or MEK (a kinase in the MAPK pathway). Cutaneous melanomas arising from mucosal, acral, chronicallysun-damaged surfaces sometimes have oncogenic mutations in KIT, against which several inhibitors have shown clinical efficacy.These findings suggest that prospective genotyping of patients with melanoma, combined with the growing availability of targetedagents, which can be used to rationally exploit these findings, should be used increasingly as we work to develop new and moreeffective treatments for this devastating disease.

Keywords: Melanoma, cell cycle, pRb pathway, p16INK4A, p14ARF, cyclins, CDKs, Apaf1, Bcl2, MAPK pathway, Ras, BRAF,PTEN, PI3K/AKTpathway, SCF, ckit, HGF, cmet, cmyc, Notch

1. Introduction

As the treatment of metastatic melanoma movesrapidly into the era of targeted therapy, there is anever growing need to untangle the underlying genet-ic complexity and cellular signalling heterogeneity ofthis highly malignant tumor. It is widely hoped thatan understanding of how genetic and signalling profilesdictate pharmacologic response will allow for the se-lection of optimal patient sub-populations for clinicaltrials.

The molecular pathology of melanoma developmentand progression is now being deciphered (Fig. 1). Al-though we have an increasingly good understanding ofthe genetics and biochemistry of these events, themech-anisms underlying their combinatorial interactions arestill poorly defined and need to be fathomed to ulti-

∗Corresponding author. E-mail: [email protected].

mately lead to more successful therapeutic strategies.The goal of this chapter, therefore, is to highlight someof the issues that have arisen. For the purpose of thischapterwe will focus on themolecular events occurringwithin the melanoma cell, especially those pathwayswhich have been the scope of intensive research over thepast decade and which are now targets of an increasingarmamentariumof novel therapeutic approaches. Here,we discuss recent findings and hypotheses on the roleof growth factor pathways in the molecular pathologyof melanoma. The matter of melanoma-stroma interac-tions is too involved to thoroughly explore here. As theliterature on molecular and cellular events in melanomadevelopment and progression is now huge and includescomprehensive reviews [1–5], we will take the libertyof a more subjective view and review molecules andpathways that now take center stage as well as dis-cuss emerging areas and interesting questions in themelanoma field.

ISSN 1574-0153/11/$27.50 2011 – IOS Press and the authors. All rights reserved

268 T. Bogenrieder and M. Herlyn / The molecular pathology of cutaneous melanoma

Fig. 1. Development of melanoma. This model implies that melanoma commonly develops and progresses in a sequence of steps from neviclesions, which can be histologically identified in approximately 35% of cases. The model also acknowledges that melanoma may also developdirectly from normal cells. The exact role of tumor-initiating or melanoma stem cells remains to be elucidated. Genetic (such as mutationsand deletions of critical genes) and epigenetic (such as DNA-methylation leading to gene silencing) alterations of critical pathways as wellas the acquired capability of evading apoptosis, progressively disturb normal skin homeostatis leading to melanoma development. Geneticchanges are expected at the transition from common acquired (benign) nevus to dysplastic nevus and primary melanoma allowing cells to persist.During melanoma invasion and metastasis increased growth, invasion and stromal ‘landscaping’ by proteolysis occurs, leading to an increasedmanipulation of the microenvironment and interactions with stromal skin cells.

2. The unmet clinical need: “Melanoma is thetumor that gives cancer a bad name”

Melanoma is the deadliest form of skin cancer andone of the most aggressive of all neoplasms. The in-cidence of cutaneous melanoma has increased rapidlyduring the past several decades, and continues to do soat an alarming rate: The National Institutes of Healthpredicts 62,480 new cases of melanoma in the year2009,with 8,420 of the patients succumbing to their dis-ease [3]. The majority of these deaths are due to distantmetastases from the primary site because melanoma isnotorious for its propensity to metastasize. Althoughlocalized melanoma is frequently curable by surgicalexcision, metastatic melanoma is inherently resistantto most systemic treatments, and survival of patientswith advanced disease has not improved in more than30 years [6,7]. Melanoma is poorly responsive to cyto-toxic chemotherapy, and thus the survival of patients isbased on screening, early detection, and wide resectionof the primary lesion. The overall survival for patients

with metastatic melanoma ranges from only 4.7 to 11months, with a median survival of 8.5 months [8].

2.1. Melanoma is characterised by the dysfunction ofthe epidermal melanin unit

Melanoma arises from the transformation of neu-ral crest-derived melanocytes, the pigment cells of theskin, which reside in the basal layer of the epidermis.We now view melanoma as a complex tissue resultingfrom disrupted skin homeostasis, rather than focusingon the melanoma cell, and the genes within it, alone.Normal skin homeostasis is maintained by dynamic in-teractions between the melanocytes and their microen-vironment, such as keratinocytes, fibroblasts, endothe-lial and immunocompetent cells, and the extracellularmatrix. Similarly, during transformation and progres-sion of melanocytes to melanoma cells, there are, how-ever deregulated, reciprocal and conspirational inter-actions between the neoplastic cells and the adjacentstromal cells [9,10].

T. Bogenrieder and M. Herlyn / The molecular pathology of cutaneous melanoma 269

Under normal physiological conditions,melanocytesand keratinocytes form the ‘melanin unit’ in the epi-dermis. In this unit, melanocytes and keratinocytes areevenly aligned along the basement membrane zone at aratio of 1:5–8. Each melanocyte extends with its den-drites into the upper layers of the epidermis, transfer-ring pigment-containingmelanosomes to approximate-ly 35 keratinocytes in its unit via dendritic process-es. The melanin contained in these melanosomes ab-sorbs and scatters damagingultraviolet radiation, there-by shielding the nucleic acids on the skin from damage(reviewed in [9–12]).

The undifferentiated keratinocytes of the basal lay-er regulate melanocyte growth, number of dendritesand expression of cell surface molecules, providing ev-idence that this highly ordered organizational patternserves as the structural basis for intercellular regulation.Information is not yet available on how melanocytesand keratinocytes can maintain this lifelong balance,which is only disturbed during transformation into amelanocytic nevus (‘mole’) or a melanoma. It is nowgenerally accepted that melanoma risk is modulatedby skin pigmentation patterns, such as those linked toMC1R polymorphisms, and early exposure to ultravio-let (UV) light (reviewed in [13]).

3. Selected components of the cell cycle machinery

Recent excellent reviews summarize our current un-derstanding of cell cycle regulation [14–16]. Mitogenicgrowth factors promote the entry of quiescent cells intothe first gap phase (G1) and initiation of DNA synthesis(S phase) of the cell cycle (Fig. 2). G1 cyclin-dependentkinases (CDKs) serve as positive regulators. D-typecyclins (D1, D2, D3) complex with CDK4 and CDK6to stimulate their kinase activities, which in turn causethe phosphorylation and inactivation of the retinoblas-toma (Rb) tumor suppressor protein. By binding toE2F, Rb recruits histone deacetylases to the promotersof E2F-responsive genes and represses their transcrip-tion. Multiple Ras effector pathways (details of whichforthcoming) appear to play an important role in theregulation of cyclin D1. Cyclin D1, in part, regulatesthe kinase activities of both CDK4 and CDK6. Thesecomplexes are formed in the cytoplasm and are trans-ported into the nucleus and undergo stimulatory modi-fications including phosphorylationby CDK-activatingkinase (CAK) to yield active holoenzymes. Further in-to G1, cyclin E complexes with CDK2 and causes ad-ditional phosphorylation and inactivation of Rb. With

sufficient phosphorylation of Rb, E2F is released andtransactivates genes required for S phase entry, includ-ing cyclins E and A. CDK inhibitors (CKIs) serve asnegative regulators of the Rb pathway. CKIs are clas-sified into two distinct families on the basis of theirstructural and functional characteristics. The mem-bers of the INK4 family of CKIs (p16Ink4a, p15Ink4b,p18Ink4c, and p19Ink4d) contain multiple ankyrin re-peats and act as negative regulators of CDK4/6 by bind-ing to the catalytic subunit and preventing formation ofthe active cyclin-CDK complex. The Cip/Kip family ofCKIs (p21Cip1/Waf1, p27Kip1, and p57Kip2) is morebroadly acting and regulates both CDK4/6 and CDK2activity. Each member of the family contains a char-acteristic motif within the amino-terminal region thatenables them to bind to both cyclin and CDK subunits.The stoichiometry between CDKs and CKIs is impor-tant and determines the activity of Rb and, ultimately,the proliferative state of cells.

3.1. Alterations of the pRb pathway in melanoma

The p16-cyclin D/Cdk4-pRb pathway is deregulatedin a large majority of human cancers, either throughloss of p16 or pRb function or through deregulated ex-pression of cyclin D or cdk4 (reviewed in [14–16]).Several mechanisms of inactivation of the pRb path-way have been noted in various malignant tumors andcell lines thereof. The most common known ways inwhich these gene products are dysregulated in neo-plasms are through gene deletion (homozygousor hem-izygous); inactivating mutations; epigenetic changessuch as promoter methylation, transcriptional repres-sion, protein sequestration and inactivation (e.g., vi-ral oncoproteins); and posttranslational modifications(e.g., inactivating phosphorylation events). The p16-cyclin D/Cdk4-pRb pathway as a functional unit is fre-quently altered in melanoma pathogenesis [14]. Infact, in one study, virtually all (96%) of investigatedmelanoma cell lines harboured alterations at the DNAlevel within CDKN2A/p16, CDKN2B/p15, CDK4 andCCND1/cyclin D1 [17]. Inactivation mutations in thepRb ‘pocket’ region, which blocks binding to E2F,are rare in melanoma. Instead, the function of pRbin melanoma is commonly inhibited by hyperphos-phorylation and, to a lesser extent, by underexpres-sion [18]. In a recent study, Curtin and colleaguesconducted a genome-wide analysis of DNA copy num-ber and mutational analysis of BRAF and NRAS in 126melanomas from individualswith varyingUV exposurehistories [19]. They found that amplification of CDK4,


Fig. 2. Cell cycle regulatory proteins. The cell cycle is controlled by a series of complex interactions among multiple proteins that interact ina precisely orchestrated sequence. Illustrated here is a highly schematic summary of the primary protein interactions, with special attention tocomponents involved in the G1 to S transition of the cell cycle. Rb is one of the targets for the enzymatic activity of cyclin-cdk complexes.p21 is transcriptionally regulated by p53 and inactivates various cdks. p27 inactivates the same complexes as p21. p16 and p15 have a morerestricted activity and form binary complexes with cdk4 and cdk6, displacing D-type cyclins. TGF-β can downregulate the cyclin E/cdk2 activityby activating the transcription of two inhibitors of cyclin-dependent protein kinases, p21 and p15. A surge in the level of p15 displaces p27 fromthe complex of cyclin D/cdk4/6 and the resulting p27 binds to the cyclin E/cdk2 complex. This inactivates its kinase activity and blocks cell cycleprogression.

was commonly seen in acral and mucosal melanomasbut not observed in tumors with activating BRAF orNRAS mutations or Cyclin D1 copy number gains. Fur-thermore, losses/deletions of the melanoma suppres-sor CDKN2A locus were also more commonly detectedin mucosal and acral melanomas, but only in sampleswithout CDK4 amplification.

3.2. p16INK4A/p14ARF in primary specimen

Since its discovery as a CKI in 1993, the tumorsuppressor p16 has received widespread attention inmelanoma research (see [14] and references therein).A plethora of studies show that in sporadic melanomas,p16 is inactivated through homozygous deletion (6–25%), mutation (0–26%) and methylation (0-10%)in roughly one forth of primary tumors. Based onan immunohistochemical analysis of 103 melanocyt-

ic lesions representing all stages in the progression ofmelanoma, Reed et al. [20] suggest that loss of p16protein expression is not necessary for melanoma initi-ation as all in situ lesions and the majority of invasivemelanomas retained expression of this protein. Loss ispotentially more related to invasiveness or the abilityto metastasize, because 52% of primary invasive tu-mors and 72% of metastatic lesions showed partial orno expression of p16. In thin to intermediate (< 4 mm)sporadic melanomas, however, the frequencies of LOHat the INK4A locus, INK4A intragenic mutations andpromoter methylation are extremely low (10%) [21].Loss of p16 expression is associated with progressionof disease [20], suggesting that other regulatory mech-anisms allow for decreased p16 in melanomas.

In this regard, the transcriptional regulatory proteinId1 (inhibitor of differentiation) has recently been iden-tified as a repressor of p16/INK4A transcription [21].


In melanomas, Id1 expression is limited to the in situcomponent of invasive melanomas and perivascularregions of metastatic tumors. Id1 expression corre-lates with decreased p16INK4A expression in RGPmelanomas, suggesting that Id1 transcriptional repres-sion may represent one of the earliest mechanisms ofp16 dysregulation inmelanoma initiation. Consequent-ly, a model for p16 inactivation via multiple steps inmelanomagenesis has been proposed [21]: Reversibletranscriptional inactivation in RGP melanomas (viaId1 or other repressors), subsequent acquired epige-netic changes (e.g., promoter methylation), and finallyirreversible genetic alterations (e.g., mutations, dele-tions) associated with melanoma invasion and metas-tasis. Taken together, these findings indeed implicatederegulation of the p16/pRb pathway as a critical earlyor initiating event in the development of melanoma.

The INK4A/ARF locus on chromosome 9p21 alsoencodes another tumor suppressor gene called p14ARF

(p19ARF in mice). Exon 1β, located approximately15kb centromeric to exon 1α of p16, is spliced on-to exons 2 and 3 of p16 to generate this alternativetranscript. Alternate promoters regulate the indepen-dent production of these 2 mRNAs and translation ofthe p14ARF mRNA occurs in a reading frame alter-nate to p16. p19ARF acts through MDM2 to inhibitp53 degradation and exon 1β is sufficient for p19ARF -mediated stabilization of p53 function (see [11,16] forreview). Thus far, there are no molecular data support-ing a tumor suppressive role for p14ARF in sporadicmelanoma. Clearly, however, homozygous deletionsat the INK4A/ARF locus have the capacity to inacti-vate both p14ARF and p16 function through the sharedsecond exon.

3.3. Expression and prognostic relevance ofCyclins/CDKs/CKIs in melanoma

As in other malignant tumors, certain cell-cycle reg-ulators are differentially expressed during melanomaprogression and are independent prognostic markers,albeit with somewhat incongruent results. Florenes andcolleagues [11,16,22] observed only rare cyclin D3-positive cells and no cyclin D1-positive cells in benignnevi. In contrast, cyclin D3/cyclin D1 was found tobe expressed by 96%/62% of primary and 97%/29%of metastatic melanomas, respectively. Kaplan-Meieranalysis revealed that high levels of cyclin D3 is an in-dicator of early relapse and decreased overall survivalfor patients with superficial spreading, but not nodu-

lar melanoma, whereas cyclin D1 does not have anyimpact on disease-free and overall survival.

p27Kip1 is strongly expressed by normal melano-cytes and benign nevi, whereas in nodular melanomas,the level of p27Kip1 was found to correlate significant-ly with the thickness of the tumor, with less proteinexpressed in thicker lesions [23]. Patients having tu-mors with fewer than 5% p27Kip1-staining cells havea significantly higher risk of early relapse of their dis-ease compared with those expressing moderate or highlevels. In contrast, the level of p27Kip1 does not corre-late with tumor thickness or disease-free survival in pa-tients with superficial spreading melanomas. Further-more, p27Kip1 does not appear to have an influence onoverall survival for either subgroup. When examinedthe combined effect of p21Cip1/Waf1 and p27Kip1 onclinical outcome, it was found that analysis of thesetwo CKIs together may have greater prognostic poten-tial than either alone. These data underscore the valueof analyzing multiple cell cycle regulatory proteins toobtain the most reliable indication of prognosis [23].

A significant increase of cyclin E and CDK2 ex-pression occurs during tumour progression in melan-omas [24]. Cyclins B1, D2, and D3 show significant-ly increased expression in metastases, but normal oreven decreased expression in primary melanomas. Incontrast, cyclins A and D1, and CDK1 and CDK4 areexpressed very weakly in situ with no significant differ-ences between nevi, primary melanomas or metastases,and histopathological staging. Approximately 30% ofprimary melanomas and 40% of metastases completelylack p21Cip1Waf1 expression. In superficial spread-ing melanomas, a significant correlation between pro-tein expression and tumor thickness was found, withthin lesions showing low protein levels. By comparingprimary and metastatic specimens obtained from thesame patient, a reduction in p21Cip1Waf1 staining wasobserved in the latter [25].

4. Selective components of the apoptotic program

Acquired resistance toward apoptosis (programmedcell death) is a hallmark of cancer [26]. The conceptthat apoptosis might influence the malignant phenotypewas first raised in 1972 by Kerr, Wyllie and Currie.However, the importance of apoptosis in the pathogen-esis of cancer remained under-appreciated for almosttwo decades. A comprehensive overview of apopto-sis is beyond the scope of this chapter (for review, seerefs [27,28]). As a general summary, most apoptotic


pathways involve a sensor that detects a death-inducingsignal, a signal transduction network and executionmachinery that actively carries out the process of celldeath. DNA damage can induce apoptosis through acentral sensor, p53, although p53-independent path-ways clearly exist (Fig. 3). Many of the signals that elic-it apoptosis converge on the mitochondria, which re-spond to pro-apoptotic signals by releasing cytochromec. Cytochrome c can interact in a multi-protein com-plex with Apaf1 and pro-caspase-9, leading to caspase-9 activation and the initiation of a protease cascade.One of the most important modulators is Bcl-2, owingto its ability to affect cytochrome c release from mi-tochondria and/or modulate the Apaf1/caspase-9 inter-action. However, neither Bcl-2 nor Bcl-xL has beenshown to be completely cytoprotective against Apaf-1/caspase-9 mediated cell death.

Bcl-2 is the founding member of an expanding genefamily that includes other anti-apoptotic moleculessuch as Bcl-xL as well as pro-apoptotic members suchas Bax. A series of enzymes known as caspases areconsidered the engine of apoptotic cell death. Cas-pases are cysteine proteases that are expressed as in-active pro-enzymes. The “signaling” caspase-9 asso-ciateswithApaf-1 (apoptotic protease activating factor-1), and oligomerization of this complex in the presenceof cytochrome c can activate the downstream “effector”caspase cascade. Thus, essential downstream compo-nents of p53-induced cell death include caspase-9 andits adapter Apaf-1 [29]. It is now becoming increas-ingly clear that mutations in many cancer-related genescan disrupt the apoptotic machinery and compellingevidence indicates that apoptotic activity is importantin tumor suppression.

4.1. Defects of the apoptotic program in melanoma

p53 was the first tumor suppressor gene linked toapoptosis. p53 mutations occur in the majority of hu-man tumors and are often associated with advanced tu-mor stage and poor patient prognosis. Moreover, sev-eral upstream and downstream components of the p53pathway (e.g. Mdm-2, p14ARF and Bax) are altered inhuman tumors [26,30,31] (Fig. 3). What triggers apop-tosis during tumor development? A variety of signalsappear important [26,31–34]. Extracellular triggers in-clude radiation, growth/survival factor depletion andloss of cell-matrix interactions and cell-cell adherence-based survival signals. In the skin, excessive exposureto UV radiation induces apoptosis, which presumablyserves to eliminate heavily damaged cells. UV irradia-

tion induces apoptosis, and loss of p53 function leadsto the survival of these damaged cells thereby initiatingtumor development. Hence, loss of apoptotic functioncan impact tumor initiation, progression and metasta-sis.

p53 mutations were found to be associated with poorprognosis, de novo or acquired resistance, and relapsein a broad field of solid and hematologic malignancies(for review, see refs [31]). Melanoma is a radiation-and chemotherapy-refractory neoplasm and there is nostandard therapy for patients with disseminated dis-ease: the commonly used anticancer drugs do not al-ter the prognosis, i.e. the fatal outcome [8], and arehighly heterogeneous in their in vitro effects. However,contrary to other malignancies, p53 mutations are veryrare in melanoma, being less than 5%. Tumor-derivedp53 does localize to the nucleus in melanoma, howev-er, its transcriptional activity is weak and overexpres-sion of wild-type p53 does not induce immediate celldeath. Another intriguing factor is that cells which car-ry mutated p53 undergo apoptosis when overexpressedwith wild-type p53 [35]. This suggests that p53 is keptin an inactive state, either by other factors (such asApaf-1) or through post-transcriptional mechanism(s),thereby functionally inactivating its pathways. Thus, inmelanoma studies fail to correlate p53 mutations withreduced toxicity to anticancer agents.

4.2. Apaf-1

Findings by Scott Lowe and colleagues at the begin-ning of the decade [36] provided exciting and promisingnew insights into mechanisms how melanomas becomeapoptosis-, and thus chemotherapy-resistant, despite afully functional p53. They show that melanoma cellsof different progression stages can avoid apoptosis byinactivating a downstream component of the apoptoticpathway, namely Apaf-1, thus disabling the p53 apop-totic program. What makes the results of Soengas et al.even more interesting is that they showed a) a deletionof one of the alleles in 42% of the specimens tested andb) a reversible “switching off” of the remaining alleleby methylation, rather than a permanent deactivationby deletion or mutation. The methylation of Apaf-1could experimentally be reversed by a methylation in-hibitor (5-aza-2dC) and thus its pro-apoptotic functionrestored. Pre-treatment assessment of Apaf-1 statusmay therefore improve the therapeutic management ofpatients with melanoma.


Fig. 3. Interrelationship between selected key molecules in the apoptotic program in melanoma. Apoptosis may involve either death-receptor-mediated (extrinsic) or mitochondria-mediated (intrinsic) events. The basic premise is that p53, although wild-type, is functionallyinactive to induce apoptosis due to silencing of Apaf-1. The apoptotic p53 pathway includes cytochrome c release from mitochondria, activationof Apaf-1, its complex formation with caspase-9, and the “effector” caspase family. Apaf-1-mediated apoptosis is inhibited by IAPs (inhibitor ofapoptosis proteins). Post-transcriptional modifications such as phosphorylation by Akt/PKB also influence the apoptotic effects of caspase-9. Inaddition, cytochrome c release from the mitochondria is inhibited by anti-apoptotic proteins such as Bcl-2/Bcl-xL.

4.3. Bcl-2 and melanoma

Bcl-2 is a strongly antiapoptotic protein that is ex-pressed in normal melanocytes, benign nevi, primarymelanoma, and melanoma metastases. At this time, theoverall importance of Bcl-2 expression by stage andlesion type remains controversial (reviewed in [37]).Bcl-2 protein is located within the inner layers of themitochondrial membrane, where it prevents apoptosisby blocking the release of cytochrome C from mito-chondria. In this role, Bcl-2 promotes resistance toanticancer therapy and could contribute to enhancedmetastatic potential. Moreover, Bcl-2 protein has beenfound in normal developing melanocytes, and the lossof Bcl-2 function in Bcl-2 knockout mice leads to ac-celerated disappearance of melanocytes. The mice areviable but rapidly lose hair color and turn from blackto white [38]. Drug resistance in melanoma has beenpartially attributed to overexpression of Bcl-2, which

blocks the release of cytochrome C [28]. Experimen-tal transfection of cells with Bcl-2 confers a multidrugresistant phenotype in both hematologic and solid tu-mor cells [39]. Conversely, pharmacologic reduction ortargeted inactivation of Bcl-2 amplifies anticancer re-sponses to chemotherapy in multiple in vivo models, in-cluding melanoma [40]. Hence, several factors supporta role for Bcl-2 antisense therapy to treat melanoma,based on evidence suggesting a critical role for Bcl-2in melanoma cell survival.

5. Mitogen-activated protein kinase (MAPK)signal transduction pathways

Mitogen-activated protein kinase (MAPK) pathwaysare cellular signalling pathways that enable cells totransduce extracellular signals to an intracellular re-sponse (Fig. 4) [41,42]. The molecular details of mam-


malian MAP kinase signal transduction pathways acti-vated by stress and inflammation have only begun to bedissected. The impact of these pathways on, amongstothers, the pathology of cancer, the side effects of can-cer therapy, chronic inflammation, embryonic develop-ment, innate and acquired immunity, is profound. Thusit is not surprising that understanding these pathwayshas attracted wide interest, and in the past 10 years,dramatic progress has been made. Accordingly, it isnow becoming possible to envisage the transition ofthese findings to the development of novel treatmentstrategies.

5.1. Ras-Raf-MEK-MAP kinase pathway

Since the discovery of the role of Ras oncogenesin carcinogenesis in the early 80s, we have witnessedan explosion of research in the signal transductionarea [41,42]. Likewise, a host of growth signal-ing pathways is now implicated in melanocyte andmelanoma biology. They include cell growth and dif-ferentiation, melanogenesis, cell fate determination,survival and apoptosis. The Ras gene product is amonomeric, membrane-localized guanosine triphos-phate (GTP)/guanosine diphosphate (GDP)-binding(G) protein of 21 kd that functions as a nodal point,a molecular “switch” converting signals from receptorand non-receptor tyrosine kinases on the cell membraneto downstream cytoplasmic or nuclear events (Fig. 4).There are multiple effects that occur after Ras activa-tion, initiating a multitude of cytoplasmic signallingcascades, leading to protein synthesis and regulationof cell survival, proliferation, and differentiation. Theprecise functions of Ras proteins continue to be elu-cidated. Ras utilizes a variety of downstream effec-tors and mediates its actions in mechanistically distinctways, depending on both species and cell-type identity.In turn, these further complicate our ability to define asimple relationship between Ras and, for example, cellcycle progression. The MAP kinase (MAPK) pathwayhas emerged as the crucial route between membrane-bound Ras and the nucleus (Fig. 4). This pathway en-compasses a cascade of phosphorylation events involv-ing three key kinases, namely Raf, MEK (MAP kinasekinase) and ERK (MAP kinase). The critical nucleartarget of the Ras-Raf-MEK-MAP kinase pathway arethe transcription factors. Another well-characterizedeffector of Ras is the phosphoinositide 3-phosphatelipid kinase (PI3K)-AKT/PKB pathway (Fig. 4).

Each human cell contains at least three distinct rasproto-oncogenes, which encode four highly related but

distinct proteins, H-ras, N-ras, and K-ras (K4A- andK4B-). Activating single-point mutations of the rasgene can result in constitutive activation of Ras pro-tein, thereby continuously stimulating cell prolifera-tion and inhibiting apoptosis. These mutated formsof Ras have impaired GTPase activity. Although theystill bind GTPase activating proteins (GAP), there is no“off” sign, since GTPase is no longer activated. Acti-vating mutations are limited to a small number of sites(codons/amino acids 12, 13, 59, and 61), all of whichabolish GAP-induced GTP hydrolysis of the Ras pro-teins. Oncogenic K-ras mutations occur frequently innon-small-cell lung, colorectal, and pancreatic carcino-mas; H-ras mutations are common in bladder, kidney,and thyroid carcinomas; and N-ras mutations are foundin hepatocellular carcinoma, hematologicmalignanciesand melanoma [41,42].

5.2. Ras genes and melanoma

Within the ras gene family, only N-ras has a signif-icant association with melanoma progression, whereasH-ras or K-ras are rarely mutated. Mutations in N-rashave since been identified in 15–20%of all melanomas,and are most commonly the result of the substitu-tion from leucine to glutamine at position 61 [43,44].The presence of N-ras mutations appears to corre-late with the vertical growth phase (VGP) and withmelanoma arising in chronically UV-exposed bodysites. Moreover, N-ras mutations are associated withnodular melanomas and to a lesser extent with lentigomaligna melanoma. Benign and dysplastic nevi didnot exhibit ras mutations, however, N-ras mutations incongenital nevi have been observed. N-ras mutationsdid not correlate with metastasis or survival parameters(reviewed in [3,4,11]).

5.3. BRAF in benign nevi and melanoma

When active in its GTP-bound state, RAS activatesa number of downstream effectors, one of which isthe RAF family of serine/threonine kinases. Thereare three isoforms of RAF, namely, A-Raf, BRAF,and CRAF (also called Raf-1) (reviewed in [3,41,42,45]). Once activated, RAF stimulates the MAPK cas-cade, resulting in the sequential activation of MEK1and MEK2, which in turn activates ERK1 and ERK2(Fig. 4). Once activated, the ERKs either activate cy-toplasmic targets or migrate to the nucleus, where theyphosphorylate transcription factors. In melanocytes,the MAPK (ERK) pathway is activated by growth fac-


Fig. 4. The Ras/RAF/MEK/ERK and the PI3K/AKT signalling pathways in melanoma. Ligand-mediated dimerization of the receptor tyrosinekinases (RTKs) and subsequent autophosphorylation or transphosphorylation leads to their association with a variety of cytoplasmic phosphoty-rosine binding proteins. This results in the initiation of a phosphorylation cascade and activation of several downstream pathways involved in cellgrowth and survival, including the Ras/Raf/MEK/ERK and PI3K/Akt pathways. Stimulation of these pathways transmits a signal to the nucleusresulting in modification of gene transcription patterns that ultimately affects processes such as cell division, apoptosis, adhesion, migration,and/or differentiation.

tors released from the local microenvironment, such asstem cell factor (SCF), α-melanocyte-stimulating hor-mone (α-MSH), and hepatocyte growth factor (HGF).Under physiological conditions, these growth factorsonly induce a weak stimulation of the MAPK (ERK)pathway that is insufficient to induce melanocyte pro-liferation. In most melanoma cells, the situation is verydifferent and it has been shown that > 90% of clinicalmelanoma specimens have continuous hyperactivity inthe MAPK (ERK) pathway.

Particularly BRAF has recently attracted enormousbiological and therapeutical interest, as the b-rafgene is found mutated in at least 60% of cutaneousmelanomas [44], and at a lower frequency in manyother human malignancies. This discovery has firmlyestablished the involvement of Raf kinases in cancer.The most common mutation (over 80%) is a V600Echange in the activation loop that induces the constitu-tive activation of catalytic activity [46]. Curiously, inmelanoma this mutation is rare in unexposed or chron-

ically sun-damaged skin, but frequent in skin with in-termittent sun exposure and often accompanied by am-plification of the mutant allele [47]. The importance oflocalization is underlined by the observation that B-Rafmutations do not occur in melanomas of the uvea [48].Furthermore, the frequency of BRAF mutations inmelanoma is positively linked with genetic variants ofthe melanocortin-1 receptor in melanocytes [49]. Ad-ditionally, melanomas with wild-typeBRAF or N-RASfrequently have increases in the number of copies ofthe genes for cyclin-dependent kinase 4 (CDK4) andcyclin D1 (CCND1), downstream components of theRAS-BRAF pathway [19]. What is more, constitutiveMAPK (ERK) activity increases cyclin D1 and down-regulates p27 expression in melanoma cells [50].

Interestingly, however, a high frequency of BRAFV600E mutations (73–82%) has also been reportedin common benign naevi, suggesting that BRAF ac-tivation might be the critical inducer of these benign‘precursor’ lesions [51–53]. Hence, acquisition of the


BRAF V600E mutation seems to be an early eventin melanoma development with a high percentage ofmelanocytic nevi also found to harbor the mutation. In-terestingly, the presence of the BRAF V600E mutationalone is not sufficient to oncogenically transform naevicells into melanoma [54]. Instead, the forced expres-sion of BRAF V600E in primary human melanocytesleads to an irreversible growth arrest characteristic ofsenescence [55]. Pathological studies have confirmedthese in vitro findings, and showed that most nevi aregrowth arrested and express many markers of senes-cence [55]. This process, termed “oncogene-inducedsenescence,” is thought to be an important mechanismin protecting cells from malignant change [3].

It has been suggested that the presence of a BRAFV600E mutation, rather than an NRAS or low-activityBRAF mutation, predicts for sensitivity to both MEKand BRAF inhibitors. [4,56,57]. Recent investigationshave also sought to determine the prognostic relevanceof BRAF mutations in melanoma. However, studiesto date have failed to identify a link between BRAFmutational status and disease outcome [58]. In anoth-er study, patients with BRAF mutation showed longerdisease-free survival (median of 12 months) than pa-tients without mutation (median of 5 months), althoughthis association was not statistically significant [59].From such tiny samplings it would be reckless to drawdefinite conclusions, but it is interesting to note thatthus far no molecule has emerged as a bona fide pre-dictive or prognostic maker in melanoma for routineclinical use.

There are also possibilities that BRAF/MAPK sig-naling may allow melanoma cells to escape immunesurveillance through suppression of their highly im-munogenic pigmentation antigens [60]. Moreover,it has been shown that treating melanoma cells withan RNAI to V600E BRAF mutation reduces the re-lease of immunosuppressive cytokines from melanomacells [61].

The melanoma MAPK (ERK) signalling pathway ishighly robust, with BRAF-V600E mutated melanomacells rapidly developing resistance to BRAF inhibitionin vitro [62]. Rather than this resulting from an ac-quired genetic change in the melanoma cells, resis-tance instead results from a simple switching withinthe MAPK pathway so that the signal is routed fromBRAF to CRAF (Fig. 4). Studies on melanoma celllines harbouring NRAS mutations have already shownthat melanomas can use CRAF as a mechanism to ac-tivate downstream MAPK signalling [63]. The pres-ence of a network connection that can easily switch

MAPK signalling from BRAF to CRAF (also calledRaf-1) is suggestive of a mechanism by which resis-tance to BRAF inhibitorsmay be acquired and could al-so potentially explain side effects observed with BRAFinhibitors [63–65].

6. PTEN and the phosphatidylinositol 3-kinase(PI3K)/Akt signaling pathway

PTEN (phosphatase and tensin homolog delet-ed on chromosome 10), also known as MMAC1 (mutated in multiple advanced cancers 1) orTEP1 (TGFβ-regulated and epithelial cell enrichedphosphatase 1) was recently identified as a tu-mor suppressor gene located at 10q23.3 and hasbeen shown to be deleted or mutated in a varietyof advanced malignant tumors (see [3–5,66,67] formore comprehensive reviews). PTEN is a dual-specificity phosphatase capable of dephosphorylatingboth tyrosine phosphate and serine/threonine phos-phate residues on proteins. It also functions as amajor lipid phosphatase, counteracting phosphatidyli-nositol 3-OH kinase (PI3 kinase) by dephosphory-lating the second messengers phosphatidylinositol-3,4,5-triphosphate (PIP3) and phosphatidylinositol-3,4-diphosphate (PIP2), which are required for the ac-tivation of PKB/Akt, an important mediator of cell sur-vival protecting cells from apoptosis (Fig. 4). PTENhas been shown to be involved in cell migration,spreading, and focal adhesion formation through di-rect dephosphorylation and inactivation of focal ad-hesion kinase (FAK). In addition, PTEN inhibitsShc phosphorylation, thus preventing association ofShc with Grb2/Sos and subsequent activation of theRas/raf/MEK1/MAPK pathway. PTEN suppresses thestabilization of hypoxia-mediated HIF-1α (hypoxia-inducible favor 1α), which when stabilized throughthe PI3K/AKTpathway, upregulatesVEGFexpression.This suggests a possible role for PTEN in angiogene-sis. While some of the functions of PTEN have beendelineated, the nature of the factors involved in PTEN’sdecision towards inducing cell-cycle arrest or apopto-sis are still elusive. Germline mutations of PTEN havebeen found in the autosomal dominant Cowden syn-drome,which is characterized bymultiple hamartomas,especially of the skin, and an increased risk of breast(25–50% of affected females) and thyroid (10% of af-fected individuals) cancers as well as, to a lesser extent,malignant melanomas.


Loss of chromosome 10q has been detected in 30to 50% of both early- and advanced-stage sporadicmelanomas, and has been associated with poor clini-cal outcome. There is a host of studies examining thegenomic status of PTEN in melanoma cell lines andnon-cultured melanomas of various stages, with incon-gruous results (reviewed in [3–5,66,67]). Overall it ap-pears that homozygous deletions and intragenic muta-tions of PTEN occur in the metastatic setting, especial-ly in cell lines. Typically, PTEN expression is lost inup to 30% of melanoma cell lines and 10% of humantumor material. There is also evidence of inactivatingPTEN mutations in > 10% of short-term melanomacell cultures [68]. One study examined the mutationalstatus of both PTEN and NRAS in 53 melanoma celllines. The authors found in 30% alterations in PTENand 21% to harbor activating NRAS mutations, withonly 1 cell line having mutations in both, suggestingthat PTEN and NRAS may function in the same path-way. PTEN can also function as a haploinsufficienttumor suppressor, and it is known that allelic loss ofPTEN occurs in at least 58% of melanoma metastases.

With the accumulating knowledge of PTEN inactiva-tion, it appears that there are several mechanismswhichlead to PTEN inactivation: Initial work on tumor celllines almost uniformly suggested that PTEN intragenicmutations, homozygous deletions, and the two struc-tural hits (one structural hit – LOH – followed by thesecond epigenetic hit) would be the rule across a largevariety of tumors. In this regard, studies on melanomashave shown that biallelic structural inactivation occursby either homozygous deletion at 10q23 or somaticintragenic PTEN mutation plus loss of the remainingwild-type allele. Zhou et al. [69] have proposed thatnon-cultured melanomas might be somewhat unique inthat, although PTEN inactivation is seen to occur byone structural hit (LOH) followed by the second epige-netic hit, PTEN protein expression in melanomas canbe biallelically silenced with relatively high frequencyby epigenetic phenomena as well: Examination of 4primary and 30 metastatic melanoma showed little tono expression of PTEN in 65% of the cases. These re-sults would support PTEN as a major tumor suppressoron 10q involved in melanoma tumorigenesis.

6.1. Activation of the PI3K/AKT pathway

Activation of theMAPK(ERK)pathway clearly doesnot account for all aspects of melanoma biology andthere is evidence that other signaling pathways may be

equally important (reviewed in [3–5,66,67]). Todate, the most widely studied of these is the phospho-inositide 3-kinase (PI3K)/protein kinase B (AKT) path-way [70,71]. Activation of the PI3K/AKT pathwayin melanoma occurs through either paracrine/autocrinegrowth factors or the loss of expression and/or muta-tion of negative pathway regulators. In particular, theinsulin-like growth factor-1 (IGF-1) is known to aid thegrowth of early-stage melanoma cells, at least in part,through the activation of PI3K/AKT [72].

The AKT family consists of three members, AKT1–3 [73], which exhibit different expression patternsdepending on the cell type. Of these, 43–50%of melanomas have selective constitutive activity inAKT3 [70]. There is evidence that overexpression ofAKT3 may occur as a result of copy number increasesin the long arm of chromosome 1 [74]. Another mech-anism for PI3K/AKT pathway activation in melanomais through the acquisition of activating E17K mutationsin AKT3, but this is thought to be relatively rare [75].AKT has a critical role in cancer development throughits ability to regulate apoptosis through the direct phos-phorylation of BAD as well as it effects many oth-er pathways, including the inhibition of forkhead sig-nalling and the inhibition of glycogen synthase kinase-3 [76]. As already mentioned, one of the most criticalregulators of AKT is the phosphatase and tensin ho-molog (PTEN), which degrades the products of PI3K,thereby preventing the activation of AKT. In summary,the mechanism by which the PI3K-AKT pathway is ac-tivated in melanoma is not yet fully elucidated, but mayinvolve the loss of expression or functional inactivationof PTEN.

6.2. Entente fatale: For melanoma progression bothPI3K/AKT and MAPK (ERK) signalling arerequired

Data are now emerging from both pre-clinical andclinical studies that the MAPK (ERK) and the PI3K-AKT pathways have overlapping functions with regardto melanoma biology (Fig. 4). Analysis of melanomalines and human melanoma samples have shown thatthose mutational profiles are favoured which simul-taneously promote both MAPK and PI3K/AKT sig-nalling [71]. As activating NRAS mutations can stim-ulate both the MAPK (ERK) and PI3K/AKT pathways,they are rarely found in concert with BRAF mutationsand these mutations are mutually exclusive [19]. Like-wise, BRAF mutations are frequently found in combi-nation with either PTEN loss/inactivation or activating


AKT3 mutations, again showing a requirement for dualMAPK (ERK) and PI3K/AKT pathway activity [68,71,75].

AKT3 seems to cooperate with the BRAF V600Emutation in melanoma initiation by suppressingMAPKsignalling activity through phosphorylation at Ser364/428 of BRAF [77]. It is therefore suggested thatAKT3’s ability to downregulate MAPK (ERK) path-way activity may allow early-stage melanoma cells toescape from oncogeneinduced senescence. To date,the strongest experimental proof of concept for thecooperation between the BRAF V600E mutation andPI3K/AKT signalling in melanoma development is de-rived from the mouse modelling work of Dankort andco-workers [78]. These authors have elegantly shownthat the conditional expression of melanocyte-specificmutated BRAF alone led to the development of be-nign melanocytic hyperplasia but not melanoma. Fulltransformation to melanoma occurred only when theBRAF-V600E mutation was activated in the presenceof PI3K/AKT activity after suppression of PTEN ex-pression.

Hence, the intracellular melanoma signalling net-work is highly interconnectedwith a great deal of func-tional redundancy built into it (Fig. 4). Studies havealready shown that single-agent MEK inhibition is rel-atively ineffective at inducing melanoma regression inboth pre-clinicalmodels [56] and in early-phase clinicaltrials (reviewed in [79]. The disappointing clinical ac-tivity seen after single-agent MEK inhibition parallelswith that observedwith single-agent sorafenib [80],andthus clearly argue in favour of a combination of severaltargeted agents. Smalley et al have identified other sitesof redundancywithin the melanoma signaling network,with cyclin D1 amplifications being found in concertwith BRAF V600E mutations in minor sub-groups ofmelanomas [81].

7. Stem cell factor (SCF)/c-kit signalling pathwayas the driving oncogenic event in distinctsubtypes of melanoma

c-KIT is a transmembrane receptor tyrosine kinase(RTK) to the PDGF/CSF-1 receptor subfamily, whoseaberrant activation is implicated in the progressionof gastrointestinal stromal tumors (GIST) and someacute myeloid leukemias (reviewed in [82]). The c-KIT receptor ligand is the glycoprotein Stem CellFactor (SCF), which is also known under a varietyof other names including steel factor (SF). c-KIT is

known to recruit and activate a number of intracellu-lar signalling pathways implicated in tumor progres-sion, such as phosphoinositide-3 kinase (PI3K)/AKT,Src, mitogen-activated protein kinase (MAPK), januskinase (JAK)/signal transducers and activators of tran-scription (STAT) and phospholipase-C (PLC)-γ [82].

The role of c-KIT signalling in melanoma has beencontroversial; although c-KIT activity is critical tomelanocyte development and migration, its expressiontends to be lost in most melanomas [83]. Progres-sion of melanoma towards the invasive and metastat-ic phenotype is associated with loss of expression ofc-kit (see [11,82,84] for review). c-Kit plays a piv-otal role in normal growth and differentiation of em-bryonic melanoblasts. SCF is highly mitogenic for hu-man melanoblasts and melanocytes in culture and formelanocyes in vivo. In mice, c-kit has been mappedto the dominant white spotting (w) locus; its ligand isthe product of the steel (sl) locus on chromosome 10.Mutations in the w or sl locus, or injection of neutral-izing anti-c-KIT antibodies into pregnant mice, resultsin the piebald phenotype, characterised by white spot-ting of the fur and attributed to a local reduction of thenumber of cutaneous melanocytes. Mutations in thec-kit receptor also have been identified in human pa-tients with piebaldism, suggesting that normal functionof c-kit is required for human melanocyte developmentand migration [11,82].

Several early pathological studies have demonstrat-ed that the progression of melanoma is associated witha loss of expression of c-kit [11,82]. The vast major-ity of metastatic lesions and cell lines do not expressdetectable levels of the c-kit receptor. Transfectionof c-kit into c-kit negative, highly metastatic humanmelanoma cells significantly inhibited tumor growthand metastasis in nude mice [85]. Exposure of c-kit-positive melanoma cells in vitro and in vivo to SCFtriggered apoptosis of these cells but not of normalmelanocytes. Loss of c-kit receptor may thus allowmelanoma cells to escape SCF/c-kit-mediated apopto-sis [85]. On the basis of the aforementioned resultsit was assumed that c-KIT was primarily a regulatorof melanocyte behavior and therefore dispensable formelanoma growth.

However, the recent publication showing the pres-ence of activating c-KIT aberrations (either amplifi-cation and/or mutation) in mucosal (39%) and acral(36%) melanomas, as well as melanomas arising onskin with chronic sun damage (28%), has renewed in-terest in c-KIT signaling in melanoma [86]. In ad-dition, subsequent studies have shown that c-KIT is


Table 1Selected genetic alterations in cutaneous melanoma

Gene Alteration frequency/type(s) References

BRAF 50–70% mutated (e.g., V600) [44]NRAS 15–30% mutated [44,71]AKT3 overexpressed;

43–50% selective constitutive activity[70]

Cyclin D1 6–44% amplified (depending on subtype: chronic sun exposure/face,acral and mucosal melanoma)

[19,118]

CDK4 amplification commonly seen in acral and mucosal melanomas [19]MITF 10–16% amplified [119]Notch overexpressed [110,112]c-met overexpression [97,98]c-kit amplification and/or mutation: 22–39% (mucosal), 36% (acral), 28%

(sun-damaged skin), 15% (anal)[86–88]

c-myc overexpression [100,120,121]CDKN2A (p16INK4A) 30–70% deleted, mutated or silenced [122]PTEN 5–20% deleted or mutated [71,101,123]APAF-1 40% silenced [36]p53 10% lost or mutated [124]

expressed in 88% of oral mucosal melanomas, andthat at least 22% of these harbored activating muta-tions [87]. In these instances, most of the c-KIT ex-pression was in the in situ component, with strong ex-pression reported in the invasive portion in only 22% ofcases. Another recent paper also reported the presenceof the activating L576P mutation in c-KIT in 15% ofanal melanomas, a mutation that the authors showedto be imatinib sensitive in vitro [88]. What’s more,work from our own laboratory has further identified asubset of melanomas that lack BRAF mutations, withconstitutive c-KIT signaling activity resulting from c-KIT/CDK4 co-overexpression [89]. Although the ini-tial clinical trials of the c-KIT inhibitor imatinib me-sylate in melanoma were negative, some dramatic re-sponses have been seen in patients with very high c-KIT expression and/or documented activating muta-tions, fostering the belief that focused studies in pa-tients selected on the basis of c-KIT mutational statuswill yield more encouraging results [82].

8. Hepatocyte growth factor (HGF)/scatter factor(SF)/c-Met tyrosine kinase receptor

c-Met is a receptor tyrosine kinase (RTK) that hasbeen known to stimulate the invasive growth of can-cer cells, increase their metastatic potential, and is al-so known to be expressed and mutated in a variety ofsolid tumors [90]. The structure of c-Met consists ofa disulfide-linked α − β heterodimer with a molecularweight of 190 kDa (see [11] for review). The physio-logical ligand for c-Met is the hepatocyte growth factor(HGF), which is also known as scatter factor. HGF is

a multifunctional cytokine acting as a mitogen, moto-gen, and morphogen for many epithelial cells. HGF isphysiologically secreted by cells of mesenchymal ori-gin and acts on neighboring epithelial cells through aparacrine loop [91]. Hepatocyte growth factor (HGF) isa multifunctional cytokine acting as mitogen, motogenand morphogen for many epithelial cells through itstyrosine kinase receptor c-Met, which is present in ep-ithelial cells and melanocytes. HGF is physiologicallysecreted by cells of mesenchymal origin and acts onneighboring epithelial cells through a paracrine loop.However, coexpression of HGF and c-Met occurs in avariety of transformed cells and tumors both in vitroand in vivo and has been shown to be involved in tumordevelopment and invasion (reviewed in [11]).

HGF/c-Met signaling has been implicated in the de-velopment of melanoma [92]. HGF stimulates prolifer-ation and motility of humanmelanocytes [93]. In trans-genic mice that ubiquitously expressed HGF, ectopiclocalization of melanocytes and hyperpigmentation inskin were observed and melanoma arose spontaneous-ly. In these mice, ultraviolet radiation-induced carcino-genesis was accelerated( reviewed in [94]). It has beensuggested that c-Met autocrine activation induced de-velopment of malignant melanoma and acquisition ofthe metastatic phenotype [95].

Manymissense mutations of c-Met have been report-ed in a variety of cancers, with the majority of themidentified in the cytoplasmic activation loop tyrosinekinase domain. c-Met activating point mutations in thekinase domain are implicated as the cause of hereditarypapillary renal carcinoma. In addition, kinase domainmutations have been observed in sporadic papillary re-


nal carcinoma, ovarian cancer, childhood hepatocellu-lar carcinoma, and gastric cancer [90].

In normal skin, c-Met is present on epithelial cellsand melanocytes, whereas HGF is produced mainly bymesenchymal cells and, consequently, interacts with itsreceptor in a paracrine manner [96]. HGF is a mitogenof human melanocytes, and overexpression of c-Metcorrelates with the invasive growth phase of melanomacells [97]. c-Met is overexpression is also associatedwith patient survival [98]. Studies by our group haveshown that most of melanoma cells, but not normalmelanocytes, produceHGF,which can induce sustainedactivation of its receptor [91]. Hence, an autocrineHGF/c-Met signaling loop may be involved in the de-velopment of melanomas. Consistent with this, pro-longed HGF stimulation induces the down-regulationof the intercellular adhesive molecule E-cadherin thatis implicated in the control of melanocyte prolifera-tion [91]. Finally, in transgenic mice that ubiquitous-ly expressed HGF, ectopic localization of melanocytesand hyperpigmentation in skin were observed, andmelanoma arose spontaneously. In these mice, UVradiation-induced carcinogenesis was accelerated [99].It is suggested that c-Met autocrine activation inducedthe development of malignantmelanomaand the acqui-sition of the metastatic phenotype [95]. More recent-ly, Puri and co-workers showed that a small-moleculetyorsine kinase inhibitor of c-met inhibits growth ofmelanoma by causing apoptosis and induces differen-tiation [90].

9. c-myc

c-myc is a gene coding for a nuclear phosphopro-tein that is involved in cell proliferation and differ-entiation. c-myc acts as a central regulator of cellu-lar proliferation and is sufficient to induce G1 entryand cell proliferation in the absence of external growthfactors (reviewed in [11]). Moreover, c-myc func-tions as an inducer of apoptotic cell death under con-ditions of growth factor deprivation or in the presenceof wild-type p53 protein. In primary melanomas, over-expression correlates with tumor thickness, the pres-ence of ulceration and disease-free survival. Overex-pression of c-myc also predicts poor outcome in bothprimary and metastatic disease [71,100–102]. Trans-fection of c-myc into a melanoma cell line led to in-creased growth rates, anchorage independent growthand directed cell movement in culture. Subcutaneouslyimplanted IGR39D melanomas highly overexpressing

c-myc spontaneously formed macroscopic metastases(lymph nodes and lung) in severe combined immun-odeficient (SCID) mice, whereas less prominent c-mycoverexpression caused the development of only lungmicrometastases [103]. These findings suggest thatconstitutive high c-myc expression in humanmelanomaresults in a more aggressive growth behavior both invitro and in vivo, and favors metastasis.

More recently, it has been suggested that one of themajor functions of c-myc overexpression in melanomaprogression is to continuously suppress BRAFV 600E

or NRASQ61R-dependent oncogene-induced senes-cence [104].

10. Notch signaling and melanoma

The Notch signaling pathway is an evolutionarilyconserved signaling cascade that affects cell fate deci-sions and many differentiation processes during bothembryonic and postnatal development (see [105] forreview). Notch signaling is critical for developing andmaintaining tissue homeostasis. Its pathway compris-es a family of transmembrane receptors and their lig-ands, negative and positive modifiers, and transcrip-tion factors. To date, 4 mammalian receptors (Notch1through Notch4) and at least 5 ligands (Delta 1, 3,and 4 and Jagged 1, 2) have been identified. Bind-ing of the ligand renders the Notch receptor suscepti-ble to metalloprotease- and γ-secretase–mediated pro-teolytic cleavage, which in turn results in the releaseof the Notch intracellular domain (NIC) from the plas-ma membrane and its subsequent translocation into thenucleus.

Notch signaling has also been implicated in a vari-ety of neoplastic malignancies reviewed in [106–108]).Notch can function as a tumor promoter or a suppres-sor depending on the cell type and context. In exper-imental mouse models of skin tumorigenesis, Notch1shows antitumor activity. Ablation of Notch1 in ker-atinocytes leads to spontaneous development of basalcell carcinoma–like tumors, likely as a result of reacti-vation of the β-catenin signaling pathway [109].

Recent data suggest that Notch activation may al-so play a role in melanoma progression. Our grouphas previously shown that activation of Notch signal-ing promotes the progression of early-stage melanomacell lines in a β-catenin–dependent manner both in vit-ro and in vivo [110]. Blocking Notch signaling sup-pressed whereas constitutive activation of the Notch1pathway enhanced primary melanoma cell growth both


in vitro and in vivo yet had little effect on metastat-ic melanoma cells. Activation of Notch1 signalingenabled primary melanoma cells to gain metastat-ic capability. Furthermore, the oncogenic effect ofNotch1 on primary melanoma cells was mediated byβ-catenin, which was upregulated following Notch1activation. Inhibiting β-catenin expression reversedNotch1-enhanced tumor growth and metastasis, there-fore suggesting a β-catenin–dependent, stage-specificrole for Notch1 signaling in promoting the progressionof primary melanoma.

What’s more, microarray profiling comparing thegene expression of normal human melanocytes to hu-man melanomas revealed up-regulation of Notch tar-get genes in melanoma cells, suggesting activationof the Notch signaling pathway in melanoma [111].More recently, an analysis of cell lines and patient le-sions showed that Notch activity is significantly higherin melanomas than their nontransformed counterparts.Notch and Notch target genes are up-regulated in bothmelanoma lesions and melanoma cell lines. Notch re-ceptors 1, 2, and 4 are overexpressed in melanoma celllines and lesions, particularly when compared againstprimary melanocytes or normal human skin [112].

Moreover,Notch1 signaling is activated inmelanomacells, but not melanocytes, and constitutive Notch1activation confers transforming properties to primarymelanocytes in vitro. The use of a constitutively active,truncated Notch transgene construct (NIC) was exploit-ed by Pinnix and co-workers to determine if Notch ac-tivation is a “driving” event in melanocytic transfor-mation or instead a “passenger” event associated withmelanomaprogression. NIC-infectedmelanocytesdis-played increased proliferative capacity and biologi-cal features more reminiscent of melanoma, such asdysregulated cell adhesion and migration. Specifi-cally, ectopic NIC expression induced gross morpho-logic changes, increased growth, adhesion, migration,and survival, and resulted in the loss of E-cadherinexpression and up-regulation of MCAM, two well-characterized events in melanoma development. In ad-dition, MCAM was identified as a direct Notch targetdue to the presence of two high-affinity CSL bindingsites present in the MCAM promoter. The NIC onco-protein conferred anchorage-independent growth, in-creased survival, and loss of contact inhibition; sup-pression of Notch signalling decreased the growth ofmelanoma cell lines, whereas primary melanocyteswere unaffected. Taken together, these data suggestthat deregulation of Notch signaling plays a specificrole in promoting a transformed phenotype in human

melanocytes and define the importance of Notch sig-naling in human melanoma.

Gene expression analyses supported these observa-tions and aided in the identification of MCAM, an ad-hesion molecule associated with acquisition of the ma-lignant phenotype, as a direct target of Notch trans-activation. NIC-positive melanocytes grew at clonaldensity, proliferated in limiting media conditions, andalso exhibited anchorage-independent growth. Hence,Notch alone is a transforming oncogene in humanmelanocytes, a phenomenon not previously describedfor any melanoma oncogene [112].

Our aforementioned microarray data describingNotch pathway activity is underscored by a recentreport by Hoek and colleagues that revealed up-regulation of Notch2 and Hey1 in a separate but dis-tinct panel of malignant melanoma cell lines, sug-gesting a role for Notch activation in the transforma-tion of melanocytes [111]. Previous immunohisto-chemical studies on early-phasemelanoma lesions haveshown overexpression of full-length Notch1 proteinin melanoma tissue compared against benign humannevi [110] and normal human skin [113].

What’s more, our current studies strongly suggestthat constitutive Notch signaling is associated withmelanocyte transformation and melanoma tumorigen-esis. Of particular significance is the ability of a γ-secretase inhibitor to selectively inhibit the growth ofmelanomacell lines. These findings are consistent witha recent report identifying a γ-secretase inhibitor thatinduced effective apoptosis in human melanoma cellswhile sparing melanocytes [114].

These findings were recently confirmed and expand-ed by Bedogni and co-workers who found that Notch1signaling is elevated in human melanoma samples andcell lines and is required for Akt and hypoxia to trans-form melanocytes in vitro [115]. Notch1 facilitatedmelanoma development in a xenograft model by main-taining cell proliferation and by protecting cells fromstress-induced cell death. Hyperactivated PI3K/Aktsignaling led to upregulation of Notch1 through NF-κB activity, while the low oxygen content normallyfound in skin increased mRNA and protein levels ofNotch1 via stabilization of HIF-1α. Taken together,these findings demonstrate that Notch1 is a key effec-tor of both Akt and hypoxia in melanoma developmentand identify the Notch signaling pathway as a potentialtherapeutic target in melanoma treatment.


11. Conclusions and outlook: A changingtreatment paradigm based on molecularclassification

Cutaneous melanoma can be perceived as a diseaseof communication between and within skin cells. Themolecular aberrations are pleiotropic, but growth factorreceptor and mitogen-activated protein kinase (MAPK)pathways feature prominently. Following the discov-ery that the overwhelming proportion of cutaneousmelanomas have constitutive activity in the extracellu-lar signal regulated kinase (ERK) pathway, there hasbeen considerable interest in pharmacologically target-ing this pathway using small molecule inhibitors [3,67]. Although there is evidence to suggest that thepresence of the BRAF V600E mutation is predictive ofresponse to BRAF and MEK inhibitors [116], recentclinical trials on MEK inhibitors have not led to theexpected favorable results [3,4]. BRAF/MAPK sig-naling may be more heterogeneous than first thoughtand locally regulated by the microenvironment [56,104]. It also is possible that other factors, such as en-hanced phosphoinositide-3-kinase/AKT signaling ac-tivity, may further influence response to BRAF/MEKinhibition [117].

As yet, very little is known about the factors un-derlying resistance to BRAF inhibition in the BRAFV600E–mutated melanoma population. A greater un-derstanding of the genetic basis of response to BRAFand other inhibitors is essential in selecting the mostappropriate patient sub-population for future clinicalstudies and developing strategies to overcome inherentresistance.

Although many melanomas harbor either activatingmutations in BRAF or NRAS, there remains a sub-stantial, yet little known, group of tumors without ei-ther mutation. We have recently suggested that co-overexpression of KIT/CDK4 is a potential mechanismof oncogenic transformation in some BRAF/NRASwild-type melanomas. This group of melanomas maybe a subpopulation for which KIT inhibitors may con-stitute optimal therapy [89].

We now also realize that there are differences in thegenetic profiles of melanomas that originate from skinthat is either chronically sun-damaged (as defined bythe appearance of solar elastosis) or skin that lackssun-induced damage [4]. Thus, melanomas that ariseon skin with chronic sun-induced damage have a lowincidence of BRAF mutations and instead showed in-creased cyclin D1 copy number. Frequent amplifica-

tions of cyclin D1 also occur in distinct histologic sub-types of melanoma [118].

As we have seen, multiple studies support the ex-istence of distinct genetic pathways associated withmelanoma development, mirroring different molecu-lar and clinical behaviour. The complexity of the sig-nalling observed inmelanoma,and outlined in this bookchapter, calls for a network-biology approach and acombination of targeted therapies. The challenge formelanoma investigators is to identify the key signalling“hubs” or relay stations in this melanoma network. Al-though we have made great strides in unravelling themysteries ofmelanoma genetics andmolecular biology,the important task now facing clinicians and researchersis to translate these discoveries into novel therapeuticstrategies that will improve patient outcomes.

During the past three decades the translation ofmolecular and immunological studies in melanoma bi-ology into useful clinical correlates or novel effica-cious therapies has been overwhelmingly disappoint-ing. Even though molecular markers and novel thera-peutic compounds look promising in small-scale stud-ies, the vast majority have failed to prove clinically use-ful in large-scale (phase III) studies. Today, in 2010, weclearly stand at the crossroads of melanoma therapy. Inview of the plethora of targeted agents currently underclinical investigation and our emerging knowledge ofmolecular sub-classifications, we may already have thepharmacologic arsenal that could control melanoma iftargeted to molecularly defined patient subgroups orif administered in biologically appropriate combina-tions. Clearly, therefore, pharmacogenomic analysisof melanoma populations may be a suitable strategyfor the further subclassification of melanoma, ultimate-ly leading to more “personalized” therapy approach-es in the future. Lastly, to improve the outcome ofmelanoma and to determine the biological mechanismsof efficacy or failure, future clinical studies with target-ed agents will also require more stringent monitoringof biological endpoints.

References

[1] A.M. DeLuca, A. Srinivas and A. Alani, BRAF kinase inmelanoma development and progression, Expert Rev MolMed 10 (2008), e6.

[2] K.S. Smalley and M. Herlyn, Integrating tumor-initiatingcells into the paradigm for melanoma targeted therapy, Int JCancer 124 (2009), 1245–1250.

[3] K.S. Smalley, Understanding Melanoma Signaling Networksas the Basis for Molecular Targeted Therapy, J Invest Der-matol (2009).


[4] K.S. Smalley, K.L. Nathanson and K.T. Flaherty, Geneticsubgrouping of melanoma reveals new opportunities for tar-geted therapy, Cancer Res 69 (2009), 3241–3244.

[5] A. Sekulic, P.Jr. Haluska, A.J. Miller et al., Malignantmelanoma in the 21st century: the emerging molecular land-scape, Mayo Clin Proc 83 (2008), 825–846.

[6] E. Bajetta, V.M. Del, C. Bernard-Marty et al., Metastaticmelanoma: chemotherapy, Semin Oncol 29 (2002), 427–445.

[7] L. Serrone, M. Zeuli, F.M. Sega et al., Dacarbazine-basedchemotherapy for metastatic melanoma: thirty-year experi-ence overview, J Exp Clin Cancer Res 19 (2000), 21–34.

[8] W. Sun and L.M. Schuchter, Metastatic melanoma, CurrTreat Options Oncol 2 (2001), 193–202.

[9] L.M. Haass and M. Herlyn, Normal human melanocytehomeostasis as a paradigm for understanding melanoma, JInvestig Dermatol Symp Proc 10 (2005), 153–163.

[10] N.K. Haass, K.S. Smalley, L. Li et al., Adhesion, migra-tion and communication in melanocytes and melanoma 56,Pigment Cell Res 18 (2005), 150–159.

[11] T. Bogenrieder, T. Elder and M. Herlyn, Molecular and cel-lular Biology, in: Cutaneous Melanoma, C.M. Balch, A.N.Houghton, A.J. Sober and S.J. Soong, eds, 2003, p. 751.

[12] K.S. Smalley, M. Lioni and M. Herlyn, Targeting the stromalfibroblasts: a novel approach to melanoma therapy, ExpertRev Anticancer Ther 5 (2005), 1069–1078.

[13] M. Thompson, M. Scolyer and M. Kefford, Cutaneousmelanoma, Lancet 365 (2005), 687–701.

[14] M. Li, A. Sanki, A. Karim et al., The role of cell cycle regu-latory proteins in the pathogenesis of melanoma, Pathology(Phila) 38 (2006), 287–301.

[15] M. Malumbres, P. Pevarello, M. Barbacid et al., CDK in-hibitors in cancer therapy: what is next? Trends PharmacolSci 29 (2008), 16–21.

[16] M. Malumbres and M. Barbacid£‹ Cell cycle, CDKs andcancer: a changing paradigm, Nat Rev Cancer 9 (2009),153–166.

[17] G.J. Walker, J.F. Flores, J.F. Glendening et al., Virtually100% of melanoma cell lines harbor alterations at the DNAlevelwithinCDKN2A,CDKN2B, or one of their downstreamtargets, Genes Chromosomes Cancer 22 (1998), 157–163.

[18] J. Bartkova, J. Lukas, J. Guldberg et al., The p16-cyclinD/Cdk4-pRb pathway as a functional unit frequently alteredin melanoma pathogenesis, Cancer Res 56 (1996), 5475–5483.

[19] J.A. Curtin, J. Fridlyand, T. Kageshita et al., Distinct sets ofgenetic alterations in melanoma, N Engl J Med 353 (2005),2135–2147.

[20] T. Reed, F. Loganzo, Jr., F. Shea, Jr. et al., Loss of expres-sion of the p16/cyclin-dependent kinase inhibitor 2 tumorsuppressor gene in melanocytic lesions correlates with in-vasive stage of tumor progression, Cancer Res 55 (1995),2713–2718.

[21] D. Polsky, A.Z. Young, A.Z. Busam et al., The transcrip-tional repressor of p16/Ink4a, Id1, is up-regulated in earlymelanomas, Cancer Res 61 (2001), 6008–6011.

[22] A.Z. Florenes, A.Z. Faye, G.M. Maelandsmo et al., Levelsof cyclin D1 and D3 in malignant melanoma: deregulatedcyclin D3 expression is associated with poor clinical outcomein superficial melanoma, Clin Cancer Res 6 (2000), 3614–3620.

[23] G.M. Florenes, G.M. Maelandsmo, G.M. Kerbel et al., Pro-tein expression of the cell-cycle inhibitor p27Kip1 in ma-lignant melanoma: inverse correlation with disease-free sur-vival, Am J Pathol 153 (1998), 305–312.

[24] J. Georgieva, P. Sinha and P. Schadendorf, Expression ofcyclins and cyclin dependent kinases in human benign andmalignant melanocytic lesions, J Clin Pathol 54 (2001), 229–235.

[25] G.M. Maelandsmo, R. Holm, O. Fodstad et al., Cyclin kinaseinhibitor p21WAF1/CIP1 in malignant melanoma: reducedexpression in metastatic lesions, Am J Pathol 149 (1996),1813–1822.

[26] D. Hanahan and R.A. Weinberg, The hallmarks of cancer,Cell 100 (2000), 57–70.

[27] R.A. Hussein, A.K. Haemel and G.S. Wood, Apoptosis andmelanoma: molecular mechanisms, J Pathol 199 (2003),275–288.

[28] M.S. Soengas and S.W. Lowe, Apoptosis and melanomachemoresistance, Oncogene 22 (2003), 3138–3151.

[29] S.W. Soengas, R.M. Alarcon, R.M. Yoshida et al., Apaf-1 andcaspase-9 in p53-dependent apoptosis and tumor inhibition,Science 284 (1999), 156–159.

[30] K.S. Smalley, K.S. Contractor, N.K. Haass et al., Anorganometallic protein kinase inhibitor pharmacologicallyactivates p53 and induces apoptosis in human melanomacells, Cancer Res 67 (2007), 209–217.

[31] M.S. Soengas and S.W. Lowe, p53 and p73: seeing double?Nat Genet 26 (2000), 391–392.

[32] M.R. Hussein, A.K. Haemel and G.S. Wood, Apoptosis andmelanoma: molecular mechanisms, J Pathol 199 (2003),275–288.

[33] S.W. Lowe and A.W. Lin, Apoptosis in cancer, Carcinogen-esis 21 (2000), 485–495.

[34] K. Satyamoorthy, T. Bogenrieder and M. Herlyn, No longera molecular black box–new clues to apoptosis and drug re-sistance in melanoma, Trends Mol Med 7 (2001), 191–194.

[35] K. Satyamoorthy, N.H. Chehab, M.J. Waterman et al., Aber-rant regulation and function of wild-type p53 in radioresistantmelanoma cells, Cell Growth Differ 11 (2000), 467–474.

[36] M.S. Soengas, P. Capodieci, D. Polsky et al., Inactivation ofthe apoptosis effector Apaf-1 in malignant melanoma, Nature409 (2001), 207–211.

[37] J.A. Bush and G. Li, The role of Bcl-2 family members in theprogression of cutaneous melanoma, Clin Exp Metastasis 20(2003), 531–539.

[38] K. Yamamura, S. Kamada, S. Ito et al., Accelerated disap-pearance of melanocytes in bcl-2-deficient mice, Cancer Res56 (1996), 3546–3550.

[39] C.A. Schmitt, C.T. Rosenthal and S.W. Lowe, Genetic anal-ysis of chemoresistance in primary murine lymphomas, NatMed 6 (2000), 1029–1035.

[40] B. Jansen, H. Schlagbauer-Wadl, B.D. Brown et al., bcl-2antisense therapy chemosensitizes humanmelanoma inSCIDmice, Nat Med 4 (1998), 232–234.

[41] A.S. Dhillon, S. Hagan, O. Rath et al., MAP kinase signallingpathways in cancer, Oncogene 26 (2007), 3279–3290.

[42] P.J. Roberts and C.J. Der, Targeting the Raf-MEK-ERKmitogen-activated protein kinase cascade for the treatment ofcancer, Oncogene 26 (2007), 3291–3310.

[43] M.S. Brose, P. Volpe, M. Feldman et al., BRAF and RASmutations in human lung cancer and melanoma, Cancer Res62 (2002), 6997–7000.

[44] H. Davies, G.R. Bignell, C. Cox et al., Mutations of theBRAF gene in human cancer, Nature 417 (2002), 949–954.

[45] N. Dhomen and R. Marais, BRAF signaling and targetedtherapies in melanoma, Hematol Oncol Clin North Am 23(2009), 529–545.


[46] P.T. Wan, M.J. Garnett, S.M. Roe et al., Mechanism of ac-tivation of the RAF-ERK signaling pathway by oncogenicmutations of B-RAF, Cell 116 (2004), 855–867.

[47] J.L. Maldonado, J. Fridlyand, H. Patel et al., Determinants ofBRAF mutations in primary melanomas, J Natl Cancer Inst95 (2003), 1878–1890.

[48] A. Weber, U.R. Hengge, D. Urbanik et al., Absence ofmutations of the BRAF gene and constitutive activation ofextracellular-regulated kinase in malignant melanomas of theuvea, Lab Invest 83 (2003), 1771–1776.

[49] M.T. Landi, J. Bauer, R.M. Pfeiffer et al., MC1R germlinevariants confer risk for BRAF-mutant melanoma, Science313 (2006), 521–522.

[50] K.V. Bhatt, L.S. Spofford, G. Aram et al., Adhesion controlof cyclin D1 and p27Kip1 levels is deregulated in melanomacells through BRAF-MEK-ERK signaling, Oncogene 24(2005), 3459–3471.

[51] P.M. Pollock, U.L. Harper, K.S. Hansen et al., High frequen-cy of BRAF mutations in nevi 308, Nat Genet 33 (2003),19–20.

[52] P. Uribe, I.I. Wistuba and S. Gonzalez, BRAF mutation: afrequent event in benign, atypical, andmalignantmelanocyticlesions of the skin, Am J Dermatopathol 25 (2003), 365–370.

[53] R. Kumar, S. Angelini, E. Snellman et al., BRAF mutationsare common somatic events in melanocytic nevi, J InvestDermatol 122 (2004), 342–348.

[54] P. Uribe, L. Andrade and S. Gonzalez, Lack of associa-tion between BRAF mutation and MAPK ERK activation inmelanocytic nevi, J Invest Dermatol 126 (2006), 161–166.

[55] C. Michaloglou, L.C. Vredeveld and M.S. Soengas et al.,BRAFE600-associated senescence-like cell cycle arrest ofhuman naevi, Nature 436 (2005), 720–724.

[56] N.K. Haass, K. Sproesser, T.K. Nguyen et al., The mitogen-activated protein/extracellular signal-regulated kinase kinaseinhibitor AZD6244 (ARRY-142886) induces growth arrest inmelanoma cells and tumor regression when combined withdocetaxel, Clin Cancer Res 14 (2008), 230–239.

[57] K.S. Smalley andM.Herlyn, Targeting intracellular signalingpathways as a novel strategy in melanoma therapeutics, AnnN Y Acad Sci 1059 (2005), 16–25.

[58] M. Shinozaki, A. Fujimoto, D.L. Morton et al., Incidence ofBRAF oncogene mutation and clinical relevance for primarycutaneous melanomas, Clin Cancer Res 10 (2004), 1753–1757.

[59] R. Kumar, S. Angelini, K. Czene et al., BRAF mutationsin metastatic melanoma: a possible association with clinicaloutcome, Clin Cancer Res 9 (2003), 3362–3368.

[60] M. Kono, I.S. Dunn, P.J. Durda et al., Role of the mitogen-activated protein kinase signaling pathway in the regulationof human melanocytic antigen expression, Mol Cancer Res4 (2006), 779–792.

[61] H. Sumimoto, F. Imabayashi, T. Iwata et al., The BRAF-MAPK signaling pathway is essential for cancer-immuneevasion in human melanoma cells, J Exp Med 203 (2006),1651–1656.

[62] C. Montagut, S.V. Sharma, T. Shioda et al., Elevated CRAFas a potential mechanism of acquired resistance to BRAFinhibition in melanoma, Cancer Res 68 (2008), 4853–4861.

[63] N. Dumaz, R. Hayward, J. Martin et al., In melanoma,RAS mutations are accompanied by switching signaling fromBRAF to CRAF and disrupted cyclic AMP signaling, CancerRes 66 (2006), 9483–9491.

[64] J.P. Arnault, J. Wechsler, B. Escudier et al., Keratoacan-thomas and squamous cell carcinomas in patients receivingsorafenib, J Clin Oncol 27 (2009), e59–e61.

[65] K. Ehrenreiter, F. Kern, V. Velamoor et al., Raf-1 addictionin Ras-induced skin carcinogenesis, Cancer Cell 16 (2009),149–160.

[66] L.A. Fecher, S.D. Cummings, M.J. Keefe et al., Towarda molecular classification of melanoma, J Clin Oncol 25(2007), 1606–1620.

[67] V. Gray-Schopfer, C. Wellbrock and R. Marais, Melanomabiology and new targeted therapy, Nature 445 (2007), 851–857.

[68] W.M. Lin, A.C. Baker, R. Beroukhim et al., Model-ing genomic diversity and tumor dependency in malignantmelanoma, Cancer Res 68 (2008), 664–673.

[69] X.P. Zhou, O. Gimm, H. Hampel et al., Epigenetic PTENsilencing in malignant melanomas without PTEN mutation,Am J Pathol 157 (2000), 1123–1128.

[70] J.M. Stahl, A. Sharma, M. Cheung et al., Deregulated Akt3activity promotes development of malignantmelanoma, Can-cer Res 64 (2004), 7002–7010.

[71] H. Tsao, V. Goel, H. Wu et al., Genetic interaction betweenNRAS and BRAF mutations and PTEN/MMAC1 inactiva-tion in melanoma, J Invest Dermatol 122 (2004), 337–341.

[72] K. Satyamoorthy, G. Li, B. Vaidya et al., Insulin-like growthfactor-1 induces survival and growth of biologically earlymelanoma cells through both the mitogen-activated proteinkinase and beta-catenin pathways, Cancer Res 61 (2001),7318–7324.

[73] D.P. Brazil, J. Park and B.A. Hemmings, PKB binding pro-teins. Getting in on the Akt, Cell 111 (2002), 293–303.

[74] B.C. Bastian, P.E. LeBoit, H. Hamm et al., Chromosomalgains and losses in primary cutaneous melanomas detected bycomparative genomic hybridization, Cancer Res 58 (1998),2170–2175.

[75] M.A.Davies, K. Stemke-Hale, C. Tellez et al., A novel AKT3mutation in melanoma tumours and cell lines, Br J Cancer99 (2008), 1265–1268.

[76] G.P. Robertson, Functional and therapeutic significance ofAkt deregulation in malignant melanoma, Cancer MetastasisRev 24 (2005), 273–285.

[77] M. Cheung, A. Sharma, S.V. Madhunapantula et al., Akt3 andmutant V600E B-Raf cooperate to promote early melanomadevelopment, Cancer Res 68 (2008), 3429–3439.

[78] D. Dankort, D.P. Curley, R.A. Cartlidge et al., Braf(V600E)cooperates with Pten loss to induce metastatic melanoma,Nat Genet 41 (2009), 544–552.

[79] K.S. Smalley and K.T. Flaherty, Integrating BRAF/MEK in-hibitors into combination therapy for melanoma, Br J Cancer100 (2009), 431–435.

[80] T. Eisen, T. Ahmad, K.T. Flaherty et al., Sorafenib in ad-vanced melanoma: a Phase II randomised discontinuationtrial analysis, Br J Cancer 95 (2006), 581–586.

[81] K.S. Smalley, M. Lioni, P.M. Dalla et al., Increased cy-clin D1 expression can mediate BRAF inhibitor resistancein BRAF V600E-mutated melanomas, Mol Cancer Ther 7(2008), 2876–2883.

[82] K.S. Smalley, V.K. Sondak and J.S. Weber, c-KIT signalingas the driving oncogenic event in sub-groups of melanomas,Histol Histopathol 24 (2009), 643–650.

[83] P.G. Natali, M.R. Nicotra, A.B. Winkler et al., Progressionof human cutaneous melanoma is associated with loss ofexpression of c-kit proto-oncogene receptor, Int J Cancer 52(1992), 197–201.


[84] J.M. Grichnik, Kit and melanocyte migration, J Invest Der-matol 126 (2006), 945–947.

[85] S.Huang, M.Luca, M.Gutman et al., Enforced c-KIT expres-sion renders highlymetastatic humanmelanoma cells suscep-tible to stem cell factor-induced apoptosis and inhibits theirtumorigenic and metastatic potential, Oncogene 13 (1996),2339–2347.

[86] J.A. Curtin, K. Busam, D. Pinkel et al., Somatic activationof KIT in distinct subtypes of melanoma, J Clin Oncol 24(2006), 4340–4346.

[87] R.S. Rivera, H. Nagatsuka, M. Gunduz et al., C-kit proteinexpression correlated with activating mutations in KIT genein oral mucosal melanoma, Virchows Arch 452 (2008), 27–32.

[88] C.R. Antonescu, K.J. Busam, T.D. Francone et al., L576PKIT mutation in anal melanomas correlates with KIT proteinexpression and is sensitive to specific kinase inhibition, Int JCancer 121 (2007), 257–264.

[89] K.S. Smalley, R. Contractor, T.K. Nguyen et al., Identifi-cation of a novel subgroup of melanomas with KIT/cyclin-dependent kinase-4 overexpression, Cancer Res 68 (2008),5743–5752.

[90] N. Puri, S. Ahmed, V. Janamanchi et al., c-Met is a potentiallynew therapeutic target for treatment of human melanoma,Clin Cancer Res 13 (2007), 2246–2253.

[91] G. Li, H. Schaider, K. Satyamoorthy et al., Downregulationof E-cadherin and Desmoglein 1 by autocrine hepatocytegrowth factor during melanoma development, Oncogene 20(2001), 8125–8135.

[92] T. Kunisada, H. Yamazaki, T. Hirobe et al., Keratinocyteexpression of transgenic hepatocyte growth factor affectsmelanocyte development, leading to dermal melanocytosis,Mech Dev 94 (2000), 67–78.

[93] R. Halaban, J.S. Rubin, Y. Funasaka et al., Met and hepato-cyte growth factor/scatter factor signal transduction in nor-mal melanocytes and melanoma cells, Oncogene 7 (1992),2195–2206.

[94] F.P. Noonan, J. Dudek, G. Merlino et al., Animal models ofmelanoma: an HGF/SF transgenic mouse model may facil-itate experimental access to UV initiating events, PigmentCell Res 16 (2003), 16–25.

[95] T. Otsuka, H. Takayama, R. Sharp et al., c-Met autocrine acti-vation induces development of malignant melanoma and ac-quisition of the metastatic phenotype, Cancer Res 58 (1998),5157–5167.

[96] M.Y. Hsu, F. Meier and M. Herlyn, Melanoma developmentand progression: a conspiracy between tumor and host, Dif-ferentiation 70 (2002), 522–536.

[97] P.G.Natali, M.R. Nicotra, M.F.Di Renzo et al., Expression ofthe c-Met/HGF receptor in human melanocytic neoplasms:demonstration of the relationship to malignant melanomatumour progression, Br J Cancer 68 (1993), 746–750.

[98] J. Cruz, J.S. Reis-Filho, P. Silva et al., Expression of c-met ty-rosine kinase receptor is biologically and prognostically rele-vant for primary cutaneous malignant melanomas, Oncology65 (2003), 72–82.

[99] F.P. Noonan, T. Otsuka, S. Bang et al., Accelerated ultravi-olet radiation-induced carcinogenesis in hepatocyte growthfactor/scatter factor transgenic mice, Cancer Res 60 (2000),3738–3743.

[100] A.J. Ramsden, R. Grover, J. Chana et al., A prospectiveanalysis of c-myc oncoprotein levels as a prognostic markerin malignant melanoma, J Plast Reconstr Aesthet Surg 60(2007), 626–630.

[101] H. Wu, V. Goel and F.G. Haluska, PTEN signaling pathwaysin melanoma, Oncogene 22 (2003), 3113–3122.

[102] D.A. Ross, J.H. Laing, R. Sanders et al., Long term follow-upof c-myc, p53 and proliferation measurements in malignantmelanoma, Eur J Surg Oncol 32 (2006), 80–84.

[103] H. Schlagbauer-Wadl, M. Griffioen, E.A. van et al., Influenceof increased c-Myc expression on the growth characteristicsof human melanoma, J Invest Dermatol 112 (1999), 332–336.

[104] D. Zhuang, S. Mannava, V. Grachtchouk et al., C-MYCoverexpression is required for continuous suppression ofoncogene-induced senescence in melanoma cells, Oncogene27 (2008), 6623–6634.

[105] G.D. Hurlbut, M.W. Kankel, R.J. Lake et al., Crossing pathswith Notch in the hyper-network, Curr Opin Cell Biol 19(2007), 166–175.

[106] E.J. Allenspach, I. Maillard, J.C. Aster et al., Notch signalingin cancer, Cancer Biol Ther 1 (2002), 466–476.

[107] C.C. Pinnix and M. Herlyn, The many faces of Notch sig-naling in skin-derived cells, Pigment Cell Res 20 (2007),458–465.

[108] B.J. Nickoloff, M.J. Hendrix, P.M. Pollock et al., Notch andNOXA-related pathways in melanoma cells, J Investig Der-matol Symp Proc 10 (2005), 95–104.

[109] M. Nicolas, A. Wolfer, K. Raj et al., Notch1 functions asa tumor suppressor in mouse skin, Nat Genet 33 (2003),416–421.

[110] K. Balint, M. Xiao, C.C. Pinnix et al., Activation of Notch1signaling is required for beta-catenin-mediated human pri-mary melanoma progression, Clin Invest 115 (2005), 3166–3176.

[111] K. Hoek, D.L. Rimm, K.R. Williams et al., Expressionprofiling reveals novel pathways in the transformation ofmelanocytes to melanomas, Cancer Res 64 (2004), 5270–5282.

[112] C.C. Pinnix, J.T. Lee, Z.J. Liu et al., Active Notch1 confersa transformed phenotype to primary human melanocytes,Cancer Res 69 (2009), 5312–5320.

[113] C.L. Curry, L.L. Reed, T.E. Golde et al., Gamma secretaseinhibitor blocks Notch activation and induces apoptosis inKaposi’s sarcoma tumor cells, Oncogene 24 (2005), 6333–6344.

[114] J.Z. Qin, L. Stennett, P. Bacon et al., p53-independentNOXA induction overcomes apoptotic resistance of malig-nant melanomas, Mol Cancer Ther 3 (2004), 895–902.

[115] B. Bedogni, J.A. Warneke, B.J. Nickoloff et al., Notch1 isan effector of Akt and hypoxia in melanoma development, JClin Invest 118 (2008), 3660–3670.

[116] D.B. Solit, L.A. Garraway, C.A. Pratilas et al., BRAF mu-tation predicts sensitivity to MEK inhibition, Nature 439(2006), 358–362.

[117] K.S. Smalley, N.K. Haass, P.A. Brafford et al., Mul

ios press the molecular pathology of cutaneous melanoma...in melanomas, id1 expression is limited to...

Documents