UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of EIF5A2 and eIF4E translation factors in human cutaneous melanoma Khosravi, Shahram 2016

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

Item Metadata

Download

Media
24-ubc_2016_may_khosravi_shahram.pdf [ 2.38MB ]
Metadata
JSON: 24-1.0229496.json
JSON-LD: 24-1.0229496-ld.json
RDF/XML (Pretty): 24-1.0229496-rdf.xml
RDF/JSON: 24-1.0229496-rdf.json
Turtle: 24-1.0229496-turtle.txt
N-Triples: 24-1.0229496-rdf-ntriples.txt
Original Record: 24-1.0229496-source.json
Full Text
24-1.0229496-fulltext.txt
Citation
24-1.0229496.ris

Full Text

THE ROLE OF EIF5A2 AND eIF4E TRANSLATION FACTORS IN HUMANCUTANEOUS MELANOMAbyShahram KhosraviA THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Experimental Medicine)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)April 2016© Shahram Khosravi, 2016iiAbstractCutaneous melanoma is a life-threatening skin cancer due to its poorly understood invasivenature and high metastatic potential. On the other hand, translational dysregulation has beenshown to have an important role in the development and progression of cancer. In this study weinvestigated the role of two translational factors called eukaryotic translation initiation factor5A2 (EIF5A2) and eukaryotic translation initiation factor 4E (eIF4E) in melanoma. Using tissuemicroarray (TMA) we showed that eIF4E expression and nuclear and cytoplasmic EIF5A2expression increased from dysplastic nevi to primary melanomas, and further increased inmetastatic melanomas. The expression of eIF4E and both EIF5A2 forms were correlated withmelanoma thickness and AJCC stages and were inversely correlated with the 5-year survival ofmelanoma patients. TMA data also revealed that nuclear EIF5A2, cytoplasmic EIF5A2 andeIF4E expression were all directly correlated with MMP-2 expression which is an importantfactor for promoting cancer cell invasion. In addition, in vitro analysis revealed that bothmelanoma cell invasion and MMP-2 activity decreased as a result of EIF5A2 or eIF4Eknockdown indicating that EIF5A2 and eIF4E may be responsible for increasing melanoma cellinvasiveness at least partly through increasing MMP-2 activity. Our experiments also showedthat EIF5A2 is a novel downstream target of p-Akt. Additionally, we indicated that both EIF5A2and eIF4E may play a role in EMT by upregulating mesenchymal markers and downregulatingepithelial markers. eIF4E knockdown also decreased cell proliferation and increased apoptosisand caused a decrease in c-myc and BCL-2 expression, and an increase in cleaved PARP andcleaved caspase-3 expression and chemosensitivity. Nuclear and cytoplasmic EIF5A2 expressionwere found to be significantly correlated, and simultaneous expression of both was significantlyiiiassociated with poor overall and disease-specific 5-year survival of melanoma patients. Thiscorrelation; however, was not perfect which may indicate the differential regulation of nuclearand cytoplasmic EIF5A2 expression in melanoma. Multivariate Cox regression analysis revealedthat eIF4E expression, nuclear and cytoplasmic EIF5A2 expression and the combination of thelast two were an adverse independent prognostic factor for melanoma patients. Therefore, theyhave the potential to be used as prognostic therapeutic markers in melanoma.ivPrefaceContributions1. A version of chapter 3 has been published in British Journal of Cancer. [Khosravi S],Wong RP, Ardekani GS, Zhang G, Martinka M, Ong CJ, Li G. Role of EIF5A2, adownstream target of Akt, in promoting melanoma cell invasion. G.L. provided theresearch facilities and materials and contributed to the design of the experiments. R.P.W,G.S.A and G.Z. provided helpful discussion on the manuscript. M.M. assisted withscoring the TMA slides. C.J.O. contributed to the design of the experiments andmanuscript preparation for revision. I contributed to designing and performing theexperiments and preparing the manuscript.2. A version of chapter 4 has been accepted for publication in Oncology Letters. [KhosraviS], Martinka M, Zhou Y, Ong CJ. Prognostic significance of the expression of nuclearEIF5A2 in human melanoma. C.J.O. contributed to the design of the experiments andmanuscript preparation. M.M. helped with scoring the TMA slides. Y.Z. provided theresearch facilities and materials. I contributed to the design and performing theexperiments and preparing the manuscript.3. A version of chapter 5 has been published in Journal of Investigative Dermatology.[Khosravi S], Tam KJ, Ardekani GS, Martinka M, McElwee KJ, Ong CJ. eIF4E is anadverse prognostic marker of melanoma patient survival by increasing melanoma cellinvasion. C.J.O. contributed to the design of the experiments and manuscript preparation.K.J.T and G.S.A. provided helpful discussion on the manuscript. M.M. assisted withvscoring the TMA slides. K.J.M. provided the research facilities and materials, helpfuldiscussion on the manuscript and assisted with preparing the manuscript.List of publications1. Khosravi S, Martinka M, Zhou Y, Ong CJ. Prognostic significance of the expression ofnuclear EIF5A2 in human melanoma. Oncol Lett. Manuscript accepted for publication.2. Khosravi S, Tam KJ, Ardekani GS, Martinka M, McElwee KJ, Ong CJ. eIF4E is anadverse prognostic marker of melanoma patient survival by increasing melanoma cellinvasion. J Invest Dermatol. 2015 May; 135(5):1358-673. Khosravi S, Wong RP, Ardekani GS, Zhang G, Martinka M, Ong CJ, Li G. Role ofEIF5A2, a downstream target of Akt, in promoting melanoma cell invasion. Br J Cancer.2014 Jan; 110(2):399-408.Ethics certificate:The use of human skin tissues in this study was approved by the Clinical Research Ethics Boardof University of British Columbia (certificate number is H09-01321).viTable of ContentsAbstract.......................................................................................................................................... iiPreface........................................................................................................................................... ivTable of Contents ......................................................................................................................... viList of Tables ................................................................................................................................ xiList of Figures.............................................................................................................................. xiiList of Abbreviations ...................................................................................................................xvAcknowledgements .................................................................................................................... xxiDedication .................................................................................................................................. xxiiChapter 1: Introduction ................................................................................................................11.1 Cutaneous melanoma ...................................................................................................... 11.1.1 Biology of melanocytes and melanoma...................................................................... 11.1.2 Melanoma etiology ..................................................................................................... 51.1.3 Staging and subtypes of cutaneous melanoma............................................................ 81.1.4 Melanoma epidemiology .......................................................................................... 101.2 Translation .................................................................................................................... 111.2.1 An introduction to translation ................................................................................... 111.2.2 Translation initiation................................................................................................. 121.2.2.1 eIF4F cap-binding complex binds the mRNA.................................................. 151.2.2.2 43S PIC formation ............................................................................................ 171.2.2.3 Recruitment of mRNA to 43S PIC ................................................................... 181.2.2.4 5′ to 3′ Scanning................................................................................................ 19vii1.2.2.5 Localization of the initiation codon .................................................................. 201.2.2.6 60S ribosome joining and establishment of 80S ribosome............................... 211.2.3 Translation elongation .............................................................................................. 221.2.4 Translation termination............................................................................................. 251.3 eIF4E............................................................................................................................. 271.3.1 eIF4E binds the cap................................................................................................... 271.3.2 eIF4E a tumourigenic translation initiation factor .................................................... 271.3.3 Regulation of eIF4E availability by 4EBPs .............................................................. 281.3.4 Regulation of eIF4E activity by phosphorylation ..................................................... 291.3.5 Regulation of eIF4E at the transcriptional level ....................................................... 301.3.6 Role of eIF4E in evasion of growth suppression ...................................................... 301.3.7 Role of eIF4E in sustaining proliferative signaling .................................................. 311.3.8 Role of eIF4E in promoting invasion and metastasis ............................................... 311.3.9 Importance of eIF4E and 4EBPs in survival and cancer recurrence ........................ 321.3.10 Therapies targeting eIF4E..................................................................................... 331.4 EIF5A............................................................................................................................ 341.4.1 An introduction to EIF5A structure and function ..................................................... 341.4.2 EIF5A isoforms......................................................................................................... 351.4.3 EIF5A and hypusination are a potential therapeutic target for cancer treatment ..... 361.5 Objective and hypotheses ............................................................................................. 37Chapter 2: Materials and methods.............................................................................................392.1 TMA construction......................................................................................................... 392.2 Immunohistochemistry ................................................................................................. 39viii2.3 Evaluation of TMA immunostaining ............................................................................ 402.4 Statistical analysis of TMA........................................................................................... 402.5 Cell culture, antibodies and drugs................................................................................. 412.6 Plasmids and siRNA transfections................................................................................ 422.7 Protein extraction and Western blot.............................................................................. 422.8 Reverse-transcription and real-time quantitative polymerase chain reaction (qPCR).. 432.9 Cell invasion assay........................................................................................................ 442.10 Zymography.................................................................................................................. 442.11 Cell proliferation assay ................................................................................................. 452.12 Fluorescence-activated cell sorting (FACS) analysis ................................................... 45Chapter 3: Role of EIF5A2, a downstream target of Akt, in promoting melanoma cellinvasion .........................................................................................................................................463.1 Background and rationale ............................................................................................. 463.2 Results........................................................................................................................... 473.2.1 EIF5A2 mRNA and protein expression in melanoma cell lines............................... 473.2.2 Correlation between cytoplasmic EIF5A2 expression and clinicopathologicparameters ............................................................................................................................. 493.2.3 Increased cytoplasmic EIF5A2 expression correlates with melanoma progression . 533.2.4 Increased cytoplasmic EIF5A2 expression correlates with poor patient survival .... 553.2.5 EIF5A2 regulates cell invasion and MMP-2 activity ............................................... 593.2.6 EIF5A2 is a downstream target of PI3K/Akt in melanoma cell invasion................. 653.2.7 EIF5A2 may induce epithelial-mesenchymal transition (EMT)............................... 693.3 Discussion ..................................................................................................................... 70ixChapter 4: Prognostic significance of the expression of nuclear EIF5A2 in humanmelanoma......................................................................................................................................754.1 Background and rationale ............................................................................................. 754.2 Results........................................................................................................................... 774.2.1 Correlation between nuclear EIF5A2 expression and clinicopathologiccharacteristics........................................................................................................................ 774.2.2 Nuclear EIF5A2 expression is increased with melanoma progression..................... 784.2.3 Nuclear EIF5A2 expression positively correlates with poor patient survival .......... 814.2.4 Nuclear EIF5A2 is an independent prognostic marker for melanoma patients ........ 844.2.5 Concurrent cytoplasmic and nuclear EIF5A2 expression is correlated with a worse5-Year survival for all and primary melanoma patients ....................................................... 864.2.6 Simultaneous expression of nuclear EIF5A2 and MMP-2 is associated with poor 5-year survival for all and primary melanoma patients............................................................ 894.3 Discussion ..................................................................................................................... 91Chapter 5: eIF4E is an adverse prognostic marker of melanoma patient survival byincreasing melanoma cell invasion .............................................................................................945.1 Background and rationale ............................................................................................. 945.2 Results........................................................................................................................... 955.2.1 eIF4E expression correlates with melanoma tumour thickness and AJCC stages.... 955.2.2 eIF4E expression is increased with melanoma progression ..................................... 985.2.3 Increased eIF4E expression is associated with poor patient survival ..................... 1015.2.4 eIF4E is an independent prognostic marker for melanoma patients....................... 1045.2.5 eIF4E regulates cell invasion via MMP-2 expression and activity......................... 108x5.2.6 eIF4E knockdown increases apoptosis and decreases cell proliferation in melanomacell lines .............................................................................................................................. 1135.2.7 eIF4E may have a role in inducing epithelial-mesenchymal transition.................. 1175.2.8 eIF4E knockdown increases chemosensitivity of melanoma cell lines .................. 1195.3 Discussion ................................................................................................................... 122Chapter 6: Conclusion...............................................................................................................1266.1 Summary of findings and implications ....................................................................... 1266.2 Limitations of this study and future directions ........................................................... 129Bibliography ...............................................................................................................................132xiList of TablesTable 3.1 Cytoplasmic EIF5A2 staining and clinicopathologic characteristics of 382 melanomas....................................................................................................................................................... 52Table 3.2 Multivariate Cox regression analysis on 5-year overall and disease-specific survival ofmelanoma patients ........................................................................................................................ 58Table 4.1 Nuclear EIF5A2 staining and clinicopathologic characteristics of 382 melanomas .... 79Table 4.2 Univariate Cox regression analysis on 5-year overall and disease-specific survival of atotal of 382 melanoma patients and 242 primary melanoma patients .......................................... 83Table 4.3 Multivariate Cox regression analysis on 5-year overall and disease-specific survival ofmelanoma patients ........................................................................................................................ 85Table 4.4 Multivariate Cox regression analysis on 5-year overall and disease-specific survival ofmelanoma patients ........................................................................................................................ 88Table 5.1 eIF4E staining and clinicopathologic characteristics of 381 melanomas ..................... 97Table 5.2 Univariate Cox regression analysis on 5-year survival of a total of 381 melanomapatients and 238 primary melanoma patients.............................................................................. 105Table 5.3 Univariate Cox regression analysis on 5-year survival of 182 primary melanomapatients with tumours ≥1mm thick ............................................................................................. 106xiiList of FiguresFigure 1.1 Biological events involved in progression of melanoma. ............................................. 4Figure 1.2 Schematic diagram summarizing the steps of the translation initiation in eukaryotes.14Figure 1.3 Schematic diagram summarizing the steps of the translation elongation in eukaryotes........................................................................................................................................................ 24Figure 1.4 Schematic diagram summarizing translation termination and recycling in eukaryotes........................................................................................................................................................ 26Figure 3.1 Increase of EIF5A2 mRNA and protein expression in melanoma cell lines comparedto normal human epithelial melanocytes and HEK293 cells. ....................................................... 48Figure 3.2 Diagram showing patient inclusion and exclusion. ..................................................... 50Figure 3.3 EIF5A2 staining by anti-EIF5A2 antibody is substantially blocked by the blockingpeptide, compared with staining with anti-EIF5A2 antibody alone. ............................................ 51Figure 3.4 EIF5A2 expression in common acquired nevi (CAN), dysplastic nevi (DN), primarymelanomas (PM), and metastatic melanomas (MM). ................................................................... 54Figure 3.5 Correlation between cytoplasmic EIF5A2 expression and 5-year patient survival. ... 56Figure 3.6 Correlation between cytoplasmic EIF5A2 expression and 5-year patient survival inlow risk melanoma patients. ......................................................................................................... 57Figure 3.7 EIF5A2 knockdown inhibits melanoma cell invasion by reducing MMP-2 activity. . 61Figure 3.8 EIF5A2 regulates melanoma cell invasion.................................................................. 62Figure 3.9 EIF5A2, a downstream target of p-Akt, regulates melanoma cell invasion................ 63Figure 3.10 EIF5A2 is a downstream target of p-Akt, regulating melanoma cell invasion. ........ 64xiiiFigure 3.11 ILK knockdown or PTEN overexpression decrease EIF5A2 expression and invasionin MMRU melanoma cells............................................................................................................ 67Figure 3.12 PTEN overexpression decreases EIF5A2 expression and invasion in A375 melanomacells. .............................................................................................................................................. 68Figure 3.13 EIF5A2 may promote melanoma cell invasion and metastasis by inducing EMT. .. 69Figure 4.1 Correlation between nuclear EIF5A2 expression, thickness, AJCC stages and differentstages of melanoma progression. .................................................................................................. 80Figure 4.2 Kaplan-Meier analyses for the correlation between nuclear EIF5A2 expression and 5-year survival of melanoma patients. ............................................................................................. 82Figure 4.3 Simultaneous nuclear and cytoplasmic EIF5A2 expression correlates with a poorer 5-year survival.................................................................................................................................. 87Figure 4.4 Simultaneous nuclear EIF5A2 expression and strong MMP-2 expression correlatewith a poorer 5-year survival. ....................................................................................................... 90Figure 5.1 Diagram showing patient inclusion and exclusion. ..................................................... 96Figure 5.2 eIF4E protein expression in different melanoma stages, and in melanoma cell linescompared to normal human melanocytes. .................................................................................... 99Figure 5.3 Changes in eIF4E protein expression during melanoma progression. ...................... 100Figure 5.4 Kaplan-Meier curves representing the correlation between eIF4E expression and 5-year survival of melanoma patients. ........................................................................................... 102Figure 5.5 Kaplan-Meier analyses comparing the survival of patients with tumours ≥1mm thickcompared to the ones with tumours <1mm thick........................................................................ 103Figure 5.6 eIF4E knockdown inhibits melanoma cell invasion by reducing MMP-2 expressionand activity.................................................................................................................................. 110xivFigure 5.7 eIF4E inhibition reduces melanoma cell invasion by decreasing MMP-2 expressionand activity.................................................................................................................................. 111Figure 5.8 Simultaneous high eIF4E expression and strong MMP-2 expression correlate with apoorer 5-year survival. ................................................................................................................ 112Figure 5.9 eIF4E knockdown promotes apoptosis and inhibits cell proliferation in melanomacells. ............................................................................................................................................ 115Figure 5.10 eIF4E inhibition promotes apoptosis and reduces cell proliferation in melanomacells. ............................................................................................................................................ 116Figure 5.11 eIF4E may promote melanoma cell invasion and metastasis by inducing EMT. ... 118Figure 5.12 Doxorubicin-induced apoptosis is enhanced in melanoma cells after eIF4Eknockdown.................................................................................................................................. 120Figure 5.13 Cisplatin-induced apoptosis is enhanced in melanoma cells after eIF4E knockdown...................................................................................................................................................... 121xvList of AbbreviationsAbbreviation Definition4E-BP Eukaryotic translation initiation factor 4E-binding protein5′ UTR 5′ untranslated regionACTH Adrenocorticotropic hormoneAJCC American Joint Committee on CancerAkt Protein kinase BALM Acral lentiginous melanomaAMD1 Adenosylmethionine decarboxylase 1Ap-1 Activator protein-1ARF Alternate reading frameA-site Aminoacyl siteASOs Antisense oligonucleotidesATP Adenosine triphosphateBCL-2 B-cell lymphoma 2BCL-XL B-cell lymphoma-extra largeBIRC2 Baculoviral inhibitor of apoptosis repeat-containing protein 2BRAF Ras associated factor BBSA Bovine serum albuminCAN Common acquired neviCDK Cyclin-dependent kinasexviCDKN2A Cyclin-dependent kinase inhibitor 2ACHO Chinese hamster ovaryc-myc Cellular myelocytomatosis viral oncogeneCRC Colorectal cancerCRM1 Chromosomal maintenance 1DEAD Aspartic acid- Glutamic acid -Alanine-Aspartic acidDHS Deoxyhypusine synthaseDMEM Dulbecco’s modified eagle mediumDMSO Dimethyl sulfoxideDN Dysplastic neviDNA Deoxyribonucleic acidDOHH Deoxyhypusine hydroxylaseE-box Enhancer boxeEF1A Eukaryotic elongation factor 1AeEF1B Eukaryotic translation elongation factor 1 betaeEF2 Eukaryotic translation elongation factor 2eIF1 Eukaryotic translation initiation factor 1eIF1A Eukaryotic translation initiation factor 1AeIF2 Eukaryotic translation initiation factor 2eIF2-TC eIF2 ternary complexeIF3 Eukaryotic translation initiation factor 3eIF4A Eukaryotic translation initiation factor 4AeIF4B Eukaryotic translation initiation factor 4BxviieIF4E Eukaryotic translation initiation factor 4EeIF4F Eukaryotic translation initiation factor 4FeIF4G Eukaryotic translation initiation factor 4GeIF5 Eukaryotic translation initiation factor 5EIF5A Eukaryotic translation initiation factor 5AeIF5B Eukaryotic translation initiation factor 5BeIF6 Eukaryotic translation initiation factor 6EMT Epithelial-mesenchymal transitioneRF1 Eukaryotic release factor 1eRF3 Eukaryotic release factor 3ERK Extracellular regulated kinaseE-site Exit siteFACS Fluorescence-activated cell sortingFBS Fetal bovine serumFOXO Forkhead box OGAP GTPase-activating proteinGAPDH Glyceraldehydes 3-phosphate dehydrogenaseGDP Guanosine diphosphateGTP Guanosine triphosphateHCC Hepatocellular carcinomaHDGF Hepatoma-derived growth factorHER2 Human epidermal growth factor receptor-2HIV Human immunodeficiency virusxviiiHNSCC Head and neck squamous cell carcinomaHR Hazard ratioIGF Insulin-like growth factorILK Integrin-linked kinaseIRESs Internal ribosome entry sitesIRS Immunoreactive scoreJNK c-Jun N-terminal kinaseKD KnockdownLDH Lactate dehydrogenaseLMM Lentigo maligna melanomaMAPK Mitogen activated protein kinaseMC MelanocyteMC1R Melanocortin-1 receptorMCL-1 Myeloid cell leukemia 1MDM2 Mouse double minute 2 homologMEK MAPK/ERK kinaseMITF Microphthalmia transcription factorMM Metastatic melanomasMMP Matrix metalloproteinaseMNK MAPK-interacting kinasemRNA Messenger RNAMT1-MMP membrane type1-MMPmTORC1 Mechanistic target of rapamycin complex 1xixNF-κB Nuclear factor-κBNM Nodular melanomaNos2 Nitric oxide synthase 2NRAS Neuroblastoma RAS viral oncogene homologNSCLC Non-small-cell lung carcinomap38 Protein 38p53 Protein 53PABP Poly(A)-binding proteinPARP Poly (ADP-ribose) polymerasePBS Phosphate buffered salinePDCD4 Programmed cell death 4PGF Placental growth factorPI3K Phosphoinositide-3-kinasePIC Preinitiation complexPIP2 Phosphatidylinositol-4,5-bisphosphatePIP3 Phosphatidylinositol-3,4,5-trisphosphatePM Primary melanomapRB Retinoblastoma proteinP-site Peptidyl sitePTEN Phosphatase and tensin homologuePVDF Polyvinylidene difluorideqPCR Quantitative PCRRB1 Retinoblastoma 1xxRNA Ribonucleic acidrRNA Ribosomal RNART-PCR Reverse transcription polymerase chain reactionshRNA Short hairpin RNAsiRNA Small interference RNASkp2 S-Phase kinase-associated protein 2snRNA Small nuclear ribonucleic acidSPSS Statistical package for the social sciencesSRB Sulforhodamine B assaySSM Superficial spreading melanomaTCA Trichloroacetic acidTGF-β Transforming growth factor βTMA Tissue microarraytRNA Transfer RNAUV UltravioletVEGF Vascular endothelial growth factorXpo4 Exportin 4Xpo4 Exportin 4YB-1 Y box-binding protein 1α-MSH α-melanocytic stimulating hormoneα-SMA α-smooth muscle actinxxiAcknowledgementsI would like to thank my supervisor Dr. Christopher J. Ong who has always been a great supportfor me in all conditions. Dr. Ong inspired me to develop critical thinking skills in science andtaught me how to pay attention to details while keeping the big picture in mind. He is a veryknowledgeable scientist in the field of cancer biology and I have been very fortunate for havingthe opportunity to learn from him during past few years. I would also like to thank mysupervisory committee members, Dr. Vincent Duronio, Dr. Kevin McElwee and Dr. AzizGhahary for their invaluable support and constructive inputs in my projects.I would like to thank Dr. Magdalena Martinka for assisting in evaluation of tissuemicroarray staining and precious advice on melanoma pathology. I also thank all the colleaguesin Cancer Biology Lab, Jack Bell Research Center and Research Pavilion for their friendlysupport and collaborations.I would like to express my appreciation for Dr. Harvey Lui, Karen Ng and other stafffrom the Department of Dermatology and Skin Sciences, and Dr. Vincent Duronio, VirginiaGrosman and Cornelia Reichelsdorfer from the Experimental Medicine Program for theirexcellent support and assistance.I am honored to be the recipients of Natural Sciences and Engineering Research Councilof Canada Industrial Postgraduate Scholarship and five Faculty of Medicine Graduate Awards.This work was supported by research grants from Canadian Institutes of Health Research,Canadian Cancer Society Research Institute and Canadian Dermatology Foundation.xxiiDedicationI dedicate this work to my beloved mother and father.1Chapter 1: Introduction1.1 Cutaneous melanoma1.1.1 Biology of melanocytes and melanomaMelanocytes are pigment producing cells of the skin, responsible for providing protection againstultraviolet radiation, and originate from pluripotent neural crest progenitors or melanoblasts thatgradually become lineage specific during development (Dorsky et al, 1998; Erickson & Reedy,1998; Ernfors, 2010). Melanoblasts proliferate and differentiate on their way to the basal layer ofepidermis and hair follicles where they develop into mature cutaneous melanocytes (Lin &Fisher, 2007; Slominski et al, 2004). The protection properties of melanocytes are attributed totheir ability in producing melanin pigments. The two types of melanin pigments are the brownand black eumelanin and the reddish and yellow pheomelanin which are responsible for differentskin and hair colors (Prota, 1980). The genotype of melanocortin-1 receptor (MC1R) genedetermines the type of melanin being produced (Rees, 2003). The MC1R gene encodes a G-protein coupled receptor that is inhibited by agouti signaling protein (Suzuki et al, 1997) andactivated by adrenocorticotropic hormone (ACTH) or α-melanocytic stimulating hormone (α-MSH).(Barsh et al, 2000; D'Orazio et al, 2006). α-MSH is a cleavage product of ACTH whichitself is a cleavage product of proopiomelanocortin (POMC) (Pritchard et al, 2002; Slominski etal, 2000).Activation of the MC1R by α-MSH and ACTH triggers tyrosinase activity byupregulating the transcription of tyrosinase gene as well as by posttranscriptional mechanisms.Upregulation of tyrosinase gene expression downstream of MC1R depends on activation of thecAMP signaling pathway and the microphthalmia transcription factor (MITF). MC1R signalingalso stimultes a cAMP-dependent increase of the melanosomal pH from acidic to near-neutral2values, improving the catalytic efficiency of tyrosinase (Cheli et al, 2010). The development ofboth melanins happens through the tyrosinase-dependent pathway. The lack of tyrosine leads tooculocutaneous albinism that happens because of the inability to produce melanin whilemelanocytes are intact (Oetting et al, 2003). α-MSH is released by keratinocytes under theinfluence of UV radiation. After production, melanin is transported from melanocytes tokeratinocytes in organelles called melanosome (Miyamura et al, 2007). In the human epidermisevery melanocyte is on average in contact with 36 keratinocytes. As a result, melanocytesconstitute about 2-4% of the total population of the human epidermal cells (Vancoillie et al,1999).Melanocyte growth is normally controlled by their neighboring keratinocytes viakeratinocyte-derived paracrine growth factors and cell-cell adhesion molecules (Haass et al,2005). Disturbance of the balance between melanoctyes and keratinocytes may cause anuncontrolled increase in melanocyte proliferation and melanoma development (Bissell &Radisky, 2001). Some of the responsible mechanisms are hypothesized to be down-regulation ofproteins such as E-cadherin that are crucial for interaction between melanocytes andkeratinocytes, up-regulation of other proteins such as N-cadherin that are important formelanoma-melanoma and melanoma-fibroblast interactions, and loss of anchorage to thebasement membrane because of an altered expression of the extracellular-matrix binding integrinfamily (Haass et al, 2005).Figure 1.1 shows the biological events involved in the progression of melanoma fromnormal melanocytes to malignant melanoma (Miller & Mihm, 2006). The first step in theprogression of melanoma is the formation of common nevi which mostly arise during the firsttwo decades of life (Bastian, 2014; Miller & Mihm, 2006). Melanocytes may proliferate and3spread in a limited amount leading to formation of nevi (Clark et al, 1984). Common nevi areflat or slightly raised lesions with well defined borders and have a uniform color or a regularpattern of dot like pigment in a dark brown or tan background (Hussein, 2005a; Miller & Mihm,2006). Although the growth control of nevi is disrupted, they have a limited growth potential andrarely develop into cancer (Clark et al, 1984). Histologically, benign nevi show an increasednumber of nested melanocytes along the basal layer (Miller & Mihm, 2006).The next step in melanoma progression is the formation of dysplastic nevi either in a newlocation or from benign nevi (Clemente et al, 1991). Dysplastic nevi are usually lesions that maybe asymmetric, have multiple colours and irregular borders or have increasing diameters. Theselesions have random and discontiguous cytologic atypia. (Halpern et al, 1991; Hussein, 2005a).The radial-growth phase of melanoma progression starts by cells gaining the capability ofproliferating intraepidermally. They occasionally look like raised lesions and showcytomorphologic cancer throughout the lesion instead of presenting random atypia. Besidesintraepidermal proliferation, single cells or small nests of cells acquire the ability to penetrate thepapillary dermis (Miller & Mihm, 2006). Cells enter the vertical-growth phase once they startinvading the dermis and widening the papillary dermis by creating an expansile nodule. Thesecells might invade into reticular dermis and fat as well (Piris & Mihm, 2009). Metastasisnormally happens when the tumour reaches the lymphatic or vascular system (Braeuer et al,2011).4Figure 1.1 Biological events involved in progression of melanoma.Modified from Miller & Mihm, 2006, with permission to reprint, Copyright Massachusetts Medical Society.51.1.2 Melanoma etiologyStudies discovering the genetic basis of familial melanomas have led to identification of differentmelanoma-predisposing genes. For example, the loss or inactivating mutation of cyclin-dependent kinase inhibitor 2A (CDKN2A), a well characterized high-penetrance gene, occurs inabout one third of familial melanoma cases (Borg et al, 2000; Thompson et al, 2005). CDKN2Aencodes two different tumor-suppressor genes named p16 (INK4a) and p14 (ARF). CDKN2Agene has 4 exons. The first exons (E1B and E1A) are spliced into alternate reading frames (ARF)of the second (E2) and third (E3) exons, allowing two different proteins to be expressed from thesame genetic locus. mRNA transcription may start at either E1B or E1A, and the site of initiationspecifies which gene will be expressed from the locus. RNA that is transcribed from either exonis spliced with the rest of the exons (E2 and E3) to make mRNA for either ARF or INK4A.However, the reading frame of the exon 2 and 3 codons that ARF uses is different from the onethat INK4A uses. (Miller & Mihm, 2006; Quelle et al, 1995).INK4A, blocks the cell cycle at the G1–S checkpoint by inhibiting the interaction ofcyclin-dependent kinase 4/6 (CDK4/6) with cyclin D and preventing phosphorylation ofretinoblastoma protein (pRB) (Sharpless & Chin, 2003). ARF is another tumour suppressor thatbinds to mouse double minute 2 homolog (MDM2) and sequesters it from p53. Therefore,MDM2 won’t be able to trigger the ubiquination of p53 and its subsequent destruction in theproteasome which leads to accumulation of p53. This allows p53 to arrest the cell cycle at G2-Mphase that permits repair of damaged DNA or induction of apoptosis (Harris & Levine, 2005;Pomerantz et al, 1998). RB1 is also a high-penetrance gene for melanoma. Individuals with amutation in this cell cycle regulator gene have 4-80 times higher chance to develop melanomacompared to people with no mutations (Braam et al, 2012). Mutations in CDK4 have also been6reported in some familial melanoma cases. As a result, INK4A loses the ability to bind to andinhibit CDK4, resulting in uncontrolled cell proliferation (Coleman et al, 1997).The most common type of mutations in melanoma occur in the extracellular regulatedkinase/mitogen activated protein kinase (ERK/MAPK) pathway which is constitutively activatedin about 90% of melanoma cases (Cohen et al, 2002). In this pathway, receptor tyrosine kinasesare activated by their ligands leading to guanosine triphosphate (GTP) loading of the RasGTPase and recruitment of Ras associated factor (Raf) kinases to the plasma membrane foractivation. Raf kinases then phosphorylate and activate MAPK/ERK kinases (MEK1 and MEK2)that subsequently activate ERK1 and ERK2 (Chang & Karin, 2001). ERK activation ultimatelyleads to phosphorylation of factors such as myelocytomatosis viral oncogene (myc), MITF andactivator protein-1 (Ap-1) (Yang et al, 2003). Neuroblastoma Ras viral oncogene homolog(NRAS) mutation rates are about 56% in congenital nevi (Papp et al, 1999), 33% in primarymelanomas, and 26% in metastatic melanomas (Demunter et al, 2001). Ras associated factor B(BRAF) mutations occur in about 50% of melanomas (Miller & Mihm, 2006) with a singlemutation resulting in substitution of glutamic acid for valine at codon 600 (BRAF V600E)accounting for 80% of these mutations (Davies et al, 2002). However, in melanocytic nevi BRAFmutations occur in more than 80% of the cases and as mentioned before NRAS mutations are alsomore prevalent in nevi (Pollock et al, 2003; Yazdi et al, 2003). Therefore, activation ofERK/MAPK pathway may be important for earlier stages of melanoma progression andadditional mutations may be necessary for progression to later stages. Accordingly, BRAFv600Emutation together with ectopic expression of MITF were shown to transform primary humanmelanocytes (Garraway et al, 2005). Similarly, activation of the Ras signaling pathway in7melanoma usually occurs in combination with other genetic changes such as phosphatase andtensin homologue (PTEN) loss (Dankort et al, 2009).The phosphoinositide-3-kinase (PI3K)-Akt signaling pathway is another pathway that hasa crucial role in development of melanoma (Dai et al, 2005; Madhunapantula & Robertson,2009). PI3K phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) to generatephosphatidylinositol-3,4,5-trisphosphate (PIP3) which acts as a second messenger and recruitsseveral downstream mediators such as Akt (Cantley, 2002). Akt is activated by phosphorylationat Thr308 and Ser473 sites leading to activation of a number of its downstream effectors such asmechanistic target of rapamycin (mTOR) and MAPK resulting in growth, proliferation, survivalsignaling and invasion (Chudnovsky et al, 2005; Hsu et al, 2002; Smalley & Herlyn, 2005).The role of PI3K in cancer and cell growth is now well known (Kang et al, 2005; Osakiet al, 2004) and supported in part by studies of its negative regulator, PTEN, that acts as atumour suppressor (Myers et al, 1998; Stambolic et al, 1998). Loss of PTEN expression inmelanoma because of chromosomal deletion and also PTEN mutation in melanoma as well asother cancers (Guldberg et al, 1997; Teng et al, 1997) has been observed before. Moreover,PTEN overexpression in melanoma cells has been shown to suppress tumour development inmice by decreasing Akt phosphorylation (Stahl et al, 2003). Overall, mutations in PI3K, loss ofPTEN function and Akt overexpression have been detected in about 3% (Omholt et al, 2006), 5-20% (Wu et al, 2003) and 60% (Stahl et al, 2004) of melanomas, respectively.The primary environmental factor for the development of melanoma is exposure toUltraviolet (UV) radiation (Oliveria et al, 2006). UV radiation can be divided into threespectrums: UVA with the wavelength of 320-400 nm and UVC with the wavelength of 200-290nm have the lowest and highest energy, respectively and UVB with the wavelength of 280-3208nm has intermediate energy (Hussein, 2005b; Platz et al, 2008; von Thaler et al, 2010). UVC iscompletely and UVB is partly absorbed by the ozone layer of the atmosphere. UVB and UVAcompose 5% and 95% of the UV rays reaching the earth’s surface, respectively. UVB; however,is the most biologically active wavelength as it is directly absorbed by DNA (Budden &Bowden, 2013; Runger, 1999). Exposure to UVB may result in the formation of cyclobutanepyrimidine dimers and 6-4 pyrimidine-pyrimidone photoproducts, leading to single-basesubstitutions of cytosine (C) for thymine (T) or Double-base changes from CC to TT if notproperly repaired via DNA repair mechanisms. Occurrence of these mutations in oncogenes ortumour suppressors may lead to cancer initiation (Budden & Bowden, 2013; Volkovova et al,2012; von Thaler et al, 2010).1.1.3 Staging and subtypes of cutaneous melanomaCutaneous melanoma is divided into four major categories named superficial spreadingmelanoma (SSM), lentigo maligna melanoma (LMM), nodular melanoma (NM), and acrallentiginous melanoma (ALM) (Clark et al, 1969; Cummins et al, 2006).SSM is the most common type of melanoma constituting approximately 70% of allmelanomas (Gray-Schopfer et al, 2007). SSM may form from an existing nevus andpredominantly affects the intermittently sun-exposed areas of body such as the upper back ofmen and women and the legs of women (Lens, 2008). The lesions average about 2 cm indiameter and are usually flat or slightly elevated and have irregular borders and variegated colour(Clark et al, 1969; Cummins et al, 2006; Porras & Cockerell, 1997).9LMM represents about 5-10% of melanomas. They are mostly seen in elderly individualsin regions of the body such neck, cheek and nose that are chronically exposed to sun. The lesionsare usually large, flat and have variable brown colours and irregular borders (MacKie, 2000).As the second most common form of melanoma, NM represents about 10-15% of allmelanomas (Lens, 2008) and is mostly seen on head, neck and trunk region (Barnhill & Mihm,1993). NM grows faster and is more invasive compared to other melanoma subtypes. As a result,it is responsible for the highest death rate in white populations (Mar et al, 2013). The lesions areusually evenly colored, have round borders, are raised and relatively small (Demierre et al, 2005;Geller et al, 2009; Porras & Cockerell, 1997).ALM is the least common type of melanoma accounting for about 5% of all melanomacases. It is not associated with UV exposure and is the type responsible for most of themelanoma cases in people with dark skin. Body regions affected by this subtype are mostlypalms of the hand, soles of the feet and nail beds.  The lesions have irregular borders andvariegated colors (Gray-Schopfer et al, 2007).American Joint Committee on Cancer (AJCC) has divided melanoma into four stagesconsidering important prognostic factors such as thickness, ulceration, lymph node involvement,site of metastasis, mitotic rate and Lactate dehydrogenase (LDH) level of serum which is used asa general marker in the prognosis of cancers (Balch et al, 2009).Stage I tumours are ≤ 2 mm thick with no ulceration or ≤ 1 mm thick with ulceration(Balch et al, 2009). Tumours > 1mm thick with ulceration or > 2mm thick without ulceration areconsidered stage II tumours (Balch et al, 2009). Stage I and II melanomas show no evidence ofmetastasis to lymph nodes or distant sites (Balch et al, 2009). Stage III melanomas spread toregional lymph nodes. The depth of the melanoma doesn’t matter anymore and there is no10evidence of distant metastasis and ulceration may or may not be present (Balch et al, 2009).Stage IV melanomas are the ones that have metastasized to distant sites. Patients with normalserum LDH levels and metastases in the skin, subcutaneous tissue or distant lymph nodes mighthave a relatively favorable outcome but the ones with metastases in visceral tissues other thanthe skin, subcutaneous tissue, distant lymph nodes and lungs, and/or high LDH levels don’tnormally show a favorable outcome (Balch et al, 2009).1.1.4 Melanoma epidemiologyMalignant melanoma is an aggressive and lethal form of skin cancer that is resistant toconventional radio- and chemo-therapy (Gray-Schopfer et al, 2007). Number of new melanomacases and deaths caused by melanoma in USA in 2015 are estimated to be 73,870 and 9,940,respectively (American Cancer Society, 2015). Melanoma is estimated to have an incidence rateof 2.8-3.1 per 100,000 in the world (Ferlay et al, 2010). The incidence and the mortality rates ofmelanoma have increased by about 2-3% per year in three decades (Perlis & Herlyn, 2004).Although melanoma constitutes less than 5% of all dermatological cancer cases, it is responsiblefor more than 80% of the deaths caused by skin cancer (Cummins et al, 2006). Patients withAJCC stage IV metastatic melanoma have a poor prognosis with a median survival of only 6-8months (Cummins et al, 2006) and about 16% of the patients with distant stage disease live morethan 5 years (American Cancer Society, 2015).Melanoma incidence rates vary widely from location to location. For example, theincidence rates for Indian women and Caucasian men in Queensland, Australia are 0.2 and 55-65per 100,000 per year, respectively (Garbe & Leiter, 2009).This could be because of thedifference in racial skin phenotype and the level of exposure to sun (Chang et al, 1998). People11with light skin have ten times higher chance of developing melanoma compared to Hispanic,Black or Asian people (Ries et al, 2000). Having a family history of melanoma, previousmelanoma or high number of nevi are also very important risk factors for melanomadevelopment (Ries et al, 2000). Other factors such as age and sex have also been shown to affectthe incidence rate of melanoma. Overall, older people and males have been shown to be moreaffected by the disease compared to younger people and females (Rigel, 2010).1.2 Translation1.2.1 An introduction to translationThe process of translation is a very important step in the biosynthesis of protein during whichmRNA genetic information is used to make a polypeptide chain. Furthermore, translationaldysregulation and abnormalities in protein synthesis have an important role in the developmentand progression of the neoplastic process (Ruggero, 2013).Protein synthesis includes initiation, elongation and termination steps. Initiation is thestep that is mostly involved in cancer development and progression and is also the mostcomplicated step because it involves the highest number of translation factors. Initiation isinvolved in mRNA recruitment to the 40S subunit, finding the start codon, and the joining of the60S subunit and subsequent formation of the 80S ribosome (Silvera et al, 2010). Duringelongation, the movement of the 80S ribosome along the mRNA leads to the translation of eachcodon to an amino acid which is then added to the growing polypeptide chain. Translationtermination occurs when the 80S ribosome encounters a stop codon, causing the release of therecently produced protein. Ribosome recycling finally liberates the mRNA, and the 80S12ribosome is again split into the 40S and 60S subunits which can be used for another round oftranslation cycle (Dever & Green, 2012).1.2.2 Translation initiationTranslation initiation is nearly all the time the rate-limiting stage of protein synthesis as well asthe most regulated one (Aitken & Lorsch, 2012; Fraser, 2009; Hinnebusch & Lorsch, 2012;Jackson et al, 2010; Sonenberg & Hinnebusch, 2009). Initiation rates may vary dramatically. Forexample, in yeast initiation rates may vary between 4-233 seconds on different mRNAs (Shah etal, 2013). This could be partly because of the regulation of the availability and activity ofinitiation factors by signaling pathways such as the PI3K/Akt/mTOR and MAPK pathways. Thedifference in initiation rates could also result from differences in mRNA regulatory features, likea highly structured 5′ untranslated region (UTR). Translation initiation has five steps: (1) mRNAbinding by the eukaryotic translation initiation factor 4F (eIF4F) cap-binding complex; (2) 43Spreinitiation complex (PIC) formation; (3) mRNA recruitment to the 40S subunit; (4) Selectionof the initiation codon; and (5) 60S ribosome joining (Aitken & Lorsch, 2012; Fraser, 2009;Hinnebusch & Lorsch, 2012; Jackson et al, 2010; Sonenberg & Hinnebusch, 2009). Figure 1.2 isa schematic diagram summarizing the steps of the translation initiation in eukaryotes (Aitken &Lorsch, 2012).1314Figure 1.2 Schematic diagram summarizing the steps of the translation initiation in eukaryotes.Translation initiation starts by joining the initiator tRNA and eIF2.GTP to form the ternary complex (TC) (1). eIFs1, 1A, 3 and 5 help to recruit the ternary complex to the 40S subunit in order to form the PIC (2). At the same timethe PABP and the eIF4 factors bind mRNA in order to make an activated messenger ribonucleoprotein (mRNP)(3a). This mRNP is next recruited to the PIC (3b). After binding to the 5′ end of the mRNA, the PIC starts to scanthe mRNA in order to find the initiation codon (AUG) (4). Recognition of the initiation codon causes the release ofeIF1 and conversion of eIF2.GTP to a GDP-bound form eIF2, making the scanning process come to a halt (5). Therelease of eIF5 and eIF2.GDP allows eIF5B to mediate the union of the 60S ribosome with the 40S ribosome (6).After joining of the ribosomal subunits eIF5B hydrolyzes GTP that results in the separation of the GDP bound formof eIF5B. eIF1A is the final factor that dissociates from the 80S initiation complex (IC) (7). Adopted from Aitken &Lorsch, 2012, with permission to reprint.151.2.2.1 eIF4F cap-binding complex binds the mRNAmRNAs translation is prompted by recognition of the 5′ 7-methyl-guanosine cap (Carroll &Borden, 2013; Topisirovic et al, 2011) and the 3′ poly(A) tail (Mangus et al, 2003; Sachs et al,1997) both known to help protect an mRNA from degradation and also promote translationinitiation (Gallie, 1991; Searfoss et al, 2001). eIF4F interacts with these features and iscomposed of eukaryotic translation initiation factor 4E (eIF4E) (cap-binding), eukaryotictranslation initiation factor 4A (eIF4A) a member of the DEAD (Aspartic acid- Glutamic acid -Alanine-Aspartic acid) box helicase family, and eukaryotic translation initiation factor 4G(eIF4G) (molecular scaffold) (Gingras et al, 1999; Grifo et al, 1983).eIF4E is the cap-binding protein that uses two conserved tryptophans located in its cap-binding pocket to sandwich and recognize the 5′ 7-methyl-guanosine cap (Marcotrigiano et al,1997; Matsuo et al, 1997). eIF4E limits mRNA recruitment to the ribosome because it is the leastabundant initiation factor (Duncan et al, 1987). eIF4E also has a high binding affinity to eIF4Gby means of a few conserved residues present on the convex dorsal surface of eIF4E on theopposite side of the cap-binding pocket (Marcotrigiano et al, 1999). This interaction keepseIF4G close to the 5′ end of the mRNA resulting in mRNA recruitment to the 40S ribosome. Theavailability of eIF4E is regulated by eIF4E-binding proteins (4E-BPs). 4E-BPs inhibit cap-dependent translation by sequestering eIF4E from eIF4G. Likewise, programmed cell death 4(PDCD4) inhibits cap-dependent translation by preventing eIF4A from its unwinding activity(Lankat-Buttgereit & Goke, 2009). Additionally, eIF4E has a role in transporting mRNAsinvolved in survival and cell cycle progression (Culjkovic et al, 2006) from the nucleus to thecytoplasm (Culjkovic et al, 2005; Rousseau et al, 1996).16eIF4G is the largest member of the eIF4F complex (Gingras et al, 1999; Hentze, 1997;Imataka et al, 1998; Keiper et al, 1999; Prevot et al, 2003; Yan et al, 1992) that acts as amolecular scaffold binding to and coordinating the activities of other components such as mRNA(Berset et al, 2003; Goyer et al, 1993; Park et al, 2011), poly(A)-binding protein (PABP)(Imataka et al, 1998; Tarun & Sachs, 1996), eIF4A (Imataka & Sonenberg, 1997; Korneeva et al,2001), eIF4E (Lamphear et al, 1995; Mader et al, 1995), eIF3 (Korneeva et al, 2001; Lamphearet al, 1995), and the MAPK-interacting kinases (MNKs) (Pyronnet et al, 1999). eIF4G may alsoplay a role in stabilizing the interaction of eIF4E with the cap and strengthening the affinity ofPABP to the poly(A) tail (Yanagiya et al, 2009). The latter may contribute to circularizing andreinitiating a terminating ribosome on the mRNA (Le et al, 1997; Wells et al, 1998).As a result of interaction with eIF3, eIF4G plays a role in connecting eIF4F-mRNAcomplex and the 43S preinitiation complex (PIC) (Hinton et al, 2007; Lamphear et al, 1995;Villa et al, 2013). Additionally, as a result of interaction between eIF4G and DEAD box helicaseeIF4A, the latter is recruited to the mRNA unwinding the secondary structure of the mRNA5′UTR in an ATP-dependent reaction in order to facilitate the binding of the mRNA to themRNA-binding site of the 40S ribosome (Lawson et al, 1989; Rogers et al, 2001; Rogers et al,1999; Rozen et al, 1990). The ATP hydrolysis and helicase activity of eIF4A are facilitated bythe interaction of eIF4A with eIF4E, eIF4G and the helicase accessory protein eIF4B(Feoktistova et al, 2013; Korneeva et al, 2005; Nielsen et al, 2011; Ozes et al, 2011; Schutz etal, 2008). eIF4E has been demonstrated to excite unwinding of mRNA secondary structure bybinding to eIF4G, relieving autoinhibition and permitting eIF4G to stimulate the eIF4A helicaseactivity; this function of eIF4E has been shown to be independent from its cap-binding functions(Feoktistova et al, 2013). However, it is not quite clear whether eIF4F components and its17binding partners remain associated during the whole process of translation initiation or not(Pestova & Kolupaeva, 2002; Poyry et al, 2004).1.2.2.2 43S PIC formationThe role of 43S PIC is to make the 40S ribosome mRNA entry channel and decoding site readyfor mRNA recruitment, scanning, and localization of initiation codon. 43S PIC consists of the40S small ribosomal subunit, the initiator methionyl tRNA (Met-tRNAi) and a group ofeukaryotic initiation factors including eIF1, eIF1A, eIF2, eIF3 and eIF5. Based on the“scanning” model, the 40S subunit moves along the mRNA in a linear fashion in order to selectan initiation codon (Jackson et al, 2010; Silvera et al, 2010).The mRNA decoding site of the 40S subunit consists of tRNA-binding sites, namely anaminoacyl (A site) site, a peptidyl (P site) site, and an exit (E site) site. eIF1 and eIF1A are thetwo initiation factors that play an important role in opening of the mRNA entry channel anddecoding site of the 40S subunit (Fraser, 2009; Passmore et al, 2007) to allow the mRNA toenter the 40S subunit for scanning and initiation to happen (Pestova et al, 1998). Methionine asthe first amino acid of the polypeptide is bound to an initiator tRNA and is transferred to the P-site as initiator methionyl tRNA (Met-tRNAi). The initiator tRNA may form a complex witheIF2 and GTP called the eIF2 ternary complex or eIF2-TC which binds the 40S subunit. Theinitiator tRNA may also bind to the 40S subunit in a complex of factors containing eIF1, eIF2-TC, eIF3 and eIF5 (Asano et al, 2000; Sokabe et al, 2012). Finally, however, the initiator tRNAresides on the 43S PIC as part of a ternary complex (Hinnebusch & Lorsch, 2012).eIF2 is composed of 3 subunits called eIF2α, eIF2β and eIF2γ. The last one binds GTP,the 40S ribosomal subunit and the initiator tRNA (Kapp & Lorsch, 2004; Schmitt et al, 2010;18Shin et al, 2011) and the first two elevate the affinity of eIF2γ for the initiator tRNA (Naveau etal, 2010). eIF2α also interacts with mRNA and 40S ribosome several times during initiation oftranslation (Naveau et al, 2010; Pisarev et al, 2006). eIF1A and mRNA stabilize the interactionbetween eIF2-TC and the 40S subunit (Passmore et al, 2007). eIF1, eIF1A and mRNA all have arole in accelerating this interaction (Algire et al, 2002; Chaudhuri et al, 1999; Fekete et al, 2005;Passmore et al, 2007). eIF3 is the scaffold that is thought to organize the 43S PIC (Damoc et al,2007). Studies have shown that eIF3 elevates the affinity of eIF2-TC to the ribosome (Benne &Hershey, 1978; Peterson et al, 1979), to eIF1 (Karaskova et al, 2012; Sun et al, 2011), eIF1A(Fraser et al, 2007; Sun et al, 2011) and the 40S subunit (Hashem et al, 2013; Siridechadilok etal, 2005).1.2.2.3 Recruitment of mRNA to 43S PICIn mammals, the interaction between the eIF4G element of the eIF4F complex and the eIF3element of the 43S PIC (De Gregorio et al, 1999; Hinton et al, 2007; Villa et al, 2013) as well asother interactions such as eIF4B/PABP (Bushell et al, 2001; Cheng & Gallie, 2010; Le et al,1997), eIF4B/eIF3 (Methot et al, 1996) and eIF4B/40S (Methot et al, 1996; Rozovsky et al,2008; Walker et al, 2013) have been shown to play a role in recruitment of the mRNA to the 43SPIC. In addition, in vivo studies indicate that mTOR complex 1 (mTORC1) stimulates theinteraction between eIF4G and eIF3 (Harris et al, 2006; Thoreen et al, 2012). By the end of therecruitment phase, mRNA together with associated initiation factors and Met-tRNAi is firmlyattached to the 40S ribosome. At this stage the 40S ribosomal subunit has to find the initiationcodon, a process necessary for the formation of the 80S ribosome and the start of elongation.191.2.2.4 5′ to 3′ ScanningAfter the mRNA is stably bound in the decoding site of the 40s ribosomal subunit, the subunitstarts scanning the mRNA in the 5′ to 3′ direction in an ATP dependent manner. In this processthe anticodon of the initiator tRNA searches for the matching codon on the mRNA (Cigan et al,1988). The mRNA must be single stranded to be able to bind in the decoding site. Two modelsare used to explain how 40S ribosome unwinds the secondary structure of the mRNA.In the “ratchet” model, eIF4B binds to the mRNA upstream of 40S subunit. This inhibitsbackwards movement (Spirin, 2009). The mRNA secondary structure is then unwound as it ispulled through the small channel of the ribosome (Aitken & Lorsch, 2012; Spirin, 2009). In thesecond model eIF4F unwinds mRNA before entering the 40S ribosomal decoding site (Aitken &Lorsch, 2012; Jackson et al, 2010; Marintchev et al, 2009). The scanning model is the mostaccepted model for the movement of ribosome during translation initiation. There are; however,other proposed models as well, such as shunting and initiation at internal ribosome entry sites(IRESs). During shunting mRNA binds to the 40S subunit near the cap but in this case theribosome shunts downstream instead of moving in a linear fashion, missing a big segment of the5′ untranslated region (5′ UTR) and initiation might start at downstream codons (Spirin, 2009).In the IRES model a group of translation initiation factors, or sometimes none, recruit the mRNAto the 40S ribosome by using an internal segment of the mRNA instead of by adhering to thecap. The shunting and IRES models have been well shown in the viral system but only a fewcases have been presented in mammals (Jackson, 2013).The binding of eIF1A and eIF1 to the 40S subunit interface seems to be very importantfor the recruitment of mRNA and its motion inside the decoding site by maintaining an openconformation of the mRNA-binding channel (Maag et al, 2005; Passmore et al, 2007; Pestova et20al, 1998). It has been shown that 43S PIC including eIF1, eIF1A, eIF2-TC and eIF3 withouteIF4F and ATP is able to recruit and identify the initiation codon in the case of a totallyunstructured 5′ UTR (Pestova & Kolupaeva, 2002). However, eIF4F seems to be necessary in thepresence of the cap structure (Mitchell et al, 2010).1.2.2.5 Localization of the initiation codonThe sequence surrounding the initiation site is critical for determining the possibility ofsuccessful initiation at the start codon. The optimal consensus sequence, called the Kozaksequence, is GCC(A/G)CCAUGG. The A of the AUG and the C before that are defined as the+1 and -1 position, respectively. The bases at the -3 and +4 positions are the most important(Kozak, 1986; Kozak, 1987). In contrast to what was thought before, in about only 25% ofmammalian mRNAs initiation starts from the first AUG codon (Ingolia et al, 2011). In the rest ofthe mRNAs potential initiation codons present either upstream or downstream of the main openreading frame could be used. Factors such as eIF1, eIF1A, eIF2 and eIF5 have been shown tocontrol the fidelity of the initiation site recognition in yeast (Lorsch & Dever, 2010). eIF1 hasalso been shown to inhibit translation initiation at codons different from AUG or AUG codons inpoor context (Pestova et al, 1998; Pestova & Kolupaeva, 2002; Pisarev et al, 2006). eIF1 bindsnear the P site and with the help of its N-terminal tail incompletely closes the P site (Lomakin etal, 2003; Rabl et al, 2011), inhibiting complete entry of the initiator tRNA into the P site (Rabl etal, 2011). This way the tRNA anticodon stem is prevented from steady base pairing with anycodons before finding a start codon (Lorsch & Dever, 2010).Recognition of a start codon results in establishment of codon–anti codon base pairing that isaccompanied by displacement of eIF1 from near the P-site (Cheung et al, 2007; Maag et al,212005; Unbehaun et al, 2004). This leads to closing of the mRNA-binding channel and helping tostop the scanning process by securing the mRNA in the decoding site (Passmore et al, 2007).This change of conformation also changes the position of eIF1A and eIF5 on the 40S subunit(Maag et al, 2006; Nanda et al, 2013). eIF5 is the GTPase-activating protein (GAP) forribosome-bound eIF2-GTP. This GTP hydrolysis function of eIF5 also aids in stopping the43S/mRNA complex from scanning at the initiation codon (Das et al, 2001; Nanda et al, 2009;Paulin et al, 2001). Premature hydrolysis of eIF2-bound GTP and subsequent phosphate release,are prevented by eIF1 (Algire et al, 2005; Unbehaun et al, 2004). This inhibitory effect of eIF1 isremoved after its dissociation following the recognition of an initiation codon. GTP hydrolysisreduces eIF2’s affinity for Met-tRNAi, leading to partial dissociation of eIF2–GDP from 40Ssubunit (Kapp & Lorsch, 2004; Pisarev et al, 2006). After dissociation, eIF2-GDP is recycledback to eIF2-GTP mediated by eIF2B, the guanine nucleotide exchange factor. eIF2-GTP canthen bind to another Met-tRNAi for the next round of translation initiation (Proud, 2005).1.2.2.6 60S ribosome joining and establishment of 80S ribosomeThe 40S ribosomal subunit has to recruit a 60S subunit to form an 80S ribosome to startelongation. After the recognition of initiation codon and before binding of 60S subunit to 40S, alleIF5 and eIF2-GDP complexes have to completely dissociate from the 40S ribosome interface.This is mediated by binding a ribosome-dependent GTPase called eIF5B to the intersubunit cleftof the 80S ribosome (Allen et al, 2005; Simonetti et al, 2008; Unbehaun et al, 2004; Unbehaunet al, 2007). Joining of ribosomal subunits and eIF5B-GTP hydrolysis are promoted (Acker et al,2006; Marintchev et al, 2003; Olsen et al, 2003) through the relocation of the C-terminal tail ofeIF1A from the P-site and its subsequent interaction with eIF5B following the recognition of22start codon (Nanda et al, 2013). eIF5B-GTP hydrolysis leads to dissociation of eIF5B from therecently formed 80S ribosome (Shin et al, 2002). eIF1A is the final factor separating from theribosome interface after taking part in nearly all steps of initiation such as the recruitment ofmRNA, recognition of initiation codon and formation of 80S ribosome (Acker et al, 2009).1.2.3 Translation elongationAt the end of the initiation step the initiator tRNA is located in the ribosomal P site, and the Asite is ready to accept the next aminoacyl-tRNA. Each course of translation elongation consistsof three phases. First the appropriate aminoacyl-tRNA must bind to the mRNA codon in the Asite. Next, peptide bond should be formed and then the tRNAs and mRNA need to be shifted byone codon relative to the ribosome (Dever & Green, 2012). Figure 1.3 is a schematic diagramsummarizing the steps of the translation elongation in eukaryotes (Dever & Green, 2012).Aminoacyl-tRNA is transferred to the A site of the 80S ribosome by GTP-boundeukaryotic elongation factor (eEF) 1A (eEF1A) leading to the formation of a base pair betweencodon and anticodon (Dever & Green, 2012; Voorhees & Ramakrishnan, 2013). This base pairformation causes a conformational change leading to activation of GTP hydrolysis which isfollowed by the release of eEF1A/GDP and phosphate from the ribosome (Schmeing et al, 2009;Voorhees et al, 2010). Subsequently, peptide bond formation is catalyzed by the peptidyltransferase centre in the 60S ribosome (Zhang et al, 2009). This involves the tRNA in the P sitereleasing its peptide onto the tRNA in the A site which leads to translocation of the tRNAs intohybrid P/E and A/P sites. GTP-bound eEF2 is the factor that completes this translocation (Gao etal, 2009). GTP hydrolysis leads to eEF2 conformational change resulting in an openconformation of the ribosome and completion of the translocation of the tRNAs (Chen et al,232012d). EIF5A has been suggested to play a role in elongation by interacting with eEF2 (Dias etal, 2012; Gregio et al, 2009; Li et al, 2010; Saini et al, 2009). EIF5A is thought to possibly assistin the translation of proteins with polyproline motifs that might be hard to translate because oftheir conformation (Doerfel et al, 2013; Gutierrez et al, 2013). eEF1B is a Guanine NucleotideExchange Factor for eEF1A that speeds up the dissociation of GDP and reproduction ofeEF1A/GTP/aminoacyl-tRNA complex for the coming rounds of elongation (Kemper et al,1976). After complete translocation, the A site is empty and ready for another aminoacyl-tRNAto decode the next codon. The P site holds the peptidyl-tRNA and the E site is occupied by thedeacylated tRNA that will soon be dismissed. The polypeptide chain grows through therepetition of this process along the mRNA’s coding sequence (Bostrom et al, 1986; Ingolia et al,2011).24Figure 1.3 Schematic diagram summarizing the steps of the translation elongation in eukaryotes.Beginning from the top of the diagram, eEF1A-teranry complex consisting of eEF1A, GTP and aminoacyl-tRNArecruits the aminoacyl-tRNA to bind the 80S ribosome by base-pairing the tRNA anticodon with the mRNAcodon in the A site of the 80S ribosome. After dissociation of the GDP bound eEF1A, the aminoacyl-tRNA isaccommodated into the A site. The exchange factor eEF1B next recycles the eEF1A.GDP to eEF1A.GTP. Thepeptidyl transferase center in the 60S ribosome catalyzes peptide bond formation leading to the addition of a newamino acid to the growing polypeptide chain.This reaction results in the transition of tRNAs from the P and A sitesinto the P/E and A/P sites, respectively, forming a hybrid state. The binding of eEF2.GTP to the 80S ribosome andhydrolyzing its GTP then results in complete translocation of tRNAs into the P and E sites. The released deacylatedtRNA from the E site and eEF2.GDP leave the ribosome which is now ready for the next codon to be translated init’s A site. In this diagram, large ribosomal subunit is drawn transparent. The red ball represents GDP and the greenball represents GTP. Adopted from Dever & Green, 2012, with permission to reprint.251.2.4 Translation terminationFigure 1.4 is a schematic diagram summarizing the steps of the translation termination andrecycling in eukaryotes (Dever & Green, 2012). The presence of a stop codon (UAA, UGA,UAG) in the A site triggers the binding of eukaryotic release factor (eRF) 1 (eRF1) to eRF3-GTP(Ito et al, 1998; Mitkevich et al, 2006; Pisareva et al, 2006) to terminate translation in a GTP-dependent manner (Alkalaeva et al, 2006; Stansfield et al, 1995). eRF1 recognizes and binds thestop codon (Song et al, 2000) followed by eRF3 hydrolyzing GTP (Frolova et al, 1996) resultingin its release and permitting complete presence of the middle domain of eRF1 in the peptidyltransferase center (Song et al, 2000). The release of the peptide is next catalyzed, stimulated byan ATP-independent activity of ABCE1 (Dever & Green, 2012). ABCE1 finally causes ribosomedissociation and factor release in an ATP dependent manner (Dever & Green, 2012). As a resultof ribosome recycling all of the constituents are released for the next round of protein synthesis(Jackson et al, 2012).26Figure 1.4 Schematic diagram summarizing translation termination and recycling in eukaryotes.When a stop codon reaches the A site of the 80S ribosome, the eRF1:eRF3:GTP ternary complex binds to the A site. Upon GTP hydrolysis eRF3 is released.eRF3 release and binding of ABCE1/Rli1 help eRF1 to accommodate into an optimally active configuration. Following peptide release from the P site tRNA,ABCE1/Rli1 facilitates ribosome dissociation and factor release in an ATP-dependent fashion. In this diagram, large ribosomal subunit is drawn transparent. Thered ball represents GDP and the green ball represents GTP. Adopted from Dever & Green, 2012, with permission to reprint.271.3 eIF4E1.3.1 eIF4E binds the capeIF4E was originally detected as a 24 KDa protein capable of binding to the cap structure ofmRNAs at their 5′ end (Sonenberg et al, 1978). eIF4E looks like a cupped hand squeezing thecap between finger and thumb (Marcotrigiano et al, 1997; Tomoo et al, 2005). eIF4E activity isrelated to its cap-binding properties. Mutation of tryptophan 56 to an alanine has been shown toimpair the oncogenic activities of eIF4E in mouse models and human cell lines by disrupting thecap binding ability of eIF4E (Topisirovic et al, 2011).1.3.2 eIF4E a tumourigenic translation initiation factoreIF4E is a translation initiation factor needed for the translation of most mRNAs but altering itsexpression doesn’t seriously affect global protein synthesis (De Benedetti & Harris, 1999;Sonenberg & Hinnebusch, 2009). However, a particular group of mRNAs with highly structured5′ UTR are very sensitive to changes in eIF4E activity and expression (De Benedetti & Graff,2004; De Benedetti & Harris, 1999). These mRNAs mostly encode cancer-promoting proteinsplaying roles in promoting proliferation, survival and succession to metastasis such as theoncogenic c-myc, different cyclins, the antiapoptotic factor B-cell lymphoma-extra large (BCL-XL), vascular endothelial growth factor (VEGF) and matrix metalloproteinases (MMPs) (Graffet al, 1995; Kevil et al, 1996; Li et al, 2003; Rosenwald et al, 1995; Rousseau et al, 1996; Westet al, 1995). Polysome profile experiments have shown that as a result of eIF4E overexpressionthe translational efficiency of these molecules increases as indicated by the increase in thenumber of ribosomes per mRNA molecule (Graff et al, 1995; Jiang & Muschel, 2002; Kevil etal, 1996; Li et al, 2003; Rosenwald et al, 1995; Rousseau et al, 1996; West et al, 1995).28However the translation of mRNAs such as glyceraldehydes 3-phosphate dehydrogenase(GAPDH) with a simple and short 5′ UTR doesn’t change significantly as a result of an increasein eIF4E expression and activity (De Benedetti & Harris, 1999).1.3.3 Regulation of eIF4E availability by 4EBPsmTORC1 regulates the 4E-BPs by phosphorylating first, Thr37/Thr46 residues and then Thr70and finally Ser65 in a hierarchical manner (Gingras et al, 2001). Phosphorylation of these sites(the hyperphosphorylated form) prevents the binding of 4E-BPs to eIF4E. The 4E-BPs preventthe formation of eIF4F complex and initiation of translation (Pause et al, 1994) by competingwith eIF4G for binding to the dorsal side of eIF4E (Mader et al, 1995). Because of their higherdependence on the activity of eIF4F complex, mRNAs with highly structured 5′ UTR structureare mostly affected by the sequestration of eIF4E by the 4E-BPs (Cawley & Warwicker, 2012;Provenzani et al, 2006).eIF4G and the 4EBPs show similar binding affinities towards eIF4E via their conservedeIF4E-binding motif (Gosselin et al, 2011; Lukhele et al, 2013; Marcotrigiano et al, 1999);however, the rates of binding and dissociation of the 4EBPs are two to three orders of magnitudefaster than that of eIF4G (Umenaga et al, 2011). This difference may be explained by the factthat, as inhibitors, the 4EBPs need quick control of eIF4E binding but eIF4G needs longer lastinginteraction with eIF4E to support translation initiation. Some studies have explained thisdifference by revealing a second eIF4E-binding site that is different between the 4EBPs andeIF4G (Mizuno et al, 2008; Umenaga et al, 2011). As inhibitors of eIF4E, the 4EBPs aretherefore considered tumour suppressors. On the other hand, mTOR is considered an oncogeneas a regulator of eIF4E via phosphorylation of the 4EBPs (Kim et al, 2009). Low levels of294EBP1 and its hyperphosphorylation have been shown to be correlated with poor prognosis indiseases such as melanoma, ovarian, prostate and breast cancers as well as childhoodrhabdomyosarcoma (Armengol et al, 2007; Graff et al, 2009; O'Reilly et al, 2009; Petricoin et al,2007; Rojo et al, 2007).1.3.4 Regulation of eIF4E activity by phosphorylationMNK 1 and 2 phosphorylate eIF4E on Ser209. MNK 1 and 2 are activated in response to cellularstress and survival signals from MEK/ERK, and p38 MAPK pathways (Buxade et al, 2008;Waskiewicz et al, 1999). Interaction of the MNKs and eIF4G is a prerequisite forphosphorylation to occur. Therefore, the eIF4F complex formation should happen before eIF4Ephosphorylation (Pyronnet et al, 1999). eIF4E phosphorylation has been shown to decrease(Scheper et al, 2002) the affinity for the cap (Scheper et al, 2002; Slepenkov et al, 2006; Zubereket al, 2003) which might help the separation of eIF4E and mRNA and the recycling of eIF4E foranother round of translation (Scheper & Proud, 2002). eIF4E phosphorylation has also beenindicated to have an important role in tumourigenesis. Abolishing eIF4E phosphorylation bymutating serine 209 or genetic ablation of the MNKs diminished the tumourigenic properties ofeIF4E in mice (Furic et al, 2010; Ueda et al, 2010). Other studies have indicated that eIF4Ephosphorylation promotes invasion and metastasis by increasing the translation ofprotumourigenic transcripts like MMP-3 or antiapoptotic factors such as (myeloid cellleukemia 1) MCL-1 and baculoviral inhibitor of apoptosis repeat-containing protein 2 (BIRC2)and others (Furic et al, 2010; Robichaud et al, 2015; Wendel et al, 2007).301.3.5 Regulation of eIF4E at the transcriptional leveleIF4E transcriptional regulation has a key role in regulating the expression level of eIF4E. The c-myc proto-oncogene is capable of upregulating eIF4E expression by binding to the twoconserved enhancer box (E-box) motifs on eIF4E promoter (Jones et al, 1996). c-myc alsoupregulates factors that promote apoptosis but eIF4E opposes this action of c-myc bytranslational upregulation of antiapoptotic factors such as MCL-1 and BCL-XL and promotingcell proliferation (Lin et al, 2008; Ruggero et al, 2004). Surprisingly, the c-myc mRNA is atranslational target of eIF4E as well. Therefore, c-myc promotes the transcription of eIF4E andeIF4E in turn promotes the translation of c-myc mRNA in a feed-forward loop (Lin et al,2008).The result is that c-myc and eIF4E increase each other’s expression and together promotegrowth, proliferation and survival of cells. Moreover, c-myc has been shown to upregulate theactivity of eIF4E via enhancing phosphorylation of the 4E-BPs by mTOR (Pourdehnad et al,2013). c-myc-driven lymphomas are very dependent on 4E-BP1 hyperphosphorylation and as aresult sensitive to mTOR inhibitors (Pourdehnad et al, 2013). Consequently, inhibiting eIF4Ecould be a critical factor in cancer therapy since myc family plays an important role in manyhuman cancers (Vita & Henriksson, 2006).1.3.6 Role of eIF4E in evasion of growth suppressioneIF4E was first known as an oncoprotein when it was shown to promote evading growthsuppression. eIF4E overexpression in NIH 3T3 cells made them capable of escaping contactinhibition and forming foci (Lazaris-Karatzas et al, 1990). eIF4E also plays a role in the nucleartransport of MDM2 that degrades the tumour suppressor p53 (Phillips & Blaydes, 2008). In31addition, eIF4E is important for the translation of different cyclins such as A, D1, D3, E1 thatcan dominate growth suppressive signals (Deffie et al, 1995; Lukas et al, 1997).1.3.7 Role of eIF4E in sustaining proliferative signalingeIF4E overexpression enhances the translation of mRNAs promoting proliferation such as theinsulin-like growth factor (IGF), hepatoma-derived growth factor (HDGF) and placental growthfactor (PGF) (Larsson et al, 2007; Mamane et al, 2007), cyclins A, D1, D3, E1 and cyclin-dependent kinases CDK 2 and 4 (Larsson et al, 2007; Larsson et al, 2012; Rousseau et al, 1996),the transcription factor c-myc (Darveau et al, 1985; Saito et al, 1983) and others. Additionally,eIF4E causes RAS hyperactivation leading to sustaining proliferative signaling (Lazaris-Karatzaset al, 1992). Inhibition of eIF4E by the 4EBPs is also one of the ways that mTOR and itsinhibitors control proliferation (Dowling et al, 2010).1.3.8 Role of eIF4E in promoting invasion and metastasiseIF4E has been demonstrated to play a role in invasion and metastasis that is one of thehallmarks of cancer (Hanahan & Weinberg, 2011). In breast cancer mouse models, eIF4Ephosphorylation and availability via mTOR activation have upregulated metastasis (Nasr et al,2013; Robichaud et al, 2015). eIF4E has also been shown to enhance the mRNA translation ofproteins important for invasion and metastasis, such as the Y box-binding protein 1 (YB-1),SNAIL, MMPs, SMAD2, vitronectin, integrins, transforming growth factor β (TGF-β)  andothers (Ghosh et al, 2009; Grzmil et al, 2011; Hsieh et al, 2012; Nasr et al, 2013; Pola et al,2013).321.3.9 Importance of eIF4E and 4EBPs in survival and cancer recurrenceeIF4E can be used as a biomarker in cancer, predicting poor prognosis. Increased expressionof eIF4E in breast cancer is associated with increased risk of recurrence and decreased survival,regardless of the nodal status (Holm et al, 2008; Li et al, 2002; Li et al, 1998; McClusky et al,2005). eIF4E has also been used as a biomarker in cancers of the prostate, head and neck,pharynx, lung, bladder, liver, esophagus and stomach (Chen et al, 2004; Crew et al, 2000; Graffet al, 2009; Nathan et al, 1999; Salehi & Mashayekhi, 2006; Seki et al, 2002; Seki et al, 2010;Wang et al, 2012; Wu et al, 2013). However, in osteosarcoma and acute myeloid leukemia, anincrease in eIF4E expression is not associated with poor prognosis.4EBP1 phosphorylation has also been shown to predict poor prognosis in breast cancerpatients (Rojo et al, 2007; Zhou et al, 2004). In a subset of receptor positive breast cancerpatients, eIF4E expression levels did not correlate with prognosis but phosphorylated 4E-BP1expression levels did (Meric-Bernstam et al, 2012). In prostate cancer, an increase inexpression of eIF4E or phosphorylated 4EBP1 correlated with poor survival independent fromeach other. However, high total 4EBP1 expression levels correlated with better survival (Graff etal, 2009). Similar results were reported about ovarian (Castellvi et al, 2006), gastrointestinal(Martin et al, 2000), and esophageal cancers (Salehi & Mashayekhi, 2006).Different results have been reported about phosphorylated eIF4E levels. High phospho-eIF4E levels in penile squamous cell carcinoma associate with recurrence and metastasis(Ferrandiz-Pulido et al, 2013) and in non-small-cell lung carcinoma (NSCLC) associate withpoor survival (Yoshizawa et al, 2010). In ovarian cancer; however, high phospho-eIF4E levelswere associated with better overall survival (Noske et al, 2008). In head and neck squamous cellcarcinoma (HNSCC) mortality is mostly because of local recurrence instead of metastasis, and33tumour resection is an effective way to prevent recurrence. As a result, determining tumourmargins is very important and eIF4E expression in margins can be utilized as an independentprognostic factor used for surgical management (Franklin et al, 1999). Despite looking normalhistologically, margins that have even 5% eIF4E positive cells indicate local recurrencecompared to the rest of the margins with eIF4E negative cells that predict increased survival(Franklin et al, 1999; Nathan et al, 1999; Nathan et al, 1997). Others have shown that mTOR isvery well activated in tumour margins in HNSCC (Nathan et al, 2004) and measuringeIF4E/4EBP1 ratio is a better way to predict recurrence rather than measuring eIF4E levels alone(Sunavala-Dossabhoy et al, 2011).1.3.10 Therapies targeting eIF4ETherapeutic targeting of eIF4E is presently being considered. The oldest of the eIF4E inhibitorsare cap analogues that compete with mRNAs for binding to eIF4E. However, using these drugsin vivo has limitations such as instability and permeability problems (Wagner et al, 2000). A newprodrug named 4Ei-1 has been designed to better handle these in vivo limitations (Ghosh et al,2009; Li et al, 2013). 4EGI-1 is another drug that prevents the interaction between eIF4G andeIF4E (Moerke et al, 2007). Other compounds such as the MNK inhibitor called cercosporamideare also used to prevent eIF4E phosphorylation (Konicek et al, 2011). Additionally, antisenseoligonucleotides (ASOs) target the eIF4E mRNA for destruction leading to repressingexpression of eIF4E-regulated proteins (Graff et al, 2007). Finally, allosteric or active siteinhibitors of mTOR such as rapamycin, PP242 and Torin 1 result in inhibition of 4EBPphosphorylation and subsequent sequesteration of eIF4E (Apsel et al, 2008; Feldman et al, 2009;Thoreen et al, 2009; Zhang et al, 2011). Recently, combination therapy is being tested in some34studies as well. An example is investigating the effect of the mTOR inhibitor, everolimus,combined with trastuzumab (a monoclonal antibody that interferes with the HER2/neu receptor)on human breast cancer stem cells in vitro and in vivo. The combination of the two drugs wasmore effective at inhibiting cell growth and tumourigenicity in vitro and reducing tumourvolume in vivo compared to when one single agent was used alone (Zhu et al, 2012b).1.4 EIF5A1.4.1 An introduction to EIF5A structure and functionEukaryotic translation initiation factor 5A (EIF5A) is the only cellular protein that contains therare amino acid called hypusine [N6-(4 amino-2-hydroxy) lysine] (Cooper et al, 1983).Hypusination happens posttranslationally that involves two steps. During the first step, the 4-aminobutyl moiety is transferred from spermidine to the ε-amino group of Lys50 leading toformation of deoxyhypusine. This reaction is catalyzed by an enzyme called deoxyhypusinesynthase (DHS). The second step is completed by another enzyme called deoxyhypusinehydroxylase (DOHH) which catalyzes hydroxylation of deoxyhypusine (Park, 2006). EIF5A is asmall protein that was originally obtained from rabbit reticulocyte ribosomes as a proteinnecessary for methionyl-puromycin synthesis in vitro (Benne et al, 1978; Kemper et al, 1976).Hypusine residue is required for EIF5A activity and this could be because of a long side chain ofthis residue that is critical for association of EIF5A to ribosomes. EIF5A isoforms have beenshown to play a part in diseases such as cancer, diabetes (Maier et al, 2010b) and HIV infection(Hoque et al, 2009). These isoforms have also been shown to be involved in cell proliferation,apoptosis, differentiation, transformation and tumourigenesis (Caraglia et al, 2013).35EIF5A was later determined to be involved in translation elongation rather thantranslation initiation. This is because depletion of EIF5A in yeast mutant strains didn’t decreasethe ratio of polysome to monosome but instead led to an increase in polysome over 80Smonosome ratio and the ribosome transit time was determined to be longer as well, which isexpected in the case of depletion of a translation elongation factor not an initiation factor (Gregioet al, 2009; Kang & Hershey, 1994; Saini et al, 2009). EIF5A has been thought to play a role inelongation by interacting with eEF2 (Dias et al, 2012; Gregio et al, 2009; Li et al, 2010; Saini etal, 2009) and to assist in the translation of proteins with polyproline motifs that might be hard totranslate because of their conformation (Doerfel et al, 2013; Gutierrez et al, 2013).1.4.2 EIF5A isoformsTwo isoforms of EIF5A called EIF5A1 and EIF5A2 exist in humans and have 83% sequencesimilarity (Clement et al, 2003). EIF5A1 is present in all tissues and is highly found in fastproliferating cells while EIF5A2 may be more expressed in a tissue-dependent manner (Clementet al, 2006; Jenkins et al, 2001). Both forms undergo hypusination in a similar manner (Clementet al, 2003) and both are overexpressed in certain cancers where they have been suggested as adiagnostic marker. The EIF5A2 gene is located at locus 3q26 which is amplified in differenthuman cancers (Guan et al, 2001). EIF5A2 was for the first time identified as an oncogene inovarian cancer because its gene was found to be amplified and overexpressed in ovarian cancercells and tissues (Guan et al, 2001; Yang et al, 2009). Later human liver cells ectopicallyoverexpressed with EIF5A2 became tumourigenic and induced tumour formation in nude mice(Guan et al, 2004). Overexpression of EIF5A2 was also associated with advanced tumour stageand poor survival in patients with bladder cancers (Luo et al, 2009), colon cancers (Tunca et al,362013), lung cancers (He et al, 2011) and hepatocellular carcinomas (HCCs) (Lee et al, 2010;Shek et al, 2012).Contrary to most available data, in one case EIF5A1 was identified as a tumoursuppressor gene in a mouse lymphoma model. In vivo screening of a short hairpin RNA (shRNA)library targeting genes deleted in human lymphoma, identified S-adenosylmethioninedecarboxylase (AMD1), deoxyhypusine synthase and EIF5A1 as tumour suppressor genes(Scuoppo et al, 2012). However, using small interference RNA (siRNA) to knockdown EIF5A1in a mouse liver cell line did not increase colony formation (Zender et al, 2008). Therefore, theeffect of EIF5A1 as a tumour suppressor in the above mentioned mouse lymphoma model maybe specific to the model.1.4.3 EIF5A and hypusination are a potential therapeutic target for cancer treatmentDeoxyhypusine synthase and deoxyhypusine hydroxylase are completely specific to EIF5Amodification, as a result they have the potential to be used as therapeutic targets in differentcancers to prevent unlimited cell growth (Park, 2006). N1-guanyl-diaminoheptane (GC7) is aspermidine analog that effectively inhibits deoxyhypusine synthase and consequently restrainsthe growth of tumours in animal models (Jasiulionis et al, 2007) as well as growth of varioushuman cancer cell lines (Jasiulionis et al, 2007; Preukschas et al, 2012). However, due totoxicity, GC7 cannot be tested in human clinical trials. Deoxyhypusine hydroxylase, on the otherhand, is an iron-dependent mono-oxygenase that can be inhibited by different iron chelators.Deferoxamine and ciclopirox (Clement et al, 2002; Guan et al, 2004; Hanauske-Abel et al, 1994)are the two iron chelators that are clinically used in humans to treat iron overload and fungal37infection, respectively and ciclopirox has been shown to behave like an antiangiogenic andantiproliferative agent in vitro (Clement et al, 2002).EIF5A expression can also be inhibited using antisense RNA, siRNA or shRNA. EIF5A-2 antisense RNA (Guan et al, 2004) and EIF5A2 siRNA (Lee et al, 2010) inhibited the growth ofa colorectal cancer (CRC) cell line called UACC-1598 and HCC cells, respectively.Additionally, in mouse models of multiple myeloma, EIF5A1 knockdown using siRNA,inhibited the growth of tumours that did not overexpress EIF5A2 (Taylor et al, 2012). AlthoughEIF5A has been demonstrated to play a role in tumour development and progression, themechanism of this process still requires more investigation.1.5 Objective and hypothesesAbnormal expression of translation factors has been detected in different types of cancer.EIF5A2 and eIF4E are the two important translation factors that have been demonstrated to beassociated with patients’ survival and cancer progression and metastasis in various malignancies.However, to our knowledge, so far no studies have fully examined the expression patterns ofEIF5A2 and eIF4E in different stages of melanoma, their possible correlation with patientsurvival, tumour progression and their biological roles in human melanoma.We hypothesized that EIF5A2 is involved in regulation of melanomagenesis. Weinvestigated the expression pattern of cytoplasmic EIF5A2 in melanocytic lesions and how it wascorrelated with patients’ survival. We also studied the role of EIF5A2 in regulating melanomacell invasion and metastasis and the signaling pathway that EIF5A2 belongs to.Mostly, the cytoplasmic form of EIF5A2 has been reported to have a role in oncogenesisand not so many studies have addressed the importance of nuclear EIF5A2 in cancer. We38hypothesized that EIF5A2 is not solely expressed in cytoplasm of melanocytic cells butnuclear EIF5A2 may also be important in regulating melanoma tumourigenesis. Weexplored the expression pattern of nuclear EIF5A2 in melanoma, its correlation with thickness,AJCC stages, patients’ survival and factors such as MMP-2 that are important for cell invasion.eIF4E is another translation factor that has been shown to be significantly involved inoncogenesis in different cancers. We hypothesized that eIF4E is important for melanomadevelopment and progression. We studied the expression pattern of eIF4E in melanoma and itsassociation with patient’s survival. We also investigated the role of eIF4E in epithelial-mesenchymal transition (EMT), melanoma cell invasion, proliferation and apoptosis.39Chapter 2: Materials and methods2.1 TMA construction713 formalin-fixed, paraffin-embedded tissues were obtained from Vancouver General Hospitalfrom the 1990 to 1998 archives of the Department of Pathology according to the Declaration ofHelsinki guidelines and approved by the Clinical Research Ethics Board of The University ofBritish Columbia. The most representative tumour regions were carefully selected and markedon the H&E-stained slide. The TMAs were assembled using a tissue-array instrument (BeecherInstruments, Silver Spring, MD), and duplicate 0.6 mm thick tissue cores were taken from eachbiopsy specimen and spotted on high density blocks. 4 μm sections were cut using a Leicamicrotome (Leica Microsystems Inc., Bannockburn, IL), and transferred to adhesive-coatedslides.2.2 ImmunohistochemistryThe dewaxation of the TMA slides was performed by heating them at 55°C for 30 minutesfollowed by three 5 minutes washes with xylene. Tissues were then rehydrated by being washedin 100%, 95%, and 80% ethanol, and distilled water for 5 minutes each. For performing antigenretrieval, the samples were heated for 30 minutes at 95°C in 10 mmol/L sodium citrate (pH 6.0).Twenty minutes incubation with 3% hydrogen peroxide was performed in order to blockendogenous peroxidase activity. Samples were next incubated for 30 minutes with universalblocking serum (DAKO Diagnostics, Mississauga, Ontario, Canada) and then incubated with therelevant primary antibody overnight at 4°C. The slides were incubated for 30 minutes with eachone of biotin-labelled secondary antibody and streptavidin-peroxidase (DAKO Diagnostics). Thesamples were then developed with 3,3-diaminobenzidine substrate (Vector Laboratories,40Burlington, Ontario, Canada) and counterstained with hematoxylin. Dehydration of the sectionswas performed using a standard procedure and the slides were then sealed with coverslips.Negative control stainings were performed by omitting the primary antibody during the primaryantibody incubation. For the blocking experiment we incubated anti-EIF5A2 antibody with 10times concentration of its synthetic immunogenic peptide (1:10, Biomatik) at 4°C the nightbefore immunohistochemical staining.2.3 Evaluation of TMA immunostainingThe EIF5A2 cytoplasmic and nuclear staining and eIF4E cytoplasmic staining in TMAs wereexamined blinded by independent observers (including one dermatopathologist), one core at atime simultaneously, and a consensus score was reached at the time of scoring each core.Staining intensity was scored as 0, 1+, 2+, and 3+. The percentage of positive staining cells wasalso scored into 4 categories: 1 (0–25%), 2 (26–50%), 3 (51–75%), and 4 (76–100%). The levelof staining was evaluated by immunoreactive score (IRS), which was calculated by multiplyingthe scores of staining intensity and the percentage of positive cells. On the basis of theimmunoreactive score, staining pattern was defined as negative (0), weak (1-3), moderate (4-6)and strong (8-12). For eIF4E we grouped negative and weak staining as low expression andmoderate and strong staining as high expression (Chen et al, 2012b). EIF5A2 staining patternwas defined as 0, negative; and 1-12, positive (Chen et al, 2012c; Jafarnejad et al, 2012).2.4 Statistical analysis of TMADifferences in demographics, clinicopathological characteristics, cytoplasmic and nuclearEIF5A2 expression and cytoplasmic eIF4E expression between patient subgroups were evaluated41by 2 test. Survival time was calculated from the date of melanoma diagnosis to the date of deathor last follow-up. Kaplan-Meier analysis and log-rank test were used to assess the correlationbetween nuclear EIF5A2 expression and patient survival. The Cox proportional hazardsregression model was performed for univariate and multivariate survival analyses. A P value of<0.05 was considered significant, and all tests were two-sided. SPSS versions 11.5 and 16.0(SPSS Inc, Chicago, IL) software were used for all analyses.2.5 Cell culture, antibodies and drugsMelanoma cell lines MMRU, MMAN, MMLH, PMWK, RPEP and RPMMC were kindlyprovided by Dr. H. R. Byers (Boston University School of Medicine, Boston, MA, USA).PMWK is a primary cutaneous melanoma cell line. RPMMC and RPEP are recurrent primarymelanoma cell lines. MMRU, MMAN and MMLH are metastatic melanoma cell lines. SK-mel-3and SK-mel-110 cell lines were kind gifts from Dr. A. P. Albino (Memorial Sloan-KetteringCancer center, New York, USA). A375 and SK-mel-31 cell lines were obtained from AmericanType Culture Collection. All cell lines except melanocytes were cultured in Dulbecco’s ModifiedEagle Media (DMEM) (Invitrogen, Burlington, ON, Canada) supplemented with 10% fetalbovine serum (FBS) (Invitrogen), 100 units/ml penicillin and 100 μg/ml streptomycin(Invitrogen) in 5% CO2 humidified atmosphere at 37°C. Human melanocytes were cultured inmelanocyte growth media supplemented with 5 μg/ml bovine pituitary extract, 1 ng/ml basicfibroblast growth factor, 5 μg/ml insulin, 0.5 μg/ml hydrocortisone, 10 ng/ml phorbol myristateacetate and 4% FBS (PromoCell, Heidelberg, Germany).Anti-actin antibody and rabbit antibody against EIF5A2 were purchased from Sigma-Aldrich (St. Louis, MO, USA); rabbit antibodies against phospho-Ser-473 of Akt, ILK, vimentin,424EBP1, phospho-4EBP1 (p-4EBP1, Ser65), p70S6K, phospho-p70S6K (p-p70S6K, Thr389),cleaved PARP and cleaved caspase-3 were purchased from Cell Signaling (Beverly, MA, USA);mouse antibodies against PTEN, c-myc, Twist, BCL2 and rabbit antibody against eIF4E werepurchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA); rabbit antibody against α-smooth muscle actin (α-SMA) was purchased from Abcam (Cambridge, MA, USA); mouseantibodies against fibronectin, E-cadherin and N-cadherin were purchased from BD Biosciences(Mississauga, ON, Canada). Cisplatin, doxorubicin and rapamycin were purchased from Sigma-Aldrich (St. Louis, MO, USA).2.6 Plasmids and siRNA transfectionsThe EIF5A2 open reading frame was subcloned from pBluescriptR (ImaGegen, Berlin,Germany) into pCDNA3.1+ using PCR primers harboring XbaI and HindIII restriction sites. Therecombinant DNA was handled in accordance with the National Institiutes of Health (NIH)guidelines. EIF5A2 and PTEN overexpression plasmids were transfected by Effectene reagent(Qiagen, Mississauga, Ontario, Canada) according to the manufacturer’s instructions. siRNAstargeting EIF5A2, MMP-2 and eIF4E were purchased from Qiagen (Mississauga, ON, Canada).siRNA was transfected to cells by silenfect transfection reagent (Bio-Rad, Mississauga, Ontario,Canada) according to the manufacturer’s protocol.2.7 Protein extraction and Western blotAfter washing with Phosphate buffered saline (PBS), cells were harvested by scraping on ice andwere next pelleted by centrifugation at 2500 g for 3 min. To extract the whole cell proteins, cellspellets were lysed in modified RIPA buffer (50 mM Tris-HCl, (pH 8.0), 150 mM NaCl, 1% NP-4340, 0.25% sodium deoxycholate, 1mM EDTA) containing freshly added protease inhibitors (100μg/ml phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin A).Next, Samples were sonicated and incubated on ice for 30 min and centrifuged at 12,000 g for 10min at 4°C and supernatant was collected.Protein concentration was determined by Bradford assay (Bio-Rad) according tomanufacturer’s instructions. Then, 50 μg protein of each sample was separated on an 8, 10 or12% SDS polyacrylamide gel (SDS-PAGE) and blotted onto a polyvinylidene difluoride (PVDF)membrane (Bio-Rad). Blocking of the membrane was done with 1.5% Bovine serum albumin(BSA) in PBST (PBS containing 0.05% Tween-20) for 1 hour at room temperature. Themembrane was then incubated with primary antibody prepared in 1.5% BSA in PBST for 1 hourat room temperature or overnight at 4°C. Blots were next washed three times, 5 minutes each inPBST and incubated with secondary infrared dye-conjugated antibodies IRDye 800 or IRDye680 (LI-COR Biosciences, Lincoln, NE, U.S.A.) at room temperature for 1 h. Blots were washedagain with PBST and then scanned on the Odyssey Infrared Imaging System to visualize proteins(LI-COR Biosciences). Immunoblotting of β-actin was used as loading control.2.8 Reverse-transcription and real-time quantitative polymerase chain reaction (qPCR)Total RNA was prepared by Trizol (Invitrogen, Burlington, ON, Canada) and RNAconcentrations were measured with a spectrometer at 260 nm. 2 μg of total RNA was reversetranscribed into cDNA with the SuperScript First-Strand Synthesis System (Invitrogen)according to the manufacturer’s protocol. Real-time PCR (RT-PCR) was performed with SYBRGreen Master mix system (Applied Biosystem, Carlsbad, CA) using a 7900HT qPCR systemthermal cycler (Applied Biosystems, Foster City, CA, USA). The sequences of EIF5A2 primers44were 5′-CCCTGCTGACAGAAACTGGT-3′ (forward) and 5′TTGCACACATGACAGACACC-3′ (reverse). The primers for β-actin were 5′-GCTCTTTTCCAGCCTTCCTT-3′ (forward) and 5′-CGGATGTCAACTTCACACTT-3′(reverse).2.9 Cell invasion assayBoyden chamber assay was used for the cell invasion analysis. The upper compartmentof 24-well Transwell culture chambers (polycarbonate membrane with 8.0 μm pores) was coatedwith 20 μl of 5 mg/ml Matrigel (BD Biosciences) in serum-free medium and was incubated in37°C and 5% CO2 for 2 hours. 50,000 melanoma cells were suspended in 250 μl of serum-freemedium and were loaded on the upper compartment and 750 μl of complete medium was addedto the lower compartment. Cells were incubated in 37°C and 5% CO2 for 24 hours and thenfixed with 10% trichloroacetic acid (TCA) at 4°C for 1 hour. Membranes were next air dried andstained with 0.5% crystal violet for 2 hours that was followed by removing non-invading cellsfrom the upper surface of the membrane with a cotton swab. The dye on the membrane wasextracted by 30% acetic acid and the absorbance was read at 590 nm.2.10 ZymographyZymography assay was used to study the activity of MMP-2. Serum-free medium was applied tocells overnight. YM-3 centricon membranes (Millipore) were utilized to concentrate the proteinsin the conditioned medium by centrifugation at 7000 g for 4 hours at 4°C. Five μg protein wasloaded on a polyacrylamide gel containing 0.2% gelatine. Following electrophoresis, the gel wasincubated in Triton X-100 exchange buffer (20 mM Tris-HCl [pH 8.0], 150 mM NaCl, 5 mM45CaCl2, and 2.5% Triton X-100) for 60 minutes and then washed with incubation buffer for(exchange buffer without Triton X-100) six times, 15 minutes each. Next, the gel was incubatedin incubation buffer at 37°C overnight. The next day the gel was stained with 0.5% Coomassieblue R250 (Sigma) for 1 hour and then destained with 30% methanol and 10% acetic acid for 1hour. Gelatinolytic activity was detected as clear bands on the gel. Recombinant MMP-2 (R&DSystems, Minneapolis, MN, USA) was used as a positive control.2.11 Cell proliferation assayCell proliferation was determined with the sulforhodamine B (SRB) assay. At various timepoints, after the medium was removed, cells were fixed with 10% trichloroacetic acid for 1 h at4°C. The cells were then washed with tap water, air dried, and stained with 0.4% SRB (dissolvedin 1% acetic acid) for 30 min at room temperature. The cells were destained with 1% acetic acidand air dried. For quantification, the cells were incubated with 10 mM Tris (pH 10.5) on a shakerfor 20 min to dissolve the bound dye followed by colorimetric determination at 550 nm.2.12 Fluorescence-activated cell sorting (FACS) analysisAfter collecting the cells by trypsinization, cell pellets were obtained by centrifugation at 500 × gfor 5 min. After washing twice with cold PBS, cells were fixed with 70% ethanol at 4°C for onehour. Next, pellets were resuspended in 1 ml of hypotonic fluorochrome buffer (0.1% Triton X-100, 0.1% sodium citrate) containing 25 µg/ml of RNase A and 50 µg/ml of propidium iodide(PI) (Sigma). After 1 hour incubation at 4°C in the dark, samples were analyzed by EPICS XL-MCL flow cytometer (Beckman Coulter, Miami, FL) to find out the percentage of subdiploidDNA. Cells in sub-G1 phase were considered apoptotic cells.46Chapter 3: Role of EIF5A2, a downstream target of Akt, in promotingmelanoma cell invasion3.1 Background and rationaleRamaswary et al showed that a 17-gene signature in  primary solid tumours is associated withtumour metastasis and poor prognosis (Ramaswamy et al, 2003). One of these genes is DHPSthat encodes for an enzyme necessary for posttranslational hypusination (Wolff et al, 1995). Itwas originally thought that only EIF5A was the substrate of DHPS (Wolff et al, 1995). However,the EIF5A2 isoform was later discovered as another candidate substrate of this enzyme (Guan etal, 2004).Immunohistochemistry analysis in CRC revealed that EIF5A2 overexpression wassignificantly correlated with tumour metastasis and was also an independent predictor ofshortened survival. Ectopic overexpression of EIF5A2 in CRC cells increased cell invasion andmotility in vitro and tumour metastasis in vivo and also induced EMT by upregulatingmesenchymal markers such as vimentin and fibronectin and downregulating epithelial markerssuch as E-cadherin and β-catenin. The depletion of EIF5A2 expression, on the other hand,inhibited cell invasion and EMT (Zhu et al, 2012a). Likewise, a TMA study showed a higherEIF5A2 protein expression in metastatic HCC tumours compared with their matched primaryHCC. Moreover, ectopic expression of EIF5A2 increased HCC cell invasion and migration invitro and tumour metastasis in vivo in a mouse model. Additionally, the same study indicatedthat EIF5A2 induced EMT by reducing the expression of E-cadherin and β-catenin andincreasing the expression of N-cadherine, vimentin and α-SMA (Tang et al, 2010).47Overexpression of EIF5A2 has also been shown to be associated with poor prognosis incolon (Xie et al, 2008; Zhu et al, 2012a), ovarian (Yang et al, 2009), lung (He et al, 2011), andbladder cancers (Chen et al, 2009). The only study regarding the role of EIF5A in melanoma wasdone by Jasiulionis et al who showed that inhibition of EIF5A suppressed melanoma growth(Jasiulionis et al, 2007), suggesting that EIF5A plays an important role in melanomadevelopment. However, to our knowledge, there is no information available on the role ofEIF5A2 isoform in melanoma.To investigate the role of EIF5A2 in melanoma development, we used TMA andimmunohistochemistry to evaluate cytoplasmic EIF5A2 expression in a large set of humanmelanocytic lesions at different stages and analyzed the correlation between cytoplasmic EIF5A2expression and clinicopathologic variables and patient survival.3.2 Results3.2.1 EIF5A2 mRNA and protein expression in melanoma cell linesTo investigate the expression pattern of EIF5A2 in melanoma, we studied the EIF5A2 mRNAlevels in eight melanoma cell lines and normal human epithelial melanocytes (MC). Sixmelanoma cell lines showed over 2-fold increase of EIF5A2 mRNA expression when comparedwith normal melanocytes (Figure 3.1A). We also examined the EIF5A2 protein level in ninemelanoma cell lines and two different batches of normal human epithelial melanocytes as well asHEK293 cells as control. Seven melanoma cell lines showed over 2-fold increase of EIF5A2protein expression compared to the controls (Figures 3.1B and C).48Figure 3.1 Increase of EIF5A2 mRNA and protein expression in melanoma cell lines compared to normalhuman epithelial melanocytes and HEK293 cells.(A) Quantitative RT-PCR analysis of EIF5A2 from total RNA extraction. *P < 0.05; **P < 0.01, Student’s t-test.(B) Western blot analysis of EIF5A2 in whole cell extracts from human melanocytes and melanoma cell lines. (C)Western blot analysis of EIF5A2 in whole cell extracts from a different batch of human melanocytes, HEK293 cellsand melanoma cell lines. β-actin was used as a loading control.493.2.2 Correlation between cytoplasmic EIF5A2 expression and clinicopathologicparametersA total of 713 melanocytic lesions were used for TMA construction. Due to loss of biopsy cores,insufficient cells present in the cores and loss of follow-up, 382 melanoma (242 cases in primarymelanoma and 140 cases in metastatic melanoma), and 77 cases of nevi (28 common acquirednevi and 49 dysplastic nevi) could be evaluated for cytoplasmic EIF5A2 staining (Figure 3.2). Toverify the specificity of the antibody used for staining, we used the synthetic immunogenicpeptide (Biomatik) against anti-EIF5A2 antibody and found that this peptide substantiallyblocked EIF5A2 staining (Figure 3.3). The clinical features of melanoma patients are listed inTable 3.1. We analyzed the expression level of cytoplasmic EIF5A2 in tumours with differentthickness and ulceration, as thickness and ulceration are well known prognostic markers forpatients with primary melanoma. While positive cytoplasmic EIF5A2 expression was detected in90% of thick melanomas (>2mm), only 70% of thin melanomas (≤2mm) showed positivecytoplasmic EIF5A2 expression (P<0.001, ² test; Table 3.1). Positive cytoplasmic EIF5A2expression was detected in 94% of melanomas with ulceration compared to 75% of melanomaswithout ulceration (P=0.005, ² test; Table 3.1). Furthermore, positive cytoplasmic EIF5A2expression was significantly higher in AJCC stage I-II melanomas compared to stage III-IVmelanomas (Table 3.1). The correlation of cytoplasmic EIF5A2 positive staining with tumourthickness and AJCC stages suggests that cytoplasmic EIF5A2 plays an important role inmelanoma invasion.50Figure 3.2 Diagram showing patient inclusion and exclusion.51Figure 3.3 EIF5A2 staining by anti-EIF5A2 antibody is substantially blocked by the blocking peptide,compared with staining with anti-EIF5A2 antibody alone.For blocking experiment, anti-EIF5A2 antibody was incubated with 10 times concentration of its syntheticimmunogenic peptide (1:10, Biomatik) at 4°C overnight before immunohistochemical staining. Bar=100 µm52Table 3.1 Cytoplasmic EIF5A2 staining and clinicopathologic characteristics of 382 melanomasEIF5A2 stainingVariables Positive Negative Total χ2 Value P valueNo. (%) No. (%)All melanoma (n=382)Age, y≤60 161 (81.3) 37 (18.7) 198 0.552 0.458>60 144 (78.3) 40 (21.7) 184SexMale 191 (83.4) 38 (16.6) 229 0.834 0.361Female 122 (79.7) 31 (20.3) 153AJCC stageI 88 (69.3) 39 (30.7) 127 4.046 0.044aII 103 (89.6) 12 (10.4) 115III 48 (88.9) 6 (11.1) 54IV 74 (86.0) 12 (14.0) 86Primary melanoma (n=242)Age, y≤61 95 (77.2) 28 (22.8) 123 0.429 0.512>61 96 (80.7) 23 (19.3) 119SexMale 105 (78.9) 28 (21.1) 133 0.000 1.000Female 86 (78.9) 23 (21.1) 109Tumor thickness (mm)≤2 94 (70.1) 40 (29.9) 134 13.90 <0.001>2 97 (89.8) 11 (10.2) 108UlcerationAbsent 146 (75.3) 48 (24.7) 194 7.911 0.005Present 45 (93.7) 3 (6.3) 48SubtypeLentigo maligna 26 (70.3) 11 (29.7) 37 2.828 0.419Superficial spreading 69 (77.5) 20 (22.5) 89Nodular 35 (85.4) 6 (14.6) 41Unspecified 61 (81.3) 14 (18.7) 75SitebSun-protected 150 (79.8) 38 (20.2) 188 0.376 0.540Sun-exposed 41 (76.0) 13 (24.0) 54Metastatic melanoma (n=140)Age, y≤59 63 (86.3) 10 (13.7) 73 0.096 0.757>59 59 (88.1) 8 (11.9) 67SexMale 86 (89.6) 10 (10.4) 96 1.624 0.203Female 36 (81.8) 8 (18.2) 44AJCC indicates American Joint Committee on Cancer.a Comparison between AJCC stage I to II and III to IV, χ2 test.b Sun-protected sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.533.2.3 Increased cytoplasmic EIF5A2 expression correlates with melanoma progressionTo investigate if cytoplasmic EIF5A2 expression is altered in melanocytic lesions,immunohistochemical staining was performed in TMA slides (Figures 3.4A-D). When wegrouped the samples into negative and positive cytoplasmic EIF5A2 staining, ² test indicated asignificant increase of cytoplasmic EIF5A2 expression as melanoma progressed from dysplasticnevi to primary melanoma and from primary melanoma to metastatic melanoma (Figure 3.4E).54Figure 3.4 EIF5A2 expression in common acquired nevi (CAN), dysplastic nevi (DN), primary melanomas(PM), and metastatic melanomas (MM).(A) and (C) Common acquired nevus with negative EIF5A2 staining. (B) and (D) Metastatic melanoma withpositive EIF5A2 staining. Bar=100 µm. (E) Positive cytoplasmic EIF5A2 expression is increased in metastaticmelanomas compared with common acquired nevi (P<0.001), dysplastic nevi (P<0.001), and primary melanomas(P=0.044). Positive cytoplasmic EIF5A2 expression is increased in primary melanomas compared with commonacquired nevi (P=0.009) and dysplastic nevi (P=0.001).553.2.4 Increased cytoplasmic EIF5A2 expression correlates with poor patient survivalKaplan-Meier survival analysis indicated that positive cytoplasmic EIF5A2 staining inverselycorrelated with overall and disease-specific 5-year survival in all melanomas (P=0.001, log ranktest; Figures 3.5A and B). Both overall and disease-specific 5-year survival are also significantlyworse for primary melanoma patients with positive cytoplasmic EIF5A2 staining compared tothose with negative cytoplasmic EIF5A2 staining (P=0.001 and P<0.001, respectively log ranktest; Figures 3.5C and D). When the primary melanoma cases were further divided into twosubgroups according to tumour thickness, we found that positive cytoplasmic EIF5A2 expressionis significantly correlated with poor patient overall and disease-specific survival in low-riskmelanoma patients (≤2.0mm, P=0.026 and 0.044, respectively log-rank test; Figures 3.5E and3.5F). Strikingly, no patients with tumours ≤2mm thick died within 5 years in negativecytoplasmic EIF5A2 staining group (Figures 3.5E and F). In order to validate this data weanalyzed another cohort of 63 low risk melanoma patients and obtained very similar results(Figure 3.6). However, cytoplasmic EIF5A2 expression did not significantly correlate withpatient survival of thick primary melanomas (>2.0mm) or metastatic melanomas (data notshown). Additionally, we performed Cox regression multivariate analysis including cytoplasmicEIF5A2 expression, age, gender, location, thickness, histological subtype and ulceration inprimary melanoma patients. We also performed Cox regression multivariate analysis to study theeffect of cytoplasmic EIF5A2 expression in patient survival together with age, sex and AJCC inall melanoma patients. In both cases cytoplasmic EIF5A2 expression was indicated to be anindependent prognostic factor of poor 5-year overall and disease-specific survival (Table 3.2).56Figure 3.5 Correlation between cytoplasmic EIF5A2 expression and 5-year patient survival.(A) and (B) Overall and disease-specific 5-year survival of all melanoma patients, respectively (P=0.001 , log ranktest). (C) and (D) Overall and disease-specific 5-year survival of primary melanoma patients (P=0.001 and P<0.001,respectively, log rank test). (E) and (F) Overall and disease-specific 5-year survival of thin (≤2mm) melanomapatients (P=0.026 and 0.044, respectively, log rank test).57Figure 3.6 Correlation between cytoplasmic EIF5A2 expression and 5-year patient survival in low riskmelanoma patients.(A) and (B) Overall and disease-specific 5-year survival of thin (≤2mm) melanoma patients (P=0.002 and 0.030,respectively, log rank test).58Table 3.2 Multivariate Cox regression analysis on 5-year overall and disease-specific survival of melanoma patientsVariables a Overall Survival Disease-specific Survival b SE HR 95% CI P  b SE HR 95% CI PAll melanoma (n=382)Cytoplasmic EIF5A2 0.547 0.268 1.729 1.02-2.92 0.041 0.556 1.203 1.744 1.02-2.10 0.044Age 0.229 0.159 1.257 0.92-1.72 0.151 0.185 1.352 1.203 0.87-1.66 0.259Sex 0.264 0.168 1.302 0.94-1.81 0.116 0.302 5.490 1.352 0.97-1.89 0.079AJCC 1.641 0.171 5.162 3.69-7.21 <0.001 1.703 0.177 5.490 3.88-7.76 <0.001Primary Melanoma (n=242)Cytoplasmic EIF5A2 1.204 0.602 3.334 1.02-10.84 0.045 1.486 0.730 4.420 1.05-18.46 0.042Age 0.397 0.296 1.487 0.83-2.66 0.180 0.219 0.304 1.245 0.68-2.25 0.472Sex 0.045 0.273 1.046 0.61-1.79 0.870 0.066 0.284 1.068 0.61-1.86 0.816Ulceration 0.951 0.293 2.589 1.46-4.60 0.001 1.073 0.305 2.924 1.60-5.31 <0.001Thickness 1.344 0.363 3.835 1.88-7.81 <0.001 1.475 0.395 4.371 2.01-9.48 <0.001Location -0.242 0.320 0.785 0.42-1.47 0.448 -0.459 0.356 0.632 0.31-1.26 0.197Subtype 0.126 0.299 1.135 0.63-2.04 0.672 0.099 0.310 1.104 0.60-2.02 0.750a Coding of variables: EIF5A2 was coded as 1 (negative) and 2 (positive). Age was coded as 1 (60 years) and 2 (>60 years) for all melanoma and 1(61 years) and 2 (>61 years) for primary melanoma. Sex was coded as 1 (male) and 2 (female). AJCC was coded as 1 (stage I to II) and 2 (stage IIIto IV). Ulceration was coded as 1 (absent) and 2 (present). Thickness was coded as 1 (≤2.00 mm) and 2 (>2.00 mm). Location was coded as 1 (sunprotected) and 2 (sun exposed). Subtype was coded as 1 (superficial spreading) and 2 (others).b: regression coefficient.NOTE: SE, standard error of ; HR, hazard ratio; CI, confidence interval; PM, primary melanoma; MM, metastatic melanoma.593.2.5 EIF5A2 regulates cell invasion and MMP-2 activityIncreased ability of cells to invade is one of the hallmarks of cancer, resulting in highermetastatic potential of melanoma and shorter survival of melanoma patients (Friedl & Wolf,2003). To study the role of EIF5A2 in melanoma cell invasion, we used EIF5A2 specific siRNAto knockdown (KD) EIF5A2 expression and found that cell invasion was significantly reduced inEIF5A2-KD MMRU and A375 cells compared with control siRNA-transfected cells (P=0.027and 0.029, respectively, t-test; Figures 3.7A and B and 3.8A, respectively). We also transfectedmelanoma A375, MMLH, SK-mel-3, SK-mel-28, SK-mel-31 (Figures 3.8B-F), MMRU (Figure3.9B) and SK-mel-93 (Figure 3.10B) cells with EIF5A2 overexpression vector and found thatcell invasion in these cell lines increased significantly compared to the vector control. Out ofthese cell lines MMRU (unpublished sequencing data from our lab), A375 (Chen et al, 2012a),SK-mel-28 and SK-mel-3 (Hao et al, 2007) carry BRAFV600E mutation and SK-mel-31 (Chowet al, 2012) carries NRASQ61K mutation.Since MMP-2 was shown to play a crucial role in cell invasion, (Li et al, 2008; Vaisanenet al, 1996) we then carried out the zymography assay to compare the activity of MMP-2 inEIF5A2-KD or EIF5A2-overexpressing MMRU cells compared with respective controls. MMP-2 gelatinolytic activity was decreased in EIF5A2-KD cells (Figure 3.7C), while increased inEIF5A2-overexpressing MMRU cells (Figure 3.9C), compared with respective controls.We previously investigated the expression of MMP-2 in melanoma using tissuemicroarray and found that MMP-2 expression is a prognostic marker for melanoma patients(Rotte et al, 2012). Since the TMA for MMP-2 study contained 369 melanoma biopsies whichwere also included in the current study, we combined the two data sets in order to analyze thecorrelation between MMP-2 and EIF5A2 expression. Our results showed a direct correlation60between the positive staining of EIF5A2 and strong expression of MMP-2 (P=0.004, 2 test;Figure 3.9D).Finally, In order to further determine the role of MMP2 in EIF5A2 induced invasion, weperformed a simultaneous knockdown of MMP-2 and overexpression of EIF5A2 in MMRUmelanoma cell lines. The results showed that relative cell invasion decreased significantlycompared to when cells were overexpressed with EIF5A2 alone (Figure 3.9E), suggesting thatEIF5A2 induced invasion may be mediated at least in part through MMP-2.61Figure 3.7 EIF5A2 knockdown inhibits melanoma cell invasion by reducing MMP-2 activity.MMRU melanoma cell lines were transfected with either control siRNA (siC) or siEIF5A2. (A) Protein extractswere prepared 72 h after transfection and analyzed for the expression of EIF5A2. The numbers below the blotindicate fold change of EIF5A2 expression level. (B) For Boyden chamber assay, 48 h after siRNA transfection,cells were suspended in serum-free medium and seeded on matrigel, incubated at 37oC for 24 h, stained with crystalviolet and quantified. *P<0.05, Student’s t-test. (C) MMP-2 activity determined by zymography which wasperformed 48 h after transfection. Recombinant MMP-2 was used as a positive control. The numbers below the gelpicture indicate fold change of MMP-2 activity.62Figure 3.8 EIF5A2 regulates melanoma cell invasion.(A) A375 melanoma cells were transfected with either siC or siEIF5A2. Protein extracts were prepared 72 h aftertransfection and analyzed for the expression of EIF5A2. Boyden chamber assay was performed 48 h aftertransfection. (B) A375 melanoma cells were transfected with vector control or EIF5A2 overexpression plasmid for48 h. Protein extracts were prepared and analyzed for the expression of EIF5A2 by Western blot analysis. Boydenchamber assay was performed 24 h after transfection. (C), (D), (E) and (F) MMLH, SK-mel-3, SK-mel-28 and SK-mel-31 melanoma cells, respectively, were transfected with vector or EIF5A2 for 48 h. Protein extracts wereprepared and analyzed for the expression of EIF5A2 by Western blot analysis. Boyden chamber assay wasperformed 24 h after transfection. *P<0.05, **P<0.01, Student’s t-test.63Figure 3.9 EIF5A2, a downstream target of p-Akt, regulates melanoma cell invasion.(A) MMRU cells were transfected with vector control or EIF5A2 overexpression plasmid, or treated with 2.5 µM ofeither API1 or DMSO for 48 h. Protein extracts were prepared and analyzed for the expression of EIF5A2, p-Aktand total Akt by Western blot analysis. (B) Boyden chamber assay was performed 24 h after transfection or API1treatment. *P<0.05, **P<0.01, Student’s t-test. (C) MMP-2 activity determined by zymography which wasperformed 48 h after transfection or API1 treatment. Recombinant MMP-2 was used as positive control. (D) Positivestaining of EIF5A2 directly correlates with strong expression of MMP-2 in human melanomas (n=369; P=0.0024, ²test). (E) MMRU cells were transfected with either siC plus empty vector or EIF5A2 overexpression plasmid or acombination of EIF5A2 overexpression plasmid and siMMP-2. Boyden chamber assay was performed 48 h aftertransfection. **P<0.01, Student’s t-test. (F) Positive staining of EIF5A2 directly correlates with strong expression ofp-Akt in human melanomas (n=123, P=0.026, ² test).64Figure 3.10 EIF5A2 is a downstream target of p-Akt, regulating melanoma cell invasion.SK-mel-93 melanoma cells were transfected with vector control or EIF5A2 overexpression plasmid, or treated withDMSO or 2.5 µM of API1 for 48 h. (A) Protein extracts were prepared and analyzed for the expression of EIF5A2,p-Akt and total Akt by Western blot analysis. (B) 24 h after transfection or API1 treatment, cells were subjected toBoyden chamber assay. 24 h later, invaded cells were quantified. *P < 0.05, **P < 0.01, Student’s t-test.653.2.6 EIF5A2 is a downstream target of PI3K/Akt in melanoma cell invasionWe and others have reported that phosphorylated Akt (p-Akt) plays an important role inmelanoma progression and invasion, and p-Akt levels inversely correlate with melanoma patientsurvival (Dai et al, 2005; Dhawan et al, 2002). Because the TMA for p-Akt study contained 123melanoma biopsies which were also included in the current study, we were able to combine thetwo data sets to analyze the correlation between p-Akt and cytoplasmic EIF5A2 expression. Ourresults showed that positive staining of cytoplasmic EIF5A2 directly correlated with strongexpression of p-Akt (In our previous study we grouped the samples into neg-mod and strong p-Akt staining and different levels of p-Akt were detected in nevi and melanoma biopsies). Thepercentage of strong p-Akt staining increased from 45% in negative cytoplasmic EIF5A2 groupto 70% in positive cytoplasmic EIF5A2 group (P=0.026, 2 test; Figure 3.9F). Based on theseresults we then examined if EIF5A2 regulates p-Akt expression or vice versa. Overexpression ofEIF5A2 in MMRU and SK-mel-93 cells resulted in no change in p-Akt expression (Figures 3.9Aand 3.10A); whereas treatment with a small molecule p-Akt inhibitor API-1, led to inhibition ofAkt phosphorylation and EIF5A2 expression, decreased melanoma cell invasion by 48% and43%, as well as a decrease in MMP-2 activity in MMRU and SK-mel-93 melanoma cell linescompared with vehicle (Figures 3.9 and 3.10). This suggests that EIF5A2 is a downstream targetof p-Akt in PI3K pathway. Since Integrine-linked kinase (ILK) has been shown to directlyphosphorylate Akt at Serine-473, we next examined whether ILK regulates EIF5A2 expressionand melanoma cell invasion. We found that a decrease in ILK expression in ILK knockdownMMRU cells led to a decrease in EIF5A2 and p-Akt expression  as well as a decrease inmelanoma cell invasion (Figures 3.11A and B), suggesting that EIF5A2 is a downstream targetof ILK as expected. Furthermore, qRT-PCR data showed that the EIF5A2 mRNA level was66significantly decreased by 61% in ILK knockdown MMRU cells compared with control (Figure3.11C), indicating that EIF5A2 expression is regulated by ILK at the mRNA level.PI3K is an upstream regulator of ILK and PI3K activity is negatively regulated by PTEN.In order to determine the relationship between EIF5A2 and PI3K, we overexpressed PTEN in thePTEN-null MMRU and A375 melanoma cell lines and found that EIF5A2 expression andmelanoma cell invasion were decreased in PTEN-overexpressing cells (Figures 3.11D-E and3.12), confirming that EIF5A2 is a downstream target of PI3K/ILK/Akt pathway. However,PTEN overexpression in MMRU cells did not change the protein expression level of ILKbecause PTEN has been shown to negatively regulate ILK activity (Persad et al, 2000) which isindicated by a decrease in p-Akt expression (Figure 3.11). Therefore, it is the activity of PI3Knot necessarily its expression level that regulates EIF5A2.67Figure 3.11 ILK knockdown or PTEN overexpression decrease EIF5A2 expression and invasion in MMRUmelanoma cells.(A) Control or shILK knockdown MMRU melanoma cells were subjected to Western blot analysis and analyzed forthe expression of EIF5A2, p-Akt, total Akt and ILK. (B) Boyden chamber assay was performed 24 h aftertransfection. *P<0.05, Student’s t-test. (C) Quantitative RT-PCR analysis of EIF5A2 from total RNA extraction incontrol and shILK knockdown MMRU cells. ***P<0.001, Student’s t-test. (D) MMRU melanoma cells weretransfected with either vector control or PTEN overexpression plasmid. Protein extracts were prepared 48 h aftertransfection and analyzed for expression of EIF5A2, p-Akt, total Akt, ILK and PTEN by Western blotting. (E)Boyden chamber assay was performed 24 h after transfection. **P<0.01, Student’s t-test.68Figure 3.12 PTEN overexpression decreases EIF5A2 expression and invasion in A375 melanoma cells.(A) A375 melanoma cells were transfected with either vector control or PTEN overexpression plasmid. Proteinextracts were prepared 48 h after transfection and analyzed for expression of EIF5A2 and PTEN by Westernblotting. (B) Boyden chamber assay was performed 24 h after transfection. **P<0.01, Student’s t-test.693.2.7 EIF5A2 may induce epithelial-mesenchymal transition (EMT)In order to investigate whether or not EIF5A2 induces EMT, Western blot analysis wasperformed. EIF5A2 overexpressing MMRU cells showed an upregulation of mesenchymalmarkers such as vimentin, fibronectin and α-SMA and a downregulation of E-cadherin epithelialmarker, compared to the vector control (Figure 3.13). These data suggest that EIF5A2 mayinduce EMT.Figure 3.13 EIF5A2 may promote melanoma cell invasion and metastasis by inducing EMT.MMRU melanoma cells were transfected with either vector control or EIF5A2 overexpression plasmid for 48 h.Protein extracts were prepared and analyzed for the expression of EIF5A2, vimentin, fibronectin, α-SMA and E-cadherin by Western blot analysis.703.3 DiscussionElevated EIF5A2 activity has been observed in different types of human cancers (Chen et al,2009; Tang et al, 2010; Xie et al, 2008; Yang et al, 2009; Zhu et al, 2011). In this study weinvestigated the potential oncogenic role of EIF5A2 in melanoma. Our results showed thatEIF5A2 mRNA and protein expression increased significantly in different melanoma cell linescompared to melanocytes (Figure 3.1) consistent with the results obtained by Tang et al whoshowed the upregulation of EIF5A2 expression in eight HCC cell lines compared toimmortalized liver cell lines (Tang et al, 2010). Cross- cancer alteration summery provided bycBioPortal also shows DNA copy number alterations as well some some somatic mutations forEIF5A2 across different cancer types (Cerami et al, 2012; Gao et al, 2013).In order to better understand the role of EIF5A2 activity in melanoma progression, weinvestigated cytoplasmic EIF5A2 expression in 459 cases of melanocytic lesions at differentstages using an antibody specific for EIF5A2. Our results demonstrated that cytoplasmic EIF5A2expression is significantly increased during melanoma progression (Figure 3.4). These stage-specific expression patterns suggest that increased cytoplasmic EIF5A2 may play a role in thetransformation from dysplastic nevi to primary melanoma as well as progression from primary tometastatic melanoma. Furthermore, cytoplasmic EIF5A2 expression significantly correlated withtumour thickness and ulceration (Table 3.1) and inversely correlated with overall and disease-specific 5-year survival of all and primary, especially low-risk (≤2.0mm) melanoma patients(Figures 3.5 and 3.6). Interestingly, in the low-risk melanoma patients, all deaths within 5 yearsoccurred only in the positive cytoplasmic EIF5A2 staining group that might be an indication ofthe clinical importance of cytoplasmic EIF5A2 for this group of patients. Multivariate Cox71regression analysis further indicated that positive cytoplasmic EIF5A2 expression was anindependent prognostic marker for melanoma.To our knowledge, this is the first study to demonstrate that EIF5A2 plays an importantrole in melanoma progression and patient survival. A few studies supported the oncogenic role ofEIF5A2 in other types of cancers. It was previously shown that overexpression of EIF5A2 maybe important in the acquisition of a metastatic phenotype of colorectal carcinoma (Xie et al,2008; Zhu et al, 2011). Similarly, in ovarian cancer, EIF5A2 protein expression was positivelycorrelated with an ascending clinical stage of the tumour (Guan et al, 2004). Additionally,Marchet et al found that upregulated expression of EIF5A2 mRNA was associated with a higherrisk of lymph node metastasis of gastric carcinomas (Marchet et al, 2007). Furthermore,overexpression of EIF5A2 in urothelial carcinomas is associated with acquisition of a poorprognostic phenotype, suggesting that the expression of EIF5A2, as detected byimmunohistochemistry, is a prognostic marker for patient survival of urothelial carcinoma (Chenet al, 2009). All these reports support the notion that increased expression of EIF5A2 may beinvolved in the invasive and metastatic processes, leading to poor patient survival of differenttypes of human cancers.Our data showed that EIF5A2 regulates melanoma cell invasion in vitro (Figures 3.7B,3.9B, 3.8 and 3.10B). We also demonstrated that EIF5A2 overexpression in melanoma cellsresults in an upregulation of mesenchymal markers such as vimentin, fibronectin and α-SMA anda downregulation of E-cadherin epithelial marker (Figure 3.13) which suggests that EIF5A2 maypromote invasion and metastasis by inducing EMT. Similarly, others demonstrated that EIF5A2plays an important role in HCC and CRC invasion and metastasis by inducing EMT,72characterized by downregulation of E-cadherin and upregulation of fibronectin, N-cadherin, α-SMA and vimentin (Tang et al, 2010; Zhu et al, 2011).Moreover, we found that EIF5A2 enhances the gelatinolytic activity of MMP-2 (Figures3.7C and 3.9C). Similar to our results, in HCC MMP-2 expression and activity were indicated tobe reduced in response to ablation of endogenous EIF5A2 (Wang et al, 2014). The stimulatoryeffect of EIF5A2 on MMP-2 activity may at least partially contribute to the increase ofmelanoma cell invasion (Figure 3.9E), and this is consistent with our TMA data showing thepositive correlation between cytoplasmic EIF5A2 expression and melanoma tumour thickness.These data may also explain our observation that increased cytoplasmic EIF5A2 expressionsignificantly correlates with a poorer 5-year survival of primary melanoma patients. However,one of the weaknesses of this study is the lack of in vivo data using pre-clinical mouse models inorder to support the in vitro data on invasion and metastasis.Our tissue microarray data showed a direct correlation between positive staining ofcytoplasmic EIF5A2 and strong expression of p-Akt (Figure 3.9F). Furthermore, the p-Aktinhibitor API1 downregulated EIF5A2 protein expression and melanoma cell invasion (Figures3.9 and 3.10), suggesting that EIF5A2 may be a downstream effector of p-Akt which plays a rolein melanoma cell invasion. Additionally, we previously found a significant correlation betweenstrong p-Akt expression and tumour invasion and a poorer 5-year melanoma patient survival(Dai et al, 2005) which is consistent with our present data for EIF5A2. The consistency betweenthe two sets of TMA data further supports the possibility of EIF5A2 and p-Akt being in the samepathway.ILK and PTEN are well-known upstream regulators of p-Akt (Dai et al, 2005). We havepreviously shown that the expression of ILK was correlated with tumour invasion in primary73melanomas (Dai et al, 2003). The expression of PTEN, a negative regulator of Akt pathway, wasreduced in melanomas, and loss of PTEN was shown to increase melanoma tumour growth invivo (Stahl et al, 2003; Tsao et al, 2003). These results are consistent with our present findingthat both ILK knockdown and PTEN overexpression lead to a decrease in EIF5A2 expression(Figures 3.11A, D and 3.12), further supporting that EIF5A2 is a downstream target of PI3K/Aktpathway.There are many factors that regulate MMP-2 expression and activity. For example, NF-κB is a transcription factor downstream of p-Akt (Dhawan et al, 2002) regulating cancerprogression and metastasis (Sun & Zhang, 2007). Activation of NF-κB induces membrane typeproteases (MT1-MMP), the activator of pro-MMP-2 which proteolytically cleaves to generatefunctionally active MMP-2 (Sato et al, 1994). Activating enhancer binding protein 2 alpha (AP-2α) is another transcription factor that deregulates several factors including MMP-2 (Qin et al,1999) and its loss predicts poor survival of melanoma patients (Karjalainen et al, 1998). S-Phasekinase-associated protein 2 (Skp2) is an oncogene that is required for the degradation of tumoursuppressor p27. Overexpression of Skp2 has been shown to upregulate the expression andactivity of MMP-2 in lung cancer cells (Hung et al, 2010). The above molecules could bepossible mediators to regulate EIF5A2-induced gelatinolytic activity of MMP-2. We also showedthat EIF5A2 mRNA expression decreased in ILK knockdown cells compared to the control(Figure 3.11C). The exact mechanism of how PI3K/p-Akt regulates the expression of EIF5A2 isnot yet clear and requires further investigation.Taken together, we showed that cytoplasmic EIF5A2 expression is significantlyincreased with melanoma progression and the increased cytoplasmic EIF5A2 expressioncorrelates with a worse 5-year survival of melanoma patients in particular those with tumour74thickness less than 2.0mm. Furthermore, our data for the first time demonstrate that EIF5A2 is adownstream target of PI3K/Akt which plays an important role in regulating melanoma cellinvasion, suggesting that targeting EIF5A2 may provide a novel therapeutic approach formelanoma.75Chapter 4: Prognostic significance of the expression of nuclear EIF5A2 inhuman melanoma4.1 Background and rationaleEIF5A has been best known for its cytoplasmic role in translation regulation (Patel et al, 2009;Saini et al, 2009) and, in mammals, is encoded by two highly related genes (EIF5A1 andEIF5A2). However, a number of studies have suggested that EIF5A1 may also have activity inthe nucleus, particularly in mammals (Hauber et al, 2005; Kruse et al, 2000; Lipowsky et al,2000). For instance, EIF5A1 has been shown to take part in the nucleocytoplasmic transport ofincompletely spliced and unspliced HIV-1 mRNAs (Hauber et al, 2005) that are translocatedacross the nuclear envelope via Chromosomal maintenance 1 (CRM1) which is a member of theimportin β family of transport receptors. The main role for CRM1 is to facilitate the translocationof rRNA, and ribosomal subunits across the nuclear envelope (Hutten & Kehlenbach, 2007).EIF5A1 has also been shown to be able to interact and colocalize with the transport receptorexportin4 which is another member of the importin β family (Lipowsky et al, 2000). It has alsobeen suggested that EIF5A1 can recognize Nos2 mRNA in the nucleus and facilitate theirtransport to the cytoplasm in a CRM1 dependent- manner (Maier et al, 2010a). Furthermore, inS. Cerevisiae, EIF5A1 seems to be involved in mRNA degradation/turnover that is a processhighly related to nucleocytoplasmic transport of mRNA (Schrader et al, 2006; Zuk & Jacobson,1998). In summary, these data demonstrate a nuclear activity for EIF5A1 that is probably relatedto cellular mRNA metabolism.EIF5A2, a phylogenetically conserved vertebrate variant of EIF5A1, was first reported tobe highly expressed in testis and colorectal adenocarcinoma and at moderate levels in the brain76(Jenkins et al, 2001). EIF5A2 shares 83% amino acid identity with EIF5A1 (Clement et al, 2003)and has been associated with different oncogenic roles such as invasion, metastasis or poorprognosis in a variety of cancers such as ovarian cancer (Guan et al, 2004), gastricadenocarcinomas (Marchet et al, 2007), colorectal cancer (Xie et al, 2008; Zhu et al, 2012a),hapatocellular carcinomas (HCC) (Tang et al, 2010), ovarian cancer (Yang et al, 2009), non–small cell lung cancer (He et al, 2011), and bladder cancer (Chen et al, 2009; Wei et al, 2014).However, none of the above studies investigated the nuclear expression or activity of EIF5A2 inthe different cancers except Zender et al (Zender et al, 2008), who addressed the importance ofnuclear EIF5A2 in HCC.We previously reported an increase in cytoplasmic expression of EIF5A2 in melanomaand its role in melanoma progression and patient survival (Khosravi et al, 2014). In the presentstudy using immunohistochemistry and TMA we investigated the status of nuclear EIF5A2expression in melanoma. The results showed that nuclear EIF5A2 is an independent prognosticmarker whose expression significantly increased during melanoma progression, and upregulationof nuclear EIF5A2 was determined to be correlated with a significantly worse 5-year survival ofall and primary melanoma patients. Moreover, simultaneous nuclear and cytoplasmic EIF5A2expression as well as concurrent nuclear EIF5A2 and MMP-2 expression were shown to beassociated with a worse 5-year patient survival.774.2 Results4.2.1 Correlation between nuclear EIF5A2 expression and clinicopathologiccharacteristicsA total of 713 melanoma patients were enrolled for TMA construction. Due to loss of biopsycores, insufficient tumour cells present in the cores or loss of follow up we evaluated 382melanoma (242 cases in primary melanoma and 140 cases in metastatic melanoma), and 77 casesof nevi (28 common acquired nevi and 49 dysplastic nevi) for nuclear EIF5A2 staining. Theclinical features of melanoma patients are listed in Table 4.1.Thickness is one of these features which is a very important prognostic marker for primarymelanoma patients and our analysis showed that in primary melanoma patients, positive nuclearEIF5A2 expression was found in 58% of melanoma patients with tumour thickness >2mm,compared to 45% of melanoma patients with tumour thickness ≤2mm (P=0.036, ² test; Table4.1 and Figure 4.1a), suggesting that in primary melanoma nuclear EIF5A2 expression might beinduced during transition from thin to thick melanoma. In all melanoma patients, expression ofnuclear EIF5A2 was detected in 85% of advanced stage melanomas (AJCC stages III and IV)compared to only 51% of early stage melanomas (AJCC stages I and II) (P<0.001, ² test; Table4.1 and Figure 4.1b). The correlation between positive nuclear EIF5A2 expression andmelanoma thickness and AJCC stages could be an indication for the involvement of nuclearEIF5A2 expression in melanoma cell invasion.784.2.2 Nuclear EIF5A2 expression is increased with melanoma progressionTo study the changes in the expression of nuclear EIF5A2 with melanoma progression,immunohistochemical staining was performed in TMA slides and samples were categorized intonegative and positive EIF5A2 staining groups (Figure 4.1c-f). Positive nuclear EIF5A2 stainingwas observed in 29% of common acquired nevi, 31% of dysplastic nevi, 51% of primarymelanomas and 85% of metastatic melanomas. Consequently, nuclear EIF5A2 expression wasfound to be significantly higher in primary melanomas compared to dysplastic nevi and normalnevi (P=0.010 and 0.026, respectively, ² test; Figure 4.1g), and in metastatic melanomascompared to primary melanomas, dysplastic nevi and common acquired nevi (P<0.001 for all, ²test; Figure 4.1g), suggesting the role of nuclear EIF5A2 in transformation form nevus tomalignant tumours and the development of melanoma metastasis. No difference in nuclearEIF5A2 expression was observed between common acquired nevi and dysplastic (P=0.852, ²test; Figure 4.1g).79Table 4.1 Nuclear EIF5A2 staining and clinicopathologic characteristics of 382 melanomasNuclear EIF5A2 stainingVariables Positive Negative Total χ2 Value P valueNo. (%) No. (%)All melanoma (n=382)Age, y≤60 122 (61.6) 76 (38.4) 198 0.533 0.465>60 120 (65.2) 64 (34.8) 184SexMale 146 (63.8) 83 (36.2) 229 0.04 0.841Female 96 (62.7) 57 (37.3) 153AJCC stageI 56 (44.1) 71 (55.9) 127 44.611 <0.001aII 67 (58.3) 48 (41.7) 115III 51 (94.4) 3 (5.6) 54IV 68 (79.1) 18 (20.9) 86Primary melanoma (n=242)Age, y≤61 58 (47.2) 65 (52.8) 123 1.349 0.245>61 65 (54.6) 54 (45.4) 119SexMale 63 (47.4) 70 (52.6) 133 1.413 0.235Female 60 (55.0) 49 (45.0) 109Tumor thickness (mm)≤2 60 (44.8) 74 (55.2) 134 4.398 0.036>2 63 (58.3) 45 (41.7) 108UlcerationAbsent 96 (49.5) 98 (50.5) 194 0.705 0.401Present 27 (56.2) 21 (43.8) 48SubtypeLentigo maligna 21 (56.8) 16 (43.2) 37 2.843 0.416Superficial spreading 39 (43.8) 50 (56.2) 89Nodular 22 (53.7) 19 (46.3) 41Unspecified 41 (54.7) 34 (45.3) 75SitebSun-protected 95 (50.5) 93 (49.5) 188 0.029 0.865Sun-exposed 28 (51.9) 26 (48.1) 54Metastatic melanoma (n=140)Age, y≤59 61 (83.6) 12 (16.4) 73 0.248 0.618>59 58 (86.6) 9 (13.4) 67SexMale 83 (86.5) 13 (13.5) 96 0.510 0.475Female 36 (81.8) 8 (18.2) 44AJCC indicates American Joint Committee on Cancer.a Comparison between AJCC stage I to II and III to IV, χ2 test.b Sun-protected sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.80Figure 4.1 Correlation between nuclear EIF5A2 expression, thickness, AJCC stages and different stages ofmelanoma progression.(a) Nuclear EIF5A2 expression was significantly higher in melanoma patients with tumour thickness >2mm,compared with melanoma patients with tumour thickness ≤2mm (P=0.036, ² test). (b) Nuclear EIF5A2 expressionwas significantly higher in advanced stage melanomas (AJCC stages III and IV) compared with early stagemelanomas (AJCC stages I and II) (P<0.001, ² test). (c-f) Representative images of nuclear EIF5A2immunohistochemical staining in melanocytic lesions at (c-d) X 100 and (e-f) X 400 magnification. (c, e) Negativenuclear EIF5A2 staining. (d, f) Positive nuclear EIF5A2 staining. (g) Nuclear EIF5A2 expression was increased inMM compared with CAN, DN, and PM (P<0.001 for all, ² test). Nuclear EIF5A2 expression was also increased inPM compared with CAN and DN (P=0.010 and 0.026, respectively, ² test). CAN, common acquired nevi; DN,dysplastic nevi; PM, primary melanoma; MM, metastatic melanoma. Nuc., nuclear814.2.3 Nuclear EIF5A2 expression positively correlates with poor patient survivalWe evaluated the correlation between nuclear EIF5A2 expression and 5-year survival of primaryand metastatic melanoma patients by constructing Kaplan–Meier survival curves. Both overalland disease-specific 5-year survival were worse for all (P<0.001 for both, log rank test; Figure4.2a-b) and primary (P=0.014 and P=0.015, respectively, log rank test; Figure 4.2c-d) melanomapatients with positive staining for nuclear EIF5A2 compared to the patients with negativestaining for nuclear EIF5A2. The results from the Kaplan-Meier survival analysis were furthersupported by univariate Cox proportional hazard regression analysis by indicating that nuclearEIF5A2 expression was a significant prognostic factor for the overall and disease-specific 5-yearsurvival of all melanoma patients (HR, 2.26, 95% CI, 1.57-3.27, P<0.001 and HR, 2.27, 95% CI,1.56-3.31, P<0.001, respectively; Table 4.2) and primary melanoma patients (HR, 1.95, 95% CI,1.14-3.36, P=0.015 and HR, 2.00, 95% CI, 1.13-3.53, P=0.017, respectively; Table 4.2).82Figure 4.2 Kaplan-Meier analyses for the correlation between nuclear EIF5A2 expression and 5-year survivalof melanoma patients.(a-b) Nuclear EIF5A2 expression is associated with a worse overall and disease-specific 5-year survival in allmelanoma patients (P<0.001, log rank test). (c-d) Nuclear EIF5A2 expression is associated with a worse overall anddisease-specific 5-year survival in primary melanoma patients (P=0.014 and 0.015, respectively, log rank test).83Table 4.2 Univariate Cox regression analysis on 5-year overall and disease-specific survival of a total of 382melanoma patients and 242 primary melanoma patientsAbbreviations: CI, confidence interval; HR, hazard ratio.ˡ Log-rank test.² Sun-protected  sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.VariablesOverall Survival Disease-specific SurvivalPatients (%) Deaths Death rate (%) HR (95% CI) P ˡ Deaths Death rate (%) HR (95% CI) P ˡAll melanoma (n=382)Nuclear EIF5A2Negative expression 140 (36.6) 37 26.4 1.00 <0.001 35 25.0 1.00 <0.001Positive expression 242 (63.4) 123 50.8 2.26 (1.57-3.27) 117 48.3 2.27 (1.56-3.31)Age (years)60 198 (51.8) 79 39.9 1.00 0.426 77 38.9 1.00 0.648>60 184 (48.2) 81 44.0 1.13 (0.83-1.55) 75 40.1 1.08 (0.78-1.48)SexMale 229 (60.0) 99 43.2 1.00 0.616 93 40.1 1.00 0.755Female 153 (40.0) 61 39.9 0.92 (0.67-1.27) 59 38.6 0.95 (0.68-1.32)AJCCI-II 242 (63.4) 58 24.0 1.00 <0.001 53 22.0 1.00 <0.001III-IV 140 (36.6) 102 72.9 5.07 (3.66-7.03) 99 70.7 5.37 (3.84-7.53)Primary melanoma (n=242)Nuclear EIF5A2Negative expression 119 (49.2) 20 16.8 1.00 0.015 18 15.1 1.00 0.017Positive expression 123 (50.8) 38 30.9 1.95 (1.14-3.36) 35 28.5 2.00 (1.13-3.53)Age (years)61 123 (50.8) 19 15.4 1.00 0.002 19 15.4 1.00 0.010>61 119 (49.2) 39 32.8 2.41 (1.39-4.17) 34 28.6 2.10 (1.20-3.69)SexMale 133 (55.0) 31 23.3 1.00 0.749 28 21.1 1.00 0.733Female 109 (45.0) 27 24.8 1.07 (0.64-1.80) 25 22.9 1.10 (0.64-1.88)UlcerationAbsent 194 (80.2) 30 15.5 1.00 <0.001 26 13.4 1.00 <0.001Present 48 (19.8) 28 58.3 5.38 (3.20-9.03) 27 56.3 6.00 (3.49-10.31)Thickness (mm)2 134 (55.4) 11 8.21 1.00 <0.001 9 6.72 1.00 <0.001>2 108 (44.6) 47 43.5 6.71 (3.48-12.9) 44 40.1 7.69 (3.75-15.78)Location ²Sun-protected 188 (77.7) 45 23.9 1.00 0.892 43 22.9 1.00 0.460Sun-exposed 54 (22.3) 13 24.1 0.96 (0.52-1.78) 10 18.5 0.77 (0.39-1.54)SubtypeOthers 153 (63.2) 42 27.5 1.00 0.129 38 24.8 1.00 0.177Superficial spreading 89 (36.8) 16 18.0 1.56 (0.88-2.78) 15 16.9 1.51 (0.83-2.74)844.2.4 Nuclear EIF5A2 is an independent prognostic marker for melanoma patientsResults from multivariate Cox regression analysis revealed that nuclear EIF5A2 was an adverseindependent prognostic marker for the overall and disease-specific 5-year survival of allmelanoma patients (HR, 1.78, 95% CI, 1.22-2.60, P=0.003 and HR, 1.77, 95% CI, 1.20-2.62,P=0.004, respectively; Table 4.3) and primary melanoma patients (HR, 1.78, 95% CI, 1.02-3.12,P=0.043 and HR, 1.90, 95% CI, 1.06-3.43, P=0.032, respectively; Table 4.3). For all melanomapatients, sex, age, AJCC and EIF5A2 expression were included and for primary melanomapatients, sex, age, tumour thickness, ulceration status, tumour location, histological subtype andEIF5A2 expression were included in the analysis.85Table 4.3 Multivariate Cox regression analysis on 5-year overall and disease-specific survival of melanoma patientsVariables a Overall Survival Disease-specific Survival b SE HR 95% CI P  b SE HR 95% CI PAll melanoma (n=382)Nuclear EIF5A2 0.576 0.194 1.778 1.22-2.60 0.003 0.575 0.199 1.777 1.20-2.62 0.004Age 0.099 0.159 1.104 0.81-1.51 0.534 0.050 0.163 1.052 0.76-1.45 0.758Sex 0.168 0.169 1.183 0.85-1.65 0.319 0.206 0.173 1.229 0.88-1.72 0.232AJCC 1.174 0.173 3.236 2.31-4.54 <0.001 1.215 0.177 3.372 2.38-4.77 <0.001Primary Melanoma (n=242)Nuclear EIF5A2 0.578 0.286 1.783 1.02-3.12 0.043 0.643 0.300 1.902 1.06-3.43 0.032Age 0.269 0.298 1.309 0.73-2.35 0.367 0.080 0.307 1.083 0.59-1.98 0.795Sex -0.107 0.276 0.899 0.52-1.55 0.699 -0.110 0.288 0.896 0.51-1.58 0.704Ulceration 1.134 0.297 3.109 1.74-5.57 <0.001 1.269 0.310 3.559 1.94-6.54 <0.001Thickness 1.410 0.362 4.095 2.01-8.33 <0.001 1.563 0.394 4.772 2.20-10.34 <0.001Location -0.335 0.322 0.715 0.38-1.34 0.298 -0.556 0.358 0.573 0.28-1.16 0.120Subtype 0.052 0.301 1.054 0.58-1.90 0.862 0.012 0.313 1.012 0.55-1.87 0.969a Coding of variables: Nuclear EIF5A2 was coded as 1 (negative) and 2 (positive). Age was coded as 1 (60 years) and 2 (>60 years) for all melanomaand 1 (61 years) and 2 (>61 years) for primary melanoma. Sex was coded as 1 (male) and 2 (female). Ulceration was coded as 1 (absent) and 2(present). Thickness was coded as 1 (≤2.00 mm) and 2 (>2.00 mm). Location was coded as 1 (sun protected) and 2 (sun exposed). Subtype was codedas 1 (superficial spreading) and 2 (others).b: regression coefficient.NOTE: SE, standard error of ; HR, hazard ratio; CI, confidence interval; PM, primary melanoma; MM, metastatic melanoma.864.2.5 Concurrent cytoplasmic and nuclear EIF5A2 expression is correlated with a worse5-Year survival for all and primary melanoma patientsWe previously investigated the expression of cytoplasmic EIF5A2 in melanoma using tissuemicroarray and found that cytoplasmic EIF5A2 expression is a prognostic marker for melanomapatients (Khosravi et al, 2014). In the present study we analyzed the correlation betweencytoplasmic and nuclear expression of EIF5A2 using the 382 melanoma cases and our resultsshowed a direct correlation between the positive staining of cytoplasmic and nuclear EIF5A2(P<0.001, 2 test; Figure 4.3a).To further examine this correlation and its effect on patient survival, we divided thesamples into three groups based on their staining: (1) negative cytoplasmic and nuclear EIF5A2;(2) negative cytoplasmic and positive nuclear EIF5A2 or positive cytoplasmic and negativenuclear EIF5A2; (3) positive cytoplasmic and nuclear EIF5A2. Based on the results fromKaplan–Meier survival analyses, overall and disease-specific 5-year survival for both all(P<0.001, log rank test; Figure 4.3b-c) and primary (P=0.002, log rank test; Figure 4.3d-e)melanoma patients, was worst for category 3 patients, best for category 1 patients andintermediate for category 2 patients. Furthermore, Multivariate Cox regression analysis showedthat the simultaneous positive expression of cytoplasmic and nuclear EIF5A2 (category 3) wasan independent prognostic factor for overall and disease-specific 5-year survival of all (HR, 1.87,95% CI, 1.31-2.68, P=0.001 and HR, 1.85, 95% CI, 1.28-2.67, P=0.001, respectively; Table 4.4)and primary (HR, 2.01, 95% CI, 1.15-3.51, P=0.014 and HR, 2.11, 95% CI, 1.17-3.78, P=0.013,respectively; Table 4.4) melanoma patients.87Figure 4.3 Simultaneous nuclear and cytoplasmic EIF5A2 expression correlates with a poorer 5-yearsurvival.(a) Nuclear EIF5A2 expression directly correlates with cytoplasmic EIF5A2 expression in human melanomas(n=382; P<0.001, ² test). (b-c) Simultaneous negative expression of nuclear and cytoplasmic EIF5A2 (category 1)was significantly associated with a better overall and disease-specific 5-year survival outcome compared withnegative nuclear EIF5A2 expression and positive cytoplasmic EIF5A2 expression, or positive nuclear EIF5A2expression and negative cytoplasmic EIF5A2 expression (category 2), or positive nuclear and cytoplasmic EIF5A2expression (category 3) in all melanoma patients (P<0.001 for both overall and disease-specific 5-year survival, logrank test) and in (d-e) primary melanoma patients (P=0.002 for both overall and disease-specific 5-year survival, logrank test). Neg, negative; Pos, positive.88Table 4.4 Multivariate Cox regression analysis on 5-year overall and disease-specific survival of melanoma patientsVariables a Overall Survival Disease-specific Survival b SE HR 95% CI P  b SE HR 95% CI PAll Melanoma (n=382)Nuc. and Cyt. EIF5A2 0.626 0.184 1.871 1.31-2.68 0.001 0.614 0.188 1.848 1.28-2.67 0.001Age 0.112 0.159 1.118 0.82-1.53 0.483 0.063 0.163 1.065 0.77-1.47 0.698Sex 0.195 0.170 1.216 0.87-1.70 0.251 0.234 0.174 1.264 0.90-1.78 0.178AJCC 1.169 0.174 3.220 2.29-4.52 <0.001 1.214 0.178 3.365 2.38-4.77 <0.001Primary Melanoma (n=242)Nuc. and Cyt. EIF5A2 0.699 0.284 2.012 1.15-3.51 0.014 0.745 0.299 2.107 1.17-3.78 0.013Age 0.328 0.299 1.388 0.77-2.50 0.273 0.132 0.309 1.141 0.62-2.09 0.670Sex -0.128 0.277 0.880 0.51-1.51 0.644 -0.125 0.289 0.882 0.50-1.55 0.665Ulceration 1.107 0.296 3.026 1.69-5.41 <0.001 1.245 0.310 3.472 1.89-6.37 <0.001Thickness 1.355 0.363 3.876 1.90-7.89 <0.001 1.505 0.395 4.506 2.08-9.77 <0.001Location -0.352 0.320 0.704 0.38-1.32 0.272 -0.573 0.356 0.564 0.28-1.13 0.108Subtype 0.054 0.301 1.056 0.59-1.90 0.857 0.015 0.312 1.015 0.55-1.87 0.962a Coding of variables: EIF5A2 was coded as 1 (categories 1 and2) and 2 (category 3). Age was coded as 1 (60 years) and 2 (>60 years) for allmelanoma and 1 (61 years) and 2 (>61 years) for primary melanoma. Sex was coded as 1 (male) and 2 (female). Ulceration was coded as 1 (absent)and 2 (present). Thickness was coded as 1 (≤2.00 mm) and 2 (>2.00 mm). Location was coded as 1 (sun protected) and 2 (sun exposed). Subtype wascoded as 1 (superficial spreading) and 2 (others).b: regression coefficient.NOTE: SE, standard error of ; HR, hazard ratio; CI, confidence interval; Nuc., nuclear; Cyt., cytoplasmic.894.2.6 Simultaneous expression of nuclear EIF5A2 and MMP-2 is associated with poor 5-year survival for all and primary melanoma patientsResults from the TMA study suggested the involvement of nuclear EIF5A2 in developingmelanoma invasion, metastasis and poor patient survival. Since cell invasion is one of thehallmarks of cancer that can lead to increased metastasis and poor patient survival, we decided toinvestigate the correlation between the expression of nuclear EIF5A2 and MMP-2 which is oneof the important factors for promoting cancer cell invasion. Our previous TMA study indicatedMMP-2 expression as a prognostic marker for melanoma (Rotte et al, 2012). As a result, in thepresent study we examined the correlation between the expression of nuclear EIF5A2 and MMP-2 using the 369 melanoma cases that were in common between the two TMA studies. The resultshowed that positive staining of nuclear EIF5A2 directly correlated with strong MMP-2expression (P=0.015, 2 test; Figure 4.4a). To more extensively study the association betweenMMP-2 and nuclear EIF5A2 expression and their effects on patient survival, we categorized thesamples into three groups based on their staining: (1) negative nuclear EIF5A2 expression andnegative-moderate MMP-2 expression; (2) either negative nuclear EIF5A2 expression and strongMMP-2 expression or positive nuclear EIF5A2 expression and negative-moderate MMP-2expression; and (3) positive EIF5A2 expression and strong MMP-2 expression. Kaplan–Meiersurvival analyses showed that overall and disease-specific 5-year survival outcomes for allmelanoma (P<0.001, log rank test; Figure 4.4b-c) and primary melanoma (P<0.001 andP=0.001, respectively log rank test; Figure 4.4d-e) patients, were most favourable for category 1and least favourable for category 3, and survival rate for patients in category 2 was in betweenthe other two categories.90Figure 4.4 Simultaneous nuclear EIF5A2 expression and strong MMP-2 expression correlate with a poorer 5-year survival.(a) Nuclear EIF5A2 expression directly correlates with strong MMP-2 expression in human melanomas (n=369;P=0.015, ² test). (b-c) Simultaneous negative nuclear EIF5A2 expression and negative-moderate MMP-2expression (category 1) was significantly associated with a better overall and disease-specific 5-year survivaloutcome compared with negative nuclear EIF5A2 and strong MMP-2 expression, or positive nuclear EIF5A2 andnegative-moderate MMP-2 expression (category 2), or positive nuclear EIF5A2 and strong MMP-2 expression(category 3) in all melanoma patients (P<0.001 for both overall and disease-specific 5-year survival, log rank test)and in (d-e) primary melanoma patients (P<0.001, and P=0.001, respectively, log rank test).914.3 DiscussionRecently we showed an increase in cytoplasmic EIF5A2 expression with melanoma progression(Khosravi et al, 2014) which prompted us to additionally investigate the nuclear expression ofEIF5A2 using our TMA data consisting of 459 melanocytic lesions. We found that, like incytoplasmic EIF5A2 expression, increase in nuclear EIF5A2 expression was significantlyassociated with melanoma progression, thickness and AJCC stages. Similarly, nuclear EIF5A2expression was significantly correlated with a worse overall and disease-specific 5-year survivalof all and primary melanoma patients that could be a result of direct correlation between nuclearEIF5A2 expression and melanoma thickness and AJCC stages which is also consistent with thenotion that EIF5A2 is an oncogene in different cancers. Interestingly, the nuclear expression andalso the combination of both nuclear and cytoplasmic expression were found to be anindependent prognostic marker for the overall and disease specific 5-year survival of all andprimary melanoma patients.Furthermore, we observed a direct association between nuclear EIF5A2 expression andstrong expression of MMP-2 which might be a reason for nuclear EIF5A2 expression beingdirectly correlated with melanoma thickness and a worse 5-year survival of melanoma patients.Moreover, we found that simultaneous negative-moderate MMP-2 expression and loss of nuclearEIF5A2 expression (category 3) was correlated with a better 5-year survival compared to eitherstrong MMP-2 expression and loss of nuclear EIF5A2 expression or positive expression ofnuclear EIF5A2 and negative-moderate expression of MMP-2 (category 2). This could possiblybe a rational for dual therapeutic targeting of nuclear EIF5A2 and MMP-2 in melanoma patientsthat requires more research in the future. The importance of MMP-2 in melanoma has beenshown by others as well. For example, MT1-MMP was demonstrated to increase melanoma cell92invasion and motility by activating its target MMP-2 (Shaverdashvili et al, 2014). Another studypointed out that expression of activated MMP-2 in a melanoma xenograft model correlates withincreased malignancy, highlighting the role of MMP-2 in melanoma invasion and metastasis(Hofmann et al, 1999).We also compared the nuclear and cytoplasmic EIF5A2 expression in the same cases ofmelanocytic lesions and found a significant correlation between positive staining of nuclear andcytoplasmic EIF5A2. This correlation; however, is not perfect which could be an indication ofthe differential regulation of nuclear and cytoplasmic EIF5A2 expression in melanoma.Interestingly, our results also showed that simultaneous expression of both forms of EIF5A2(category 3) was associated with the worst survival outcome as compared to expression of onlyone form (category 2) which was associated with an intermediate survival outcome, suggestingthat oncogenic properties of nuclear and cytoplasmic EIF5A2 may at least partly differ from eachother in melanoma. Therefore, concurrent expression of both forms (category 3) could havesynergistic or additive effects on melanoma progression, metastasis and patient survival. Ourprevious in vitro results indicated that EIF5A2 promoted melanoma cell invasion partly viaincreasing the activity of MMP-2 (Khosravi et al, 2014). However, further investigation isrequired to determine which form of EIF5A2 and to what extent is responsible for this function.To our knowledge we are the first to report the nuclear expression of EIF5A2 and itsimportance in melanoma. However, the presence of EIF5A2 in nuclei of HCC cells has beenshown before (Zender et al, 2008). Exportin 4 (Xpo4) is a tumour suppressor that belongs to theimportin-β family of nuclear transporters and EIF5A is known to be a substrate of Xpo4(Kurisaki et al, 2006; Lipowsky et al, 2000). Knockdown of Xpo4 in murine hepatoma cells ledto nuclear accumulation of EIF5A1 and EIF5A2 which significantly increased the in vitro93proliferation of murine liver progenitor cells (Zender et al, 2008). In a human HCC cell line,XPO4 inactivation was shown to contribute to tumour maintenance. In the same study, EIF5A2was indicated to be needed for efficient proliferation in cells lacking XPO4, suggesting theimportance of nuclear accumulation of EIF5A2 in mediating the oncogenic effects associatedwith XPO4 loss (Zender et al, 2008). Similar phenomenon is seen when an increase or decreasein the nuclear accumulation of β-catenin and (Forkhead box O) FOXO influences tumourigenesisas a result of deregulation of the WNT and AKT signaling pathways, respectively (Kau et al,2004). Nuclear expression of EIF5A2 has also been shown in the human bladder carcinoma cellline 5637 (Wei et al, 2014).In conclusion, in this study we examined the expression profile of nuclear EIF5A2 inmelanoma and found that nuclear EIF5A2 expression is increased during melanoma progressionand is correlated with a worse 5-year survival of all and primary melanoma patients. We alsodiscovered that nuclear EIF5A2 was an independent prognostic factor for the 5-year survival ofall and primary melanoma patients, suggesting that nuclear EIF5A2 could be a potential targetcandidate for melanoma therapy. Simultaneous expression of both cytoplasmic and nuclearEIF5A2 was associated with the worst survival outcome as well. Additional investigation isrequired to further determine the biological functions of nuclear EIF5A2 and its role intumourigenesis in melanoma.94Chapter 5: eIF4E is an adverse prognostic marker of melanoma patientsurvival by increasing melanoma cell invasion5.1 Background and rationaleIn addition to EIF5A2, eIF4E is another translation factor that is considered a potent oncogeneand has elevated expression in many human cancers, including carcinomas of the breast,prostate, lung, head and neck, bladder, cervix, nasopharynx, as well as in many leukemias andlymphomas (De Benedetti & Graff, 2004; Hariri et al, 2013; Wu et al, 2013). eIF4E elevationleads to increased proliferation, evasion of apoptosis, oncogenic transformation, tumour invasionand metastases (Borden & Culjkovic-Kraljacic, 2010; De Benedetti & Rhoads, 1990; Graff et al,1995; Lazaris-Karatzas et al, 1990; Polunovsky et al, 1996).In tumours, eIF4E concentrations are elevated by the activation of mTOR pathway.mTOR phosphorylates 4EBP1 which leads to disengagement of 4EBP1 from eIF4E anddisinhibition of translation by increasing the availability of eIF4E (Sun et al, 2005). eIF4Edepletion in cancer cells, on the other hand, leads to cell cycle arrest and decreasedtumourigenicity (Assouline et al, 2009; Graff et al, 2007; Kentsis et al, 2004; Oridate et al,2005). In melanoma, immunohistochemistry using the tissue array research program (TARP)TMA revealed that eIF4E was elevated in 59% of the 23 melanoma tissue cores that were usedfor the analysis (Yang et al, 2007).In this study we investigated the role of eIF4E in melanoma including its significance inpromoting melanoma cell invasion. In addition we examined the expression pattern of eIF4E inmelanoma using immunohistochemistry in a large set of melanocytic lesions.955.2 Results5.2.1 eIF4E expression correlates with melanoma tumour thickness and AJCC stagesMelanocytic lesions in a total of 713 patients were used to construct the TMA. Due to the loss ofbiopsy cores or insufficient tumour cells present in the cores, 402 melanoma (250 primary and152 metastatic melanoma cases), and 67 nevi cases (28 common acquired and 39 dysplastic nevi)could be evaluated for eIF4E staining (Figure 5.1). Table 5.1 includes a list of the clinicalfeatures of the melanoma patients. Since thickness is a well known prognostic marker forprimary melanoma patients, we analyzed eIF4E expression level in tumours with differentthicknesses. High eIF4E expression was found in 86.3% of thick melanomas (≥1mm) comparedto 75% of thin melanomas (˂1mm) (P=0.046, ² test; Table 5.1), indicating that within primarymelanoma eIF4E expression is likely induced during the transition from thin to thick melanoma.In addition, high eIF4E expression was significantly more common in advanced stagemelanomas (AJCC stages III and IV) compared to early-stage melanomas (AJCC stages I and II)(P=0.002, ² test; Table 5.1). The correlation of high eIF4E expression with tumour thicknessand AJCC stages suggests that eIF4E may play an important role in melanoma invasion.96Figure 5.1 Diagram showing patient inclusion and exclusion.97Table 5.1 eIF4E staining and clinicopathologic characteristics of 381 melanomaseIF4E stainingVariables Low High Total χ2 Value P valueNo. (%) No. (%)All melanoma (n=381)Age, y≤60 21 (10.8) 173 (89.2) 194 0.835 0.360>60 26 (13.9) 161 (86.1) 187SexMale 28 (12.3) 200 (87.7) 228 0.002 0.964Female 19 (12.4) 134 (87.6) 153AJCC stageI 21 (17.6) 98 (82.4) 119 9.621 0.002aII 18 (15.1) 101 (84.9) 119III 5 (8.8) 52 (91.2) 57IV 3 (3.5) 83 (96.5) 86Primary melanoma (n=238)Age, y≤62 18 (14.9) 103 (85.1) 121 0.410 0.521>62 21 (17.9) 96 (82.1) 117SexMale 23 (17.7) 107 (82.3) 130 0.356 0.551Female 16 (14.9) 92 (85.2) 108Tumor thickness (mm)˂1 14 (25.0) 42 (75.0) 56 3.97 0.046≥1 25 (13.7) 157 (86.3) 182UlcerationAbsent 32 (17.1) 155 (82.9) 187 0.335 0.563Present 7 (13.7) 44 (86.3) 51SubtypeLentigo maligna 12 (33.3) 24 (66.7) 36 20.406 ˂0.001Superficial spreading 6 (7.0) 80 (93.0) 86Nodular 3 (6.80) 41 (93.2) 44Others 18 (25.0) 54 (75.0) 72SitebSun-protected 28 (15.4) 154 (84.6) 182 0.567 0.451Sun-exposed 11 (19.6) 45 (80.4) 56Metastatic melanoma (n=143)Age, y≤59 5 (6.7) 70 (93.3) 75 0.343 0.558>59 3 (4.4) 65 (95.6) 68SexMale 5 (5.1) 93 (94.9) 98 0.143 0.705Female 3 (6.7) 42 (93.3) 45a Comparison between AJCC stage I to II and III to IV, χ2 test.b Sun-protected sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.985.2.2 eIF4E expression is increased with melanoma progressionWe performed immunohistochemical staining in TMA slides (Figure 5.2a-h) and categorized thesamples into low and high eIF4E expression. A significant increase in the expression of eIF4Ewas detected with the progression of melanoma from dysplastic nevi to primary melanoma andfrom primary melanoma to metastatic melanoma (P<0.001 and P=0.008, respectively, ² test;Figure 5.2i and Figure 5.3) suggesting that increased eIF4E expression might be a criticalrequirement for the transformation from nevus to malignant tumour and also for the developmentof melanoma metastasis. We also examined the expression pattern of eIF4E in six melanoma celllines compared to melanocytes. At least in five of the cell lines eIF4E protein expressionincreased compared to melanocytes (Figure 5.2j).99Figure 5.2 eIF4E protein expression in different melanoma stages, and in melanoma cell lines compared tonormal human melanocytes.Representative images of eIF4E immunohistochemical staining in melanocytic lesions. (a-d) Bar=100 µm. (e-h)Bar=10 µm. (a, e) Negative eIF4E staining. (b, f) Weak eIF4E staining. (c, g) Moderate eIF4E staining. (d, h) StrongeIF4E staining. (i) eIF4E expression is increased in MM compared with CAN (P<0.001, ² test), DN (P<0.001, ²test), and PM (P=0.008, ² test). eIF4E expression is also increased in PM compared with CAN and DN (P<0.001,² test). (j) Western blot analysis of eIF4E in whole cell extracts from melanoma cell lines and melanocytes. CAN,common acquired nevi; DN, dysplastic nevi; PM, primary melanoma; MM, metastatic melanoma; MC, melanocyte.100Figure 5.3 Changes in eIF4E protein expression during melanoma progression.CAN, common acquired nevi; DN, dysplastic nevi; PM, primary melanoma; MM, metastaticmelanoma; IRS, immunoreactive score.1015.2.3 Increased eIF4E expression is associated with poor patient survivalKaplan-Meier survival analysis showed that relative to low eIF4E expression, high eIF4Eexpression inversely correlated with overall and disease-specific 5-year survival in all (P<0.001,log rank test; Figure 5.4a-b) and in primary melanoma patients (P=0.027 and 0.022, respectively,log rank test; Figure 5.4c-d). Since the expression of eIF4E is increased from thin (<1mm) tothick (≥1mm) melanomas (Table 5.1), the difference in patient survival between the two groupswas examined next. Figure 5.5 shows that primary melanoma patients with tumours ≥1mm thickhad a significantly less overall and disease-specific 5-year survival compared to patients withtumours <1mm thick (P=0.001, log rank test). When eIF4E expression was included in theanalysis, a significant difference was observed in survival of patients with tumours ≥1mm thick,with high eIF4E expression being associated with a poorer survival outcome (P=0.021 and 0.024for overall and disease-specific survival, respectively, log rank test; Figure 5.4e-f). However, nocorrelation between eIF4E expression and patient survival was observed for tumours <1mm thick(data not shown).102Figure 5.4 Kaplan-Meier curves representing the correlation between eIF4E expression and 5-year survivalof melanoma patients.(a, b) Increased eIF4E expression is associated with poor overall and disease-specific 5-year survival in allmelanoma patients (P<0.001, log rank test) and (c, d) primary melanoma patients (P=0.027 and 0.022, respectively,log rank test). (e, f) Increased eIF4E expression is associated with poor overall and disease-specific 5-year survivalin primary melanoma patients with tumours ≥1mm thick (P=0.021 and 0.024, respectively, log rank test). Cum.,cumulative.103Figure 5.5 Kaplan-Meier analyses comparing the survival of patients with tumours ≥1mm thick compared tothe ones with tumours <1mm thick.In primary melanoma patients tumours ≥1mm thick (n=182) were significantly associated with a poorer (a) overall,and (b) disease-specific 5-year survival compared to tumours <1mm thick (n=56, P=0.001 for both, log-rank test).1045.2.4 eIF4E is an independent prognostic marker for melanoma patientsUnivariate Cox proportional hazard regression analysis supported the results from the Kaplan-Meier survival analysis by showing that high eIF4E expression was a significant prognosticfactor for the overall and disease-specific 5-year survival of all melanoma patients (HR, 3.20,95% CI, 1.63-6.26, P=0.001 and HR, 4.33, 95% CI, 1.91-9.80, P<0.001, respectively; Table 5.2),primary melanoma patients (HR, 2.69, 95% CI, 1.08-6.70, P=0.033 and HR, 3.59, 95% CI, 1.12-11.52, P=0.032, respectively; Table 5.2) and primary melanoma patients with tumours ≥1mmthick (HR, 3.59, 95% CI, 1.12-11.47, P=0.031 and HR, 4.44, 95% CI, 1.08-18.29, P=0.039,respectively; Table 5.3).Further, multivariate Cox regression analysis indicated that eIF4E was an adverseindependent prognostic factor for the overall and disease-specific 5-year survival of allmelanoma patients (HR, 2.39, 95% CI, 1.21-4.69, P=0.012 and HR, 3.09, 95% CI, 1.36-7.03,P=0.007, respectively; Table 5.4), primary melanoma patients (HR, 2.72, 95% CI, 1.08-6.87,P=0.034 and HR, 3.47, 95% CI, 1.06-11.29, P=0.039, respectively; Table 5.4) and primarymelanoma patients with tumours ≥1mm thick (HR, 3.85, 95% CI, 1.19-12.45, P=0.024 and HR,4.59, 95% CI, 1.10-19.17, P=0.037, respectively; Table 5.4).105Table 5.2 Univariate Cox regression analysis on 5-year survival of a total of 381 melanoma patients and 238 primary melanoma patientsVariablesOverall Survival Disease-specific SurvivalPatients (%) Deaths Death rate (%) HR (95% CI) P ˡ Deaths Death rate (%) HR (95% CI) P ˡAll melanoma (n=381)eIF4ELow expression 47 (12.3) 9 19.1 1.00 0.001 6 12.8 1.00 <0.001High expression 334 (87.7) 161 48.2 3.19 (1.63-6.26) 146 43.7 4.32 (1.91-9.79)Age (years)60 194 (50.9) 84 43.3 1.00 0.524 78 40.2 1.00 0.888>60 187 (49.1) 86 46.0 1.10 (0.81-1.49) 74 39.6 1.02 (0.74-1.40)SexMale 228 (59.8) 103 45.2 1.00 0.700 92 40.4 1.00 0.841Female 153 (40.2) 66 43.1 1.06 (0.78-1.44) 60 39.2 1.03 (0.74-1.43)AJCCI-II 238 (62.5) 65 27.3 1.00 <0.001 51 21.4 1.00 <0.001III-IV 143 (37.5) 105 73.4 4.49 (3.28-6.15) 101 70.6 5.47 (3.89-7.68)Primary melanoma (n=238)eIF4ELow expression 39 (16.4) 5 12.8 1.00 0.033 3 7.6 1.00 0.032High expression 199 (83.6) 60 30.2 2.69 (1.08-6.70) 48 24.1 3.58 (1.11-11.52)Age (years)62 121 (50.8) 22 18.2 1.00 0.001 18 14.9 1.00 0.008>62 117 (49.2) 43 36.8 2.32 (1.38-3.88) 33 28.2 2.18 (1.23-3.88)SexMale 130 (54.6) 34 26.2 1.00 0.659 26 20.0 1.00 0.561Female 108 (45.4) 31 28.7 0.89 (0.55-1.45) 25 23.1 0.85 (0.49-1.47)UlcerationAbsent 187 (78.6) 36 19.3 1.00 <0.001 26 13.9 1.00 <0.001Present 51 (21.4) 29 56.9 4.05 (2.47-6.61) 25 49.0 4.86 (2.80-8.44)Thickness (mm)<1 56 (23.5) 6 10.7 1.00 0.003 3 5.4 1.00 0.003≥1 182 (76.5) 59 32.4 3.57 (1.54-8.28) 48 26.4 5.83 (1.81-18.73)Location ²Sun-protected 182 (76.5) 50 27.5 1.00 0.835 42 23.1 1.00 0.238Sun-exposed 56 (23.5) 15 26.8 0.94 (0.52-1.67) 9 16.1 0.67 (0.32-1.38)SubtypeOthers 152 (63.9) 48 31.6 1.00 0.073 37 24.3 1.00 0.159Superficial spreading 86 (36.1) 17 19.8 0.60 (0.34-1.04) 14 16.3 0.64 (0.34-1.18)Abbreviations: CI, confidence interval; HR, hazard ratio.ˡ Log-rank test.² Sun-protected  sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.106Table 5.3 Univariate Cox regression analysis on 5-year survival of 182 primary melanoma patients with tumours ≥1mm thickVariablesOverall Survival Disease-specific SurvivalPatients (%) Deaths Death rate (%) HR (95% CI) P ˡ Deaths Death rate (%) HR (95% CI) P ˡPM with tumours ≥1mm thick(n=182)eIF4ELow expression 25 (13.7) 3 12.0 1.00 0.031 2 8.0 1.00 0.039High expression 157 (86.3) 56 35.7 3.58 (1.12-11.46) 46 29.3 4.44 (1.07-18.29)Age (years)65 91 (50.0) 24 26.4 1.00 0.095 21 23.1 1.00 0.267>65 91 (50.0) 35 38.5 1.55 (0.92-2.61) 27 29.7 1.38 (0.78-2.44)SexMale 98 (53.8) 31 31.6 1.00 0.809 25 25.5 1.00 0.782Female 84 (46.2) 28 33.3 0.93 (0.56-1.56) 23 27.4 0.92 (0.52-1.62)UlcerationAbsent 132 (72.5) 31 23.5 1.00 <0.001 24 18.2 1.00 <0.001Present 50 (27.5) 28 56.0 3.15 (1.89-5.27) 24 48.0 3.52 (1.99-6.21)Thickness (mm)4 123 (67.6) 30 14.4 1.00 0.0006 23 18.7 1.00 0.004>4 59 (32.4) 29 49.2 2.03 (1.22-3.39) 25 42.4 2.30 (1.30-4.05)Location ²Sun-protected 140 (76.9) 41 29.3 1.00 0.755 36 25.7 1.00 0.433Sun-exposed 42(23.1) 15 35.7 1.09 (0.61-1.97) 9 21.4 0.74 (0.36-1.54)SubtypeOthers 126 (69.2) 43 34.1 1.00 0.211 33 26.2 1.00 0.520Superficial spreading 56 (30.8) 13 23.2 0.68 (0.37-1.24) 12 21.4 0.81 (0.42-1.53)Abbreviations: CI, confidence interval; HR, hazard ratio; PM, primary melanoma.ˡ Log-rank test.² Sun-protected sites: trunk, arm, leg and feet; sun-exposed sites: head and neck.107Table 5.4 Multivariate Cox regression analysis on 5-year overall and disease-specific survival of melanoma patientsVariables a Overall Survival Disease-specific Survival b SE HR 95% CI P  b SE HR 95% CI PAll melanoma (n=381)eIF4E 0.869 0.345 2.385 1.21-4.69 0.012 1.129 0.419 3.093 1.36-7.03 0.007Age 0.218 0.154 1.244 0.92-1.68 0.157 0.158 0.163 1.171 0.85-1.61 0.333Sex -0.222 0.161 0.801 0.58-1.09 0.166 -0.282 0.169 0.754 0.54-1.05 0.095AJCC 1.494 0.164 4.456 3.23-6.14 <0.001 1.686 0.178 5.397 3.80-7.64 <0.001PM (n=238)eIF4E 1.002 0.472 2.724 1.08-6.86 0.034 1.243 0.603 3.465 1.06-11.29 0.039Age 0.456 0.280 1.578 0.91-2.73 0.103 0.321 0.314 1.379 0.74-2.55 0.307Sex 0.083 0.254 1.087 0.66-1.78 0.743 0.093 0.285 1.098 0.62-1.92 0.744Ulceration 1.071 0.274 2.919 1.70-4.99 <0.001 1.243 0.307 3.465 1.90-6.32 <0.001Thickness 0.665 0.453 1.944 0.80-4.72 0.142 1.089 0.621 2.972 0.88-10.03 0.079Location -0.133 0.304 0.875 0.48-1.58 0.660 -0.420 0.376 0.657 0.31-1.37 0.264Subtype -0.388 0.291 0.679 0.38-1.20 0.183 -0.329 0.322 0.719 0.38-1.35 0.307PM with tumours≥1mm thick (n=182)eIF4E 1.349 0.598 3.854 1.19-12.44 0.024 1.523 0.730 4.585 1.09-19.17 0.037Age 0.086 0.284 1.090 0.62-1.90 0.763 -0.072 0.313 0.931 0.50-1.71 0.819Sex 0.076 0.267 1.079 0.63-1.82 0.777 0.102 0.295 1.107 0.62-1.97 0.730Ulceration 1.004 0.293 2.728 1.53-4.84 0.001 1.138 0.327 3.121 1.64-5.92 0.001Thickness 0.337 0.280 1.401 0.81-2.42 0.228 0.442 0.313 1.557 0.84-2.87 0.158Location 0.096 0.306 1.100 0.60-2.00 0.755-0.249 0.378 0.780 0.37-1.63 0.510Subtype-0.362 0.315 0.696 0.37-1.28 0.249-0.215 0.333 0.806 0.42-1.55 0.519a Coding of variables: eIF4E was coded as 1 (low) and 2 (high). Age was coded as 1 (60 years) and 2 (>60 years) for all melanoma, 1 (62 years)and 2 (>62 years) for primary melanoma and 1 (65 years) and 2 (>65 years) for primary melanoma with tumours ≥1mm thick. Sex was coded as 1(male) and 2 (female). AJCC was coded as 1 (stage I to II) and 2 (stage III to IV). Ulceration was coded as 1 (absent) and 2 (present). Thickness wascoded as 1 (<1mm) and 2 (≥1mm) for primary melanoma and 1 (4mm) and 2 (>4mm) for primary melanoma with tumours ≥1mm thick. Locationwas coded as 1 (sun protected) and 2 (sun exposed). Subtype was coded as 1 (superficial spreading) and 2 (others).b: regression coefficient.NOTE: SE, standard error of ; HR, hazard ratio; CI, confidence interval; PM, primary melanoma.1085.2.5 eIF4E regulates cell invasion via MMP-2 expression and activityOur TMA results suggested the importance of high eIF4E expression for the development ofmelanoma metastasis and also indicated an inverse correlation between high eIF4E expressionand patient survival. Consequently, we decided to study the role of eIF4E in increasingmelanoma cell invasion which is one of the hallmarks of cancer and leads to higher metastaticpotential of melanoma and decreased survival of melanoma patients (Friedl & Wolf, 2003).We used eIF4E specific siRNA to knockdown (KD) eIF4E expression and found that cellinvasion was significantly reduced in eIF4E-KD MMRU and MMLH cells compared withcontrol siRNA-transfected cells (P=0.022 and 0.026, respectively, t-test; Figure 5.6a and c,respectively). Likewise, the use of a second eIF4E specific siRNA (sieIF4E-1) in MMRU cellsconfirmed our results (P=0.017, t-test; Figure 5.7a).Since MMP-2 is known to play an important role in cell invasion, (Li et al, 2008;Vaisanen et al, 1996) we next performed zymography assay and Western blot analysis andobserved a decrease in the gelatinolytic activity and expression of MMP-2 in eIF4E-KD MMRUand MMLH cells compared with respective controls (Figure 5.6b and d, respectively). A similarresult was obtained using sieIF4E-1 in MMRU cell line (Figure 5.7b).In order to further support our findings, we used an mTOR inhibitor called rapamycinwhich is able to suppress 4EBP1 phosphorylation and consequently lead to inhibition of eIF4E(Yellen et al, 2011). MMRU cells were exposed to 0% serum and treated with DMSO or 2 µMrapamycin (Yellen et al, 2011) for a total of 36 hours. 12 hours after the start of the treatment,cells were subjected to Boyden chamber assay. A significant reduction in invasion was observedin rapamycin treated cells compared to DMSO treated cells (Figure 5.7c). Zymography assay and109Western blot analysis also showed a decrease in MMP-2 activity and expression, respectively inrapamycin treated cells compared to the control group (Figure 5.7d).Our previous TMA study examining the expression of MMP-2 in melanoma showed thatMMP-2 expression is a prognostic marker for melanoma patients (Rotte et al, 2012). Analysis ofthe 372 melanoma biopsies in common between the MMP-2 study and the current study,indicated a direct association between high eIF4E and strong MMP-2 expression (P<0.001, 2test; Figure 5.8a). We further divided the samples into four groups based on their staining: (1)low eIF4E expression and negative-moderate MMP-2 expression; (2) either low eIF4Eexpression and strong MMP-2 expression or high eIF4E expression and negative-moderateMMP-2 expression; and (3) high eIF4E expression and strong MMP-2 expression. For both all(Figure 5.8b-c) and primary melanoma (Figure 5.8d-e) patients, those classified as category 1had the most and the ones classified as category 3 had the least favorable survival outcome, andsurvival rate for patients in category 2 was between the other two categories (P<0.001 for overalland disease-specific 5-year survival of all and primary melanoma patients, log rank test; Figure5.8b-e).110Figure 5.6 eIF4E knockdown inhibits melanoma cell invasion by reducing MMP-2 expression and activity.MMRU (left column) and MMLH (right column) melanoma cell lines were transfected with sieIF4E or controlsiRNA for two consecutive times with the second time being 48 h after the first. (a, c) For Boyden chamber assay,72 h after the first transfection, cells were suspended in serum-free medium, seeded on matrigel, incubated at 37oCfor 24 h, stained and quantified. Top, representative images of invaded cells in the insets of Transwell chambers.Bottom, quantification of cell invasion. *P<0.05, Student’s t-test. (b, d) Protein extracts were prepared 96 h after thefirst transfection and analyzed by Western blot. MMP-2 activity was determined by zymography 96 h after the firsttransfection.111Figure 5.7 eIF4E inhibition reduces melanoma cell invasion by decreasing MMP-2 expression and activity.(a, b) MMRU cells were transfected with sieIF4E-1 or control siRNA for two consecutive times with the secondtime being 48 h after the first. (c, d) MMRU cells were exposed to 0% serum and treated with DMSO or 2 µMrapamycin for a total of 36 h. (a, c) For Boyden chamber assay, 72 h after the first siRNA transfection or 12 h afterthe start of rapamycin treatment, cells were suspended in serum-free medium, seeded on matrigel, incubated at 37oCfor 24 h, stained with crystal violet and quantified. *P<0.05, **P<0.01, Student’s t-test. (b, d) Protein extracts wereprepared 96 h after the first siRNA transfection or after 36 h of rapamycin treatment and were subjected to Westernblot analysis. MMP-2 activity was determined by zymography which was performed 96 h after the first siRNAtransfection or after 36 h of rapamycin treatment.112Figure 5.8 Simultaneous high eIF4E expression and strong MMP-2 expression correlate with a poorer 5-yearsurvival.(a) High eIF4E expression directly correlates with strong MMP-2 expression in human melanomas (n=372;P<0.001, ² test). (b, c) Simultaneous low eIF4E expression and negative-moderate MMP-2 expression (category 1)was significantly associated with a better overall and disease-specific 5-year survival outcome compared with loweIF4E and strong MMP-2 expression, or high eIF4E and negative-moderate MMP-2 expression (category 2), or higheIF4E and strong MMP-2 expression (category 3) in all melanoma patients and in (d, e) primary melanoma patients(P<0.001 for both overall and disease-specific 5-year survival of all and primary melanoma patients, log rank test).1135.2.6 eIF4E knockdown increases apoptosis and decreases cell proliferation in melanomacell linesTo knockdown eIF4E we transiently transfected MMRU and MMLH melanoma cells withsieIF4E or siC twice consecutively with the second transfection time being 48 hours after thefirst one. At 96 hours after first transfection cells were harvested for Western blot analysis.Consistent with findings in prostate cancer (Graff et al, 2009), reduced eIF4E protein expressionresulted in suppression of BCL-2 expression, which is an eIF4E-regulated protein, and anincrease in the expression of other apoptosis markers such as cleaved caspase-3 and cleavedPARP in both MMRU and MMLH melanoma cell lines (Figure 5.9a and d, respectively). Theexpression level of c-myc, which is a known eIF4E regulated protein and controls bothproliferation and apoptosis (Jamerson et al, 2000), was also decreased along with reduced eIF4Eexpression in MMRU and MMLH cell lines (Figure 5.9a and d, respectively).We next evaluated the biological consequences of reducing eIF4E expression inmelanoma cells. At 48, 72 and 96 hours after first transfection, cells were harvested forproliferation assay. eIF4E knockdown led to a continuous significant decrease in cellproliferation in MMRU and MMLH melanoma cell lines, from 48 hours to 96 hours after thefirst transfection, as indicated by sulforhodamine B cell proliferation assay (Figure 5.9b and e,respectively). Since the biggest difference in cell proliferation happened 96 hours after the firsttransfection, this time point was chosen for apoptosis assay. eIF4E knockdown induced apoptosisas evidenced by significantly increased sub-G1 populations in the knockdown groups comparedwith the control groups in MMRU and MMLH melanoma cell lines (P=0.003 and P<0.001,respectively, t-test; Figure 5.9c and f, respectively). All of the above results were confirmed byusing sieIF4E-1 in MMRU melanoma cell line (Figure 5.10a-c). Treatment of MMRU cells with114rapamycin, as explained before, also resulted in a decrease in the expression of c-myc and BCL-2 and an increase in the expression of cleaved caspase-3 and cleaved PARP in the rapamycintreated cells compared to the control (Figure 5.10d). Additionally, rapamycin treatment ofMMRU cells led to a significant decrease and increase in cell proliferation and apoptosis,respectively (Figure 5.10e-f).115Figure 5.9 eIF4E knockdown promotes apoptosis and inhibits cell proliferation in melanoma cells.MMRU (left column) and MMLH (right column) melanoma cells were transfected with sieIF4E or control siRNAfor two consecutive times with the second time being 48 h after the first. (a, d) 96 h after the first transfectionprotein extracts were analyzed by Western blot. (b, e) 48, 72 and 96 h after the first transfection, cell proliferationwas analyzed by sulforhodamine B assay. *P<0.05, **P<0.01, Student’s t-test. (c, f) 96 h after the first transfection,cells were stained with propidium iodide and the percentage of apoptotic (sub-G1) cells was measured by flowcytometry. **P<0.01, ***P<0.001, Student’s t-test.116Figure 5.10 eIF4E inhibition promotes apoptosis and reduces cell proliferation in melanoma cells.(a-c) MMRU melanoma cells were transfected with sieIF4E-1 or control siRNA for two consecutive times with thesecond time being 48 h after the first. (d-f) MMRU cells were exposed to 0% serum and treated with DMSO or 2µM rapamycin for a total of 36 h. (a, d) 96 h after the first siRNA transfection or after 36 h of rapamycin treatment,protein extracts were analyzed by Western blot. (b, e) 48, 72 and 96 h after the first siRNA transfection or after 36 hof rapamycin treatment, cell proliferation was analyzed by sulforhodamine B assay. *P<0.05, **P<0.01, Student’s t-test. (c, f) 96 h after the first siRNA transfection or after 36 h of rapamycin treatment, cells were stained withpropidium iodide and the percentage of apoptotic (sub-G1) cells was measured by flow cytometry. *P<0.05,***P<0.001 Student’s t-test.1175.2.7 eIF4E may have a role in inducing epithelial-mesenchymal transitionIn order to investigate the role of eIF4E in inducing EMT, Western blot analysis was performed.eIF4E knockdown resulted in a downregulation of mesenchymal markers such as vimentin, N-cadherin and α-SMA and an upregulation of E-cadherin epithelial marker, compared to thevector control, in MMRU and MMLH melanoma cells. eIF4E knockdown also led to adownregulation of Twist, the EMT inducing transcription factor, in MMRU cells, but Twist wasnot detected in MMLH melanoma cells (Figure 5.11a-b, respectively). These findings suggestthat eIF4E may play a role in the induction of EMT, which supports the possible role of eIF4E ininvasion and metastasis.118Figure 5.11 eIF4E may promote melanoma cell invasion and metastasis by inducing EMT.(a) MMRU and (b) MMLH melanoma cell lines were transiently transfected with sieIF4E or control siRNA for twoconsecutive times with the second time being 48 h after the first time. Protein extracts were prepared 96 h after thefirst transfection and analyzed for the expression of E-cadherin, N-cadherin, vimentin, α-SMA, Twist, eIF4E andactin by Western blot analysis.1195.2.8 eIF4E knockdown increases chemosensitivity of melanoma cell linesA major obstacle of melanoma treatment is its resistance to drug-induced apoptosis. Sinceincreased eIF4E expression strongly correlates with poor patient survival, we investigated theinvolvement of eIF4E in chemoresistance of melanoma cells. As above, we twice transfectedMMRU and MMLH melanoma cell lines with sieIF4E or control siRNA. At 72 hours after thefirst transfection, cells were treated with 0.25 µg/ml doxorubicin for 24 hours (Lin et al, 2009).Proliferation of cells after simultaneous eIF4E knockdown and doxorubicin treatment wassignificantly lower than those of cells simultaneously treated with control siRNA anddoxorubicin in MMRU and MMLH cells (P=0.015 and P<0.001, respectively, t-test; Figure5.12a-b, respectively).We next performed flow cytometry analysis and showed that drug treated cells witheIF4E knockdown have higher sub-G1 populations compared to the cells treated with controlsiRNA and doxorubicin in MMRU and MMLH cells (P=0.031 and 0.021, respectively, t-test;Figure 5.12c-d, respectively). Similarly, cisplatin treated eIF4E knockdown MMRU cellsshowed higher level of apoptosis and lower level of proliferation compared to cellssimultaneously treated with control siRNA and cisplatin (Figure 5.13). This might be anindication that perhaps eIF4E downregulation in melanoma cells increases drug-inducedapoptosis, suggesting that melanoma resistance to chemotherapy may in part be due to anincrease in eIF4E protein expression.120Figure 5.12 Doxorubicin-induced apoptosis is enhanced in melanoma cells after eIF4E knockdown.(a, c) MMRU and (b, d) MMLH cells were transiently transfected with sieIF4E or control siRNA for twoconsecutive times with the second time being 48 h after the first time. 72 h after the first transfection, cells weretreated with 0.25 µg/ml doxorubicin for 24 h. Additionally, cells treated with siC alone or doxorubicin alone wereincluded as controls. (a, b) The cells were fixed with 10% trichloroacetic acid for 1 h and cell proliferation wasquantitated by sulforhodamine B staining or (c, d) cells were stained with propidium iodide and the percentage ofapoptotic (sub-G1) cells was determined by flow cytometry. *P<0.05, Student’s t-test. dox, doxorubicin.121Figure 5.13 Cisplatin-induced apoptosis is enhanced in melanoma cells after eIF4E knockdown.MMRU cells were transiently transfected with sieIF4E or control siRNA for two consecutive times with the secondtime being 48 h after the first time. 48 h after the first transfection, cells were treated with 2 µM cisplatin for 48 h.Additionally, cells treated with siC alone or cisplatin alone were included as controls. (a) The cells were fixed with10% trichloroacetic acid for 1 h and cell proliferation was quantitated by sulforhodamine B staining or (b) cells werestained with propidium iodide and the percentage of apoptotic (sub-G1) cells was determined by flow cytometry.*P<0.05, Student’s t-test. cis, cisplatin.1225.3 DiscussionSimilar to our results, colon and prostate TMAs showed eIF4E as an indicator of tumourprogression and poor prognosis of patients (Graff et al, 2009; Niu et al, 2014). One plausibleexplanation for the negative correlation between eIF4E expression and 5-year survival ofmelanoma patients could be a role for eIF4E in regulating the proliferation and apoptosis ofmelanoma cells. Inhibition of eIF4E by siRNA or rapamycin in our study substantially inhibitedthe proliferation of melanoma cells and c-myc expression which further suggests a role for eIF4Ein increasing melanoma cell proliferation as c-myc has been shown to be important for cellproliferation and growth (Jamerson et al, 2000). Consistent with our findings, overexpression ofeIF4E in chinese hamster ovary (CHO) or cloned rat embryo fibroblast (CREF) cells yieldedincreased c-myc expression, (De Benedetti & Harris, 1999). Additionally, by looking at thesubG1 populations, we found that eIF4E inhibition induced apoptosis in melanoma cells. Ourresults further supported this by showing a downregulation in BCL-2 expression and an increasein cleaved caspase-3 and cleaved PARP expression as a result of eIF4E inhibition. Consistently,it has been demonstrated that reduced eIF4E expression in prostate and breast cancer cell linesand xenografts repressed expression of eIF4E-regulated proteins such as c-myc and BCL-2,reduced cellular proliferation, inhibited the growth of cancer xenografts in mice with minimaltoxicity, and induced apoptosis, as evidenced by an increase in cleaved caspase-3, and an overallreduction in cell number (Graff et al, 2009; Graff et al, 2007). Similarly, elevated expression ofcleaved PARP has previously been detected in NSCLC cells transfected with eIF4E siRNA (Li etal, 2012). The above suggests that downregulation of eIF4E may inhibit the growth of cancercells through both proliferation downregulation and apoptosis upregulation. Because of the lackof information about the genetic mutations of our cohort, we were unfortunately unable to123correlate eIF4E expression with mutations such as BRAF or NRAS. However, the unpublishedsequencing data from our lab shows that the MMRU melanoma cell line carries the BRAFV600Emutation.Furthermore, eIF4E inhibition led to a decrease in the expression and gelatinolyticactivity of MMP-2 which may be a contributing factor to the elevated invasiveness of melanomacells. This data is supported by observing the direct correlation between high expression ofeIF4E and strong expression of MMP-2 in our TMA which also helps to explain the TMA resultsshowing a correlation between melanoma thickness and eIF4E expression and consequently acorrelation between eIF4E expression and reduced 5-year survival of primary melanomapatients. Additionally, Our finding that concurrent higher expression of both eIF4E and MMP-2(category 3) was associated with the worst survival outcome compared to lower expression ofeither eIF4E or MMP-2 (category 2) (Figure 5.8b-e) provides a rationale for dual therapeutictargeting of eIF4E and MMP-2 in melanoma.MMP-2 has previously been demonstrated to play a role in melanoma progression. Forexample, the increase in atypia in different stages of melanoma is accompanied by an increase inthe number of MMP-2 positive cells (Vaisanen et al, 1996). Changes in MMP-2 expression andactivity can be controlled through different ways. For instance, the MAPK signaling pathway hasbeen shown to regulate MMP2 expression via p38 MAPK, ERK1/2, and JNK/c-Jun (Johanssonet al, 2000; Simon et al, 1998; Westermarck & Kahari, 1999; Zhu et al, 2013). Skp2 is anotheroncogene whose mRNA translation depends on eIF4E and is required for oncogenesis andexiting contact inhibition (Nogueira et al, 2012). MMP-2 activity and expression was shown toincrease because of an increase in Skp2 expression in lung cancer (Hung et al, 2010).124Confirming our TMA and in vitro results, eIF4E has been shown to promote invasion andEMT in other cancers, as well. For example, Shankar et al reported the transcriptome/proteomeanalysis of six metastatic cell lines of varying tumour origin, including breast, prostate,fibrosarcoma, and glioma, andproteins whose expression is critical for pseudopod protrusion,actin cytoskeleton dynamics, and tumour cell migration and invasion. Interestingly, loss of eIF4Ereversed EMT in these highly metastatic cell lines and resulted in cell-cell interaction, expressionof E-cadherin, and reduced migration and invasion (Shankar et al, 2010). Others have attemptedto design compounds to inhibit eIF4E and prevent it from triggering EMT (Ghosh et al, 2009).Effect of eIF4E on vimentin protein expression has also been observed before. Hsieh et algenerated a human prostate cancer cell line that stably expresses a mutant form of 4EBP1 whichbinds to eIF4E and decreases its hyperactivation. Their experiments showed a decrease invimentin protein upon induction of this mutant (Hsieh et al, 2012). However, our study needsfurther support from in vivo animal studies to fully delineate the role of eIF4E in the multi-stepinvasion-metastasis cascade.Melanoma is particularly resistant to conventional therapies (Tarhini & Agarwala, 2006).We showed that eIF4E knockdown increases chemosensitivity of melanoma cells to doxorubicinand cisplatin. This could be because many chemotherapeutic agents induce apoptosis whiledisruption of apoptosis during tumour evolution can promote drug resistance (Johnstone et al,2002; Minotti et al, 2004). As we and others have shown, eIF4E has anti-apoptotic activities(Lazaris-Karatzas et al, 1990; Polunovsky et al, 2000); therefore, eIF4E ablation might restoresensitivity to chemotherapy. The role of eIF4E in increasing chemoresistance has beenpreviously demonstrated. In the E-myc transgenic B cell lymphoma mouse model, eIF4Eoverexpression accelerates lymphomagenesis (Ruggero et al, 2004; Wendel et al, 2004) and125causes resistance to doxorubicin (Wendel et al, 2004). In lung cancer, however, elevated eIF4Econtributes to development of erlotinib resistance, likely through positive regulation of c-Metexpression (Li et al, 2012).In summary, we found that increased eIF4E expression is significantly correlated withmelanoma progression and a worse 5-year survival of patients, especially those with tumourthickness of ≥1mm. Furthermore, our data suggests that eIF4E is an independent prognosticfactor that plays a role in EMT and regulating melanoma cell invasion, proliferation, apoptosisand chemoresistance. All these plus the fact that reducing eIF4E expression in other cancers hasbeen shown to suppress tumour growth without having any deleterious effects on normal tissues(Graff et al, 2007), makes eIF4E a potentially very good target candidate for melanoma therapy.126Chapter 6: Conclusion6.1 Summary of findings and implicationsMelanoma is the deadliest form of skin cancer with a continuously increasing incidence (Dauda& Shehu, 2005; Thompson et al, 2005). The fact that metastatic melanoma is resistant tochemotherapy, radiotherapy and immunotherapy (Gray-Schopfer et al, 2007), makes this diseaseeven more complicated and harder to defeat. Consequently, identifying novel therapeutic targetsis essential for conquering this devastating disease. On the other hand, abnormalities in proteinsynthesis and translation have been shown to have an important role in development andprogression of different cancers (Ruggero, 2013). In this work we investigated the role of twoimportant translation factors in melanoma progression and development.In chapters 3 and 4, we studied cytoplasmic and nuclear EIF5A2 expression using tissuemicroarray and found a significantly direct correlation between the expression of both EIF5A2forms and melanoma progression from dysplastic nevi to primary melanomas and metastaticmelanomas. This suggests the importance of EIF5A2 expression in transformation form nevus tomalignant tumours and the development of melanoma metastasis. Nuclear and cytoplasmicEIF5A2 expressions were demonstrated to be significantly correlated with thickness and AJCCstages. Nuclear and cytoplasmic EIF5A2 expressions were significantly higher in patients withtumour thickness of >2mm compared to patients with tumour thickness of ≤2mm suggesting thatin primary melanoma EIF5A2 expression might be induced during transition from thin to thickmelanoma. The expression of both EIF5A2 forms was also associated with poor patient survival.Interestingly, none of the patients with tumours ≤2mm thick died within 5 years in thecytoplasmic EIF5A2 negative staining group, suggesting the importance of cytoplasmic EIF5A2expression in this group of patients.127Moreover, our in vitro analysis demonstrated that EIF5A2 may play a role in inducingEMT and also promoting melanoma cell invasion by increasing MMP-2 activity which could atleast partially explain the association between EIF5A2 expression, melanoma tumour thickness,progression and poor patient survival. Confirming the in vitro results, the TMA data also showeda direct association between the expression of cytoplasmic and nuclear EIF5A2 and strongMMP-2 expression. The TMA results and also our in vitro experiments suggested that EIF5A2 isa downstream target of p-Akt. This makes sense, because the same as EIF5A2, p-Akt has alsobeen shown to have an important role in promoting melanoma progression, invasion, anddecreasing melanoma patient survival rate (Dai et al, 2005; Dhawan et al, 2002).We also found a significant correlation between nuclear and cytoplasmic EIF5A2expression; however, this correlation is not perfect which could provide evidence for thedifferential regulation of nuclear and cytoplasmic EIF5A2 expression in melanoma.Additionally, simultaneous expression of both forms of EIF5A2 was associated with the worstsurvival outcome compared to expression of only one form, suggesting that oncogenic propertiesof the two forms of EIF5A2 may at least partly differ from each other leading to concurrentexpression of both forms having synergistic or additive effects on melanoma patient survival.Finally, expression of both EIF5A2 forms was shown to be an independent prognostic factor forthe survival of melanoma patients, highlighting the potential of EIF5A2 being a prognostictherapeutic marker in melanoma.In chapter 5 we found that high eIF4E expression was significantly higher in melanomas≥1mm thick compared to the ones ˂1mm thick suggesting the induction of EIF4E expressionduring transition from thin to thick melanoma. High eIF4E expression was also shown to becorrelated with AJCC stages. Correlation between high eIF4E expression and tumour thickness128and AJCC stages are an indication for eIF4E playing an important role in melanoma cellinvasion which was confirmed by our in vitro assays determining the role of eIF4E in invasionby increasing the expression and activity of MMP-2. In addition concurrent higher expression ofboth eIF4E and MMP-2 was correlated with the worst survival outcome compared tosimultaneous lower expression of one and higher expression of the other one providing arationale for dual therapeutic targeting of eIF4E and MMP-2 in melanoma. We also showed thepossible role of eIF4E in developing metastasis by finding a significant increase in the highexpression of eIF4E from dysplastic nevi to primary and metastatic melanomas. Accordingly,eIF4E was demonstrated to induce EMT in melanoma cell lines.Our results also clarify that besides upregulating apoptosis and decreasing cellproliferation eIF4E knockdown increases melanoma chemosensitivity when these cells aresubjected to chemotherapy at the same time, which might be a partial solution to the problem ofmelanoma cells becoming resistant to drug-induced apoptosis. Consistent with all of the abovedata increased eIF4E expression was shown to be associated with poor patient survival assupported by both the results from the Kaplan-Meier survival analysis and univariate Coxproportional hazard regression analysis. Furthermore, eIF4E was proven to be an adverseindependent prognostic factor for the survival of melanoma patients that makes it a potentiallyimportant target for drug development.In summary, our TMA-based analyses together with our in vitro assays identified twopromising translation factors with a significant prognostic value in human cutaneous melanomaand a potential to serve as therapeutic targets.1296.2 Limitations of this study and future directionsWe demonstrated that EIF5A2 expression is influenced by the treatment of cells with a p-Aktinhibitor or as a result of PTEN overexpression. The exact mechanism of how PI3K/p-Aktregulates the expression of EIF5A2 is not yet clear and requires further investigation. However,it is possible that some transcription factors are responsible for the transcriptional upregulation ofEIF5A2. NF-κB, for example, is a transcription factor downstream of p-Akt (Dhawan et al,2002). It will be important to determine if p-Akt controls the transcription of EIF5A2 throughNF-κB or not.Using our TMA results we also pointed out the possible importance of dual therapeutictargeting of eIF4E and MMP-2 or EIF5A2 and MMP-2 in melanoma. This should be tested invitro by doing a simultaneous knockdown of both EIF5A2 and MMP-2 or eIF4E and MMP-2 inmelanoma cell lines and checking the effects on parameters such as cell invasion compared towhen either EIF5A2, eIF4E, MMP2 or none have been knocked down in the same cell lines. Theresults should be further tested in an in vivo animal model.The MAPK pathway is one of the most deregulated pathways in melanoma (Omholt et al.2003; Haluska et al. 2006; Saldanha et al. 2006; Gray-Schopfer et al. 2007; Platz et al. 2008);however, because of the lack of information about the genetic mutations of our cohort, we wereunfortunately unable to correlate eIF4E and EIF5A2 expression with mutations such as BRAF orNRAS. Future experiments with a cohort of patients with known mutations can provide moreinformation.Skp2 is another oncogene whose mRNA translation depends on eIF4E and is required foroncogenesis and exiting contact inhibition (Nogueira et al, 2012). MMP-2 activity andexpression was shown to increase because of an increase in Skp2 expression in lung cancer130(Hung et al, 2010). Skp2 protein expression has also been shown to increase in melanoma casesand is associated with poor survival (Rose AE, 2011). It will be important to see if eIF4Epossibly affects MMP-2 expression and activity through Skp2 or not.One of the weaknesses of this study is the lack of in vivo data using pre-clinical mousemodels. In order to better support the data on invasion and metastasis, it will be appropriate toutilize a mouse model of metastasis such as tail vein injection or intra-cardiac injection withmelanoma cells that are overexpressing either EIF5A2 or eIF4E compared to control melanomacells. In order to confirm that EIF5A2 is a downstream target of p-Akt we should also inject themice with melanoma cells concurrently overexpressing p-Akt and deficient in EIF5A2 orotherwise just overexpressing p-Akt or EIF5A2 alone and check invasion and metastasis. Similarin vivo experiments need to be performed to indicate that MMP-2 is a downstream target ofeIF4E and EIF5A2.A study has shown that xenografts from EIF5A2 expressing esophageal squamous cellcarcinoma cells formed tumours faster than xenografts from cells that expressed the vectorcontrol only. They also expressed higher levels of HIF1α and vascular endothelial growth factor,and formed more microvessels than controls (Li et al, 2014). Since angiogenesis is one of thehallmarks of cancer, it will be important to test the effect of EIF5A2 expression on angiogenesisin melanoma using assays such as tube formation assay and also check the effect of EIF5A2expression on factors such as HIF1α and vascular endothelial growth factor in vitro andfurthermore confirm the results using in vivo analysis.Taken together, this study highlighted the importance of EIF5A2 and eIF4E in melanomaas oncogenes. We demonstrated the role of these two markers in the development andprogression of melanoma through controlling invasion and metastasis. Our results indicated that131these two translation factors may serve as ideal prognostic markers and their ablation may be anovel strategy for melanoma treatment.132BibliographyAcker MG, Shin BS, Dever TE, Lorsch JR (2006) Interaction between eukaryotic initiationfactors 1A and 5B is required for efficient ribosomal subunit joining. J Biol Chem 281(13):8469-75Acker MG, Shin BS, Nanda JS, Saini AK, Dever TE, Lorsch JR (2009) Kinetic analysis of latesteps of eukaryotic translation initiation. J Mol Biol 385(2): 491-506Aitken CE, Lorsch JR (2012) A mechanistic overview of translation initiation in eukaryotes. NatStruct Mol Biol 19(6): 568-76Algire MA, Maag D, Lorsch JR (2005) Pi release from eIF2, not GTP hydrolysis, is the stepcontrolled by start-site selection during eukaryotic translation initiation. Mol Cell 20(2): 251-62Algire MA, Maag D, Savio P, Acker MG, Tarun SZ, Jr., Sachs AB, Asano K, Nielsen KH, OlsenDS, Phan L, Hinnebusch AG, Lorsch JR (2002) Development and characterization of areconstituted yeast translation initiation system. RNA 8(3): 382-97Alkalaeva EZ, Pisarev AV, Frolova LY, Kisselev LL, Pestova TV (2006) In vitro reconstitutionof eukaryotic translation reveals cooperativity between release factors eRF1 and eRF3. Cell125(6): 1125-36Allen GS, Zavialov A, Gursky R, Ehrenberg M, Frank J (2005) The cryo-EM structure of atranslation initiation complex from Escherichia coli. Cell 121(5): 703-12American Cancer Society (2015) Cancer Facts and Figures 2015. Atlanta:AmericanCancer Society: 1-52Apsel B, Blair JA, Gonzalez B, Nazif TM, Feldman ME, Aizenstein B, Hoffman R, WilliamsRL, Shokat KM, Knight ZA (2008) Targeted polypharmacology: discovery of dual inhibitors oftyrosine and phosphoinositide kinases. Nat Chem Biol 4(11): 691-9Armengol G, Rojo F, Castellvi J, Iglesias C, Cuatrecasas M, Pons B, Baselga J, Ramon y Cajal S(2007) 4E-binding protein 1: a key molecular "funnel factor" in human cancer with clinicalimplications. Cancer Res 67(16): 7551-5Asano K, Clayton J, Shalev A, Hinnebusch AG (2000) A multifactor complex of eukaryoticinitiation factors, eIF1, eIF2, eIF3, eIF5, and initiator tRNA(Met) is an important translationinitiation intermediate in vivo. Genes Dev 14(19): 2534-46133Assouline S, Culjkovic B, Cocolakis E, Rousseau C, Beslu N, Amri A, Caplan S, Leber B, RoyDC, Miller WH, Jr., Borden KL (2009) Molecular targeting of the oncogene eIF4E in acutemyeloid leukemia (AML): a proof-of-principle clinical trial with ribavirin. Blood 114(2): 257-60Balch CM, Gershenwald JE, Soong SJ, Thompson JF, Atkins MB, Byrd DR, Buzaid AC,Cochran AJ, Coit DG, Ding S, Eggermont AM, Flaherty KT, Gimotty PA, Kirkwood JM,McMasters KM, Mihm MC, Jr., Morton DL, Ross MI, Sober AJ, Sondak VK (2009) Finalversion of 2009 AJCC melanoma staging and classification. J Clin Oncol 27(36): 6199-206Barnhill RL, Mihm MC, Jr. (1993) The histopathology of cutaneous malignant melanoma. SeminDiagn Pathol 10(1): 47-75Barsh G, Gunn T, He L, Schlossman S, Duke-Cohan J (2000) Biochemical and genetic studies ofpigment-type switching. Pigment Cell Res 13 Suppl 8: 48-53Bastian BC (2014) The molecular pathology of melanoma: an integrated taxonomy ofmelanocytic neoplasia. Annu Rev Pathol 9: 239-71Benne R, Brown-Luedi ML, Hershey JW (1978) Purification and characterization of proteinsynthesis initiation factors eIF-1, eIF-4C, eIF-4D, and eIF-5 from rabbit reticulocytes. J BiolChem 253(9): 3070-7Benne R, Hershey JW (1978) The mechanism of action of protein synthesis initiation factorsfrom rabbit reticulocytes. J Biol Chem 253(9): 3078-87Berset C, Zurbriggen A, Djafarzadeh S, Altmann M, Trachsel H (2003) RNA-binding activity oftranslation initiation factor eIF4G1 from Saccharomyces cerevisiae. RNA 9(7): 871-80Bissell MJ, Radisky D (2001) Putting tumours in context. Nat Rev Cancer 1(1): 46-54Borden KL, Culjkovic-Kraljacic B (2010) Ribavirin as an anti-cancer therapy: acute myeloidleukemia and beyond? Leuk Lymphoma 51(10): 1805-15Borg A, Sandberg T, Nilsson K, Johannsson O, Klinker M, Masback A, Westerdahl J, Olsson H,Ingvar C (2000) High frequency of multiple melanomas and breast and pancreas carcinomas inCDKN2A mutation-positive melanoma families. J Natl Cancer Inst 92(15): 1260-6Bostrom K, Wettesten M, Boren J, Bondjers G, Wiklund O, Olofsson SO (1986) Pulse-chasestudies of the synthesis and intracellular transport of apolipoprotein B-100 in Hep G2 cells. JBiol Chem 261(29): 13800-6Braam KI, Overbeek A, Kaspers GJ, Ronckers CM, Schouten-van Meeteren AY, Van Dulmen-Den Broeder E, Veening MA (2012) Malignant melanoma as second malignant neoplasm inlong-term childhood cancer survivors: a systematic review. Pediatr Blood Cancer 58(5): 665-74134Braeuer RR, Zigler M, Villares GJ, Dobroff AS, Bar-Eli M (2011) Transcriptional control ofmelanoma metastasis: the importance of the tumor microenvironment. Semin Cancer Biol 21(2):83-8Brina D, Grosso S, Miluzio A, Biffo S (2011) Translational control by 80S formation and 60Savailability: the central role of eIF6, a rate limiting factor in cell cycle progression andtumorigenesis. Cell Cycle 10(20): 3441-6Budden T, Bowden NA (2013) The Role of Altered Nucleotide Excision Repair and UVB-Induced DNA Damage in Melanomagenesis. Int J Mol Sci 14(1): 1132-51Bushell M, Wood W, Carpenter G, Pain VM, Morley SJ, Clemens MJ (2001) Disruption of theinteraction of mammalian protein synthesis eukaryotic initiation factor 4B with the poly(A)-binding protein by caspase- and viral protease-mediated cleavages. J Biol Chem 276(26): 23922-8Buxade M, Parra-Palau JL, Proud CG (2008) The Mnks: MAP kinase-interacting kinases (MAPkinase signal-integrating kinases). Front Biosci 13: 5359-73Cantley LC (2002) The phosphoinositide 3-kinase pathway. Science 296(5573): 1655-7Caraglia M, Park MH, Wolff EC, Marra M, Abbruzzese A (2013) eIF5A isoforms and cancer:two brothers for two functions? Amino Acids 44(1): 103-9Carroll M, Borden KL (2013) The oncogene eIF4E: using biochemical insights to target cancer.J Interferon Cytokine Res 33(5): 227-38Castellvi J, Garcia A, Rojo F, Ruiz-Marcellan C, Gil A, Baselga J, Ramon y Cajal S (2006)Phosphorylated 4E binding protein 1: a hallmark of cell signaling that correlates with survival inovarian cancer. Cancer 107(8): 1801-11Cawley A, Warwicker J (2012) eIF4E-binding protein regulation of mRNAs with differential 5'-UTR secondary structure: a polyelectrostatic model for a component of protein-mRNAinteractions. Nucleic Acids Res 40(16): 7666-75Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, Jacobsen A, Byrne CJ, HeuerML, Larsson E, Antipin Y, Reva B, Goldberg AP, Sander C, Schultz N (2012) The cBio cancergenomics portal: an open platform for exploring multidimensional cancer genomics data. CancerDiscov 2(5): 401-4Chang AE, Karnell LH, Menck HR (1998) The National Cancer Data Base report on cutaneousand noncutaneous melanoma: a summary of 84,836 cases from the past decade. The AmericanCollege of Surgeons Commission on Cancer and the American Cancer Society. Cancer 83(8):1664-78135Chang L, Karin M (2001) Mammalian MAP kinase signalling cascades. Nature 410(6824): 37-40Chaudhuri J, Chowdhury D, Maitra U (1999) Distinct functions of eukaryotic translationinitiation factors eIF1A and eIF3 in the formation of the 40 S ribosomal preinitiation complex. JBiol Chem 274(25): 17975-80Cheli Y, Ohanna M, Ballotti R, Bertolotto C (2010) Fifteen-year quest for microphthalmia-associated transcription factor target genes. Pigment Cell Melanoma Res 23(1): 27-40Chen B, Tardell C, Higgins B, Packman K, Boylan JF, Niu H (2012a) BRAFV600E negativelyregulates the AKT pathway in melanoma cell lines. PLoS One 7(8): e42598Chen CN, Hsieh FJ, Cheng YM, Lee PH, Chang KJ (2004) Expression of eukaryotic initiationfactor 4E in gastric adenocarcinoma and its association with clinical outcome. J Surg Oncol86(1): 22-7Chen G, Cheng Y, Tang Y, Martinka M, Li G (2012b) Role of Tip60 in human melanoma cellmigration, metastasis, and patient survival. J Invest Dermatol 132(11): 2632-41Chen G, Li M, Wu D, Tang H, Tang D (2012c) Primary splenic angiosarcoma in a 2.5-year-oldboy with hepatic metastasis. Pediatr Surg Int 28(11): 1147-50Chen J, Tsai A, O'Leary SE, Petrov A, Puglisi JD (2012d) Unraveling the dynamics of ribosometranslocation. Curr Opin Struct Biol 22(6): 804-14Chen W, Luo JH, Hua WF, Zhou FJ, Lin MC, Kung HF, Zeng YX, Guan XY, Xie D (2009)Overexpression of EIF-5A2 is an independent predictor of outcome in patients of urothelialcarcinoma of the bladder treated with radical cystectomy. Cancer Epidemiol Biomarkers Prev18(2): 400-8Cheng S, Gallie DR (2010) Competitive and noncompetitive binding of eIF4B, eIF4A, and thepoly(A) binding protein to wheat translation initiation factor eIFiso4G. Biochemistry 49(38):8251-65Cheung YN, Maag D, Mitchell SF, Fekete CA, Algire MA, Takacs JE, Shirokikh N, Pestova T,Lorsch JR, Hinnebusch AG (2007) Dissociation of eIF1 from the 40S ribosomal subunit is a keystep in start codon selection in vivo. Genes Dev 21(10): 1217-30Chow HY, Jubb AM, Koch JN, Jaffer ZM, Stepanova D, Campbell DA, Duron SG, O'Farrell M,Cai KQ, Klein-Szanto AJ, Gutkind JS, Hoeflich KP, Chernoff J (2012) p21-Activated kinase 1 isrequired for efficient tumor formation and progression in a Ras-mediated skin cancer model.Cancer Res 72(22): 5966-75136Chudnovsky Y, Khavari PA, Adams AE (2005) Melanoma genetics and the development ofrational therapeutics. J Clin Invest 115(4): 813-24Cigan AM, Feng L, Donahue TF (1988) tRNAi(met) functions in directing the scanningribosome to the start site of translation. Science 242(4875): 93-7Clark WH, Jr., Elder DE, Guerry Dt, Epstein MN, Greene MH, Van Horn M (1984) A study oftumor progression: the precursor lesions of superficial spreading and nodular melanoma. HumPathol 15(12): 1147-65Clark WH, Jr., From L, Bernardino EA, Mihm MC (1969) The histogenesis and biologicbehavior of primary human malignant melanomas of the skin. Cancer Res 29(3): 705-27Clement PM, Hanauske-Abel HM, Wolff EC, Kleinman HK, Park MH (2002) The antifungaldrug ciclopirox inhibits deoxyhypusine and proline hydroxylation, endothelial cell growth andangiogenesis in vitro. Int J Cancer 100(4): 491-8Clement PM, Henderson CA, Jenkins ZA, Smit-McBride Z, Wolff EC, Hershey JW, Park MH,Johansson HE (2003) Identification and characterization of eukaryotic initiation factor 5A-2. EurJ Biochem 270(21): 4254-63Clement PM, Johansson HE, Wolff EC, Park MH (2006) Differential expression of eIF5A-1 andeIF5A-2 in human cancer cells. FEBS J 273(6): 1102-14Clemente C, Cochran AJ, Elder DE, Levene A, MacKie RM, Mihm MC, Rilke F, Cascinelli N,Fitzpatrick TB, Sober AJ (1991) Histopathologic diagnosis of dysplastic nevi: concordanceamong pathologists convened by the World Health Organization Melanoma Programme. HumPathol 22(4): 313-9Cohen C, Zavala-Pompa A, Sequeira JH, Shoji M, Sexton DG, Cotsonis G, Cerimele F,Govindarajan B, Macaron N, Arbiser JL (2002) Mitogen-actived protein kinase activation is anearly event in melanoma progression. Clin Cancer Res 8(12): 3728-33Coleman KG, Wautlet BS, Morrissey D, Mulheron J, Sedman SA, Brinkley P, Price S, WebsterKR (1997) Identification of CDK4 sequences involved in cyclin D1 and p16 binding. J BiolChem 272(30): 18869-74Cooper HL, Park MH, Folk JE, Safer B, Braverman R (1983) Identification of the hypusine-containing protein hy+ as translation initiation factor eIF-4D. Proc Natl Acad Sci U S A 80(7):1854-7Crew JP, Fuggle S, Bicknell R, Cranston DW, de Benedetti A, Harris AL (2000) Eukaryoticinitiation factor-4E in superficial and muscle invasive bladder cancer and its correlation withvascular endothelial growth factor expression and tumour progression. Br J Cancer 82(1): 161-6137Culjkovic B, Topisirovic I, Skrabanek L, Ruiz-Gutierrez M, Borden KL (2005) eIF4E promotesnuclear export of cyclin D1 mRNAs via an element in the 3'UTR. J Cell Biol 169(2): 245-56Culjkovic B, Topisirovic I, Skrabanek L, Ruiz-Gutierrez M, Borden KL (2006) eIF4E is a centralnode of an RNA regulon that governs cellular proliferation. J Cell Biol 175(3): 415-26Cummins DL, Cummins JM, Pantle H, Silverman MA, Leonard AL, Chanmugam A (2006)Cutaneous malignant melanoma. Mayo Clin Proc 81(4): 500-7D'Orazio JA, Nobuhisa T, Cui R, Arya M, Spry M, Wakamatsu K, Igras V, Kunisada T, GranterSR, Nishimura EK, Ito S, Fisher DE (2006) Topical drug rescue strategy and skin protectionbased on the role of Mc1r in UV-induced tanning. Nature 443(7109): 340-4Dai DL, Makretsov N, Campos EI, Huang C, Zhou Y, Huntsman D, Martinka M, Li G (2003)Increased expression of integrin-linked kinase is correlated with melanoma progression and poorpatient survival. Clin Cancer Res 9(12): 4409-14Dai DL, Martinka M, Li G (2005) Prognostic significance of activated Akt expression inmelanoma: a clinicopathologic study of 292 cases. J Clin Oncol 23(7): 1473-82Damoc E, Fraser CS, Zhou M, Videler H, Mayeur GL, Hershey JW, Doudna JA, Robinson CV,Leary JA (2007) Structural characterization of the human eukaryotic initiation factor 3 proteincomplex by mass spectrometry. Mol Cell Proteomics 6(7): 1135-46Dankort D, Curley DP, Cartlidge RA, Nelson B, Karnezis AN, Damsky WE, Jr., You MJ,DePinho RA, McMahon M, Bosenberg M (2009) Braf(V600E) cooperates with Pten loss toinduce metastatic melanoma. Nat Genet 41(5): 544-52Darveau A, Pelletier J, Sonenberg N (1985) Differential efficiencies of in vitro translation ofmouse c-myc transcripts differing in the 5' untranslated region. Proc Natl Acad Sci U S A 82(8):2315-9Das S, Ghosh R, Maitra U (2001) Eukaryotic translation initiation factor 5 functions as aGTPase-activating protein. J Biol Chem 276(9): 6720-6Dauda MM, Shehu SM (2005) Malignant melanoma: a review. Niger Postgrad Med J 12(2):125-30Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, Teague J, Woffendin H, GarnettMJ, Bottomley W, Davis N, Dicks E, Ewing R, Floyd Y, Gray K, Hall S, Hawes R, Hughes J,Kosmidou V, Menzies A, Mould C, Parker A, Stevens C, Watt S, Hooper S, Wilson R,Jayatilake H, Gusterson BA, Cooper C, Shipley J, Hargrave D, Pritchard-Jones K, Maitland N,Chenevix-Trench G, Riggins GJ, Bigner DD, Palmieri G, Cossu A, Flanagan A, Nicholson A,Ho JW, Leung SY, Yuen ST, Weber BL, Seigler HF, Darrow TL, Paterson H, Marais R,138Marshall CJ, Wooster R, Stratton MR, Futreal PA (2002) Mutations of the BRAF gene in humancancer. Nature 417(6892): 949-54De Benedetti A, Graff JR (2004) eIF-4E expression and its role in malignancies and metastases.Oncogene 23(18): 3189-99De Benedetti A, Harris AL (1999) eIF4E expression in tumors: its possible role in progression ofmalignancies. Int J Biochem Cell Biol 31(1): 59-72De Benedetti A, Rhoads RE (1990) Overexpression of eukaryotic protein synthesis initiationfactor 4E in HeLa cells results in aberrant growth and morphology. Proc Natl Acad Sci U S A87(21): 8212-6De Gregorio E, Preiss T, Hentze MW (1999) Translation driven by an eIF4G core domain invivo. EMBO J 18(17): 4865-74Deffie A, Hao M, Montes de Oca Luna R, Hulboy DL, Lozano G (1995) Cyclin E restores p53activity in contact-inhibited cells. Mol Cell Biol 15(7): 3926-33Demierre MF, Chung C, Miller DR, Geller AC (2005) Early detection of thick melanomas in theUnited States: beware of the nodular subtype. Arch Dermatol 141(6): 745-50Demunter A, Stas M, Degreef H, De Wolf-Peeters C, van den Oord JJ (2001) Analysis of N- andK-ras mutations in the distinctive tumor progression phases of melanoma. J Invest Dermatol117(6): 1483-9Dever TE, Green R (2012) The elongation, termination, and recycling phases of translation ineukaryotes. Cold Spring Harb Perspect Biol 4(7): a013706Dhawan P, Singh AB, Ellis DL, Richmond A (2002) Constitutive activation of Akt/proteinkinase B in melanoma leads to up-regulation of nuclear factor-kappaB and tumor progression.Cancer Res 62(24): 7335-42Dias CA, Gregio AP, Rossi D, Galvao FC, Watanabe TF, Park MH, Valentini SR, Zanelli CF(2012) eIF5A interacts functionally with eEF2. Amino Acids 42(2-3): 697-702Doerfel LK, Wohlgemuth I, Kothe C, Peske F, Urlaub H, Rodnina MV (2013) EF-P is essentialfor rapid synthesis of proteins containing consecutive proline residues. Science 339(6115): 85-8Dorsky RI, Moon RT, Raible DW (1998) Control of neural crest cell fate by the Wnt signallingpathway. Nature 396(6709): 370-3Dowling RJ, Topisirovic I, Alain T, Bidinosti M, Fonseca BD, Petroulakis E, Wang X, LarssonO, Selvaraj A, Liu Y, Kozma SC, Thomas G, Sonenberg N (2010) mTORC1-mediated cellproliferation, but not cell growth, controlled by the 4E-BPs. Science 328(5982): 1172-6139Duncan R, Milburn SC, Hershey JW (1987) Regulated phosphorylation and low abundance ofHeLa cell initiation factor eIF-4F suggest a role in translational control. Heat shock effects oneIF-4F. J Biol Chem 262(1): 380-8Erickson CA, Reedy MV (1998) Neural crest development: the interplay between morphogenesisand cell differentiation. Curr Top Dev Biol 40: 177-209Ernfors P (2010) Cellular origin and developmental mechanisms during the formation of skinmelanocytes. Exp Cell Res 316(8): 1397-407Fekete CA, Applefield DJ, Blakely SA, Shirokikh N, Pestova T, Lorsch JR, Hinnebusch AG(2005) The eIF1A C-terminal domain promotes initiation complex assembly, scanning and AUGselection in vivo. EMBO J 24(20): 3588-601Feldman ME, Apsel B, Uotila A, Loewith R, Knight ZA, Ruggero D, Shokat KM (2009) Active-site inhibitors of mTOR target rapamycin-resistant outputs of mTORC1 and mTORC2. PLoSBiol 7(2): e38Feoktistova K, Tuvshintogs E, Do A, Fraser CS (2013) Human eIF4E promotes mRNArestructuring by stimulating eIF4A helicase activity. Proc Natl Acad Sci U S A 110(33): 13339-44Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM (2010) Estimates of worldwideburden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 127(12): 2893-917Ferrandiz-Pulido C, Masferrer E, Toll A, Hernandez-Losa J, Mojal S, Pujol RM, Ramon y CajalS, de Torres I, Garcia-Patos V (2013) mTOR signaling pathway in penile squamous cellcarcinoma: pmTOR and peIF4E over expression correlate with aggressive tumor behavior. JUrol 190(6): 2288-95Franklin S, Pho T, Abreo FW, Nassar R, De Benedetti A, Stucker FJ, Nathan CO (1999)Detection of the proto-oncogene eIF4E in larynx and hypopharynx cancers. Arch OtolaryngolHead Neck Surg 125(2): 177-82Fraser CS (2009) The molecular basis of translational control. Prog Mol Biol Transl Sci 90: 1-51Fraser CS, Berry KE, Hershey JW, Doudna JA (2007) eIF3j is located in the decoding center ofthe human 40S ribosomal subunit. Mol Cell 26(6): 811-9Friedl P, Wolf K (2003) Tumour-cell invasion and migration: diversity and escape mechanisms.Nat Rev Cancer 3(5): 362-74140Frolova L, Le Goff X, Zhouravleva G, Davydova E, Philippe M, Kisselev L (1996) Eukaryoticpolypeptide chain release factor eRF3 is an eRF1- and ribosome-dependent guanosinetriphosphatase. RNA 2(4): 334-41Furic L, Rong L, Larsson O, Koumakpayi IH, Yoshida K, Brueschke A, Petroulakis E,Robichaud N, Pollak M, Gaboury LA, Pandolfi PP, Saad F, Sonenberg N (2010) eIF4Ephosphorylation promotes tumorigenesis and is associated with prostate cancer progression. ProcNatl Acad Sci U S A 107(32): 14134-9Gallie DR (1991) The cap and poly(A) tail function synergistically to regulate mRNAtranslational efficiency. Genes Dev 5(11): 2108-16Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, Sun Y, Jacobsen A, Sinha R,Larsson E, Cerami E, Sander C, Schultz N (2013) Integrative analysis of complex cancergenomics and clinical profiles using the cBioPortal. Sci Signal 6(269): pl1Gao YG, Selmer M, Dunham CM, Weixlbaumer A, Kelley AC, Ramakrishnan V (2009) Thestructure of the ribosome with elongation factor G trapped in the posttranslocational state.Science 326(5953): 694-9Garbe C, Leiter U (2009) Melanoma epidemiology and trends. Clin Dermatol 27(1): 3-9Garraway LA, Widlund HR, Rubin MA, Getz G, Berger AJ, Ramaswamy S, Beroukhim R,Milner DA, Granter SR, Du J, Lee C, Wagner SN, Li C, Golub TR, Rimm DL, Meyerson ML,Fisher DE, Sellers WR (2005) Integrative genomic analyses identify MITF as a lineage survivaloncogene amplified in malignant melanoma. Nature 436(7047): 117-22Geller AC, Elwood M, Swetter SM, Brooks DR, Aitken J, Youl PH, Demierre MF, Baade PD(2009) Factors related to the presentation of thin and thick nodular melanoma from a population-based cancer registry in Queensland Australia. Cancer 115(6): 1318-27Ghosh B, Benyumov AO, Ghosh P, Jia Y, Avdulov S, Dahlberg PS, Peterson M, Smith K,Polunovsky VA, Bitterman PB, Wagner CR (2009) Nontoxic chemical interdiction of theepithelial-to-mesenchymal transition by targeting cap-dependent translation. ACS Chem Biol4(5): 367-77Gingras AC, Raught B, Gygi SP, Niedzwiecka A, Miron M, Burley SK, Polakiewicz RD,Wyslouch-Cieszynska A, Aebersold R, Sonenberg N (2001) Hierarchical phosphorylation of thetranslation inhibitor 4E-BP1. Genes Dev 15(21): 2852-64Gingras AC, Raught B, Sonenberg N (1999) eIF4 initiation factors: effectors of mRNArecruitment to ribosomes and regulators of translation. Annu Rev Biochem 68: 913-63141Gosselin P, Oulhen N, Jam M, Ronzca J, Cormier P, Czjzek M, Cosson B (2011) Thetranslational repressor 4E-BP called to order by eIF4E: new structural insights by SAXS. NucleicAcids Res 39(8): 3496-503Goyer C, Altmann M, Lee HS, Blanc A, Deshmukh M, Woolford JL, Jr., Trachsel H, SonenbergN (1993) TIF4631 and TIF4632: two yeast genes encoding the high-molecular-weight subunitsof the cap-binding protein complex (eukaryotic initiation factor 4F) contain an RNA recognitionmotif-like sequence and carry out an essential function. Mol Cell Biol 13(8): 4860-74Graff JR, Boghaert ER, De Benedetti A, Tudor DL, Zimmer CC, Chan SK, Zimmer SG (1995)Reduction of translation initiation factor 4E decreases the malignancy of ras-transformed clonedrat embryo fibroblasts. Int J Cancer 60(2): 255-63Graff JR, Konicek BW, Lynch RL, Dumstorf CA, Dowless MS, McNulty AM, Parsons SH,Brail LH, Colligan BM, Koop JW, Hurst BM, Deddens JA, Neubauer BL, Stancato LF, CarterHW, Douglass LE, Carter JH (2009) eIF4E activation is commonly elevated in advanced humanprostate cancers and significantly related to reduced patient survival. Cancer Res 69(9): 3866-73Graff JR, Konicek BW, Vincent TM, Lynch RL, Monteith D, Weir SN, Schwier P, Capen A,Goode RL, Dowless MS, Chen Y, Zhang H, Sissons S, Cox K, McNulty AM, Parsons SH, WangT, Sams L, Geeganage S, Douglass LE, Neubauer BL, Dean NM, Blanchard K, Shou J, StancatoLF, Carter JH, Marcusson EG (2007) Therapeutic suppression of translation initiation factoreIF4E expression reduces tumor growth without toxicity. J Clin Invest 117(9): 2638-48Gray-Schopfer V, Wellbrock C, Marais R (2007) Melanoma biology and new targeted therapy.Nature 445(7130): 851-7Gregio AP, Cano VP, Avaca JS, Valentini SR, Zanelli CF (2009) eIF5A has a function in theelongation step of translation in yeast. Biochem Biophys Res Commun 380(4): 785-90Grifo JA, Tahara SM, Morgan MA, Shatkin AJ, Merrick WC (1983) New initiation factoractivity required for globin mRNA translation. J Biol Chem 258(9): 5804-10Grzmil M, Morin P, Jr., Lino MM, Merlo A, Frank S, Wang Y, Moncayo G, Hemmings BA(2011) MAP kinase-interacting kinase 1 regulates SMAD2-dependent TGF-beta signalingpathway in human glioblastoma. Cancer Res 71(6): 2392-402Guan XY, Fung JM, Ma NF, Lau SH, Tai LS, Xie D, Zhang Y, Hu L, Wu QL, Fang Y, Sham JS(2004) Oncogenic role of eIF-5A2 in the development of ovarian cancer. Cancer Res 64(12):4197-200Guan XY, Sham JS, Tang TC, Fang Y, Huo KK, Yang JM (2001) Isolation of a novel candidateoncogene within a frequently amplified region at 3q26 in ovarian cancer. Cancer Res 61(9):3806-9142Guldberg P, thor Straten P, Birck A, Ahrenkiel V, Kirkin AF, Zeuthen J (1997) Disruption of theMMAC1/PTEN gene by deletion or mutation is a frequent event in malignant melanoma. CancerRes 57(17): 3660-3Gutierrez E, Shin BS, Woolstenhulme CJ, Kim JR, Saini P, Buskirk AR, Dever TE (2013) eIF5Apromotes translation of polyproline motifs. Mol Cell 51(1): 35-45Haass NK, Smalley KS, Li L, Herlyn M (2005) Adhesion, migration and communication inmelanocytes and melanoma. Pigment Cell Res 18(3): 150-9Halpern AC, Guerry Dt, Elder DE, Clark WH, Jr., Synnestvedt M, Norman S, Ayerle R (1991)Dysplastic nevi as risk markers of sporadic (nonfamilial) melanoma. A case-control study. ArchDermatol 127(7): 995-9Hanahan D, Weinberg RA (2011) Hallmarks of cancer: the next generation. Cell 144(5): 646-74Hanauske-Abel HM, Park MH, Hanauske AR, Popowicz AM, Lalande M, Folk JE (1994)Inhibition of the G1-S transition of the cell cycle by inhibitors of deoxyhypusine hydroxylation.Biochim Biophys Acta 1221(2): 115-24Hao H, Muniz-Medina VM, Mehta H, Thomas NE, Khazak V, Der CJ, Shields JM (2007)Context-dependent roles of mutant B-Raf signaling in melanoma and colorectal carcinoma cellgrowth. Mol Cancer Ther 6(8): 2220-9Hariri F, Arguello M, Volpon L, Culjkovic-Kraljacic B, Nielsen TH, Hiscott J, Mann KK,Borden KL (2013) The eukaryotic translation initiation factor eIF4E is a direct transcriptionaltarget of NF-kappaB and is aberrantly regulated in acute myeloid leukemia. Leukemia 27(10):2047-55Harris SL, Levine AJ (2005) The p53 pathway: positive and negative feedback loops. Oncogene24(17): 2899-908Harris TE, Chi A, Shabanowitz J, Hunt DF, Rhoads RE, Lawrence JC, Jr. (2006) mTOR-dependent stimulation of the association of eIF4G and eIF3 by insulin. EMBO J 25(8): 1659-68Hashem Y, des Georges A, Dhote V, Langlois R, Liao HY, Grassucci RA, Hellen CU, PestovaTV, Frank J (2013) Structure of the mammalian ribosomal 43S preinitiation complex bound tothe scanning factor DHX29. Cell 153(5): 1108-19Hauber I, Bevec D, Heukeshoven J, Kratzer F, Horn F, Choidas A, Harrer T, Hauber J (2005)Identification of cellular deoxyhypusine synthase as a novel target for antiretroviral therapy. JClin Invest 115(1): 76-85143He LR, Zhao HY, Li BK, Liu YH, Liu MZ, Guan XY, Bian XW, Zeng YX, Xie D (2011)Overexpression of eIF5A-2 is an adverse prognostic marker of survival in stage I non-small celllung cancer patients. Int J Cancer 129(1): 143-50Hentze MW (1997) eIF4G: a multipurpose ribosome adapter? Science 275(5299): 500-1Hinnebusch AG, Lorsch JR (2012) The mechanism of eukaryotic translation initiation: newinsights and challenges. Cold Spring Harb Perspect Biol 4(10)Hinton TM, Coldwell MJ, Carpenter GA, Morley SJ, Pain VM (2007) Functional analysis ofindividual binding activities of the scaffold protein eIF4G. J Biol Chem 282(3): 1695-708Hofmann UB, Westphal JR, Waas ET, Zendman AJ, Cornelissen IM, Ruiter DJ, van Muijen GN(1999) Matrix metalloproteinases in human melanoma cell lines and xenografts: increasedexpression of activated matrix metalloproteinase-2 (MMP-2) correlates with melanomaprogression. Br J Cancer 81(5): 774-82Holm N, Byrnes K, Johnson L, Abreo F, Sehon K, Alley J, Meschonat C, Md QC, Li BD (2008)A prospective trial on initiation factor 4E (eIF4E) overexpression and cancer recurrence in node-negative breast cancer. Ann Surg Oncol 15(11): 3207-15Hopkins AL, Groom CR (2002) The druggable genome. Nat Rev Drug Discov 1(9): 727-30Hoque M, Hanauske-Abel HM, Palumbo P, Saxena D, D'Alliessi Gandolfi D, Park MH, Pe'eryT, Mathews MB (2009) Inhibition of HIV-1 gene expression by Ciclopirox and Deferiprone,drugs that prevent hypusination of eukaryotic initiation factor 5A. Retrovirology 6: 90Hsieh AC, Liu Y, Edlind MP, Ingolia NT, Janes MR, Sher A, Shi EY, Stumpf CR, ChristensenC, Bonham MJ, Wang S, Ren P, Martin M, Jessen K, Feldman ME, Weissman JS, Shokat KM,Rommel C, Ruggero D (2012) The translational landscape of mTOR signalling steers cancerinitiation and metastasis. Nature 485(7396): 55-61Hsu MY, Meier F, Herlyn M (2002) Melanoma development and progression: a conspiracybetween tumor and host. Differentiation 70(9-10): 522-36Hung WC, Tseng WL, Shiea J, Chang HC (2010) Skp2 overexpression increases the expressionof MMP-2 and MMP-9 and invasion of lung cancer cells. Cancer Lett 288(2): 156-61Hussein MR (2005a) Melanocytic dysplastic naevi occupy the middle ground between benignmelanocytic naevi and cutaneous malignant melanomas: emerging clues. J Clin Pathol 58(5):453-6Hussein MR (2005b) Ultraviolet radiation and skin cancer: molecular mechanisms. J CutanPathol 32(3): 191-205144Hutten S, Kehlenbach RH (2007) CRM1-mediated nuclear export: to the pore and beyond.Trends Cell Biol 17(4): 193-201Imataka H, Gradi A, Sonenberg N (1998) A newly identified N-terminal amino acid sequence ofhuman eIF4G binds poly(A)-binding protein and functions in poly(A)-dependent translation.EMBO J 17(24): 7480-9Imataka H, Sonenberg N (1997) Human eukaryotic translation initiation factor 4G (eIF4G)possesses two separate and independent binding sites for eIF4A. Mol Cell Biol 17(12): 6940-7Ingolia NT, Lareau LF, Weissman JS (2011) Ribosome profiling of mouse embryonic stem cellsreveals the complexity and dynamics of mammalian proteomes. Cell 147(4): 789-802Ito K, Ebihara K, Nakamura Y (1998) The stretch of C-terminal acidic amino acids oftranslational release factor eRF1 is a primary binding site for eRF3 of fission yeast. RNA 4(8):958-72Jackson RJ (2013) The current status of vertebrate cellular mRNA IRESs. Cold Spring HarbPerspect Biol 5(2)Jackson RJ, Hellen CU, Pestova TV (2010) The mechanism of eukaryotic translation initiationand principles of its regulation. Nat Rev Mol Cell Biol 11(2): 113-27Jackson RJ, Hellen CU, Pestova TV (2012) Termination and post-termination events ineukaryotic translation. Adv Protein Chem Struct Biol 86: 45-93Jafarnejad SM, Ardekani GS, Ghaffari M, Martinka M, Li G (2012) Sox4-mediated Dicerexpression is critical for suppression of melanoma cell invasion. OncogeneJamerson MH, Johnson MD, Dickson RB (2000) Dual regulation of proliferation and apoptosis:c-myc in bitransgenic murine mammary tumor models. Oncogene 19(8): 1065-71Jasiulionis MG, Luchessi AD, Moreira AG, Souza PP, Suenaga AP, Correa M, Costa CA, CuriR, Costa-Neto CM (2007) Inhibition of eukaryotic translation initiation factor 5A (eIF5A)hypusination impairs melanoma growth. Cell Biochem Funct 25(1): 109-14Jenkins ZA, Haag PG, Johansson HE (2001) Human eIF5A2 on chromosome 3q25-q27 is aphylogenetically conserved vertebrate variant of eukaryotic translation initiation factor 5A withtissue-specific expression. Genomics 71(1): 101-9Jiang Y, Muschel RJ (2002) Regulation of matrix metalloproteinase-9 (MMP-9) by translationalefficiency in murine prostate carcinoma cells. Cancer Res 62(6): 1910-4Johansson N, Ala-aho R, Uitto V, Grenman R, Fusenig NE, Lopez-Otin C, Kahari VM (2000)Expression of collagenase-3 (MMP-13) and collagenase-1 (MMP-1) by transformed145keratinocytes is dependent on the activity of p38 mitogen-activated protein kinase. J Cell Sci 113Pt 2: 227-35Johnstone RW, Ruefli AA, Lowe SW (2002) Apoptosis: a link between cancer genetics andchemotherapy. Cell 108(2): 153-64Jones RM, Branda J, Johnston KA, Polymenis M, Gadd M, Rustgi A, Callanan L, Schmidt EV(1996) An essential E box in the promoter of the gene encoding the mRNA cap-binding protein(eukaryotic initiation factor 4E) is a target for activation by c-myc. Mol Cell Biol 16(9): 4754-64Kang HA, Hershey JW (1994) Effect of initiation factor eIF-5A depletion on protein synthesisand proliferation of Saccharomyces cerevisiae. J Biol Chem 269(6): 3934-40Kang S, Bader AG, Zhao L, Vogt PK (2005) Mutated PI 3-kinases: cancer targets on a silverplatter. Cell Cycle 4(4): 578-81Kapp LD, Lorsch JR (2004) GTP-dependent recognition of the methionine moiety on initiatortRNA by translation factor eIF2. J Mol Biol 335(4): 923-36Karaskova M, Gunisova S, Herrmannova A, Wagner S, Munzarova V, Valasek L (2012)Functional characterization of the role of the N-terminal domain of the c/Nip1 subunit ofeukaryotic initiation factor 3 (eIF3) in AUG recognition. J Biol Chem 287(34): 28420-34Karjalainen JM, Kellokoski JK, Eskelinen MJ, Alhava EM, Kosma VM (1998) Downregulationof transcription factor AP-2 predicts poor survival in stage I cutaneous malignant melanoma. JClin Oncol 16(11): 3584-91Kau TR, Way JC, Silver PA (2004) Nuclear transport and cancer: from mechanism tointervention. Nat Rev Cancer 4(2): 106-17Keiper BD, Gan W, Rhoads RE (1999) Protein synthesis initiation factor 4G. Int J Biochem CellBiol 31(1): 37-41Kemper WM, Berry KW, Merrick WC (1976) Purification and properties of rabbit reticulocyteprotein synthesis initiation factors M2Balpha and M2Bbeta. J Biol Chem 251(18): 5551-7Kentsis A, Topisirovic I, Culjkovic B, Shao L, Borden KL (2004) Ribavirin suppresses eIF4E-mediated oncogenic transformation by physical mimicry of the 7-methyl guanosine mRNA cap.Proc Natl Acad Sci U S A 101(52): 18105-10Kevil CG, De Benedetti A, Payne DK, Coe LL, Laroux FS, Alexander JS (1996) Translationalregulation of vascular permeability factor by eukaryotic initiation factor 4E: implications fortumor angiogenesis. Int J Cancer 65(6): 785-90146Khosravi S, Wong RP, Ardekani GS, Zhang G, Martinka M, Ong CJ, Li G (2014) Role ofEIF5A2, a downstream target of Akt, in promoting melanoma cell invasion. Br J Cancer 110(2):399-408Kim YY, Von Weymarn L, Larsson O, Fan D, Underwood JM, Peterson MS, Hecht SS,Polunovsky VA, Bitterman PB (2009) Eukaryotic initiation factor 4E binding protein family ofproteins: sentinels at a translational control checkpoint in lung tumor defense. Cancer Res69(21): 8455-62Konicek BW, Stephens JR, McNulty AM, Robichaud N, Peery RB, Dumstorf CA, Dowless MS,Iversen PW, Parsons S, Ellis KE, McCann DJ, Pelletier J, Furic L, Yingling JM, Stancato LF,Sonenberg N, Graff JR (2011) Therapeutic inhibition of MAP kinase interacting kinase blockseukaryotic initiation factor 4E phosphorylation and suppresses outgrowth of experimental lungmetastases. Cancer Res 71(5): 1849-57Korneeva NL, First EA, Benoit CA, Rhoads RE (2005) Interaction between the NH2-terminaldomain of eIF4A and the central domain of eIF4G modulates RNA-stimulated ATPase activity. JBiol Chem 280(3): 1872-81Korneeva NL, Lamphear BJ, Hennigan FL, Merrick WC, Rhoads RE (2001) Characterization ofthe two eIF4A-binding sites on human eIF4G-1. J Biol Chem 276(4): 2872-9Kozak M (1986) Point mutations define a sequence flanking the AUG initiator codon thatmodulates translation by eukaryotic ribosomes. Cell 44(2): 283-92Kozak M (1987) At least six nucleotides preceding the AUG initiator codon enhance translationin mammalian cells. J Mol Biol 196(4): 947-50Kruse M, Rosorius O, Kratzer F, Bevec D, Kuhnt C, Steinkasserer A, Schuler G, Hauber J(2000) Inhibition of CD83 cell surface expression during dendritic cell maturation byinterference with nuclear export of CD83 mRNA. J Exp Med 191(9): 1581-90Kurisaki A, Kurisaki K, Kowanetz M, Sugino H, Yoneda Y, Heldin CH, Moustakas A (2006)The mechanism of nuclear export of Smad3 involves exportin 4 and Ran. Mol Cell Biol 26(4):1318-32Lamphear BJ, Kirchweger R, Skern T, Rhoads RE (1995) Mapping of functional domains ineukaryotic protein synthesis initiation factor 4G (eIF4G) with picornaviral proteases.Implications for cap-dependent and cap-independent translational initiation. J Biol Chem270(37): 21975-83Lankat-Buttgereit B, Goke R (2009) The tumour suppressor Pdcd4: recent advances in theelucidation of function and regulation. Biol Cell 101(6): 309-17147Larsson O, Li S, Issaenko OA, Avdulov S, Peterson M, Smith K, Bitterman PB, Polunovsky VA(2007) Eukaryotic translation initiation factor 4E induced progression of primary humanmammary epithelial cells along the cancer pathway is associated with targeted translationalderegulation of oncogenic drivers and inhibitors. Cancer Res 67(14): 6814-24Larsson O, Morita M, Topisirovic I, Alain T, Blouin MJ, Pollak M, Sonenberg N (2012) Distinctperturbation of the translatome by the antidiabetic drug metformin. Proc Natl Acad Sci U S A109(23): 8977-82Lawson TG, Lee KA, Maimone MM, Abramson RD, Dever TE, Merrick WC, Thach RE (1989)Dissociation of double-stranded polynucleotide helical structures by eukaryotic initiation factors,as revealed by a novel assay. Biochemistry 28(11): 4729-34Lazaris-Karatzas A, Montine KS, Sonenberg N (1990) Malignant transformation by a eukaryoticinitiation factor subunit that binds to mRNA 5' cap. Nature 345(6275): 544-7Lazaris-Karatzas A, Smith MR, Frederickson RM, Jaramillo ML, Liu YL, Kung HF, SonenbergN (1992) Ras mediates translation initiation factor 4E-induced malignant transformation. GenesDev 6(9): 1631-42Le H, Tanguay RL, Balasta ML, Wei CC, Browning KS, Metz AM, Goss DJ, Gallie DR (1997)Translation initiation factors eIF-iso4G and eIF-4B interact with the poly(A)-binding protein andincrease its RNA binding activity. J Biol Chem 272(26): 16247-55Lee NP, Tsang FH, Shek FH, Mao M, Dai H, Zhang C, Dong S, Guan XY, Poon RT, Luk JM(2010) Prognostic significance and therapeutic potential of eukaryotic translation initiation factor5A (eIF5A) in hepatocellular carcinoma. Int J Cancer 127(4): 968-76Lens M (2008) Current clinical overview of cutaneous melanoma. Br J Nurs 17(5): 300-5Li BD, Gruner JS, Abreo F, Johnson LW, Yu H, Nawas S, McDonald JC, DeBenedetti A (2002)Prospective study of eukaryotic initiation factor 4E protein elevation and breast cancer outcome.Ann Surg 235(5): 732-8; discussion 738-9Li BD, McDonald JC, Nassar R, De Benedetti A (1998) Clinical outcome in stage I to III breastcarcinoma and eIF4E overexpression. Ann Surg 227(5): 756-6l; discussion 761-3Li CH, Ohn T, Ivanov P, Tisdale S, Anderson P (2010) eIF5A promotes translation elongation,polysome disassembly and stress granule assembly. PLoS One 5(4): e9942Li J, Martinka M, Li G (2008) Role of ING4 in human melanoma cell migration, invasion andpatient survival. Carcinogenesis 29(7): 1373-9Li S, Jia Y, Jacobson B, McCauley J, Kratzke R, Bitterman PB, Wagner CR (2013) Treatment ofbreast and lung cancer cells with a N-7 benzyl guanosine monophosphate tryptamine148phosphoramidate pronucleotide (4Ei-1) results in chemosensitization to gemcitabine and inducedeIF4E proteasomal degradation. Mol Pharm 10(2): 523-31Li S, Takasu T, Perlman DM, Peterson MS, Burrichter D, Avdulov S, Bitterman PB, PolunovskyVA (2003) Translation factor eIF4E rescues cells from Myc-dependent apoptosis by inhibitingcytochrome c release. J Biol Chem 278(5): 3015-22Li Y, Fan S, Koo J, Yue P, Chen ZG, Owonikoko TK, Ramalingam SS, Khuri FR, Sun SY(2012) Elevated expression of eukaryotic translation initiation factor 4E is associated withproliferation, invasion and acquired resistance to erlotinib in lung cancer. Cancer Biol Ther13(5): 272-80Li Y, Fu L, Li JB, Qin Y, Zeng TT, Zhou J, Zeng ZL, Chen J, Cao TT, Ban X, Qian C, Cai Z,Xie D, Huang P, Guan XY (2014) Increased expression of EIF5A2, via hypoxia or geneamplification, contributes to metastasis and angiogenesis of esophageal squamous cellcarcinoma. Gastroenterology 146(7): 1701-13 e9Lin CJ, Cencic R, Mills JR, Robert F, Pelletier J (2008) c-Myc and eIF4F are components of afeedforward loop that links transcription and translation. Cancer Res 68(13): 5326-34Lin H, Wong RP, Martinka M, Li G (2009) Loss of SNF5 expression correlates with poor patientsurvival in melanoma. Clin Cancer Res 15(20): 6404-11Lin JY, Fisher DE (2007) Melanocyte biology and skin pigmentation. Nature 445(7130): 843-50Lipowsky G, Bischoff FR, Schwarzmaier P, Kraft R, Kostka S, Hartmann E, Kutay U, Gorlich D(2000) Exportin 4: a mediator of a novel nuclear export pathway in higher eukaryotes. EMBO J19(16): 4362-71Lomakin IB, Kolupaeva VG, Marintchev A, Wagner G, Pestova TV (2003) Position ofeukaryotic initiation factor eIF1 on the 40S ribosomal subunit determined by directed hydroxylradical probing. Genes Dev 17(22): 2786-97Lorsch JR, Dever TE (2010) Molecular view of 43 S complex formation and start site selectionin eukaryotic translation initiation. J Biol Chem 285(28): 21203-7Lukas J, Herzinger T, Hansen K, Moroni MC, Resnitzky D, Helin K, Reed SI, Bartek J (1997)Cyclin E-induced S phase without activation of the pRb/E2F pathway. Genes Dev 11(11): 1479-92Lukhele S, Bah A, Lin H, Sonenberg N, Forman-Kay JD (2013) Interaction of the eukaryoticinitiation factor 4E with 4E-BP2 at a dynamic bipartite interface. Structure 21(12): 2186-96149Luo JH, Hua WF, Rao HL, Liao YJ, Kung HF, Zeng YX, Guan XY, Chen W, Xie D (2009)Overexpression of EIF-5A2 predicts tumor recurrence and progression in pTa/pT1 urothelialcarcinoma of the bladder. Cancer Sci 100(5): 896-902Maag D, Algire MA, Lorsch JR (2006) Communication between eukaryotic translation initiationfactors 5 and 1A within the ribosomal pre-initiation complex plays a role in start site selection. JMol Biol 356(3): 724-37Maag D, Fekete CA, Gryczynski Z, Lorsch JR (2005) A conformational change in the eukaryotictranslation preinitiation complex and release of eIF1 signal recognition of the start codon. MolCell 17(2): 265-75MacKie RM (2000) Malignant melanoma: clinical variants and prognostic indicators. Clin ExpDermatol 25(6): 471-5Mader S, Lee H, Pause A, Sonenberg N (1995) The translation initiation factor eIF-4E binds to acommon motif shared by the translation factor eIF-4 gamma and the translational repressors 4E-binding proteins. Mol Cell Biol 15(9): 4990-7Madhunapantula SV, Robertson GP (2009) The PTEN-AKT3 signaling cascade as a therapeutictarget in melanoma. Pigment Cell Melanoma Res 22(4): 400-19Maier B, Ogihara T, Trace AP, Tersey SA, Robbins RD, Chakrabarti SK, Nunemaker CS, StullND, Taylor CA, Thompson JE, Dondero RS, Lewis EC, Dinarello CA, Nadler JL, Mirmira RG(2010a) The unique hypusine modification of eIF5A promotes islet beta cell inflammation anddysfunction in mice. J Clin Invest 120(6): 2156-70Maier B, Tersey SA, Mirmira RG (2010b) Hypusine: a new target for therapeutic intervention indiabetic inflammation. Discov Med 10(50): 18-23Mamane Y, Petroulakis E, Martineau Y, Sato TA, Larsson O, Rajasekhar VK, Sonenberg N(2007) Epigenetic activation of a subset of mRNAs by eIF4E explains its effects on cellproliferation. PLoS One 2(2): e242Mangus DA, Evans MC, Jacobson A (2003) Poly(A)-binding proteins: multifunctional scaffoldsfor the post-transcriptional control of gene expression. Genome Biol 4(7): 223Mar V, Roberts H, Wolfe R, English DR, Kelly JW (2013) Nodular melanoma: a distinct clinicalentity and the largest contributor to melanoma deaths in Victoria, Australia. J Am Acad Dermatol68(4): 568-75Marchet A, Mocellin S, Belluco C, Ambrosi A, DeMarchi F, Mammano E, Digito M, Leon A,D'Arrigo A, Lise M, Nitti D (2007) Gene expression profile of primary gastric cancer: towardsthe prediction of lymph node status. Ann Surg Oncol 14(3): 1058-64150Marcotrigiano J, Gingras AC, Sonenberg N, Burley SK (1997) Cocrystal structure of themessenger RNA 5' cap-binding protein (eIF4E) bound to 7-methyl-GDP. Cell 89(6): 951-61Marcotrigiano J, Gingras AC, Sonenberg N, Burley SK (1999) Cap-dependent translationinitiation in eukaryotes is regulated by a molecular mimic of eIF4G. Mol Cell 3(6): 707-16Marintchev A, Edmonds KA, Marintcheva B, Hendrickson E, Oberer M, Suzuki C, Herdy B,Sonenberg N, Wagner G (2009) Topology and regulation of the human eIF4A/4G/4H helicasecomplex in translation initiation. Cell 136(3): 447-60Marintchev A, Kolupaeva VG, Pestova TV, Wagner G (2003) Mapping the binding interfacebetween human eukaryotic initiation factors 1A and 5B: a new interaction between old partners.Proc Natl Acad Sci U S A 100(4): 1535-40Martin ME, Perez MI, Redondo C, Alvarez MI, Salinas M, Fando JL (2000) 4E binding protein1 expression is inversely correlated to the progression of gastrointestinal cancers. Int J BiochemCell Biol 32(6): 633-42Matsuo H, Li H, McGuire AM, Fletcher CM, Gingras AC, Sonenberg N, Wagner G (1997)Structure of translation factor eIF4E bound to m7GDP and interaction with 4E-binding protein.Nat Struct Biol 4(9): 717-24McClusky DR, Chu Q, Yu H, Debenedetti A, Johnson LW, Meschonat C, Turnage R, McDonaldJC, Abreo F, Li BD (2005) A prospective trial on initiation factor 4E (eIF4E) overexpression andcancer recurrence in node-positive breast cancer. Ann Surg 242(4): 584-90; discussion 590-2Meric-Bernstam F, Chen H, Akcakanat A, Do KA, Lluch A, Hennessy BT, Hortobagyi GN,Mills GB, Gonzalez-Angulo A (2012) Aberrations in translational regulation are associated withpoor prognosis in hormone receptor-positive breast cancer. Breast Cancer Res 14(5): R138Methot N, Song MS, Sonenberg N (1996) A region rich in aspartic acid, arginine, tyrosine, andglycine (DRYG) mediates eukaryotic initiation factor 4B (eIF4B) self-association and interactionwith eIF3. Mol Cell Biol 16(10): 5328-34Miller AJ, Mihm MC, Jr. (2006) Melanoma. N Engl J Med 355(1): 51-65Miluzio A, Beugnet A, Volta V, Biffo S (2009) Eukaryotic initiation factor 6 mediates acontinuum between 60S ribosome biogenesis and translation. EMBO Rep 10(5): 459-65Minotti G, Menna P, Salvatorelli E, Cairo G, Gianni L (2004) Anthracyclines: molecularadvances and pharmacologic developments in antitumor activity and cardiotoxicity. PharmacolRev 56(2): 185-229151Mitchell SF, Walker SE, Algire MA, Park EH, Hinnebusch AG, Lorsch JR (2010) The 5'-7-methylguanosine cap on eukaryotic mRNAs serves both to stimulate canonical translationinitiation and to block an alternative pathway. Mol Cell 39(6): 950-62Mitkevich VA, Kononenko AV, Petrushanko IY, Yanvarev DV, Makarov AA, Kisselev LL(2006) Termination of translation in eukaryotes is mediated by the quaternaryeRF1*eRF3*GTP*Mg2+ complex. The biological roles of eRF3 and prokaryotic RF3 areprofoundly distinct. Nucleic Acids Res 34(14): 3947-54Miyamura Y, Coelho SG, Wolber R, Miller SA, Wakamatsu K, Zmudzka BZ, Ito S, Smuda C,Passeron T, Choi W, Batzer J, Yamaguchi Y, Beer JZ, Hearing VJ (2007) Regulation of humanskin pigmentation and responses to ultraviolet radiation. Pigment Cell Res 20(1): 2-13Mizuno A, In Y, Fujita Y, Abiko F, Miyagawa H, Kitamura K, Tomoo K, Ishida T (2008)Importance of C-terminal flexible region of 4E-binding protein in binding with eukaryoticinitiation factor 4E. FEBS Lett 582(23-24): 3439-44Moerke NJ, Aktas H, Chen H, Cantel S, Reibarkh MY, Fahmy A, Gross JD, Degterev A, Yuan J,Chorev M, Halperin JA, Wagner G (2007) Small-molecule inhibition of the interaction betweenthe translation initiation factors eIF4E and eIF4G. Cell 128(2): 257-67Myers MP, Pass I, Batty IH, Van der Kaay J, Stolarov JP, Hemmings BA, Wigler MH, DownesCP, Tonks NK (1998) The lipid phosphatase activity of PTEN is critical for its tumor supressorfunction. Proc Natl Acad Sci U S A 95(23): 13513-8Nanda JS, Cheung YN, Takacs JE, Martin-Marcos P, Saini AK, Hinnebusch AG, Lorsch JR(2009) eIF1 controls multiple steps in start codon recognition during eukaryotic translationinitiation. J Mol Biol 394(2): 268-85Nanda JS, Saini AK, Munoz AM, Hinnebusch AG, Lorsch JR (2013) Coordinated movements ofeukaryotic translation initiation factors eIF1, eIF1A, and eIF5 trigger phosphate release fromeIF2 in response to start codon recognition by the ribosomal preinitiation complex. J Biol Chem288(8): 5316-29Nasr Z, Robert F, Porco JA, Jr., Muller WJ, Pelletier J (2013) eIF4F suppression in breast canceraffects maintenance and progression. Oncogene 32(7): 861-71Nathan CO, Amirghahari N, Abreo F, Rong X, Caldito G, Jones ML, Zhou H, Smith M,Kimberly D, Glass J (2004) Overexpressed eIF4E is functionally active in surgical margins ofhead and neck cancer patients via activation of the Akt/mammalian target of rapamycin pathway.Clin Cancer Res 10(17): 5820-7Nathan CO, Franklin S, Abreo FW, Nassar R, De Benedetti A, Glass J (1999) Analysis ofsurgical margins with the molecular marker eIF4E: a prognostic factor in patients with head andneck cancer. J Clin Oncol 17(9): 2909-14152Nathan CO, Liu L, Li BD, Abreo FW, Nandy I, De Benedetti A (1997) Detection of the proto-oncogene eIF4E in surgical margins may predict recurrence in head and neck cancer. Oncogene15(5): 579-84Naveau M, Lazennec-Schurdevin C, Panvert M, Mechulam Y, Schmitt E (2010) tRNA bindingproperties of eukaryotic translation initiation factor 2 from Encephalitozoon cuniculi.Biochemistry 49(40): 8680-8Nielsen KH, Behrens MA, He Y, Oliveira CL, Jensen LS, Hoffmann SV, Pedersen JS, AndersenGR (2011) Synergistic activation of eIF4A by eIF4B and eIF4G. Nucleic Acids Res 39(7): 2678-89Niu Z, Wang J, Muhammad S, Niu W, Liu E, Peng C, Liang B, Sun Q, Obo S, He Z, Liu S, ZouX, Niu J (2014) Protein expression of eIF4E and integrin alphavbeta6 in colon cancer can predictclinical significance, reveal their correlation and imply possible mechanism of interaction. CellBiosci 4: 23Nogueira V, Sundararajan D, Kwan JM, Peng XD, Sarvepalli N, Sonenberg N, Hay N (2012)Akt-dependent Skp2 mRNA translation is required for exiting contact inhibition, oncogenesis,and adipogenesis. EMBO J 31(5): 1134-46Noske A, Lindenberg JL, Darb-Esfahani S, Weichert W, Buckendahl AC, Roske A, Sehouli J,Dietel M, Denkert C (2008) Activation of mTOR in a subgroup of ovarian carcinomas:correlation with p-eIF-4E and prognosis. Oncol Rep 20(6): 1409-17O'Reilly KE, Warycha M, Davies MA, Rodrik V, Zhou XK, Yee H, Polsky D, Pavlick AC,Rosen N, Bhardwaj N, Mills G, Osman I (2009) Phosphorylated 4E-BP1 is associated with poorsurvival in melanoma. Clin Cancer Res 15(8): 2872-8Oetting WS, Fryer JP, Shriram S, King RA (2003) Oculocutaneous albinism type 1: the last 100years. Pigment Cell Res 16(3): 307-11Oliveria SA, Saraiya M, Geller AC, Heneghan MK, Jorgensen C (2006) Sun exposure and riskof melanoma. Arch Dis Child 91(2): 131-8Olsen DS, Savner EM, Mathew A, Zhang F, Krishnamoorthy T, Phan L, Hinnebusch AG (2003)Domains of eIF1A that mediate binding to eIF2, eIF3 and eIF5B and promote ternary complexrecruitment in vivo. EMBO J 22(2): 193-204Omholt K, Krockel D, Ringborg U, Hansson J (2006) Mutations of PIK3CA are rare incutaneous melanoma. Melanoma Res 16(2): 197-200153Oridate N, Kim HJ, Xu X, Lotan R (2005) Growth inhibition of head and neck squamouscarcinoma cells by small interfering RNAs targeting eIF4E or cyclin D1 alone or combined withcisplatin. Cancer Biol Ther 4(3): 318-23Osaki M, Oshimura M, Ito H (2004) PI3K-Akt pathway: its functions and alterations in humancancer. Apoptosis 9(6): 667-76Ozes AR, Feoktistova K, Avanzino BC, Fraser CS (2011) Duplex unwinding and ATPaseactivities of the DEAD-box helicase eIF4A are coupled by eIF4G and eIF4B. J Mol Biol 412(4):674-87Papp T, Pemsel H, Zimmermann R, Bastrop R, Weiss DG, Schiffmann D (1999) Mutationalanalysis of the N-ras, p53, p16INK4a, CDK4, and MC1R genes in human congenitalmelanocytic naevi. J Med Genet 36(8): 610-4Park EH, Walker SE, Lee JM, Rothenburg S, Lorsch JR, Hinnebusch AG (2011) Multipleelements in the eIF4G1 N-terminus promote assembly of eIF4G1*PABP mRNPs in vivo. EMBOJ 30(2): 302-16Park MH (2006) The post-translational synthesis of a polyamine-derived amino acid, hypusine,in the eukaryotic translation initiation factor 5A (eIF5A). J Biochem 139(2): 161-9Passmore LA, Schmeing TM, Maag D, Applefield DJ, Acker MG, Algire MA, Lorsch JR,Ramakrishnan V (2007) The eukaryotic translation initiation factors eIF1 and eIF1A induce anopen conformation of the 40S ribosome. Mol Cell 26(1): 41-50Patel PH, Costa-Mattioli M, Schulze KL, Bellen HJ (2009) The Drosophila deoxyhypusinehydroxylase homologue nero and its target eIF5A are required for cell growth and the regulationof autophagy. J Cell Biol 185(7): 1181-94Paulin FE, Campbell LE, O'Brien K, Loughlin J, Proud CG (2001) Eukaryotic translationinitiation factor 5 (eIF5) acts as a classical GTPase-activator protein. Curr Biol 11(1): 55-9Pause A, Belsham GJ, Gingras AC, Donze O, Lin TA, Lawrence JC, Jr., Sonenberg N (1994)Insulin-dependent stimulation of protein synthesis by phosphorylation of a regulator of 5'-capfunction. Nature 371(6500): 762-7Perlis C, Herlyn M (2004) Recent advances in melanoma biology. Oncologist 9(2): 182-7Persad S, Attwell S, Gray V, Delcommenne M, Troussard A, Sanghera J, Dedhar S (2000)Inhibition of integrin-linked kinase (ILK) suppresses activation of protein kinase B/Akt andinduces cell cycle arrest and apoptosis of PTEN-mutant prostate cancer cells. Proc Natl Acad SciU S A 97(7): 3207-12154Pestova TV, Borukhov SI, Hellen CU (1998) Eukaryotic ribosomes require initiation factors 1and 1A to locate initiation codons. Nature 394(6696): 854-9Pestova TV, Kolupaeva VG (2002) The roles of individual eukaryotic translation initiationfactors in ribosomal scanning and initiation codon selection. Genes Dev 16(22): 2906-22Peterson DT, Merrick WC, Safer B (1979) Binding and release of radiolabeled eukaryoticinitiation factors 2 and 3 during 80 S initiation complex formation. J Biol Chem 254(7): 2509-16Petricoin EF, 3rd, Espina V, Araujo RP, Midura B, Yeung C, Wan X, Eichler GS, Johann DJ, Jr.,Qualman S, Tsokos M, Krishnan K, Helman LJ, Liotta LA (2007) Phosphoprotein pathwaymapping: Akt/mammalian target of rapamycin activation is negatively associated with childhoodrhabdomyosarcoma survival. Cancer Res 67(7): 3431-40Phillips A, Blaydes JP (2008) MNK1 and EIF4E are downstream effectors of MEKs in theregulation of the nuclear export of HDM2 mRNA. Oncogene 27(11): 1645-9Piris A, Mihm MC, Jr. (2009) Progress in melanoma histopathology and diagnosis. HematolOncol Clin North Am 23(3): 467-80, viiiPisarev AV, Kolupaeva VG, Pisareva VP, Merrick WC, Hellen CU, Pestova TV (2006) Specificfunctional interactions of nucleotides at key -3 and +4 positions flanking the initiation codonwith components of the mammalian 48S translation initiation complex. Genes Dev 20(5): 624-36Pisareva VP, Pisarev AV, Hellen CU, Rodnina MV, Pestova TV (2006) Kinetic analysis ofinteraction of eukaryotic release factor 3 with guanine nucleotides. J Biol Chem 281(52): 40224-35Platz A, Egyhazi S, Ringborg U, Hansson J (2008) Human cutaneous melanoma; a review ofNRAS and BRAF mutation frequencies in relation to histogenetic subclass and body site. MolOncol 1(4): 395-405Pola C, Formenti SC, Schneider RJ (2013) Vitronectin-alphavbeta3 integrin engagement directshypoxia-resistant mTOR activity and sustained protein synthesis linked to invasion by breastcancer cells. Cancer Res 73(14): 4571-8Pollock PM, Harper UL, Hansen KS, Yudt LM, Stark M, Robbins CM, Moses TY, Hostetter G,Wagner U, Kakareka J, Salem G, Pohida T, Heenan P, Duray P, Kallioniemi O, Hayward NK,Trent JM, Meltzer PS (2003) High frequency of BRAF mutations in nevi. Nat Genet 33(1): 19-20Polunovsky VA, Gingras AC, Sonenberg N, Peterson M, Tan A, Rubins JB, Manivel JC,Bitterman PB (2000) Translational control of the antiapoptotic function of Ras. J Biol Chem275(32): 24776-80155Polunovsky VA, Rosenwald IB, Tan AT, White J, Chiang L, Sonenberg N, Bitterman PB (1996)Translational control of programmed cell death: eukaryotic translation initiation factor 4E blocksapoptosis in growth-factor-restricted fibroblasts with physiologically expressed or deregulatedMyc. Mol Cell Biol 16(11): 6573-81Pomerantz J, Schreiber-Agus N, Liegeois NJ, Silverman A, Alland L, Chin L, Potes J, Chen K,Orlow I, Lee HW, Cordon-Cardo C, DePinho RA (1998) The Ink4a tumor suppressor geneproduct, p19Arf, interacts with MDM2 and neutralizes MDM2's inhibition of p53. Cell 92(6):713-23Porras BH, Cockerell CJ (1997) Cutaneous malignant melanoma: classification and clinicaldiagnosis. Semin Cutan Med Surg 16(2): 88-96Pourdehnad M, Truitt ML, Siddiqi IN, Ducker GS, Shokat KM, Ruggero D (2013) Myc andmTOR converge on a common node in protein synthesis control that confers synthetic lethalityin Myc-driven cancers. Proc Natl Acad Sci U S A 110(29): 11988-93Poyry TA, Kaminski A, Jackson RJ (2004) What determines whether mammalian ribosomesresume scanning after translation of a short upstream open reading frame? Genes Dev 18(1): 62-75Preukschas M, Hagel C, Schulte A, Weber K, Lamszus K, Sievert H, Pallmann N, Bokemeyer C,Hauber J, Braig M, Balabanov S (2012) Expression of eukaryotic initiation factor 5A andhypusine forming enzymes in glioblastoma patient samples: implications for new targetedtherapies. PLoS One 7(8): e43468Prevot D, Darlix JL, Ohlmann T (2003) Conducting the initiation of protein synthesis: the role ofeIF4G. Biol Cell 95(3-4): 141-56Pritchard LE, Turnbull AV, White A (2002) Pro-opiomelanocortin processing in thehypothalamus: impact on melanocortin signalling and obesity. J Endocrinol 172(3): 411-21Prota G (1980) Recent advances in the chemistry of melanogenesis in mammals. J InvestDermatol 75(1): 122-7Proud CG (2005) eIF2 and the control of cell physiology. Semin Cell Dev Biol 16(1): 3-12Provenzani A, Fronza R, Loreni F, Pascale A, Amadio M, Quattrone A (2006) Global alterationsin mRNA polysomal recruitment in a cell model of colorectal cancer progression to metastasis.Carcinogenesis 27(7): 1323-33Pyronnet S, Imataka H, Gingras AC, Fukunaga R, Hunter T, Sonenberg N (1999) Humaneukaryotic translation initiation factor 4G (eIF4G) recruits mnk1 to phosphorylate eIF4E. EMBOJ 18(1): 270-9156Qin H, Sun Y, Benveniste EN (1999) The transcription factors Sp1, Sp3, and AP-2 are requiredfor constitutive matrix metalloproteinase-2 gene expression in astroglioma cells. J Biol Chem274(41): 29130-7Quelle DE, Zindy F, Ashmun RA, Sherr CJ (1995) Alternative reading frames of the INK4atumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell83(6): 993-1000Rabl J, Leibundgut M, Ataide SF, Haag A, Ban N (2011) Crystal structure of the eukaryotic 40Sribosomal subunit in complex with initiation factor 1. Science 331(6018): 730-6Ramaswamy S, Ross KN, Lander ES, Golub TR (2003) A molecular signature of metastasis inprimary solid tumors. Nat Genet 33(1): 49-54Rees JL (2003) Genetics of hair and skin color. Annu Rev Genet 37: 67-90Ries LA, Wingo PA, Miller DS, Howe HL, Weir HK, Rosenberg HM, Vernon SW, Cronin K,Edwards BK (2000) The annual report to the nation on the status of cancer, 1973-1997, with aspecial section on colorectal cancer. Cancer 88(10): 2398-424Rigel DS (2010) Epidemiology of melanoma. Semin Cutan Med Surg 29(4): 204-9Robichaud N, del Rincon SV, Huor B, Alain T, Petruccelli LA, Hearnden J, Goncalves C,Grotegut S, Spruck CH, Furic L, Larsson O, Muller WJ, Miller WH, Sonenberg N (2015)Phosphorylation of eIF4E promotes EMT and metastasis via translational control of SNAIL andMMP-3. Oncogene 34(16): 2032-42Rogers GW, Jr., Richter NJ, Lima WF, Merrick WC (2001) Modulation of the helicase activityof eIF4A by eIF4B, eIF4H, and eIF4F. J Biol Chem 276(33): 30914-22Rogers GW, Jr., Richter NJ, Merrick WC (1999) Biochemical and kinetic characterization of theRNA helicase activity of eukaryotic initiation factor 4A. J Biol Chem 274(18): 12236-44Rojo F, Najera L, Lirola J, Jimenez J, Guzman M, Sabadell MD, Baselga J, Ramon y Cajal S(2007) 4E-binding protein 1, a cell signaling hallmark in breast cancer that correlates withpathologic grade and prognosis. Clin Cancer Res 13(1): 81-9Rosenwald IB, Kaspar R, Rousseau D, Gehrke L, Leboulch P, Chen JJ, Schmidt EV, SonenbergN, London IM (1995) Eukaryotic translation initiation factor 4E regulates expression of cyclinD1 at transcriptional and post-transcriptional levels. J Biol Chem 270(36): 21176-80Rotte A, Martinka M, Li G (2012) MMP2 expression is a prognostic marker for primarymelanoma patients. Cell Oncol (Dordr)157Rousseau D, Kaspar R, Rosenwald I, Gehrke L, Sonenberg N (1996) Translation initiation ofornithine decarboxylase and nucleocytoplasmic transport of cyclin D1 mRNA are increased incells overexpressing eukaryotic initiation factor 4E. Proc Natl Acad Sci U S A 93(3): 1065-70Rozen F, Edery I, Meerovitch K, Dever TE, Merrick WC, Sonenberg N (1990) BidirectionalRNA helicase activity of eucaryotic translation initiation factors 4A and 4F. Mol Cell Biol 10(3):1134-44Rozovsky N, Butterworth AC, Moore MJ (2008) Interactions between eIF4AI and its accessoryfactors eIF4B and eIF4H. RNA 14(10): 2136-48Ruggero D (2013) Translational control in cancer etiology. Cold Spring Harb Perspect Biol 5(2)Ruggero D, Montanaro L, Ma L, Xu W, Londei P, Cordon-Cardo C, Pandolfi PP (2004) Thetranslation factor eIF-4E promotes tumor formation and cooperates with c-Myc inlymphomagenesis. Nat Med 10(5): 484-6Runger TM (1999) Role of UVA in the pathogenesis of melanoma and non-melanoma skincancer. A short review. Photodermatol Photoimmunol Photomed 15(6): 212-6Sachs AB, Sarnow P, Hentze MW (1997) Starting at the beginning, middle, and end: translationinitiation in eukaryotes. Cell 89(6): 831-8Saini P, Eyler DE, Green R, Dever TE (2009) Hypusine-containing protein eIF5A promotestranslation elongation. Nature 459(7243): 118-21Saito H, Hayday AC, Wiman K, Hayward WS, Tonegawa S (1983) Activation of the c-myc geneby translocation: a model for translational control. Proc Natl Acad Sci U S A 80(24): 7476-80Salehi Z, Mashayekhi F (2006) Expression of the eukaryotic translation initiation factor 4E(eIF4E) and 4E-BP1 in esophageal cancer. Clin Biochem 39(4): 404-9Sato H, Takino T, Okada Y, Cao J, Shinagawa A, Yamamoto E, Seiki M (1994) A matrixmetalloproteinase expressed on the surface of invasive tumour cells. Nature 370(6484): 61-5Scheper GC, Proud CG (2002) Does phosphorylation of the cap-binding protein eIF4E play arole in translation initiation? Eur J Biochem 269(22): 5350-9Scheper GC, van Kollenburg B, Hu J, Luo Y, Goss DJ, Proud CG (2002) Phosphorylation ofeukaryotic initiation factor 4E markedly reduces its affinity for capped mRNA. J Biol Chem277(5): 3303-9Schmeing TM, Voorhees RM, Kelley AC, Gao YG, Murphy FVt, Weir JR, Ramakrishnan V(2009) The crystal structure of the ribosome bound to EF-Tu and aminoacyl-tRNA. Science326(5953): 688-94158Schmitt E, Naveau M, Mechulam Y (2010) Eukaryotic and archaeal translation initiation factor2: a heterotrimeric tRNA carrier. FEBS Lett 584(2): 405-12Schrader R, Young C, Kozian D, Hoffmann R, Lottspeich F (2006) Temperature-sensitive eIF5Amutant accumulates transcripts targeted to the nonsense-mediated decay pathway. J Biol Chem281(46): 35336-46Schutz P, Bumann M, Oberholzer AE, Bieniossek C, Trachsel H, Altmann M, Baumann U(2008) Crystal structure of the yeast eIF4A-eIF4G complex: an RNA-helicase controlled byprotein-protein interactions. Proc Natl Acad Sci U S A 105(28): 9564-9Scuoppo C, Miething C, Lindqvist L, Reyes J, Ruse C, Appelmann I, Yoon S, Krasnitz A,Teruya-Feldstein J, Pappin D, Pelletier J, Lowe SW (2012) A tumour suppressor network relyingon the polyamine-hypusine axis. Nature 487(7406): 244-8Searfoss A, Dever TE, Wickner R (2001) Linking the 3' poly(A) tail to the subunit joining stepof translation initiation: relations of Pab1p, eukaryotic translation initiation factor 5b (Fun12p),and Ski2p-Slh1p. Mol Cell Biol 21(15): 4900-8Seki N, Takasu T, Mandai K, Nakata M, Saeki H, Heike Y, Takata I, Segawa Y, Hanafusa T,Eguchi K (2002) Expression of eukaryotic initiation factor 4E in atypical adenomatoushyperplasia and adenocarcinoma of the human peripheral lung. Clin Cancer Res 8(10): 3046-53Seki N, Takasu T, Sawada S, Nakata M, Nishimura R, Segawa Y, Shibakuki R, Hanafusa T,Eguchi K (2010) Prognostic significance of expression of eukaryotic initiation factor 4E and 4Ebinding protein 1 in patients with pathological stage I invasive lung adenocarcinoma. LungCancer 70(3): 329-34Shah P, Ding Y, Niemczyk M, Kudla G, Plotkin JB (2013) Rate-limiting steps in yeast proteintranslation. Cell 153(7): 1589-601Shankar J, Messenberg A, Chan J, Underhill TM, Foster LJ, Nabi IR (2010) Pseudopodial actindynamics control epithelial-mesenchymal transition in metastatic cancer cells. Cancer Res 70(9):3780-90Sharpless E, Chin L (2003) The INK4a/ARF locus and melanoma. Oncogene 22(20): 3092-8Shaverdashvili K, Wong P, Ma J, Zhang K, Osman I, Bedogni B (2014) MT1-MMP modulatesmelanoma cell dissemination and metastasis through activation of MMP2 and RAC1. PigmentCell Melanoma Res 27(2): 287-96Shek FH, Fatima S, Lee NP (2012) Implications of the Use of Eukaryotic Translation InitiationFactor 5A (eIF5A) for Prognosis and Treatment of Hepatocellular Carcinoma. Int J Hepatol2012: 760928159Shin BS, Kim JR, Walker SE, Dong J, Lorsch JR, Dever TE (2011) Initiation factor eIF2gammapromotes eIF2-GTP-Met-tRNAi(Met) ternary complex binding to the 40S ribosome. Nat StructMol Biol 18(11): 1227-34Shin BS, Maag D, Roll-Mecak A, Arefin MS, Burley SK, Lorsch JR, Dever TE (2002)Uncoupling of initiation factor eIF5B/IF2 GTPase and translational activities by mutations thatlower ribosome affinity. Cell 111(7): 1015-25Silvera D, Formenti SC, Schneider RJ (2010) Translational control in cancer. Nat Rev Cancer10(4): 254-66Simon C, Goepfert H, Boyd D (1998) Inhibition of the p38 mitogen-activated protein kinase bySB 203580 blocks PMA-induced Mr 92,000 type IV collagenase secretion and in vitro invasion.Cancer Res 58(6): 1135-9Simonetti A, Marzi S, Myasnikov AG, Fabbretti A, Yusupov M, Gualerzi CO, Klaholz BP(2008) Structure of the 30S translation initiation complex. Nature 455(7211): 416-20Siridechadilok B, Fraser CS, Hall RJ, Doudna JA, Nogales E (2005) Structural roles for humantranslation factor eIF3 in initiation of protein synthesis. Science 310(5753): 1513-5Slepenkov SV, Darzynkiewicz E, Rhoads RE (2006) Stopped-flow kinetic analysis of eIF4E andphosphorylated eIF4E binding to cap analogs and capped oligoribonucleotides: evidence for aone-step binding mechanism. J Biol Chem 281(21): 14927-38Slominski A, Tobin DJ, Shibahara S, Wortsman J (2004) Melanin pigmentation in mammalianskin and its hormonal regulation. Physiol Rev 84(4): 1155-228Slominski A, Wortsman J, Luger T, Paus R, Solomon S (2000) Corticotropin releasing hormoneand proopiomelanocortin involvement in the cutaneous response to stress. Physiol Rev 80(3):979-1020Smalley KS, Herlyn M (2005) Targeting intracellular signaling pathways as a novel strategy inmelanoma therapeutics. Ann N Y Acad Sci 1059: 16-25Sokabe M, Fraser CS, Hershey JW (2012) The human translation initiation multi-factor complexpromotes methionyl-tRNAi binding to the 40S ribosomal subunit. Nucleic Acids Res 40(2): 905-13Sonenberg N, Hinnebusch AG (2009) Regulation of translation initiation in eukaryotes:mechanisms and biological targets. Cell 136(4): 731-45160Sonenberg N, Morgan MA, Merrick WC, Shatkin AJ (1978) A polypeptide in eukaryoticinitiation factors that crosslinks specifically to the 5'-terminal cap in mRNA. Proc Natl Acad SciU S A 75(10): 4843-7Song H, Mugnier P, Das AK, Webb HM, Evans DR, Tuite MF, Hemmings BA, Barford D(2000) The crystal structure of human eukaryotic release factor eRF1--mechanism of stop codonrecognition and peptidyl-tRNA hydrolysis. Cell 100(3): 311-21Spirin AS (2009) How does a scanning ribosomal particle move along the 5'-untranslated regionof eukaryotic mRNA? Brownian Ratchet model. Biochemistry 48(45): 10688-92Stahl JM, Cheung M, Sharma A, Trivedi NR, Shanmugam S, Robertson GP (2003) Loss ofPTEN promotes tumor development in malignant melanoma. Cancer Res 63(11): 2881-90Stahl JM, Sharma A, Cheung M, Zimmerman M, Cheng JQ, Bosenberg MW, Kester M,Sandirasegarane L, Robertson GP (2004) Deregulated Akt3 activity promotes development ofmalignant melanoma. Cancer Res 64(19): 7002-10Stambolic V, Suzuki A, de la Pompa JL, Brothers GM, Mirtsos C, Sasaki T, Ruland J, PenningerJM, Siderovski DP, Mak TW (1998) Negative regulation of PKB/Akt-dependent cell survival bythe tumor suppressor PTEN. Cell 95(1): 29-39Stansfield I, Jones KM, Kushnirov VV, Dagkesamanskaya AR, Poznyakovski AI, Paushkin SV,Nierras CR, Cox BS, Ter-Avanesyan MD, Tuite MF (1995) The products of the SUP45 (eRF1)and SUP35 genes interact to mediate translation termination in Saccharomyces cerevisiae.EMBO J 14(17): 4365-73Sun C, Todorovic A, Querol-Audi J, Bai Y, Villa N, Snyder M, Ashchyan J, Lewis CS, HartlandA, Gradia S, Fraser CS, Doudna JA, Nogales E, Cate JH (2011) Functional reconstitution ofhuman eukaryotic translation initiation factor 3 (eIF3). Proc Natl Acad Sci U S A 108(51):20473-8Sun SY, Rosenberg LM, Wang X, Zhou Z, Yue P, Fu H, Khuri FR (2005) Activation of Akt andeIF4E survival pathways by rapamycin-mediated mammalian target of rapamycin inhibition.Cancer Res 65(16): 7052-8Sun XF, Zhang H (2007) NFKB and NFKBI polymorphisms in relation to susceptibility oftumour and other diseases. Histol Histopathol 22(12): 1387-98Sunavala-Dossabhoy G, Palaniyandi S, Clark C, Nathan CO, Abreo FW, Caldito G (2011)Analysis of eIF4E and 4EBP1 mRNAs in head and neck cancer. Laryngoscope 121(10): 2136-41Suzuki I, Tada A, Ollmann MM, Barsh GS, Im S, Lamoreux ML, Hearing VJ, Nordlund JJ,Abdel-Malek ZA (1997) Agouti signaling protein inhibits melanogenesis and the response ofhuman melanocytes to alpha-melanotropin. J Invest Dermatol 108(6): 838-42161Tang DJ, Dong SS, Ma NF, Xie D, Chen L, Fu L, Lau SH, Li Y, Guan XY (2010)Overexpression of eukaryotic initiation factor 5A2 enhances cell motility and promotes tumormetastasis in hepatocellular carcinoma. Hepatology 51(4): 1255-63Tarhini AA, Agarwala SS (2006) Cutaneous melanoma: available therapy for metastatic disease.Dermatol Ther 19(1): 19-25Tarun SZ, Jr., Sachs AB (1996) Association of the yeast poly(A) tail binding protein withtranslation initiation factor eIF-4G. EMBO J 15(24): 7168-77Taylor CA, Liu Z, Tang TC, Zheng Q, Francis S, Wang TW, Ye B, Lust JA, Dondero R,Thompson JE (2012) Modulation of eIF5A expression using SNS01 nanoparticles inhibits NF-kappaB activity and tumor growth in murine models of multiple myeloma. Mol Ther 20(7):1305-14Teng DH, Hu R, Lin H, Davis T, Iliev D, Frye C, Swedlund B, Hansen KL, Vinson VL,Gumpper KL, Ellis L, El-Naggar A, Frazier M, Jasser S, Langford LA, Lee J, Mills GB,Pershouse MA, Pollack RE, Tornos C, Troncoso P, Yung WK, Fujii G, Berson A, Steck PA, etal. (1997) MMAC1/PTEN mutations in primary tumor specimens and tumor cell lines. CancerRes 57(23): 5221-5Thompson JF, Scolyer RA, Kefford RF (2005) Cutaneous melanoma. Lancet 365(9460): 687-701Thoreen CC, Chantranupong L, Keys HR, Wang T, Gray NS, Sabatini DM (2012) A unifyingmodel for mTORC1-mediated regulation of mRNA translation. Nature 485(7396): 109-13Thoreen CC, Kang SA, Chang JW, Liu Q, Zhang J, Gao Y, Reichling LJ, Sim T, Sabatini DM,Gray NS (2009) An ATP-competitive mammalian target of rapamycin inhibitor revealsrapamycin-resistant functions of mTORC1. J Biol Chem 284(12): 8023-32Tomoo K, Matsushita Y, Fujisaki H, Abiko F, Shen X, Taniguchi T, Miyagawa H, Kitamura K,Miura K, Ishida T (2005) Structural basis for mRNA Cap-Binding regulation of eukaryoticinitiation factor 4E by 4E-binding protein, studied by spectroscopic, X-ray crystal structural, andmolecular dynamics simulation methods. Biochim Biophys Acta 1753(2): 191-208Topisirovic I, Svitkin YV, Sonenberg N, Shatkin AJ (2011) Cap and cap-binding proteins in thecontrol of gene expression. Wiley Interdiscip Rev RNA 2(2): 277-98Tsao H, Mihm MC, Jr., Sheehan C (2003) PTEN expression in normal skin, acquiredmelanocytic nevi, and cutaneous melanoma. J Am Acad Dermatol 49(5): 865-72162Tunca B, Tezcan G, Cecener G, Egeli U, Zorluoglu A, Yilmazlar T, Ak S, Yerci O, Ozturk E,Umut G, Evrensel T (2013) Overexpression of CK20, MAP3K8 and EIF5A correlates with poorprognosis in early-onset colorectal cancer patients. J Cancer Res Clin Oncol 139(4): 691-702Ueda T, Sasaki M, Elia AJ, Chio, II, Hamada K, Fukunaga R, Mak TW (2010) Combineddeficiency for MAP kinase-interacting kinase 1 and 2 (Mnk1 and Mnk2) delays tumordevelopment. Proc Natl Acad Sci U S A 107(32): 13984-90Umenaga Y, Paku KS, In Y, Ishida T, Tomoo K (2011) Identification and function of the secondeIF4E-binding region in N-terminal domain of eIF4G: comparison with eIF4E-binding protein.Biochem Biophys Res Commun 414(3): 462-7Unbehaun A, Borukhov SI, Hellen CU, Pestova TV (2004) Release of initiation factors from 48Scomplexes during ribosomal subunit joining and the link between establishment of codon-anticodon base-pairing and hydrolysis of eIF2-bound GTP. Genes Dev 18(24): 3078-93Unbehaun A, Marintchev A, Lomakin IB, Didenko T, Wagner G, Hellen CU, Pestova TV (2007)Position of eukaryotic initiation factor eIF5B on the 80S ribosome mapped by directed hydroxylradical probing. EMBO J 26(13): 3109-23Vaisanen A, Tuominen H, Kallioinen M, Turpeenniemi-Hujanen T (1996) Matrixmetalloproteinase-2 (72 kD type IV collagenase) expression occurs in the early stage of humanmelanocytic tumour progression and may have prognostic value. J Pathol 180(3): 283-9Vancoillie G, Lambert J, Nayaert JM (1999) Melanocyte biology and its implications for theclinician. Eur J Dermatol 9(3): 241-51Villa N, Do A, Hershey JW, Fraser CS (2013) Human eukaryotic initiation factor 4G (eIF4G)protein binds to eIF3c, -d, and -e to promote mRNA recruitment to the ribosome. J Biol Chem288(46): 32932-40Vita M, Henriksson M (2006) The Myc oncoprotein as a therapeutic target for human cancer.Semin Cancer Biol 16(4): 318-30Volkovova K, Bilanicova D, Bartonova A, Letasiova S, Dusinska M (2012) Associationsbetween environmental factors and incidence of cutaneous melanoma. Review. Environ Health11 Suppl 1: S12von Thaler AK, Kamenisch Y, Berneburg M (2010) The role of ultraviolet radiation inmelanomagenesis. Exp Dermatol 19(2): 81-8Voorhees RM, Ramakrishnan V (2013) Structural basis of the translational elongation cycle.Annu Rev Biochem 82: 203-36163Voorhees RM, Schmeing TM, Kelley AC, Ramakrishnan V (2010) The mechanism foractivation of GTP hydrolysis on the ribosome. Science 330(6005): 835-8Wagner CR, Iyer VV, McIntee EJ (2000) Pronucleotides: toward the in vivo delivery of antiviraland anticancer nucleotides. Med Res Rev 20(6): 417-51Walker SE, Zhou F, Mitchell SF, Larson VS, Valasek L, Hinnebusch AG, Lorsch JR (2013)Yeast eIF4B binds to the head of the 40S ribosomal subunit and promotes mRNA recruitmentthrough its N-terminal and internal repeat domains. RNA 19(2): 191-207Wang FW, Cai MY, Mai SJ, Chen JW, Bai HY, Li Y, Liao YJ, Li CP, Tian XP, Kung HF, GuanXY, Xie D (2014) Ablation of EIF5A2 induces tumor vasculature remodeling and improvestumor response to chemotherapy via regulation of matrix metalloproteinase 2 expression.Oncotarget 5(16): 6716-33Wang XL, Cai HP, Ge JH, Su XF (2012) Detection of eukaryotic translation initiation factor 4Eand its clinical significance in hepatocellular carcinoma. World J Gastroenterol 18(20): 2540-4Waskiewicz AJ, Johnson JC, Penn B, Mahalingam M, Kimball SR, Cooper JA (1999)Phosphorylation of the cap-binding protein eukaryotic translation initiation factor 4E by proteinkinase Mnk1 in vivo. Mol Cell Biol 19(3): 1871-80Wei JH, Cao JZ, Zhang D, Liao B, Zhong WM, Lu J, Zhao HW, Zhang JX, Tong ZT, Fan S,Liang CZ, Liao YB, Pang J, Wu RH, Fang Y, Chen ZH, Li B, Xie D, Chen W, Luo JH (2014)EIF5A2 predicts outcome in localised invasive bladder cancer and promotes bladder cancer cellaggressiveness in vitro and in vivo. Br J Cancer 110(7): 1767-77Wells SE, Hillner PE, Vale RD, Sachs AB (1998) Circularization of mRNA by eukaryotictranslation initiation factors. Mol Cell 2(1): 135-40Wendel HG, De Stanchina E, Fridman JS, Malina A, Ray S, Kogan S, Cordon-Cardo C, PelletierJ, Lowe SW (2004) Survival signalling by Akt and eIF4E in oncogenesis and cancer therapy.Nature 428(6980): 332-7Wendel HG, Silva RL, Malina A, Mills JR, Zhu H, Ueda T, Watanabe-Fukunaga R, Fukunaga R,Teruya-Feldstein J, Pelletier J, Lowe SW (2007) Dissecting eIF4E action in tumorigenesis.Genes Dev 21(24): 3232-7West MJ, Sullivan NF, Willis AE (1995) Translational upregulation of the c-myc oncogene inBloom's syndrome cell lines. Oncogene 11(12): 2515-24Westermarck J, Kahari VM (1999) Regulation of matrix metalloproteinase expression in tumorinvasion. FASEB J 13(8): 781-92164Wolff EC, Lee YB, Chung SI, Folk JE, Park MH (1995) Deoxyhypusine synthase from rat testis:purification and characterization. J Biol Chem 270(15): 8660-6Wu H, Goel V, Haluska FG (2003) PTEN signaling pathways in melanoma. Oncogene 22(20):3113-22Wu M, Liu Y, Di X, Kang H, Zeng H, Zhao Y, Cai K, Pang T, Wang S, Yao Y, Hu X (2013)EIF4E over-expresses and enhances cell proliferation and cell cycle progression innasopharyngeal carcinoma. Med Oncol 30(1): 400Xie D, Ma NF, Pan ZZ, Wu HX, Liu YD, Wu GQ, Kung HF, Guan XY (2008) Overexpressionof EIF-5A2 is associated with metastasis of human colorectal carcinoma. Hum Pathol 39(1): 80-6Yan R, Rychlik W, Etchison D, Rhoads RE (1992) Amino acid sequence of the human proteinsynthesis initiation factor eIF-4 gamma. J Biol Chem 267(32): 23226-31Yanagiya A, Svitkin YV, Shibata S, Mikami S, Imataka H, Sonenberg N (2009) Requirement ofRNA binding of mammalian eukaryotic translation initiation factor 4GI (eIF4GI) for efficientinteraction of eIF4E with the mRNA cap. Mol Cell Biol 29(6): 1661-9Yang GF, Xie D, Liu JH, Luo JH, Li LJ, Hua WF, Wu HM, Kung HF, Zeng YX, Guan XY(2009) Expression and amplification of eIF-5A2 in human epithelial ovarian tumors andoverexpression of EIF-5A2 is a new independent predictor of outcome in patients with ovariancarcinoma. Gynecol Oncol 112(2): 314-8Yang SH, Sharrocks AD, Whitmarsh AJ (2003) Transcriptional regulation by the MAP kinasesignaling cascades. Gene 320: 3-21Yang SX, Hewitt SM, Steinberg SM, Liewehr DJ, Swain SM (2007) Expression levels of eIF4E,VEGF, and cyclin D1, and correlation of eIF4E with VEGF and cyclin D1 in multi-tumor tissuemicroarray. Oncol Rep 17(2): 281-7Yazdi AS, Palmedo G, Flaig MJ, Puchta U, Reckwerth A, Rutten A, Mentzel T, Hugel H,Hantschke M, Schmid-Wendtner MH, Kutzner H, Sander CA (2003) Mutations of the BRAFgene in benign and malignant melanocytic lesions. J Invest Dermatol 121(5): 1160-2Yellen P, Saqcena M, Salloum D, Feng J, Preda A, Xu L, Rodrik-Outmezguine V, Foster DA(2011) High-dose rapamycin induces apoptosis in human cancer cells by dissociating mTORcomplex 1 and suppressing phosphorylation of 4E-BP1. Cell Cycle 10(22): 3948-56Yoshizawa A, Fukuoka J, Shimizu S, Shilo K, Franks TJ, Hewitt SM, Fujii T, Cordon-Cardo C,Jen J, Travis WD (2010) Overexpression of phospho-eIF4E is associated with survival throughAKT pathway in non-small cell lung cancer. Clin Cancer Res 16(1): 240-8165Zender L, Xue W, Zuber J, Semighini CP, Krasnitz A, Ma B, Zender P, Kubicka S, Luk JM,Schirmacher P, McCombie WR, Wigler M, Hicks J, Hannon GJ, Powers S, Lowe SW (2008) Anoncogenomics-based in vivo RNAi screen identifies tumor suppressors in liver cancer. Cell135(5): 852-64Zhang W, Dunkle JA, Cate JH (2009) Structures of the ribosome in intermediate states ofratcheting. Science 325(5943): 1014-7Zhang YJ, Duan Y, Zheng XF (2011) Targeting the mTOR kinase domain: the second generationof mTOR inhibitors. Drug Discov Today 16(7-8): 325-31Zhou X, Tan M, Stone Hawthorne V, Klos KS, Lan KH, Yang Y, Yang W, Smith TL, Shi D, YuD (2004) Activation of the Akt/mammalian target of rapamycin/4E-BP1 pathway by ErbB2overexpression predicts tumor progression in breast cancers. Clin Cancer Res 10(20): 6779-88Zhu W, Cai MY, Tong ZT, Dong SS, Mai SJ, Liao YJ, Bian XW, Lin MC, Kung HF, Zeng YX,Guan XY, Xie D (2011) Overexpression of EIF5A2 promotes colorectal carcinoma cellaggressiveness by upregulating MTA1 through C-myc to induce epithelial-mesenchymaltransition. GutZhu W, Cai MY, Tong ZT, Dong SS, Mai SJ, Liao YJ, Bian XW, Lin MC, Kung HF, Zeng YX,Guan XY, Xie D (2012a) Overexpression of EIF5A2 promotes colorectal carcinoma cellaggressiveness by upregulating MTA1 through C-myc to induce epithelial-mesenchymaltransition. Gut 61(4): 562-75Zhu Y, Zhang X, Liu Y, Zhang S, Liu J, Ma Y, Zhang J (2012b) Antitumor effect of the mTORinhibitor everolimus in combination with trastuzumab on human breast cancer stem cells in vitroand in vivo. Tumour Biol 33(5): 1349-62Zhu YH, Fu L, Chen L, Qin YR, Liu H, Xie F, Zeng T, Dong SS, Li J, Li Y, Dai Y, Xie D, GuanXY (2013) Downregulation of the novel tumor suppressor DIRAS1 predicts poor prognosis inesophageal squamous cell carcinoma. Cancer Res 73(7): 2298-309Zuberek J, Wyslouch-Cieszynska A, Niedzwiecka A, Dadlez M, Stepinski J, Augustyniak W,Gingras AC, Zhang Z, Burley SK, Sonenberg N, Stolarski R, Darzynkiewicz E (2003)Phosphorylation of eIF4E attenuates its interaction with mRNA 5' cap analogs by electrostaticrepulsion: intein-mediated protein ligation strategy to obtain phosphorylated protein. RNA 9(1):52-61Zuk D, Jacobson A (1998) A single amino acid substitution in yeast eIF-5A results in mRNAstabilization. EMBO J 17(10): 2914-25

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0229496/manifest

Comment

Related Items