UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Role of tumor suppressor p33ING2 inhuman cutaneous melanoma Lu, Fuqu 2006

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

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

Item Metadata


831-ubc_2007-0173a.pdf [ 4.31MB ]
JSON: 831-1.0100906.json
JSON-LD: 831-1.0100906-ld.json
RDF/XML (Pretty): 831-1.0100906-rdf.xml
RDF/JSON: 831-1.0100906-rdf.json
Turtle: 831-1.0100906-turtle.txt
N-Triples: 831-1.0100906-rdf-ntriples.txt
Original Record: 831-1.0100906-source.json
Full Text

Full Text

R O L E O F T U M O R SUPPRESSOR P33ING2 IN H U M A N C U T A N E O U S M E L A N O M A by F U Q U L U B . M . , Harbin Medical University, 2000 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Experimental Medicine) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A December 2006 © Fuqu L u , 2006 A B S T R A C T P33JNG2 was cloned through a homology search with p33INGlb, the founding member of ING family proteins in 1998. There have been several studies indicating that p331NG2 is a tumor suppressor candidate since it is involved in the regulation of transcription, apoptosis and senescence. People in our lab already revealed that p33ING2 plays an essential role in cellular stress response against U V irradiation ( U V R ) either by enhancing nucleotide excision repair (NER) or promoting apoptosis in melanoma cell line. So far, reduced I N G family proteins have been reported in different tumor types including the loss of nuclear expression of p33INGlb in melanoma. Since p33ING2 is highly homologous to p331NGlb, we hypothesized that aberrant expression of p331NG2 may play a role in melanoma tumorigenesis. To test this hypothesis, we used tissue microarray ( T M A ) and immunohistochemistry to evaluate the expression pattern of p33ING2 in different stages of melanocyte lesions. We found that nuclear 1NG2 expression is significantly reduced in melanomas compared with dysplastic nevi. Moreover, there is no correlation between TNG2 nuclear expression and patient clinicopathological parameters or between 1NG2 nuclear expression and patient's 5-year survival in primary melanomas and metastatic melanomas. We then investigated i f gene mutation is the reason for reduced ING2 nuclear expression in melanomas. Our results showed that no mutation is found in melanoma cell lines or melanoma tissue samples. At last, we restored the expression of p33TNG2 in melanomas by establishing the 1NG2 stable cell line, and c D N A microarray analysis was performed to identify the most abundantly induced or suppressed genes by overexpressed ING2 or U V R . We found that U V R can enhance expressions of two relatively new genes BTG2 and PLK3 while stably expression of p33.ING2 significantly decreased R A R R E S 1 (TIG 1) expression. ii T A B L E O F C O N T E N T S A B S T R A C T ii T A B L E O F C O N T E N T S i i i L I ST O F T A B L E S v L I ST O F F I G U R E S vi L I ST O F A B B R E V I A T I O N S vi i A C K N O W L E D G E M E N T S v i i i C H A P T E R 1. I N T R O D U C T I O N 1 1.1 H u m a n M e l a n o m a 1 1.1.1 M e l a n o m a Incidence and Mor ta l i t y Rate 1 1.1.2 R isk Factors 2 1.1.3 Character ist ics o f Di f ferent Stages o f M e l a n o c y t e Les ions 2 1.1.4 Mo lecu l a r Changes in M e l a n o m a 3 1.1.5 M e l a n o m a Treatment and Prognosis 5 1.2 Inhibitor o f G rowth ( ING) F a m i l y 6 1.2.1 ING1 7 1.2.2 P 3 3 I N G 2 10 Gene and Protein Structures 10 Express ion Profi les 11 Funct ions o f P 3 3 I N G 2 , 12 Modes o f P 3 3 I N G 2 Ac t i on 14 1.3 Object ive 14 C H A P T E R 2. M A T E R I A L S A N D M E T H O D S 16 2.1 T M A Construct ion 16 2.2 Immunohis tochemist ry o f T M A 16 2.3 Eva luat ion o f Immunosta in ing 17 2.4 Statistical Ana l ys i s 18 2.5 C e l l L ines and C e l l Cul ture 18 2.6 N o r m a l and M e l a n o m a Tissues 1.9 2.7 U V Irradiation 19 2.8 P lasmids , Transfect ion and I N G 2 Stable C lone Generat ion 19 2.9 Western B lot Ana lys i s 20 2.10 S R B Ce l l Surv iva l Assay 20 2.11 D N A and R N A Extract ion 21 2.12 Polymerase Cha in React ion 21 2.13 D N A Sequencing 22 2.14 c D N A M ic roa r ray 22 i n 2.15 Quantitat ive Real T i m e P C R 23 C H A P T E R 3. N U C L E A R I N G 2 E X P R E S S I O N IS R E D U C E D IN H U M A N C U T A N E O U S M E L A N O M A S 24 3.1 Rat ionale and Hypothesis 24 3.2 Results 25 3.2.1 C l in i copa tho log i ca l F ind ings 25 3.2.2 Reduced I"NG2 Nuc lear Express ion in Human Me l anomas 25 3.2.3 Corre lat ion between I N G 2 Nuc lea r Express ion and C l in i copa tho log i ca l Parameters or 5-year Patient Surv iva l 26 3.3 Discuss ion 32 C H A P T E R 4. M U T A T I O N A L S T A T U S O F P33ING2 I N M E L A N O M A S 36 4.1 Rat ionale and Hypothesis 36 4.2 Results 37 4.2.1 Al terat ions o f P 3 3 I N G 2 Gene in M e l a n o m a C e l l L i n e 37 4.2.2 Alterat ions o f the N L S and P H D D o m a i n o f P 3 3 I N G 2 Gene in M e l a n o m a 37 4.3 D iscuss ion 40 C H A P T E R 5. I D E N T I F I C A T I O N O F G E N E S T R A N S C R I P T I O N A L L Y R E G U L A T E D B Y U V I R R A D I A T I O N O R P 3 3 I N G 2 43 5.1 Rat ionale and Hypothesis 43 5.2 Results 45 5.2.1 G rowth Inhibit ion o f P 3 3 I N G 2 45 5.2.2 Gene Express ion Ana l ys i s by c D N A Mic roar ray 45 5.2.3 Rea l t ime P C R Ana lys i s 46 5.3 D i scuss ion 53 C H A P T E R 6. G E N E R A L C O N C L U S I O N S 57 6.1 Summary 57 6.2 Future Direct ions 58 R E F E R E N C E S 60 iv LIST O F T A B L E S Table 3.1 C l in i copa tho log i ca l parameters o f 122 cases o f melanomas 27 Table 4.1 JNG2 alterations in normal melanocyte and me lanoma cel l lines 38 Table 5.1 Top 10 genes induced by U V irradiat ion in M M R U 49 Table 5.2 Top 10 genes suppressed by U V irradiat ion in M M R U 50 Table 5.3 T o p 10 genes induced by p331NG2 51 Table 5.4 Top 10 genes suppressed by p 3 3 I N G 2 52 v L I S T O F F I G U R E S F i g u r e 1.1 Structural features of the TNG family proteins 15 F i g u r e 3 . 1 Representative images of 1NG2 immunohistochemical staining in human melanocytic lesions 28 F i g u r e 3 . 2 ING2 nuclear expressions at different stages of melanocytic lesions 29 F i g u r e 3 . 3 No correlation was found between ING2 nuclear expression and tumor thickness or tumor ulceration of primary melanomas 30 F i g u r e 3 . 4 1NG2 nuclear expression and 5-year patient survival 31 F i g u r e 4.1 Representative sequence chromatograms of ING2 mutations in Sk-mel-110 melanoma cell line 39 F i g u r e 5.1 ING2 is overexpressed in stable clones 47 F i g u r e 5.2 Cell survival rate by S R B assay of UVB-irradiated M M R U cells and ING2 stable cell line 48 v i L I S T O F A B B R E V I A T I O N S A J C C Amer i can Joint Commit tee on Cancer B C C Basal cel l ca rc inoma C D K 4 Cycl in-dependent kinase 4 C D K N 2 A Cycl in-dependent kinase inhibitor 2 A C T The cyc le number o f threshold E C M Extrace l lu lar matrix H A T H i stone acetyltransferase H D A C Histone.deacetylase I N G Inhibitor o f growth L O H Loss o f heterozygosity M D M 2 Mouse double minute 2 M M P - 7 Mat r ix metal loproteinase 7 N E R Nuc leot ide exc is ion repair N L S Nuc lear loca l izat ion signal N M u M G M o u s e mammary epithelial cel l P B S Phosphate-buffered saline P C N A Prol i ferat ing cel lu lar nuclear antigen P H D Plant homeodomain PIP PCNA- in te rac t ing protein domain P T E N Phosphatase and tensin homologue Rb Ret inoblastoma protein R G P Rad ia l growth phase pr imary melanoma S R B Su l forhodamine B T C A Trichloroacetate TIG1 Tazarotene induced gene 1 T M A Tissue microarray T S G T u m o r suppressor gene U V R Ul t rav io let irradiation V G P Ver t ica l growth phase pr imary melanoma vu A C K N O W L E D G E M E N T S I wou ld l ike to express my sincere thanks to my supervisor D r . G a n g L i , for prov id ing me the opportunity to study in your laboratory and set up m y first step in research. To Dr. B i l l Sa lh , Dr. Y o u w e n Z h o u , Dr. K e v i n J M c E l w e e , and Dr . Torsten N ie l sen , I want to thank you for be ing my committee members and for your suggestions on m y study. I w o u l d l ike to thank Dr. Magda lena Mar t inka and Derek L. Da i for the technical assistance in tissue microarray staining evaluation and statistical analysis; D r . Co l l een Ne lson and Anne Haegert for the assistance in c D N A microarray studies; M i n w a n Su and Jun L i for assistance in tissue sample co l lec t ion and D N A sequencing. I wou ld also l ike to thank the members o f L i lab and m y fami l y , your encouragement, support and help are greatly appreciated. v i n C H A P T E R 1. INTRODUCTION 1.1. Human Melanoma Human cutaneous malignant melanoma originates from the melanocytes of the skin or melanocyte lesions including common acquired nevus, dysplastic nevus, congenital nevus and cellular blue nevus (Balch et al, 2001). Although it only accounts for 4 percent of all dermatologic cancers, melanoma is responsible for 80 percent o f deaths from skin cancer. Only 14 percent of patients with metastatic melanoma survive for 5 years (Cancer facts & figures, 2003. Atlanta: American Cancer Society, 2003). The intractability o f melanoma promotes the need to learn more about the changes that facilitate melanoma genesis, progression and metastasis. 1.1.1 Melanoma Incidence and Mortality Rate Once considered as an uncommon disease, the incidence of malignant melanoma has doubled within the last 10 years in the United States (Balch et al, 2001). It is estimated that the chance of an American developing melanoma during his/her lifetime leaped from 1 in 1500 in 1960, to 1 in 68 in 2000, and is projected to increase to 1 in 50 by the year 2010 (Rigel, 2002; Dunlap et al, 2004). Primary melanoma usually progresses from the nonmetastatic radial growth phase to the vertical growth phase, in which the malignant cells invade the dermis and develop the ability to metastasize (Elder, 1999). The large majority of patients with early stages o f melanoma can be cured by surgical removal while 50% of patients with distant metastasis die within the first 6 months after the diagnosis (Balch et al, 2001; Schoenlaub et al, 2001). 1.1.2 Risk Factors The strongest risk factors for melanoma are family history of melanoma, previous melanoma and multiple benign or atypical nevi. Additional factors include immunosuppression and ultraviolet radiation exposure. Each risk factor reflects a genetic predisposition or an environmental stressor that favors melanomagenesis. For example, the mutation of cyclin-dependent kinase inhibitor 2 A ( C D K N 2 A ) was observed in 33/36 cases of melanoma in nine families and a few rare kindreds have mutations in cyclin-dependent kinase 4 (CDK4) (Hussussian et al, 2004; Zuo et al, 1996). U V R is the major environmental risk factor for melanoma. It is well documented that U V R promotes the malignant change in the skin by having direct mutagenic effects on D N A , by stimulating the cellular components to produce growth factors and by reducing cutaneous immune defenses (Gilchrest et al, 1999; Thompson et al, 2005). On the other hand, the tanning response is a defensive mechanism in which melanocytes produce melanin and deliver it to neighboring keratinocytes. Melanin then forms a protective cap over the outer part of keratinocyte nuclei where it absorbs and dissipates ultraviolet energy (Gilchrest et al, 1999; Ortonne, 2002). 1.1.3 Characteristics of Different Stages of Melanocytic Lesions Normal nevi are the first phenotypic changes in melanocytes which derive from the proliferation of structurally normal melanocytes. Clinically, these nevi present as flat or slightly raised lesions with either uniform coloration or a regular pattern of dot-like pigment in a tan or dark brown background. Atypical nevi, also referred to as dysplastic nevi, are moles that develop from aberrant growth of preexisting benign nevi or in a new location. Clinically, such lesions may be asymmetric, have irregular borders, contain multiple colors or have increasing diameters. The clinical importance of dysplastic nevi lies in their association with an increased risk of malignant melanoma which is supported by cohort and case-control studies (Hussein, 2005). The characteristic of the radial growth phase (RGP) primary melanoma is that the malignant cells grow only within or in close proximity to the epidermis and they do not have competence for metastasis (Hsu et al, 1998). Clinically, they sometimes present as raised lesions. These lesions no longer display random atypia and instead show cytomorphologic cancer throughout the lesion (Urso, 2004). Lesions that progress to the vertical growth phase ( V G P ) acquire the ability to invade the dermis. The melanoma cells in V G P are capable of growth in soft agar, and have the capacity to form tumor nodules when implanted to nude mice. The conversion of primary melanomas from R G P to V G P is the most critical step in melanoma progression and ultimately in disease outcome (Erhard et al, 1997; Hsu et al, 1998). The final step melanoma metastasis is the spread of malignant cells to other parts of skin and other organs. These cells can grow in soft agar and when implanted in nude mice can form tumor nodules and metastasize. A t this stage, cells usually spread to nearby lymph node and distant metastases typically are found in the skin, liver, lung, bone, and brain. 1.1.4 Molecular Changes in Melanoma The histological changes of melanocytic lesions relate to particular gene mutations and these mutations further affect molecular signaling and are responsible for the progression from normal melanocyte to melanoma (Mil ler and M i h m , 2006). P53, a central sensor linking D N A damage to apoptosis, plays an essential role in tumor suppression and chemosensitivity in many tumor types (L i et al, 1998; Raffo et al, 2000; Fridman and Lowe, 2003). However, mutational analysis reveals that the p53 gene is altered in only approximately 11% of melanomas (Hussein, 2004). The low mutation rate of p53 suggests that other genes may play important roles in pathogenesis of melanoma. N - R A S , B R A F , and M A P K Growth Factor Signaling Pathway Activation of this pathway is due to the somatic mutations o f NRAS, which occur in up to 30% cases of cutaneous malignant melanoma or BRAF, which are found in 59% melanoma patients (Omholt et al, 2002; Omholt et al, 2003). Most of NRAS mutations are at codon 61 (Q61R and Q61K) and result in the expression of p21RAS oncoprotein, which remains constitutively G T P bound and active (Malumbres and Barbacid, 2003); A l l the B R A F mutations were found within the kinase domain, with a single substitution (T to A ) of glutamate for valine at codon 600 (V600E) being responsible for 90% of the observed mutations (Davies et al, 2002). NRAS and BRAF mutations, which occur exclusively of each other, cause constitutively active expression of the serine-threonine kinase in the E R K - M A P K pathway. Similar frequency of BRAF mutations in benign nevi, primary and metastatic melanoma suggests that nevi must acquire additional molecular changes to free themselves of growth constraint and become malignant. For instance, in zebrafish, melanocyte specific expression o f mutant B R A F protein leads to proliferation of melanocytes, analogous to human nevi. However, the combination of BRAF mutation and p53 inactivation causes melanocytes to transform and become melanoma (Patton et al, 2005). CDKN2A Cyclin-dependent kinase inhibitor ( C D K N ) 2 A is located on chromosome 9p21. Germline mutations of CDKN2A gene have been reported in numerous melanoma-prone families or melanoma cases selected because of young age or multiple primary tumors (Kamb et al, 1994; Ruas and Peters, 1998). Alternative splicing of various exons within CDKN2A yields two distinct tumor suppressor proteins, I N K 4 A ( p l 6 I N K 4 A ) and alternate reading frame (ARF) . I N K 4 A contributes to cell cycle arrest at the G l - S check point by inhibiting cyclin-dependent kinases 4 and 6. I N K 4 A suppresses the proliferation of cells with damaged D N A or activated oncogenes and also acts when cells are old or crowded (Sharpless and Chin, 2003). I N K 4 A deficient mice appear normal but are more susceptible to carcinogens and easily develop various tumors (Serrano et al, 1996). The INK4A" /"mice develop spontaneous cutaneous melanomas when combined with melanocyte-specific expression o f activated H-ras after a short latency and with high penetrance (Chin et al, 1997). A R F participates in the core regulatory process that controls the levels of p53 by binding to mouse double minute 2 ( M D M 2 ) protein, sequestering it from p53 and therefore causing p53 to accumulate. P53 can arrest the cells at G 2 - M phase, allowing for repairing of damaged D N A or inducing apoptosis (Harris and Levine, 2005). In animals, A R F deficient shortens the time required to develop melanoma after exposure to U V light. When both CDKN2A product ( p l 6 I N K 4 A and A R F ) are deficient, the latent time is even shorter (Recio et al, 2002). 1.1.5 Melanoma Treatment and Prognosis To date, the success of systemic therapy for melanoma remains unsatisfactory. Melanoma is associated with a very high mortality rate, especially for advanced disease. Treatment options and follow up intervals for melanoma patients vary with the depth of primary lesion and stage. Surgical excision is the primary therapeutic approach for > melanoma, which is almost 100% effective when treating an early tumor. Chemotherapy and adjuvant therapy including biologic response modifiers and vaccines, provide supplemental treatment for more advanced tumors (Rigel and Carucci, 2000). In general, the prognosis for the melanoma depends on the stage at diagnosis. For patients with American Joint Committee on Cancer (AJCC) , the overall survival is 70% for A J C C stages I and II, 30% for A J C C stage III patients and only 10% for stage I V patients (Balch et al, 2001). For certain stage primary melanoma, the dominant predictors for prognosis are lesion thickness, ulceration and lymph node involvement. Factors such as age, sex, anatomic location and satellite/in transit lesions are important. Moreover, for the metastatic melanoma, the most important prognostic factors include sites of metastasis and the presence of increased serum lactic dehydrogenase (Homsi et al, 2005). 1.2 Inhibitor of G r o w t h ( ING) Fami ly The I N G tumor suppressor family consists of five members including I N G 1 , p33ING2, p47ING3, p29ING4 and p28ING5. The ING1 gene, due to alternative splicing of its m R N A product, encodes three isoforms p47INGla , p33INGlb , p24DSfGlc (Garkavtsev et al, 1999; Gunduz et al, 2000). It is not clear whether the other I N G family members also contain multiple splicing isoforms, but structure analysis shows they all share a nuclear localization signal (NLS) and a plant homeodomain (PHD) type zinc finger which indicate I N G family members may have similar biochemical functions (Fig. 1.1) (Gong et al, 2005). N L S domain indicates the nuclear localization of I N G family proteins and mutations of this domain may disrupt interactions between I N G intracellular trafficking proteins resulting in a loss of nuclear I N G expression and affects their normal functions. In addition, P H D domains of ING1 and ING2 can bind to phosphoinositides both in vitro and in vivo and this interaction is crucial for ING2 to activate p53 and p53 dependent apoptotic pathways (Gozani et al, 2003). On the other hand, proteins with P H D domain are intimately associated with chromatin remodeling and subsequent transcriptional regulation of specific genes. A recent article published in Nature reported that the I N G P H D domains are specific and highly robust binding modules for di- and tri-methylated lysine 4 of histone H3 (H3K4me2 and H3K4me3) and this association is important for ING2-mediated cellular responses to genotoxic insults (Shi et al, 2006). I N G proteins have been found to be differentially contributed to other biologic activities including cell cycle arrest, apoptosis, D N A repair, senescence, anti-angiogenesis (Campos et al, 2004; Shi and Gozani, 2005). The mechanisms underlying these functions are partly due to the ability of I N G proteins associated with different histone acetyltransferases (HATs) and histone deacetylase ( H D A C ) to posttranslationally modify the core histones and the p53 tumor suppressor (Kuzmichev et al, 2002; Vieyra et al, 2002; Doyon et al, 2006). 1.2.1 ING1 ING1, the founding member of the I N G tumor suppressor family, was discovered in 1996 through suppressive subtractive hybridization between normal human mammary epithelial cells and breast cancer cell line, and further in vivo selection of genetic suppressor elements (Garkavtsev et al, 1996). Genetic mapping disclosed that ING1 gene is located on the long arm of chromosome 13 in which genetic deletions have been reported in a variety of human tumors (Garkavtsev etal, 1997). Among all three isoforms of ING1 , p33INGlb is the predominant form expressed in normal tissues, with various expression levels in different types o f tissues (Saito et al, 2000). ING1 expression screening showed that down-regulation of ING1 is a frequent event in different tumor types including breast (Garkavtsev et al, 1996; Toyama et al, 1999), gastric (Oki et al, 1999), esophageal (Chen et al, 2001), blood (Ito et al, 2002), lung (Kameyama et al, 2003) and brain (Tallen et al, 2003). The mechanisms resulting in decreased I N G expression are unknown and currently under active investigation. Lines of evidence suggest that ING1 can influence cell cycle progression and are actively involved in cellular checkpoints. One candidate mechanism is the cooperation of p33INGlb with p53 tumor suppressor to enhance the transcription of cyclin dependent kinase inhibitor (CDKI) p21 (Garkavtsev et al, 1998). P21 affects the G l cellular checkpoint by binding and inactivating cyc l in -CDK complexes, therefore leading to reduced phosphorylation of retinoblastoma protein (Rb), transcriptional E2F sequestration and G l - S cell cycle arrest (Harper et al, 1993). Moreover, c D N A microarray analysis relates p33INGb induced G 2 - M arrest to cyclin B l (Takahashi et al, 2002), which accumulates during G 2 - M phase of the cell cycle and associates with cdc2 for mitotic initiation (Elledge, 1996). The study is performed using mouse mammary epithelial cells ( N M u M G ) which are transformed with antisense ING1 and presented 14 upregulated genes including cyclin Bl, proto-oncogene DEK, and osteopontin while overexpressed ING1 in N M u M G cells resulted in down-regulation of cyclin Bl, DEK, and osteopontin (). Consistent with the c D N A microarray finding, later studies revealed that overexpression of p33INGlb was able to enhance adriamycin-induced G2 arrest in the H1299 non-small cell lung carcinoma cell line (Tsang et al, 2003). In addition to its involvement in cell cycle regulation, ING1 is wel l documented to be able to induce apoptosis. Initial evidence of this function came from the observation of prominent expression of ING1 in regressing tail o f Xenopus tadpoles while absence from growing hind limbs as well as the induced expression of ING1 in serum-starved P I9 teratocarcinoma cells (Helbing et al, 1997; Wagner et al, 2001). Cooperation with p53 was quickly recognized since ING1 expressing fibroblasts had little effect on cell survival in the absence of p53, while both ING1 and p53 were required to suppress colony formation (Garkavtsev et al, 1998). Moreover, people in our lab reported that overexpressed ING1 was able to increase UV-induced apoptosis in p53 wild-type M M R U melanoma cell line but not in p53 mutant M E W O cell line and we further found overexpression of ING1 enhances endogenous bax expression and alter the mitochondrial membrane potential (Cheung and L i , 2002). In addition, one study demonstrated that the ectopic expression of p33INGlb rather than p 4 7 I N G l a sensitized early passage of fibroblast to apoptosis (Vieyra et al, 2002). This difference reflects the different N-terminal sequence o f I N G proteins since p 3 3 I N G l , but not p47INGla , contains a PCNA-interacting protein (PIP) domain, which permits p33INGlb to physically interact with P C N A following U V exposure and induce apoptosis (Scott et al, 2001). Previous studies have shown that ING1 physically binds to p53 and shares similar functions with p53. People in our lab therefore hypothesized that I N G 1 , like p53, may play an important role in D N A repair. They found that U V irradiation could induce cell type specific p3 3 I N G l b expression in a time- and dose-dependent manner (Cheung et al, 2000) and overexpression of ING1 enhanced nucleotide excision repair (NER) of UVC-damaged exogenous plasmid D N A and UVB-damaged genomic D N A . As expected, p33INGlb mediated enhancement of D N A repair was also dependent on the existence o f functional p53. They further demonstrated that there is a weak association between p33 INGlb and G A D D 4 5 , but not other repair proteins like X P A and X P B (Cheung et al, 2001). Several other studies also clarified the involvement of p 3 3 I N G l b in D N A repair machinery, however, the mechanism underlying p33INGlb enhanced D N A repair is still not very clear. 1.2.2 P33ING2 P33ING2, also known as I N G 1 L (" INGl- l ike molecule"), was first isolated by Shimada and his colleagues through a homology search from a private Otsuka c D N A data base (Shimada et al, 1998). P33ING2 has been shown to negatively regulate cell growth, enhance nucleotide excision repair upon U V R as well as regulate the onset of replicative senescence. A l l these functions require the presence o f functional p53 (Chin et al, 2005; Pedeux et al, 2005; Wang J et al, 2006). Gene and Protein Structures The full-length p33ING2 c D N A contains an 840bp length open reading frame, encoding a 280 amino acid protein with a 32.8 k D a molecular weight. The gene product of p33ING2 shows 58.9% identity with p33 INGlb , while nucleotide identities between the two genes are 60% (Shimada et al, 1998). Fluorescence in situ hybridization and radiation-hybrid analysis assigned p33ING2 to chromosome 4 (Shimada et al, 1998). Besides the common P H D zinc finger motif and N L S domain, ING2 also contains a unique leucine zipper domain which is thought to mediate hydrophobic protein-protein interaction (Feng et al, 2002). ING2 is mainly localized in the nucleus with 74% in the chromatin/nuclear matrix and 9% in the nucleoplasm in HT1080 fibrosarcoma cells (Gozani et al, 2003). - 1 0 - Expression Profiles Northern blot analysis revealed ubiquitous expression of p33ING2 in various normal tissues with highest expression detected in testis. Subsequent expression screening of 20 colon tumors showed higher p33ING2 expression level in all colon cancers compared with the normal tissues acquired at the same surgical sites. The importance and involved mechanism of elevated p33ING2 expression in colon cancers was still unknown but the p53 abnormalities were found to occur frequently in this type of cancer (Shimada et al, 1998). Nagashima et al also detected p33ING2 expression in 12 different types of human cell lines. The level o f p33ING2 was highly variable among the cell lines with no visible expression in 5 of the 12 cell lines and the expression status did not correlate with the mutational status of p53 (Nagashima et al, 2001). A recent study detected the expression of p33ING2 in lung cancers and found decreased ING2 expression in 6 of 7 lung cancer cell lines with mutant p53 which suggests ING2 gene may be an independent tumor suppressor candidate on p53 in this type of cancer. In addition, they detected p33ING2 mutations in 30 human lung cancer cell lines and 31 primary lung cancer tumors but failed to detect any mutation (Okano et al, 2006). Furthermore, studies on loss of heterozygosity (LOH) in sporadic basal cell carcinomas (BCC) reported a high frequency of L O H (30%) on chromosome 4q32-35, which is mapped to p33ING2 and SAP30, both of which are believed to be involved in chromatin remodeling and gene regulation (Sironi et al, 2004). The importance o f these reported abnormal expressions o f p33ING2 in cancer cells is not clear. However, since p33ING2 is capable of activating both the Bax and p 2 1 W a f l promoters, loss of p33ING2 may impair the proper regulation of cell cycle and apoptosis and lead to cellular transformation and tumorigenesis (Nagashima et al, 2001). Functions of P33 ING2 The first evidence supports p33ING2 as a tumor suppressor gene is based on the finding that overexpressed p33ING2 can strongly inhibit colony formation in wi ld type p53 containing colorectal carcinoma R K O cells, but not as completely as in R K O E6 cells which contain inactive p53. The tumor suppression function o f p33ING2 is further supported by the finding that its expression is specifically induced by D N A damage agents including etoposide and neocarzinostatin, but not by y irradiation, doxorubicin, bleomycin or cw-platinum. The molecular mechanism underlying this function is due to the ability o f p33ING2 to increase the acetylation of p53, therefore stabilizes p53 and further enhances the promoter activities of p 2 1 W a f l and Bax (Nagashima et al, 2001). Consistent with this, people in our lab also found that overexpression of p33ING2 induces more apoptosis in UVB-irradiated and non-irradiated melanoma M M R U cells and the enhancement of apoptosis requires the existence of functional p53. Furthermore, we found overexpressed p33ING2 significantly downregulates antiapoptotic molecule Bcl-2 expression, resulting in an increased Bax/Bcl-2 ratio. Moreover, p33ING2 can promote the translocation of Bax to mitochondria, alter the mitochondrial membrane potential and initiate mitochondrion-mediated apoptotic pathway (Chin et al, 2005). The severity of D N A damage decides whether the cells w i l l go through D N A repair or apoptosis. Our previous study has shown that overexpressed p 3 3 I N G l b increases nucleotide excision repair (NER) o f UV-damaged D N A in a p53 dependent manner (Cheung et al, 2001). The structural and functional similarities between p33INGlb and p33ING2 prompted us to analyze the role of p33ING2 in N E R . We found that p33ING2 significantly enhances N E R in melanoma cell line in a p53 dependent manner by rapidly inducing the acetylation of histone H4, chromatin relaxation and facilitating the recruitment of the photolesion-recognition protein X P A - 12 -to the D N A damage site (Wang J et al, 2006). The recruitment o f X P A to the D N A damage site is considered to be a rate-limiting process for N E R (Thoma et al, 2003). Cellular senescence, a process originally described by Hayflick in 1965, prevents normal human fibroblasts from growing indefinitely in culture (Hayflick, 1965). The fact that p53 acetylation, which is required for its full activity, increases in replicative senescence and the capability of p33ING2 to increase p53 acetylation prompted people to investigate the involvement of p33ING2 in replicative senescence (Pearson et al, 2000; Nagashima et al, 2001). Their studies demonstrated that expression levels of ING2 and p53 lysine 382 acetylation are elevated during replicative senescence in human fibroblasts. Theoretically, p33ING2 can enhance the binding of p53 to histone acetyltransferase p300 and acts as a cofactor for p300-mediated p53 acetylation. In addition, p33ING2 also regulates the onset of replicative senescence since overexpressed ING2 in young fibroblasts can induce premature senescence in a p53-dependent manner while down-regulation of p33ING2 decreases p53 acetylation and delays the onset of replicative senescence (Pedeux et al, 2005). Gozani and colleagues first reported that ING2 functions as a nuclear phophoinositide (PtdlnsP) receptor through its P H D domain by performing a library expression screening using PtdlnsP-affmity resins and found p33ING2 can bind to both the PtdIns(3)P and PtdIns(5)P. Furthermore, they demonstrated that the ING2 P H D finger interacts with PtdIns(5)P in vivo and this interaction regulates the ability o f p33ING2 to activate p53 and p53-dependent apoptosis. Moreover, the P H D finger is found to be a general PtdlnsP bindng molecule (Gozoni et al, 2003). -13 - Modes of P33 ING2 Ac t ion Except for promoting p53 posttranslational modification that activates the p53 tumor suppressor protein, ING2 is a stable component of a S i n 3 A - H D A C complex (Doyon et al, 2006) and it can also binds to histone acetyl transferase p300. Recent studies found that the I N G P H D domains are specific binding modules for di- or tri-methylated lysine 4 o f histone H3 . P33ING2 especially binds with high affinity to the trimethylated species. The interaction is pivotal to stabilize the S i n 3 A - H D A C complex at the promoters of proliferation genes, leading to active gene repression in response to D N A damage (Shi et al, 2006). 1.3 Objective The primary objective of this study is to understand the role of tumor suppressor p33ING2 in melanoma. We, therefore, evaluated the expression of p33ING2 in different stages of melanocytic lesions and correlated the expression pattern to various clinicopathological parameters and 5-year patient survival. Moreover, we also evaluated the mutational status of p33ING2 gene using the c D N A from melanoma cell lines and genomic D N A from formalin-fixed, paraffin-embedded normal tissues as well as melanoma tissues. A t last, c D N A microarray was performed to analyze the transcriptional changes in M M R U melanoma cell line treated with U V R or stably transfected p33ING2. - 14-Figure 1.1 Structure features of the ING family proteins P33INGlb PIP NLS P H D ING2 Leucine zipper NLS P H D ING3 NLS P H D ING4 NLS P H D NLS ING5 NLS P H D NLS - 15 -C H A P T E R 2 . M A T E R I A L S AND M E T H O D S 2.1 T M A Construction Formalin-fixed, paraffin-embedded tissue blocks containing normal nevi, dysplastic nevi, primary melanomas and metastatic melanomas were used in this study. Sample collection was approved by the University of British Columbia and was performed in accordance with the Declaration of Helsinki Guidelines. A l l the tissue samples were from 1990 to 1997 archives of the Department o f Pathology, Vancouver General Hospital. For each case, the most representative lesion area was selected and marked on haematoxylin-eosin-stained slides. Taking into account the limitations of the representative areas of the tumor, duplicate 0.6-mm-diameter tissue cores were taken from each biopsy and the T M A s were assembled using a tissue array instrument (Beecher Instruments, Silver Spring, M D , U S A ) . Mult iple 4-um sections were cut with a Leica microtome and then transferred to adhesive-coated slides using routine histology procedures. One section from each T M A was routinely stained with haematoxylin and eosin. The remaining sections were stored at room temperature for immunohistochemistry staining. 2.2 Immunohistochemistry of T M A The T M A slides were deparaffmed by heating at 55°C for 30 min following by three washes with xylene, 5 min each. Tissues were then rehydrated in a series of 5 min washes in 100, 90 and 70% ethanol and rinsed with phosphate-buffered saline (PBS). Antigen retrieval was performed by heating the samples at 95°C for 30 min in 10 m M sodium citrate (pH 6.0). Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide for 20 min and nonspecific binding was blocked by universal blocking serum ( D A K O Diagnostics, Mississauga, Ontario, - 1 6 -Canada) for 30 min. The primary polyclonal rabbit anti-ING2 antibody Ping2, a kind gift of Dr. C C . Harris (NIH, Bethesda, M D , U S A ) , was diluted 1: 250 and incubated at 4°C overnight. After three washes, 2 min each with P B S , the sections were incubated with a biotinylated goat anti-rabbit secondary antibody for 30 min (Santa Cruz Biotechnology, Santa Cruz, C A , U S A ) . After three washes with P B S for 2 min each, horseradish peroxidase-streptavidin (Santa Cruz Biotechnology) was added to the section for 30 min, followed by another three washes, 2 min each with P B S . The samples were developed with 3,3'-diaminobenzidine substrate (Vector Laboratories, Burlington, Ontario, Canada) for 7 min and counterstained with haemotoxylin. Then the slides were dehydrated following a standard procedure and sealed with coverslips. Negative controls were performed by omitting ING2 antibody during the primary antibody incubation. 2.3 Evaluation of Immunostaining ING2 staining intensity was evaluated blinded by three independent observers (including one dermatopathologist) simultaneously, and a consensus score was reached for each core. In all, 11 normal nevi and 57 dysplastic nevi were evaluated for ING2 staining, and informative tumor staining and complete clinicopathological information were obtained in 79 primary melanoma cases and 43 metastatic melanoma cases. The staining intensity was scored as negative (0), weak (1), moderate (2) and strong (3). For ING2 staining intensity, 87% o f the biopsies have uniform staining between different cores. In the 13% cases with a discrepancy between duplicated cores, the average score was obtained. In addition, the percentage of cells showing staining in the nucleus was assessed by counting a minimal 400 cells per tissue core and the average percentage of duplicate biopsy cores was calculated. The multiplication of the average intensity score (0-3) - 17-and the average percentage (0-100%) was used as the final staining score for statistical analysis. The range for the final score was from 0 to 3. 2.4 Statistical Analysis Nuclear ING2 expression among different melanocyte lesions as well as its correlation with clinicopathological parameters of the melanoma patients, including age, gender, tumor thickness, location, histological subtype and tumor ulceration status was evaluated by Mann-Whitney test or Kruskal-Wallis test. Survival curves were plotted according to Kaplan-Meier method and the comparison of survival curves was performed with the log-rank test. The Mann-Whitney and Kruskal-Wallis tests were performed by using GraphPad Prism and the survival analysis was performed with SPSS 11.5 (SPSS, Chicago, IL, U S A ) . A P-value <0.05 was considered significant. 2.5 Cell Lines and Cell Culture Ten melanoma cell lines were used for this study. The M M A N , M M R U and P M W K cell lines were kind gifts from Dr H.R. Byers, Boston University, School of Medicine, Boston, M A ) . The M E W O , Sk-mel-3, Sk-mel-24, Sk-mel-93, Sk-mel-110, K Z - 2 , K Z - 2 8 cell lines were kind gifts from Dr. A . P . Albino (Memorial Sloan Kettering Cancer Center, New York, U . S . A . ) . Normal melanoma cell line was purchased from Clonetics (Walkersville, M D , U . S . A . ) . A l l melanoma cell lines were maintained in Dulbecco's modified Eagle's medium (Invitrogen), supplied with 10% fetal bovine serum, 100 units/ml of penicillin, 100 ug /ml of streptomycin in a 5% C O 2 atmosphere at 37°C. Normal melanocytes were cultured in melanocyte growth medium (Clonetics) at 37°C in a 5% C O 2 atmosphere. - 18-2.6 Normal and Melanoma Tissues Most normal tissue and tumor tissue (n=15) sections (12-um) were obtained from formalin-fixed, paraffin embedded blocks from the 1990-1998 archives of the Department of Pathology, Vancouver General Hospital. The tissues were dissected under a microscope and the paraffin was removed through three washes, 5 min each of xylene followed by 100% ethanol dehydration prior to D N A extraction. In addition, normal tissue samples N M - 3 C , N M N - 2 B and tumor tissue samples M M - 4 , M M - 3 A and M M - 2 B were kind gifts from Dr. Youwen Zhou and all these samples were stored in liquid nitrogen before use. Fresh melanoma tissue sample S-9650069 was also obtained from the Department of Pathology, Vancouver General Hospital. 2.7 U V Irradiation Medium was removed and the cells (at 80% confluency) were rinsed with P B S and exposed to U V B (280-320nm) using a bank of four unfiltered FS40 sunlamps (Westinghouse, Bloomfield, NJ). The Petri-dish cover was left on to filter possible U V C emissions from the U V B bulb. Medium was replaced and cells were incubated in a 5% C O 2 incubator at 37°C for desired time periods after U V B irradiation. The intensity of the U V light was measured by the IL 700 radiometer fitted with a W N 320 filter and an A127 quartz diffuser (International Light, Newburyport, M A ) . 2.8 Plasmids, Transfection and P33ING2 Stable Clone Generation The p c D N A 3 - I N G 2 plasmid (kind gift from Dr. C C . Harris, National Cancer Institute, National Institutes of Health, Bethesda, M D ) was transfected into the M M R U cells using Effectene reagent (Qiagen, Mississauga, O N , Canada). For the ING2 stable clone generation, transfected - 19-M M R U cells were incubated at 37°C for 48h followed by a 14-day selection in the culture medium supplemented with 1 mg/ml of G418 (Sigma). Single clones were then picked up and maintained in the culture medium containing 500 ug/ml of G418. 2.9 Western Blot Analysis Cel l pellets were lysed and extracted in 50 ul of triple detergent buffer (50 m M Tr i s -HCl [pH8.0], 150 m M N a C l , 0.02 % N a N 3 , 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate) containing freshly added protease inhibitors (100 ug /ml of phenylmethylsulfonyl fluoride, 1 ug/ml of aprotinin, 1 ug/ml of leupeptin, 1 ug/ml of pepstatin A ) . The protein concentration was determined by a Bradford assay, and Western blot was performed as previously described (Chin et al, 2005). The primary antisera included anti-p33ING2 rat monoclonal antibody (a kind gift from Dr. O. Gozani and Dr. J. Yuan, Harvard Medical School, Boston, M A ) and anti-/?-actin mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, C A ) . 2.10 S R B C e l l Surv iva l Assay M M R U cell line and ING2 stable cell line 17 were grown in 24-well plates at 50% confluency. They were irradiated with U V B at 300 J/m 2 and 600 J/m 2 24h after seeding the cells. Cel l survival was determined with the sulforhodamine B (SRB) (Sigma) assay 24 hours after U V R . S R B is a pink dye with two sulfonic groups that bind to basic amino acid residues in cells fixed with trichloroacetate ( T C A ) . S R B provides a sensitive index of cellular protein content. Briefly, the medium was removed and the cells were fixed with 500 ul of 10% T C A for 2 h at 4°C after treatment. The cells were then washed five times with tap water and the excess water was removed by flicking. The cells were air-dried and then stained with 500 ul of 0.4% S R B - 2 0 -(dissolved in 1% acetic acid) for 30 min at room temperature, washed four times with 1% acetic acid, and air-dried. The cells were then incubated with 500 ul of 10 m M Tris (PH 10.5) on a shaker for 20 min to solubilize the bound dye. Spectrophotometric readings were then taken at 550nm for 100 ul aliquots. 2.11 DNA and R N A Extraction Total R N A was prepared by TRIzol extraction (Invitrogen) and reverse transcribed into c D N A with the Superscript First-Strand Synthesis Systerm (Invitrogen) according to the manufacture's protocol. For the genomic D N A extraction, the tissues was incubated with lysis buffer (10 m M Tris.CI, 0.1 M N a C l , 1 m M E D T A ) containing freshly added proteinase K (20mg/ml) (Invitrogen) and followed by phenol chloroform extraction using standard protocols. 2.12 Polymerase Chain Reaction Polymerase chain reaction (PCR) was performed to amplify the whole p33ING2 coding sequence from c D N A of melanoma cell lines and normal melanocyte using following primers: forward 5'-C G C G G A T C G G C A G G A T G T T A - 3 ' and reverse 5' - T G G A T G G C C T T T A C T A C C T C - 3 ' ; The primers used to amplify the N L S and P H D domain of p33ING2 from genomic D N A of the normal and melanoma tissues are: forward 5 ' - A A A A T C G G G C A A G A C A A A T G - 3 ' and reverse 5 ' - T G T G G A T G G C C T T T A C T A C C T C - 3 ' . Hotstart P C R was performed with Taq D N A polymerase reaction system (Qiagen). Amplification was carried out as follows: (i) initial denaturation at 96°C for 3 min; (ii) denaturation at 96°C for 1 min; (iii) annealing at 55°C for 1 min; (iv) polymerization at 72°C for 2 min; (v) repeats of step i i - iv for 35 cycles; (vi)fmal -21 -polymerization at 72°C for 10 min. Samples were then electrophoresed on 2% agarose gels containing 0.5 pg /ml of ethidium bromide and visualized under U V irradiation. 2.13 D N A Sequencing P C R products were purified using a Qiagen P C R Purification K i t , re-amplified at following conditions using forward and reverse primers respectively: (i) 96°C 1 min; (ii) 96°C 10 sec; (iii) 50°C 5 sec; (iv) 60°C 4 min; (v) repeats of step i i - iv for 25 cycles and then re-purified using 100% ethanol and 3 M sodium acetate (PH 5.2) for direct sequencing. Sequencing was performed using a B i g Dye Terminator K i t (ABI) and analyzed on an A B I P R I S M 310 Genetic Analyzer. Sequence chromatograms were scrutinized by eye to confirm the results. 2.14 c D N A microarray The whole c D N A microarray analysis was done by cooperating with the Microarray Facility of The Prostate Center at Vancouver General Hospital. R N A s were extracted from M M R U and ING2 stable cell lines which treated with or without 100 J /m 2 U V R using the standard Trizol method. The quality and quantity of R N A was measured using the Agilent 2100 bioanalyzer and R N A 6000 N A N O kit (Agilent Technologies, Palo Alto , C A ) . Human Operon v.2.1 (2IK) glass arrays were produced (based on human 70 mers from Operon, Huntsville, A L ) . Total R N A from test samples and universal human reference R N A (Stratagene, Cedar Creek, T X ) were differentially labeled with Cy5 and Cy3 respectively with the 3 D N A array detection 350 kit (Genisphere, Hatfield, P A ) and cohybridized to c D N A microarrays as previous described method (Kojima et al, 2006). Briefly, reverse transcription incorporates a specific sequence present at the 5' end of the R T primer. Then the c D N A was hybridized to the array overnight at 42°C. After - 2 2 -stringent washing, the fluorescent 3 D N A reagent, which includes a "capture sequence" complementary to the sequence at the 5' end of the R T primer, was hybridized to the c D N A (47°C for 2-3 h). Following further washing, the arrays were immediately scanned on a Scan Array Express Scanner (PerkinElmer, Boston, M A ) . Image analysis and quantification were conducted with commercial software (Imagene 6.0 software:Biodiscovery Inc, E l Segundo, C A ) . After grid assignment, the adjusted intensity for each gene was calculated by subtracting the background median from the signal median. This value was then used as the input for the Genespring 7.2 program (Silicon Genetics, Redwood City, C A , U S A ) , which allows multiple further comparisons using data from different experiments. List o f differentially expressed genes with greater than 2 fold expression were generated and unpaired t-tests were subsequently performed. 2.15 Quantitative Real Time P C R The same R N A samples that had served for the microarray analysis were used here. The primer sequences for A T F 3 are 5' -C A G G T C T C T G C C T C G G A A G T - 3 ' (forward) and 5'-C A A A G G G C G T C A G G T T A G C A - 3 ' (reverse); TIG1 are 5 ' - G C C G C G C G T C C A T T A A T - 3 ' (forward) and 5 ' - C G T C C C T C A C C T T C C T G A A G - 3 ' (reverse) and P-actin are 5'-G C T C T T T T C C A G C C T T C C T T - 3 ' (forward) and 5' C G G A T G T C A A C G T A C C A C T T - 3 ' (reverse) . A l l the P C R reactions were performed in duplicated in a total of 25 pi reaction mix including Platinum S Y B R Green q P C R SuperMix-UDG with Rox (Invitrogen), primers and a portion of each c D N A . P C R was carried out at 50°C for 2 min, then 95°C for 10 min, followed by 40 cycles at 95 °C for 15 sec and 60°C for 1 min. The cycle number of threshold (CT) was recorded for each reaction and the d value of A T F - 3 was normalized to that of (3-actin. -23 -C H A P T E R 3. N U C L E A R ING2 EXPRESSION IS R E D U C E D IN H U M A N C U T A N E O U S M E L A N O M A S 3.1 Rationale and Hypothesis Cutaneous malignant melanoma is highly resistant to conventional radio- and chemo-therapy. Dacarbazine (DTIC), the only F D A approved drug for treatment of melanoma, yields a response rate of only 16% (Atallah and Flaherty, 2005). Therefore, better understanding of the molecular mechanism of melanoma progression and chemoresistance is needed for designing novel treatment regimes. P53, a central sensor linking D N A damage to apoptosis, plays an essential role in tumor suppression and chemosensitivity in many tumor types (L i et al, 1998; Raffo et al, 2000; Fridman et al, 2003). However, mutational analysis reveals that the p53 gene is altered in only approximately 11% of melanomas (Hussein, 2004). The low mutation rate oip53 suggests that other tumor suppressor genes may play important roles in pathogenesis of melanoma. Recent studies suggest that I N G family proteins function as tumor suppressors. Five members of I N G proteins have so far been identified and they all share a conserved P H D domain in the C-terminus (Campos et al, 2004). P 3 3 I N G l b is the founding member which has been shown to be able to induce cell cycle arrest, enhance D N A repair and promote apoptosis after D N A damage events (Garkavtsev et al, 1998; Cheung et al, 2001; Cheung and L i , 2002). ING2 is cloned through a homology search with p33INGlb and is found to be located to human chromosome 4 (Shimada et al, 1998). Downregulated expressions of I N G proteins have been reported in several tumor types including the loss of nuclear expression of p33INGlb in melanoma. A s ING2 exhibits 58.9% homology with p33INGlb , we first want to know i f aberrant expression of ING2 is involved in melanomagenesis. - 2 4 -3.2 Results 3.2.1 Clinicopathological Findings The clinicopathological features of the melanomas for this study are summarized in Table 3.1. In primary melanomas, 79 cases (50 male and 29 female) were available for the evaluation of ING2 staining. The median age of the patients was 58 y ranging from age 25 to 92. There were 12 cases of R G P and 67 cases of V G P . For the thickness of these primary melanomas, 23 were <1.0 mm, 27 were 1.01-2.0 mm, 15 were 2.01-4.0 mm, and 14 were >4 mm. For the tumor subtype, superficial spreading melanoma accounted for 36 cases, lentigo maligna melanoma 13 cases, acrolentigous melanoma 2 cases, nodular melanoma 13 cases, and the remaining 15 cases were unspecified. The majority of the melanomas located in sun-protected sites (65 cases, trunk, arm, leg and feet) while 14 located in sun-exposed sites (head and neck). Tumor ulceration was found in 16 patients. Forty-three out o f 50 metastatic melanoma cases were available for ING2 staining. There are 29 male and 14 female, with age ranging from 27 to 89 (median 59 y). Clinically, dysplastic nevi can be identified as mild, moderate, or severe. While mildly and moderately dysplastic nevi can be closely observed, severely dysplastic nevi should certainly be surgically removed. In this study, information on 53 of 57 dysplastic nevi are available for sub-categorization: 22 mild, 20 moderate, and 11 severe. 3.2.2 Reduced ING2 Nuclear Expression in Human Melanomas We examined ING2 nuclear expression in normal nevi, dysplastic nevi, primary melanomas and metastatic melanomas by immunohistochemistry (Figure 3.1). There is no significant difference in ING2 nuclear expression between normal nevi and dysplastic nevi (P>0.05, Mann-Whitney test). No difference in ING2 expression was found among mildly, moderately, or severely -25 -dysplastic nevi (data not shown). However, reduced ING2 expression was observed in R G P and V G P primary melanomas as wel l as metastatic melanomas compared with dysplastic nevi (P=0.029, P=0.0001 and.P=0.0286 respectively, Mann-Whitney test). There are no significant differences in ING2 nuclear expression among R G P , V G P , and metastatic melanomas (P>0.05, Kruskal-Wallis test) (Figure 3.2). 3.2.3 Correlation between ING2 Nuclear Expression and Clinicopathological Parameters or 5-year Patient Survival Tumor thickness and ulceration are well-known indicators for melanoma prognosis. However, no correlation was found between ING2 nuclear staining and these parameters (P>0.05 for both) (Figure 3.3). In addition, no association was found between ING2 nuclear expressions and other clinicopathological parameters including age, gender, subtype and location of tumors (data not shown). To investigate whether ING2 expression was correlated with patient survival, Kaplan-Meier survival curves were plotted to see i f there was a relationship between ING2 nuclear expression and five-year patient survival in primary melanomas or metastatic melanomas. We defined the staining as strong i f the score is between 1.51 and 3.0 or weak i f the score is <1.5. Our results showed that ING2 nuclear expression did not significantly correlate with both 5-year overall and disease-specific patient survival in primary and metastatic melanomas (.P>0.05, log-rank test) (Figure 3.4). - 2 6 -Table 3.1 Clinicopathological parameters of 122 cases of melanomas No. of Patient % Primary melanomas Age <58 38 48 >58 41 52 Gender Male 50 63 Female 29 37 Tumor thickness (mm) <1 23 29 1.01-2 27 34 2.01-4 15 20 >4 14 17 Ulceration Absent 63 80 Present 16 20 Tumor subtype Superficial spreading melanoma 36 46 Lentigo maligna melanoma 13 16 Acrolentigous melanoma 2 3 Nodular melanoma 13 16 Unspecified 15 19 Tumor growth phase Radial growth phase 12 15 Vertical growth phase 67 85 Sitea Sun-protected 65 82 Sun-exposed 14 18 Metastatic melanomas Age <59 22 51 >59 21 49 Gender Male 29 67 Female 14 33 Sun-protected sites: trunk, arm, leg and feet. Sun-exposed sites: head and neck. -27-F i g u r e 3.1 Representative images of ING2 immunohistochemical staining in human melanocyte lesions. Strong ING2 expression in adjacent normal epidermis (A ) , normal nevi (B), dysplastic nevi (C ) , and weak ING2 staining in primary melanoma (D) and metastatic melanoma (E) . Arrows indicate strong staining in melanocyte. Magnification, X 4 0 0 . -28 -Figure 3.2 ING2 nuclear expression at different stages of melanocyte lesions. There are less ING2 nuclear expression in R G P , V G P and metastatic melanomas compared with dysplastic nevi (P=0.029, 0.0001 and 0.0286, respectively, Mann-Whitney test). e n cs as o =3 C CM (3 3 - 1 5 2 N N ***** —f— DN R G P VGP MM - 2 9 -Figure 3.3 N o correlation was found between ING2 nuclear expression and tumor thickness (P>0.05, Kruskal-Wallis test) (A) or tumor ulceration (P>0.05, Mann-Whitney test) (B) of primary melanomas. A 3-c "c •5 2 • tn co J£ o c <M 1 • O B 3-H ;1 1-2 2-4 Tumour thickness (mm) >4 "c "5 2 « 3 e 1 <3 Absent Present Ulceration - 3 0 -Figure 3.4 ING2 nuclear expression and 5-year patient survival. ING2 nuclear expression is not correlated with 5-year overall (A, C ) and disease-specific survival ( B , D ) in primary melanoma patients (A, B ) or metastatic melanoma patients (C, D ) . B 100 CO > w E o 90 80 70 60 P= 0.1556 0-1.5 -1.51-3 — i ( 1 1 1 1 10 20 30 40 50 60 Time (months) Time (months) Time (months) i 1 1 r 1 1 10 20 30 40 50 60 Time (months) - 3 1 -3.3 Discussion The main purpose of this study is to investigate i f the novel tumor suppressor ING2 is aberrantly expressed in human cutaneous melanomas. Using tissue microarray technology and immunohistochemistry, we for the first time demonstrated that ING2 nuclear expression is reduced in human melanomas compared to dysplastic nevi (Fig.3. 2). Although a number of studies indicated that ING2 possesses tumor suppressive functions, such as inducing growth arrest, senescence, apoptosis, and enhancing D N A repair (Nagashima et al, 2001; Gozani et al, 2003; Pedeux et al, 2005; Chin et al, 2005; Wang J et al, 2006), there is limited information on ING2 expression level in human cancers. There is only one recent report by Okano et al (2006) showing that ING2 expression is reduced in 6 of 7 lung cancer cell lines. However, these authors did not detect any ING2 mutation in 31 human lung cancer cell lines and 30 lung cancer biopsies. Although the reason for reduced ING2 expression in human melanomas is unclear, we cannot rule out mutation o f the ING2 gene in melanoma because different tumor types may have different mechanisms for gene inactivation. For example, reduced ING1 expression was found in 73% of non-small cell lung cancer biopsies (Kameyama et al, 2003) and 44% of breast cancer primaries (Toyama et al, 1999), while no missense mutation was found in these lung cancer biopsies and only 1 missense mutation was found in 377 breast cancer carcinomas (0.27%). On the other hand, aberrant I N G l expression was associated with a much higher mutation rate of the ING1 gene in human primary melanomas (Campos et al, 2004). In addition, loss of heterozygosity of the region 4q32 in the long arm of chromosome 4, which includes ING2 and SAP30, was found in 20% o f basal cell carcinomas (Sironi et al, 2004), suggesting that genetic alteration of the ING2 gene does occur in human tumors. Future studies on ING2 expression and - 3 2 -gene mutation in different cancer types w i l l provide further evidence on the important role of this tumor suppressor in the pathogenesis of human cancers. Similar levels of ING2 nuclear expression in melanomas regardless of the growth phases (RGP vs V G P ) (Fig. 3.2), tumor thickness or ulceration (Fig. 3.3) suggest that reduced ING2 expression may be involved in the initiation, rather than progression of melanoma. We have recently shown that ING2 plays an essential role for maintaining genomic stability upon U V irradiation. ING2 acts as a D N A damage sensor for nucleotide excision repair, as physiological level of ING2 is required for rapid induction of histone H4 acetylation, chromatin relaxation, and the recruitment of repair recognition factor X P A to the photolesion sites (Wang J et al, 2006). ING2 can also enhance the removal of D N A damage by triggering the apoptosis process when the D N A damage is too severe. We have found that overexpression o f ING2 significantly enhances UV-induced apoptosis by downregulating the expression of Bcl -2 , promoting the translocation of the Bax protein to the mitochondria, resulting in the mitochondrial membrane potential change, release of cytochrome c and activation of caspases-9 and -3 (Chin et al, 2005). Since U V radiation is the main environment factor for melanoma formation, reduced ING2 expression would impair the removal of UV-damaged D N A , leading to gene mutation and malignant cell transformation. Our data that no significant correlation between ING2 expression and melanoma thickness supports a multiple genetic change model during melanoma pathogenesis. Many different molecules in the apoptosis and survival pathways have been shown to be involved in melanoma initiation and progression. It appears that defects in the apoptosis pathway mostly contribute to the initiation step in melanoma pathogenesis, while activated survival pathway often correlates with tumor progression. For example, the expression of pro-apoptotic factor Apaf-1, a p53 -33 -downstream effector, which links the release of cytochrome c to the activation of caspase-9, is reduced in human cutaneous melanomas compared with normal nevi (Dai et al, 2004). The expression of another p53 downstream effector P U M A is significantly reduced in melanoma compared to dysplastic nevi (Karst et al, 2005). Furthermore, the expression of proapototic protein X A F 1 which blocks the XIAP-mediated inhibition of caspase-3 (Liston et al, 2001) is also reduced in human melanomas (Ng et al, 2004). However, the expression of Apaf-1, P U M A and X A F 1 expression did not correlate with tumor thickness. On the other hand, increased expression of integrin-linked kinase and phospho-Akt in the PI3 kinase survival pathway has been observed in melanoma and correlated with tumor thickness and 5-year patient survival (Dai et al, 2003; 2005). Furthermore, reduced expression of P T E N , the negative regulator of the PI3 kinase, was also found to be reduced in melanoma and significantly associated with tumor thickness (Goel et al, 2006). However, some factors are involved in more than one step in melanoma pathogenesis. For instance, although P U M A expression does not correlate with tumor thickness in primary melanoma, it is further reduced in metastatic melanoma and weaker P U M A expression is correlated with a poorer 5-year patient survival (Karst et al, 2005). Based on the complexity of the apoptotic and survival pathways which govern the fate of a cell, additional studies on the timing of the gene inactivation/overexpression in these pathways from the same set of tumor biopsies and the interdependence among these events w i l l provide a more complete picture of the molecular changes during melanoma pathogenesis. Our data that reduced ING2 expression does not significantly correlate with 5-year patient survival (Fig. 3.4) is consistent with the findings that ING2 reduction is an early event in the R G P and does not correlate with tumor progression (Fig. 3.2, 3.3). On the other hand, although it is not statistically significant, we observed that patients with weaker I N G 2 expression have a - 3 4 -trend towards better prognosis for both overall and disease-specific 5-year patient survival in primary melanomas (Figure 3.4). Interestingly, decreased expression o f p33 INGlb was also associated with better prognosis in childhood acute lymphoblastic leukemia patients as well as in invasive bladder cancer (Nouman et al, 2002; Sanchez et al, 2003). Although the mechanism underlying these phenomena is unclear, we propose that the enhancement of D N A repair by ING2 (Wang J et al, 2006) may account for a better trend of 5-year patient survival in weaker ING2 expression group because enhanced D N A repair have been shown to reduce the cytotoxicity of anticancer drugs (Siddik et al, 2003). Future studies on the correlation between ING2 expression and patient survival in a large set of melanoma biopsies from patients with and without chemotherapy w i l l provide a clearer answer. In summary, we found that ING2 nuclear expression is reduced in human cutaneous melanomas compared to normal or dysplastic nevi. These data, together with our previous findings that ING2 plays essential roles in cellular stress responses (Chin et al, 2005; Wang J et al, 2006), suggest the importance o f this tumor suppressor in the pathogenesis of melanoma. -35 -C H A P T E R 4. M U T A T I O N A L S T A T U S O F P33ING2 I N M E L A N O M A S 4.1 Rationale and Hypothesis Melanomagenesis is a multistep progress which requires the activation of oncogenes and the inactivation of the tumor suppressor genes. The molecular abnormalities attributable to the progression from dysplastic nevi to malignant melanoma affect cell growth, D N A repair, and susceptibility to cell death. For instance, in 25 to 50 percent of cases of non-familial melanoma, the tumor suppressor gene, phosphatase and tensin homologue (PTEN), is inactivated by mutation (L i et al, 1997; Steck et al, 1997). In murine models of melanoma, mutation of either CDKN2A or PTEN alone fails to cause melanoma, but when combined with each other or with mutations in other genes, melanomas arise (You et al, 2002). U V R is the primary environmental cause for melanoma formation. Our group has shown that p33ING2 is capable o f regulating melanoma cellular stress response to U V R either by enhancing nucleotide excision repair or promoting apoptosis (Chin et al, 2005; Wang J et al, 2006). We also found that reduced p33ING2 is important for the progression from the dysplastic nevi to malignant melanoma (Lu et al, 2006). Considering the importance of p33ING2 in melanoma, we therefore, investigated the mutational status of the p33ING2 gene in human cutaneous melanomas. - 3 6 -4.2 Results 4.2.1 Alterations of P33ING2 Gene in Melanoma Cell Line We searched for mutations in the whole coding regions of the p33ING2 in ten melanoma cell lines by direct sequencing of R T - P C R products. We also sequenced a normal melanocyte cell line to determine whether the variant was tumor specific. We found except for normal melanocytes, K Z - 2 and Sk-mel-110 melanoma cell lines, all other eight melanoma cell lines contained the same C - » T mutation which located in the codon 13 in the leucine zipper domain. However, the mutation did not change the amino acid (Ala—>Ala). The c D N A from normal melanocytes did not have this mutation indicating it is tumor specific. Surprisingly, one melanoma cell line Sk-mel-110 contains another twelve mutations including three mutations residing in the P H D domain o f p33ING2. A l l the mutated nucleotides located at the third nucleotide of the codon and thus did not change the coded amino acid (Table 4.1; Figure 4.1). U V signature C - » T transitions were found at codon 121, 142, 165. The other changes are transitions A—»G; G—>A; T—»C and transversions C—>A; A—>C. A l l o f the point mutations were confirmed by sequencing from both ends. 4.2.2 Alterations of the NLS and PHD Domain of P33ING2 Gene in Melanoma Taking into account of the limited numbers of the melanoma cell lines, we further examined the mutational status of p33ING2 gene in twelve melanoma biopsies and eight normal tissue biopsies. The primers we used amplified around 400 base pairs including both the N L S and P H D domain of p33ING2. We found that genomic D N A from both the normal tissues and melanoma tissues did not show any mutations. - 3 7 -Table 4.1 ING2 alterations in normal melanocyte and melanoma cell lines Cell Line Mutation Type Location Normal Melanocyte N o KZ-2 N o KZ-28 Silent mutation G C C - G C T Codon 13, L E L domain M E W O Silent mutation G C C - G C T Codon 13, L E L domain M M A N Silent mutation G C C - G C T Codon 13, L E L domain M M R U Silent mutation G C C - G C T Codon 13, L E L domain P M W K Silent mutation G C C - G C T Codon 13, L E L domain Sk-mel-3 Silent mutation G C C - G C T Codon 13, L E L domain Sk-mel-28 Silent mutation G C C - G C T Codon 13, L E L domain Sk-mel-93 Silent mutation G C C - G C T Codon 13, L E L domain Sk-mel-110 Silent mutation C A C - C A T Codon 121 Sk-mel-110 Silent mutation C G A - C G G Codon 133 Sk-mel-110 Silent mutation T C C - T C T Codon 142 Sk-mel-110 Silent mutation A G G - A G A Codon 154 N L S domain Sk-mel-110 Silent mutation C A C - C A T Codon 165 N L S domain Sk-mel-110 Silent mutation A T T - A T C Codon 170 N L S domain Sk-mel-110 Silent mutation A A A - A A G Codon 188 N L S domain Sk-mel-110 Silent mutation A A T - A A C Codon 210 Sk-mel-110 Silent mutation T G T - T G C Codon 242 P H D domain Sk-mel-110 Silent mutation C T T - C T C Codon 245 P H D domain Sk-mel-110 Silent mutation G A T - G A C Codon 261 P H D domain Sk-mel-110 Silent mutation A C A - A C C Codon 265 - 3 8 -F i g 4.1 Representative sequence chromatograms of ING2 mutations in Sk-mel-110 melanoma cell line. Arrows indicate the locations of the mutation. 10 8330 8340 8350 83G0 8370 838C 'T G T C C A A T T f T A A T G G T T T C f f C T T T T C A T G C G T T T C A C T C f r C C T A T A A A C T A A A G G G G A A A" TGTCCRRTTG RRTGGTTTCRC TTTTCRTGCGTTTCRCTCRCCTRT RRRC G RRRGGGC R R R 1 A — p r o m TED m G C A J T G G G A G A C A A T G A G A A A A C C f f T G G A C A A A A i T T A C T G A A A G C RG GG G RG RC RR T G RGRRRRC CRT GGRC Rflfl AG T RC T G flflfl - 3 9 -4.3 Discussion P33ING2 is a tumor suppressor candidate. It can negatively regulate cell growth in a p53-dependent manner through induction of G l cell cycle arrest and apoptosis (Nagashima et al, 2001). It also interacts with phosphoinositides, Ptdlns(3)P and Ptdlns(5)P through its P H D zinc finger, which plays a crucial role in D N A damage-initiated stress signaling (Gozani et al, 2003). Moreover, significantly reduced p33ING2 expression has been reported in melanomas compared to dysplastic nevi (Lu et al, 2006). To further reveal the role of p33ING2 in melanoma formation, in this study we screened for p33ING2 mutations in 10 melanoma cell lines and 12 melanoma biopsies. We did not find any missense mutations in this gene which is consistent with a previous finding that no mutation of ING2 was detected in 30 human lung cancer cell lines and 31 primary lung cancer tumors (Okano et al, 2006). We observed same C —»T synonymous variant which is located in codon 13 of the L E L domain in eight of ten melanoma cell lines (Table 4.1). Okano et al also showed same locus variant existing in six of thirty-one lung cancer tissues (Okano et al, 2006). However, these authors reported a T to C change which is incorrect in our opinion. The sequence from PubMed (NM_001564) shows cytosine instead of thymidine in this location suggesting twenty-five instead of six lung cancer tissues contain this variant in the study by Okano et al. The L E L domain of p33ING2 is important for interacting with other L E L domain containing proteins and hydrophobic interactions and a previous study from our group also indicates that the L E L domain is important for the proper functions of p33ING2 in D N A repair, apoptosis and chromatin remodeling after U V irradiation (Wang Y et al, 2006). However, we believe this variant is most probably a common polymorphism in melanoma as it is a synonymous variant and C - » T mutation is a U V signature mutation. Especially, the mutation rate at this location is - 4 0 -quite high (80%). Previous studies indicated that within coding sequences the silent or synonymous polymorphisms are much more common than are changes that result in amino acid substitution, presumably because many amino acid changes interfere with normal function of the protein and are eliminated by natural selection. However, different alternative codons for the same amino acid may differ in speed and accuracy of transcription, and the m R N A corresponding to different alternative codons may have different accuracy and speed of translation. Although Sk-mel-110 melanoma cell line does not contain the silent mutation in codon 13, it contains another twelve synonymous variants in the whole coding sequence of p33JNG2. The specialty of Sk-mel-110 melanoma cell line has been reported before since it has five missense mutations of p53 although the mutation of p53 is rare in melanoma (Albino et al, 1994; Hussein, 2004). Besides the melanoma cell lines, we also analyzed the mutational status of p33ING2 using melanoma biopsies. We focused our studies on the P H D domain of p33ING2 since it is the most widely studied domain and it is pivotal for the normal function of p33ING2 in different experimental models (Gozani et al, 2003; Shi et al, 2006). However, we failed to detect any mutations of p33ING2 P H D domain in the tissue samples. The limited tissues presented here are due to the difficulty of obtaining high quality D N A from the formalin-fixed, paraffin-embedded tissues. Although some literatures suggest that high-temperature heating may improve D N A extraction from archival formalin-fixed, paraffin-embedded tissues (Coombs et al, 1999; Shi et al, 2002), we found the best method is still the traditional phenol-chloroform extraction followed by ethanol precipitation method. We also found that the genomic D N A quality to a larger degree depends on the size of the tissues and the presence of melanin or not. Normally, it is easier to -41 -obtain good quantity and quality of D N A from larger tissue biopsies while the presence of melanin w i l l affect the P C R reaction and lead to failure of the experiment. We also applied Sephadex 25 in our experiments, but it did not help get rid of the melanin from the D N A samples. In summary, our results suggest that p33ING2, like other I N G family members, is rarely mutated in human cancers. Other mechanisms may be responsible for the reduced p33ING2 expression in melanoma. First, since p33INGlb gene and flanking regions are highly G C rich and p33ING2 shows 58.9% identity with p33INGlb (Shimada et al, 1998; Gunduz et al, 2000), we postulate that epigenetic modification of p33ING2 promoter which can cause reduced expression of corresponding protein may be the reason for the loss o f p33ING2 expression in melanoma. Second, abnormal subcellular translocation of p33ING2 may happen during melanomagenesis. Aberrant cytoplasmic expression of p 3 3 I N G l b has been reported to be associated with malignancy in melanocyte lesions and the failure o f induction of cyclin dependent kinase inhibitor p 2 1 W A F 1 upon D N A damage stress (Nouman et al, 2002; Gong et al, 2006). In addition, we recently found that the nuclear p29ING4 is translocated to cytoplasm in melanoma (unpublished data). These results indicate that translocation of I N G family proteins is likely a common phenomenon in melanoma and more studies are required to clarify this possibility for p33ING2 in the future. - 4 2 -C H A P T E R 5. IDENTIFICATION O F GENES T R A N S C R I P T I O N A L L Y R E G U L A T E D B Y U V IRRADIATION OR P33ING2 5.1 Rational and Hypothesis U V R is the main environmental factor causing melanoma. Studies of U V R responses at the molecular level have advanced most quickly using primary melanocyte cultures and more readily melanoma cells. For instance, using normal melanocytes, it was found that U V R exposure induces a G l arrest in melanocytes that is at least partially attributable to both p53 (Barker et al, 1995) and p l 6 I N K 4 a (Pavey et al, 1999). Although much of the understanding of U V R responses in melanocytes are based on single-gene/pathway approaches, Valery et al undertook a 9,000-transcript human c D N A microarray approach to study U V R in melanocytes at the transcriptional level. They successfully identified 198 out of 9,000 genes which were shown to be altered (>1.9-fold). Altogether, 117 clones and 81 clones were suppressed and induced by U V R , respectively (Valery et al, 2001). On the other hand, deregulated apoptosis is believed to be another key factor contributing to melanomagenesis. I N G family proteins, as candidate tumor suppressors, are aberrantly expressed in different tumor types including melanoma (Nouman et al, 2002; Kameyama et al, 2003; Garkavtsev et al, 2004; L u et al, 2006). B y using c D N A microarray analysis in mouse mammary epithelial cells ( N M u M G ) after antisense ING1 -induced transformation, a Japanese group analyzed expression profiles of 2304 genes and their studies provided evidence that overexpression of antisense ING1 stimulated expression of 14 genes including cyclin B l and proto-oncogene DEK (Takahashi et al, 2002). -43 -We already found that ING2 nuclear expression is significantly reduced in melanomas compared with dysplastic nevi (Lu et al, 2006). In this study, we restored ING2 expression in melanomas by establishing ING2 stable clones and further identified the genes transcriptionally regulated by overexpessed ING2 or U V R using the M M R U melanoma cell line as an experimental model. -44-5.2 Results 5.2.1 Growth Inhibition of P33ING2 We first generated stable clones overexpressing p33ING2 in M M R U melanoma cells. Figure 5.1-A showed that compared with the parental M M R U cells, ING2 stable clones (SC) have much higher ING2 expression and S C I 6 and S C I 7 showed the highest expressions among all the clones. Consistent with previous finding (Wang Y and L i , 2006), we also noticed that the cells stably expressed p33ING2 are slightly bigger and less dendritic compared with the parental M M R U cells (Fig. 5.1B).We next showed that ING2 SC17 had reduced cell survival after various doses of U V B irradiation (Fig. 5.2). We thus chose S C I 7 for the following studies. 5.2.2 Gene Expression Analysis by cDNA Microarray The target samples we used in this study include parental M M R U cells and ING2 SCI7 , both were treated with or without U V B irradiation. The U V dose we used was 100 J /m 2 and the cells were harvested 6 hours after irradiation. The c D N A s derived from the target samples were fluorescently labeled with Cye5 which would emit red fluorescence and the c D N A s from human universal reference R N A were labeled with Cye3 which would emit green color. Then, the Cyt5 labeled target c D N A and Cyt3 labeled reference c D N A were applied simultaneously on the same microarray and the two fluorescent images were scanned with a fluorescence laser-scanning device. Each signal was normalized so that the Cyt5:Cyt3 intensity ratio of the housekeeping gene signal like ji-actin was 1. The red and green fluorescent signals indicated genes whose expression levels were relatively higher in target samples or in human universal reference R N A while yellow signals indicate genes with equal expression. First, we compared the gene expression differences between M M R U cells and M M R U cells which are treated with U V B -45 -irradiation. The final score is acquired by comparing the two Cyt5:Cyt3 ratio deriving from the untreated and treated M M R U cells. We considered genes that exhibited a final score >2 as significantly different genes. Overall, 361 genes were shown to be altered with 143 genes and 218 genes induced and suppressed by U V R , respectively. The top 10 induced genes are listed in Table 5.1. Among them, genes ATF-3 and CDKN1A (p2I CIP) are well known to be able to be induced by U V R . Table 5.2 listed top 10 genes suppressed by U V R . Except Smad 3, a member of Smad family and a key factor in wound healing and fibrogenesis, all the other genes are rarely studied. Similarly, we also evaluated gene expression changes between M M R U parental cells and ING2 stable transfected cells. We found that 17 genes are suppressed by overexpressed p33ING2 while 57 genes are induced. The top 10 induced and suppressed genes are listed in Table 5.3 and Table 5.4, respectively. We found that stably overexpressed p33ING2 significantly induced the expressions of genes like MMP-7 and NDRG-1 while reduced expression of genes including RARRES1 (TIG1). 5.2.3 Real T ime P C R Analysis To examine the reliability of the expression changes detected by the profiling analysis using the c D N A microarray, the real time P C R analysis with the same R N A samples that had served for the microarray analysis was performed to detect the most abundantly changes genes ATF-3 and RARRES1. The real time P C R results are consistent with the c D N A microarray analysis, although the induction of ATF-3 detected by real P C R is more dramatic (34.58 fold) than by using c D N A microarray analysis (18.03). - 4 6 -F i g 5.1 ING2 was overexpressed in stable clones. Overexpressed p33ING2 was detected in stable clones by western blot (A) and microscopic images were taken from M M R U and ING2 stable clone (B). - 4 7 -Fig 5.2 Cel l survival rate by S R B assay of UVB-irradiated M M R U cells and ING2 stable cell line. Columns, mean from duplicates; bars, SD. -48 -Table 5.1 Top 10 genes induced by U V irradiation in M M R U Gene Fold Induction PubMed Accession Number Description ATF-3 18.03 N M 001674, N M 004024 Cyclic-AMP-dependent transcription factor ATF-3 (Activating transcription factor 3 HEXJM1 11.59 NM_006460 HMBA-inducible; likely ortholog of mouse cardiac lineage protein 1; menage a quatre 1; hexamethylene bisacetamide-inducible protein GPR3 11.2 NM_005281 Probable G protein-coupled receptor GPR3 (ACCA orphan receptor). IL6 11.15 NM_000600 Interleukin-6 precursor (IL-6) (B-cell stimulatory factor 2) (BSF-2) (Interferon beta-2) (Hybridoma growth factor) (CTL differentiation factor) (CDF) BTG2 8.63 NM_006763 BTG2 protein (NGF-inducible anti-proliferative protein PC3) PLK3 6.65 NM_004073 Serine/threonine-protein kinase PLK3 (EC (Polo-like kinase 3) (PLK-3) (Cytokine-inducible serine/threonine-protein kinase) (FGF- inducible kinase) (Proliferation-related kinase) DLL1 6.39 NM_005618 Delta-like protein 1 precursor (Drosophila Delta homolog 1) (Delta 1) (H-Delta-1) FOSB 6.25 N M 006732 Protein fosB (G0/G1 switch regulatory protein 3). CDKN1A (P21CIP) 5.95 NM_078467 Cyclin-dependent kinase inhibitor 1 (p21) (CDK-interacting protein 1) (Melanoma differentiation associated protein 6) (MDA-6) CXCL2 5.48 NM_002089 Macrophage inflammatory protein-2-alpha precursor (MIP2-alpha) (CXCL2) (Growth regulated protein beta) - 4 9 -Table 5.2 Top 10 genes suppressed by UV irradiation in M M R U Gene Fold Suppression PubMed Accession Number Description LHFPL2 7.14 N M 005779 Lipoma HMGIC fusion partner-like 2 PUM2 6.76 N M 015317 Pumilio homolog 2 (Pumilio2) THRAP2 6.26 NM_015335 Thyroid hormone receptor associated protein 2; protein similar to TRAP240 PUM1 5.31 N M 014676 Pumilio homolog 1 (Pumilio 1) (HsPUM). NAV3 5.17 NM_014903 Neuron navigator 3; pore membrane and/or filament interacting like protein 1; steerin 3 BTEB1 4.95 NM_001206 Transcription factor BTEB1 (Basic transcription element binding protein 1) (BTE-binding protein 1) (GC box binding protein 1) (Krueppel-like factor 9) ASXL2 4.90 NM_018263 Additional sex combs like 2; polycomb group protein ASXH2 SASH1 4.59 N M 015278 S A M and SH3 domains containing protein 1 AMOTL1 4.58 NM_130847 Angiomotin like 1; junction-enriched and associated protein SMAD3 4.31 NM_005902 M A D , mothers against decapentaplegic homolog 3; mad protein homolog; mad homolog JV15-2; SMA-and MAD-related protein - 5 0 -Table 5.3 Top 10 genes induced by p33ING2 Gene Fold Induction PubMed Accession Number Description ILIA 6.94 N M 000575 Interleukin-1 alpha precursor (IL-1 alpha) MMP7 6.13 N M 002423 Matrilysin precursor GPR51 5.46 NM_005458 Gamma-am inobutyric acid type B receptor, subunit 2 precursor PAPPA 4.74 N M 002581 Pappalysin-1 precursor DLL1 3.97 N M 005618 Delta-like protein 1 precursor PLAC8 3.83 N M 016619 Placenta-specific gene 8 protein (CI5 protein) IGFBP3 3.61 NM_000598 Insulin-like growth factor binding protein 3 precursor (IGFBP-3) (IBP- 3) (IGF-binding protein 3). NDRG1 3.52 NM_006096 NDRG1 protein (N-myc downstream regulated gene 1 protein) (Differentiation-related gene 1 protein) (DRG1) (Reducing agents and tunicamycin-responsive protein) (RTP) (Nickel-specific induction protein Cap43) RAMP1 3.41 N M 005855 Receptor activity-modifying protein 1 precursor EFNB2 2.98 N M 004093 Ephrin-B2 precursor -51 -Table 5.4 Top 10 genes suppressed by I N G 2 Genes Fold Suppresstion PubMed Accession Number Description RARRES 1 15.54 NM_002888 Retinoic acid receptor responder protein 1 RPS4Y1 6.46 N M 001008 40S ribosomal protein S4, Y isoform (PR02646) LASS4 2.91 N M 024552 LAG1 longevity assurance homolog 4 IGFBP5 2.64 NM_000599 Insulin-like growth factor binding protein 5 precursor (IGFBP-5) (IBP- 5) (IGF-binding protein 5) RPL39L 2.60 N M 052969 60S ribosomal protein L39-like (L39-2). HAMP 2.36 NM_021175 Hepcidin precursor (Liver-expressed antimicrobial peptide) (LEAP-1) TBX2 2.27 NM_005994 T-box transcription factor TBX2 (T-box protein 2) CTGF 2.14 NM_001901 Connective tissue growth factor precursor (Hypertrophic chondrocyte- specific protein 24). MICA 2.13 N M 000247 M H C class I chain-related gene A protein APOB 2.03 N M 000384 Apolipoprotein B-100 precursor (Apo B-100) - 5 2 -5.3 Discussion With the availability of much higher density microarrays, a more thorough annotation of the human genome and a stronger technical and technological platform to analyze expression data, we undertook a global view of UV-response or ING2-response gene expressions to generate novel hypothesis regarding the effects of either U V R or ING2 on melanoma cells. In our analysis, we focused on the most significant gene expression changes modulated by U V R or overexpression of p33ING2. The relatively low dose of U V B irradiation (100 J/m 2) allows us to identify genes more related to D N A repair rather than apoptosis. However, the use of a 6-hour time point of analysis limited our analysis to enrich for early-gene responses. A T F - 3 and C D K N 1 A (P21CIP1) are well known to be able to work as stress inducible factors (Yang et al, 2006). In addition, A T F - 3 has also been shown to participate in axonal growth and neural development (Pearson et al, 2003). It has long been observed that melanocytes w i l l form dendrites in the direction of light when exposed to U V R (Yamashita et al, 2005). Therefore, it is possible that melanocytes coordinate critical development genes from its neuroectodermal lineage in order to regulate dendrite outgrowth in a targeted orientation. The lack of induction of p53 R N A by U V R is consistent with the known UVR-mediated post-transcriptional stabilization of p53 protein rather than an effect on p53 transcript levels. Moreover, we also found that U V R can enhance the expression of GADD45A, another well known U V R induced gene, by 3.82 fold compared with control. A t the same time, we revealed the induction of two new genes by U V irradiation including BTG2 and PIK3. PIK3 is reported to prime phosphorylation of Chk2 and mediate its full activation in response to D N A damage (Bahassi et al, 2006). The identification of U V R induced B T G 2 expression is worthy of mention. B T G 2 belongs to a newly identified family of structurally related genes whose other known human members are B T G 1 , B T G 3 and -53 -Tob. However, BTG2 is the only p53 transciptional target gene (Cortes et al, 2000; Boiko et al, 2006). In addition, B T G expression was found to be significantly reduced in a large proportion of human kidney and breast carcinomas and it works as a tumor suppressor that links both the p53 and Rb pathways in human tumorigenesis (Boiko et al, 2006). The study on the role of B T G 2 in melanoma is still lacking and we anticipate that B T G 2 may have similar functions as I N G family proteins in cancer. Although it is not the most abundantly induced gene, we did find increased p33ING2 c D N A expression (2.49 fold) in ING2 stable cell line when comparing with the parental M M R U cell. It further supports the successful establishment of our stable clone. Among the top 10 genes induced by p33ING2, one gene is MMP-7. This finding is surprising since p33ING2 is believed to work as a tumor suppressor while overexpressed M M P - 7 has been found in different types of tumors (Shiomi et al, 2003; Lee et al, 2006). There is substantial evidence that overexpression of M M P - 7 correlates with a more aggressive phenotype of tumor cells and poorer prognosis (Shiomi et al, 2003). Moreover, M M P s not only degrade extracellular matrix ( E C M ) , but degrade n o n - E C M proteins including insulin-like growth factor binding protein (IGFBP) resulting in increased bioavailability of insulin-like growth factors and enhancing cellular proliferation (Ii et al, 2006). However, we found the induction instead of suppression of IGFBP3 in ING2 stable cell line (Table 5.3) suggesting the existence o f other genes which can compete with M M P - 7 on the regulation of IGFBP3 . On the other hand, we also observed the upregulated NDRG1 gene which is necessary for p53 mediated apoptosis and overexpression of this gene, leads to the inhibition of growth in colon cancer cells as well as suppression of metastasis in prostate, colon and breast cancer cell lines (Stein et al, 2004; Maruyama et al, 2006). The contrary functions of M M P - 7 and N D R G 1 indicate the complexity of transcriptional changes - 5 4 -resulting from overexpressed ING2 in melanoma and we thus hypothesize the function of p33ING2 in melanoma invasion and metastasis depends on the balance o f these molecules. Overexpression of p33ING2 most significantly suppressed the expression of RARRES1 gene. R A R R E S 1 also called tazarotene induced gene 1 (TIG1) and it is a retinoid acid response gene. So far all the published literature focused on clarifying the mechanism of TIG1 inactivation in different tumor types and their results support that hypermethylation of TIG1 gene promoter contributing to the inactivation of it (Zhang et al, 2004; Shutoh et al, 2005). Although ING2 can recognize H3 lysine methylation and activate gene suppression (Shi et al, 2006), no correlation between ING2 and gene methylation has been reported. Therefore, it would be interesting to examine the expression levels of TIG1 gene in melanoma in future studies and investigate i f epigenetic modification w i l l lead to gene inactivation and whether p33ING2 plays an important role in this process. P33ING2 plays a crucial role in D N A repair (Wang J et al, 2006), we therefore further compared the gene expression patterns between U V treated M M R U cell line and U V treated LNG2 stable clone. However, the difference between the two cell lines doesn't seem to be affected by U V R . The most abundantly modified genes are still MMP-7, NDRG1 and RARRESl(TIGl)(data not shown). A s reported previously, the importance of p331NG2 on D N A repair depends on its involvement in U V induced rapid H4 acetylation and the recruitment of damage recognition protein xeroderma pigmentosum group A protein (Wang J et al, 2006) which could not be reflected by our c D N A microarray analysis. We propose other studies like CHIP maybe more powerful to identify the molecular changes contributing to p33ING2 mediated D N A repair. -55 -In summary, the application of c D N A microarray in this study is a effective approach to identify novel genes which are involved in the same signaling pathway as ING2 or U V R in melanoma cells. - 5 6 -C H A P T E R 6. G E N E R A L C O N C L U S I O N S 6.1 Summary P33ING2, as a tumor suppressor candidate, has been shown to share similar biological functions with its homologue p 3 3 I N G l b in cell cycle arrest, apoptosis, D N A repair and senescence. A l l the functions require the presence of functional p53. In addition, both p 3 3 I N G l b and p33ING2 are stable components of S i n 3 - H D A C indicating the involvement of them in transcriptional regulation of genes. Several studies have provided evidence that aberrant expressions of p3 3 I N G l b were found in different tumor types including the loss of p3 3 I N G l b in melanoma. Considering the highly homology between p33INGlb and p33ING2, in this thesis, we started with investigating the expression patterns of p33ING2 in different melanocytic lesions using tissue microarray and immunohistochemistry. We found significantly reduced ING2 nuclear expressions in primary melanomas and metastatic melanomas compared with dysplastic nevi. However, there was no difference of ING2 nuclear expression among R G P melanomas, V G P melanomas and metastatic melanomas. Furthermore, there was no correlation between ING2 nuclear expression and patient clinicopathological parameters or between ING2 nuclear expression and 5-year patient survival. The above results indicate that the loss o f nuclear expression o f p33ING2 may only be involved in the initiation of melanoma rather than the progression of melanoma. Loss of heterozygosity (LOH) , mutation and epigenetic modification are three main mechanisms causing decreased tumor suppressor gene expressions in tumor samples. We therefore investigated the mutational status of p33ING2 in melanoma cell lines and melanoma tissue samples. However, no mutation of p33ING2 was found in both the melanoma cell lines - 5 7 -and tissue samples suggesting that mutation is not the reason causing reduced ING2 expression in melanomas. Inactivation of p33ING2 may be due to transcriptional or post-transcriptional modification. We also investigated the transcriptional changes caused by U V R or stably transfected p33ING2 by performing c D N A microarray analysis. We not only confirmed a subset of U V -induced genes like ATF-3 and P21, but also unveiled some new genes including BTG2 and PLK3. The effect of overexpressed p33ING2 on M M R U is quite complicated because it activates both the tumor suppressor genes and oncogenes. The simultaneous expressions of these ostensibly antagonistic processes may be due to the unknown adjunctive functions of these genes and more studies are required to find the real functions of p33ING2 in melanoma. Taken together, we have elucidated in this thesis the expression and mutational status of p33ING2 in melanoma. We also described the transcriptional changes caused by overexpressed p33ING2. Together with further exploration, our knowledge of this gene may lead to more effective prevention and treatment for human melanoma. 6.2 Future Directions In this thesis, our studies showed that mutation of p33ING2 gene is not the mechanism contributing to the significantly reduced nuclear expression of p33ING2 in melanomas. Since the p331NGlb gene and flanking regions are highly G C rich and p33ING2 shows 58.9% identity with p33INGlb , we anticipate that epigenetic modification of p33ING2 promoter which can cause reduced expression of corresponding protein may be the reason for the loss of p33ING2 expression in melanoma. - 5 8 -Using ING2 stable clones and c D N A microarrays, we identified that the expression of R A R R E S 1 (TIG1) is negatively regulated by overexpressed p33ING2. This finding is interesting since the inactivation of T I G 1 has been reported in different types of tumors and it is believed to act as a tumor suppressor gene. To confirm this result, s i R N A can be used to knock down the expression of p33ING2 in M M R U cell line and see i f we can restore the expression of TIG 1. In addition, similar studies are needed to be done using normal melanocyte and different origins of melanoma cell lines to exclude the possibility of cell type specific regulation of p33ING2 on TIG1. After all these studies, we believe the confirmation of the relationship between p33ING2 and TIG1 wi l l generate new hypothesis about the role of p33ING2 in melanomas. - 5 9 -R E F E R E N C E S Albino A P , Vida l M J , McNutt N S , Shea C R , Prieto V G , Nanus D M , Palmer J M , Hayward N K . Mutation and expression of the p53 gene in human malignant melanoma. Melanoma Res 1994, 4:35-45. Atallah E , Flaherty L . Treatment of metastatic malignant melanoma. Curr Treat Options Oncol 2005,6:185-193. Balch C M , Soong SJ, Gershenwald JE, Thompson JF, Reintgen D S , Cascinelli N , Urist M , McMasters K M , Ross M I , Ki rkwood J M , Atkins M B , Thompson J A , Coit D G , Byrd D , Desmond R, Zhang Y , L i u P Y , Lyman G H , Morabito A . Prognostic factors analysis of 17 600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol200\, 19: 3622-3634. Bahassi el M , Myer D L , McKenney R J , Hennigan R F , Stambrook PJ . Priming phosphorylation of Chk2 by polo-like kinase 3 (Plk3) mediates its full activation by A T M and a downstream checkpoint in response to D N A damage. Mutat Res 2006, 596:166-176. Barker D , Dixon K , Medrano E E , Smalara D , Im S, Mitchell D , Babcock G , Abdel-Malek Z A . Comparison of the responses of human melanocytes with different melanin contents to ultraviolet B irradiation. Cancer Res 1995, 55:4041^1046. - 6 0 -Boiko A D , Porteous S, Razorenova O V , Krivokrysenko V I , Wil l iams B R , Gudkov A V . A systematic search for downstream mediators of tumor suppressor function of p53 reveals a major role of B T G 2 in suppression of Ras-induced transformation. Genes Dev 2006, 20:236-252. Campos EI, Chin M Y , Kuo W H , L i G . Biological functions of the I N G family tumor suppressors. Cell Mol Life Sci 2004, 61:2597-2613. Campos EI, Martinka M , Mitchel l D L , Dai D L , L i G . Mutations of the ING1 tumor suppressor gene detected in human melanoma abrogate nucleotide excision repair. Int J Oncol 2004, 25:73-80. Cancer facts & figures. Atlanta: American Cancer Society 2003. Chen L , Matsubara N , Yoshino T, Nagasaka T, Hoshizima N , Shirakawa Y , Naomoto Y , Isozaki H , Riabowol K , Tanaka N . Genetic alterations of candidate tumor suppressor ING1 in human esophageal squamous cell cancer. Cancer Research 2001, 61:4345-4349. Cheung K J Jr, Bush J A , Jia W , L i G . Expression of the novel tumor suppressor p33(INGl)is independent of p53. Br J Cancer 2000, 83:1468-1472. Cheung K J Jr, L i G . p33( INGl) enhances UVB-induced apoptosis in melanoma cells. Exp Cell Res 2002,279: 291-298. -61 -Cheung K J Jr, Mitchel l D , L i n P, L i G . The tumor suppressor candidate p33(INGl) mediates repair of UV-damaged D N A . Cancer Res 2001, 61:4974-4977. Chin L , Pomerantz J, Polsky D , Jacobson M , Cohen C, Cordon-Cardo C, James W. Horner J.W, DePinho R . A Cooperative effects of INK4a and ras in melanoma susceptibility in vivo. Genes Dev 1997, 11: 2822-2834. Chin M Y , N g K C , L i G . The novel tumor suppressor p33ING2 enhances UVB-induced apoptosis in human melanoma cells. Exp Cell Res 2005, 304:531-543. Coombs N J , Gough A C , Primrose J N . Optimisation of D N A and R N A extraction from archival formalin-fixed tissue. Nucleic Acids Res 1999, 27:12-17. Cortes U , Moyret-Lalle C, Falette N , Duriez C, Ghissassi F E , Barnas C, Morel A P , Hainaut P, Magaud JP, Puisieux A . B T G gene expression in the p5 3-dependent and -independent cellular response to D N A damage. Mol Carcinog 2000, 27:57-64. Dai D , Makretsov N , Campos EI, Huang C, Zhou Y , Huntsman D , Martinka M , L i G . Increased expression of integrin-linked kinase is correlated with melanoma progression. Clin Cancer Res 2003, 9: 4409-4414. Dai D L , Martinka M , Bush J A , L i G . Reduced Apaf-1 expression in human cutaneous melanoma. Br J Cancer 2004, 91: 1089-1095. - 6 2 -Dai D L , Martinka M , L i G . Prognostic significance of activated A k t expression in melanoma: A clinicopathological study of 292 cases. J Clin Oncol 2005, 23: 1473-1482. Davies H , Bignell G R , Cox C, Stephens P, Edkins S, Clegg S. Mutations of the B R A F gene in human cancer. Nature 2002, 417:949-954. Doyon Y , Cayrou C, Ul lah M , Landry A J , Cote V , Selleck W , Lane W S , Tan S, Yang X J , Cote J. I N G tumor suppressor proteins are critical regulators of chromatin acetylation required for genome expression and perpetuation. Mol Cell. 2006, 21:51-64. Dunlap S, Y u X , Cheng L , C i v i n CI , Alani R M . High-efficiency stable gene transduction in primary human melanocytes using a lentiviral expression system. J Invest Dermatol 2004, 122: 549-551. Elder D. Tumor progression, early diagnosis and prognosis of melanoma. Acta Oncol 1999, 38: 535-547. Elledge SJ. Cel l cycle checkpoints: preventing an identity crisis. Science 1996, 274: 1664-1672. -63 -Erhard H , Rietveld FJ , van Altena M C , Brocker E B , Ruiter D J , de Waal R M . Transition of horizontal to vertical growth phase melanoma is accompanied by induction of vascular endothelial growth factor expression and angiogenesis. Melanoma Res 1997, Suppl 2: SI9-26. Feng X , Hara Y , Riabowol K . Different H A T S of the ING1 gene family. Trends Cell Biol 2002, 12: 532-538. Fridman JS, Lowe SW. Control of apoptosis by p53. Oncogene 2003, 22: 9030-9040. Garkavtsev I, Demetrick D , Riabowol K . Cellular localization and chromosome mapping of a novel candidate tumor suppressor gene (ING1). Cytogenet Cell Genet 1997, 76:176-178. Garkavtsev I, Grigorian IA, Ossovskaya V S , Chernov M V , Chumakov P M , Gudkov A V . The candidate tumor suppressor p 3 3 I N G 1 cooperates with p53 in cell growth control. Nature 1998, 391: 295-298. Garkavtsev I, Kazarov A , Gudkov A,Riabowol K . Correction and addition to I N G sequences. Nature Genetics 1999, 23: 373. Garkavtsev I, Kazarov A , Gudkov A , Riabowol K . Suppression of the novel growth inhibitor p33INGl promotes neoplastic transformation, Nature Genetics 1996,14: 415-420. - 6 4 -Garkavtsev I, K o z i n S V , Chernova O, X u L , Winkler F, Brown E , Barnett G H , Jain R K . The candidate tumour suppressor protein ING4 regulates brain tumour growth and angiogenesis. Nature 2004, 428:328-332 Gilchrest B A , Eller M S , Geller A C , Yaar M . The pathogenesis of melanoma induced by ultraviolet radiation. N EnglJ Med 1999, 340:1341-1348. Goel V K , Lazar A J , Warneke C L , Redston M S , Haluska F G Examination of mutations in B R A F , N R A S , and P T E N in primary cutaneous melanoma. J Invest Dermatol 2006, 126: 154-160. Gong W , Suzuki K , Russell M , Riabowol K . Function of the I N G family of P H D proteins in cancer. Int J Biochem Cell Biol 2005, 37:1054-1065. Gong W , Russell M , Suzuki K , Riabowol K . Subcellular targeting of p3 3 I N G l b by phosphorylation-dependent 14-3-3 binding regulates p 2 1 W A F l expression. Mol Cell Biol 2006, 26: 2947-2954. - 6 5 -Gozani 0, Karuman P, Jones D R , Ivanov D , Cha J, Lugovskoy A A , Baird C L , Zhu H , Field SJ, Lessnick SL , Villasenor J, Mehrotra B , Chen J, Rao V R , Brugge JS, Ferguson C G , Payrastre B , Myszka D G , Cantley L C , Wagner G , Divecha N , Prestwich G D , Yuan J. The P H D finger of the chromatin-associated protein ING2 fuctions as a nuclear phosphoinositide receptor. Cell 2003, 11:99-111. Gunduz M , Ouchida M , Fukushima K , Hanafusa H , Etani T, Nishioka S, Nishizaki K , Shimizu K . Genomic structure of the human ING1 gene and tumor-specific mutations detected in head and neck squamous cell carcinomas. Cancer Res 2000, 60: 3143-3146. Harper J. W. , Adami G.R., Wei N . , Keyomarsi K , Elledge S. J. The p21 Cdk-interacting protein C i p l is a potent inhibitor of G l cyclin-dependent kinases. Cell. 1993, 75: 805-816. Harris S L , Levine A J . The p53 pathway: positive and negative feedback loops. Oncogene 2005, 24:2899-2908. Hayflick L . The limited in vitro lifetime of human diploid cell strains. Exp Cell Res 1965, 37: 614-636. Helbing C C , Veillette C, Riabowol K , Johnston R N , Garkavtsev I. A novel candidate tumor suppressor, ING1 , is involved in the regulation of apoptosis. Cancer Res 1997, 57:1255-1258. - 6 6 -Homsi J, Kashani-Sabet M , Messina JL , Daud A . Cutaneous melanoma: prognostic factors. Cancer Control 2005, 12:223-229. Hsu M Y , Shih D T , Meier F E , V a n Belle P, Hsu J Y , Elder D E , Buck C A , Herlyn M . Adenoviral Gene Transfer of B3 Integrin Subunit Induces Conversion from Radial to Vertical Growth Phase in Primary Human Melanoma. Am J Pathol 1998, 53:1435-1442. Hussein M R . Melanocyte dysplastic naevi occupy the middle ground between Benign melanocyte naevi and cutaneous malignant melanomas: emerging clues. J Clin Pathol 2005, 58:453-456. Hussein M R . The TP53 tumor suppressor gene and melanoma tumorigenesis: is there a relationship? Tumor Biol 2004, 25: 200-207. Hussussian C J , Struewing JP, Goldstein A M . Germline p i 6 mutations in familial melanoma. Nat Genet 1994, 8: 15-21. Ii M , Yamamoto H , Adachi Y , Maruyama Y , Shinomura Y . Role of matrix metalloproteinase-7 (matrilysin) in human cancer invasion, apoptosis, growth, and angiogenesis. Exp Biol Med 2006,231:20-27. - 6 7 -Ito K , Kinjo K , Nakazato T, Ikeda Y , K izak i M . Expression and sequence analyses of p33(INGl) gene in myeloid. Expression and sequence analyses of p33( INGl) gene in myeloid leukemia. American Journal of Hematology 2002, 69: 141-143. Kamb A , Shattuck-Eidens D , Eeles R, L i u Q, Gruis N . A . , Ding W,Hussey C , Tran T, M i k i Y , Weaver-Feldhaus J, McClure M , Aitken J, Anderson D E , Bergman W , Frants R, Goldgar D E , Green A , MacLennan R, Martin N G , Meyer L J , Y o u l P, Zone JJ, Skolnick M H , Cannon-Albright L A . Analysis of the p i 6 gene ( C D K N 2 ) as a candidate for the chromosome 9p melanoma susceptibility locus. Nat Genet 1994, 8: 23-26. Kameyama K , Huang C L , L i u D , Masuya D , Nakashima T, Sumitomo S, Takami Y , Kinoshita M , Yokomise H . Reduced I N G l b gene expression plays an important role in carcinogenesis of non-small cell lung cancer patients. Clin Cancer Res 2003, 9: 4926^-934. Karst A M , Dai D L , Martinka M , L i G . P U M A expression is significantly reduced in human cutaneous melanomas. Oncogene 2005, 24: 111 1-1116. Keller J M , Maize JC. The clinical and histological differential diagnosis of malignant melanoma. Semin Oncol 1996, 23: 693-702. - 6 8 -Kojima S, Mulholland D J , Ettinger S, Fazli L , Nelson C C , Gleave ME.Differential regulation of IGFBP-3 by the androgen receptor in the lineage-related androgen-dependent L N C a P and androgen-independent C4-2 prostate cancer models. Prostate 2006, 66: 971-986 Kuzmichev A , Zhang Y , Erdjument-Bromage H , Tempst P, Reinberg D . Role of the Sin3-histone deacetylase complex in growth regulation by the candidate tumor suppressor p 3 3 I N G 1 . Molecular and Cellular Biology 2002, 22: 835-848. L i G, Tang L , Zhou X , Tron V , Ho V . Chemotherapy-induced apoptosis in melanoma cells is p53 dependent. Melanoma Res 1998, 8: 17-23. L i J, Yen C, L iaw D , Podsypanina K , Bose S, Wang SI, Puc J, Miliaresis C , Rodgers L , McCombie R, Bigner S H , Giovanella B C , Ittmann M , Tycko B , Hibshoosh H , Wigler M H , Parsons R. P T E N , a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science 1997, 275: 1943-1947. Liston P, Fong W G , Ke l ly N L , Toji S, Miyazaki T, Conte D , Tamai K , Craig C G , McBurney M W , Korneluk R G . Identification of X A F 1 as an antagonist of X I A P anti-Caspase activity. Nat Cell Biol 2001, 3: 128-133. L u F, Dai> D L , Martinka M , Ho V , L i G . Nuclear ING2 expression is reduced in human cutaneous melanomas. Br J Cancer 2006, 95: 80-86. - 6 9 -Malumbres M . and Barbacid M . R A S oncogenes: the first 30 years. Nat Rev Cancer 2003, 3: 4 5 9 ^ 6 5 . Marrett L D , K i n g W D , Walter SD, From L . Use of host factors to identify people at high risk for cutaneous malignant melanoma CMAJ 1992,147: 445-453. Maruyama Y , Ono M , Kawahara A , Yokoyama T, Basaki Y , Kage M , Aoyagi S, Kinoshita H , Kuwano M . Tumor growth suppression in pancreatic cancer by a putative metastasis suppressor gene Cap43/NDRG1/Drg-1 through modulation of angiogenesis. Cancer Res 2006, 66:6233-6242. Metcalf JS. Melanoma: criteria for histological diagnosis and its reporting. Semin Oncol 1996, 23: 688-692. Mi l le r Ar lo J. and M i h m Martin C. Melanoma. NEnglJMed 2006, 355:51-65. Nagashima M , Shiseki M , Miura K , Hagiwara K , Linke SP, Pedeux R, Wang X W , Yokota J, Riabowol K , Harris C C . D N A damage-inducible gene p33ING2 negatively regulates cell proliferation through acetylation of p53. Proc Natl Acad Sci U SA 2001, 98: 9671-9676. N g K P , Campos EI, Martinka M , L i G . X A F 1 expression is significantly reduced in human malignant melanoma. J Invest Dermatol 2004, 123: 1127-1134. - 7 0 -Nouman GS , Anderson JJ, Wood K M , Lunec J, Hal l A G , Reid M M , Angus B . Loss of nuclear expression of the p33(INGlb) inhibitor of growth protein in childhood acute lymphoblastic leukaemia. J Clin Pathol 2002, 55: 596-601. Okano T, Gemma A , Hosoya Y , Hosomi Y , Nara M , Kokubo Y , Yoshimura A , Shibuya M , Nagashima M , Harris C C , Kudoh S. Alterations in novel candidate tumor suppressor genes, ING1 and ING2 in human lung cancer. Oncol Rep 2006, 15: 545-549. Oki E , Maehara Y , Tokunaga E , Kakeji Y , Sugimachi K . Reduced expression of p33(INGl) and the relationship with p53 expression in human gastric cancer, Cancer Letter 1999, 147: 157-162. Omholt K , Karsberg S, Platz A , Kanter L , Ringborg U , Hansson J. Screening of N-ras codon 61 mutations in paired primary and metastatic cutaneous melanomas: mutations occur early and persist throughout tumor progression. Clin.Cancer Res 2002, 8: 3468-3474. Omholt K , Platz A , Kanter L , Ringborg U , Hansson J. NRAS and BRAF Mutations Arise Early during Melanoma Pathogenesis and Are Preserved throughout Tumor Progression. Clin. Cancer Res 2003, 9: 6483-6488. Ortonne JP. Photoprotective properties of skin melanoma. Br J Dermatol. 2002, 61:7-10. Patton E E , Widlund H R , Kutok JL. B R A F mutations are sufficient to promote nevi formation and cooperate with p53 in the genesis of melanoma. Curr Biol 2005, 15: 249-254. -71 -Pavey S, Conroy S, Russell T, Gabrielli B . Ultraviolet radiation induces p l 6 C D K N 2 A expression in human skin. Cancer Res 1999, 59: 4185^4189. Pearson A G , Gray C W , Pearson JF, Greenwood J M , During M J , Dragunow M . A T F 3 enhances c-Jun-mediated neurite sprouting. Brain Res Mol Brain Res 2003, 120: 38^15. Pearson M , Carbone R, Sebastiani C, Cioce M , Fagioli M , Saito S, Higashimoto Y , Appella E , Minucci S, Pandolfi PP, Pelicci P G . P M L regulates p53 acetylation and premature senescence induced by oncogenic Ras. Nature 2000, 406: 207-210 Pedeux R, Sengupta S, Shen J C , Demidov O N , Saito S, Onogi H , Kumamoto K , Wincovitch S, Garfield S H , McMenamin M , Nagashima M , Grossman SR, Appel la E , Harris C C . ING2 regulates the onset of replicative senescence by induction of p300-dependent p53 acetylation. Mol Cell Biol 2005, 25: 6639-6648. Raffo A J , K i m A L , Fine R L Formation of nuclear Bax/p53 complexes is associated with chemotherapy induced apoptosis. Oncogene 2000, 19: 6216-6228. Recio J A , Noonan FP, Takayama H , Anver M R , Duray P, Rush W L , Lindner G , De Fabo E C , DePinho R A , Merlino G . Ink4a/arf deficiency promotes ultraviolet radiation-induced melanomagenesis. Cancer Res 2002, 62: 6724-6730. - 7 2 -Rigel DS.and Carucci J A . Malignant melanoma: prevention, early detection, and treatment in the 21st century. CA.A Cancer Journal for Clinicians 2000, 50:2 15-236. Rigel DS. The effect of sunscreen on melanoma risk. Dermatol Clin 2002, 20: 601-606. Ruas M and Peters G . Biochim. Biophys. Acta 1998, 1378: 115-177. Saito A , Furukawa T, Fukushige S, Koyama S, Hoshi M , Hayashi Y , Hor i i A . p 2 4 / I N G l - A L T l and p 4 7 / I N G l - A L T 2 , distinct alternative transcripts of p 3 3 / I N G l . J Hum Genet. 2000, 45: 177-181. Sanchez-Carbayo M , Socci N D , Lozano JJ, L i W , Charytonowicz E , Belbin TJ , Prystowsky M B , Ortiz A R , Childs G , Cordon-Cardo C. Gene discovery in bladder cancer progression using c D N A microarrays. J Pathol 2003,163: 505-516. Schoenlaub P, Sarraux A , Grosshans E , Heid E , Cribier B . Survival after cutaneous metastasis: a study of 200 cases. Ann Dermatol Venereol 2001, 128: 1310-1315. Serrano M , Lee H , Chin L , Cordon-Cardo C, Beach D , DePinho R A . Role of the INK4a locus in tumor suppression and cell mortality. Cell 1996, 85: 27-37. Sharpless E , Chin L . The I N K 4 a / A R F locus and melanoma. Oncogene 2003, 22: 3092-3098. -73 -Shi SR, Cote RJ , W u L , L i u C, Datar R, Shi Y , L i u D , L i m H , Taylor C R . D N A extraction from archival formalin-fixed, paraffin-embedded tissue sections based on the antigen retrieval principle: heating under the influence of pH. J Histochem Cytochem 2002, 50:1005-1011. Shi X , Gozani O. Fellowships of the INGs. J Cell Biochem 2005, 96: 1127-1136. Shi X , Hong T, Walter K L , Ewalt M , Michishita E , Hung T, Carney D , Pena P, Lan F, Kaadige M R , Lacoste N , Cayrou C, Davrazou F, Saha A , Cairns B R , Ayer D E , Kutateladze T G , Shi Y , Cote J, Chua K F , Gozani O. ING2 P H D domain links histone H3 lysine 4 methylation to active gene repression. Nature, 2006, 442: 96-99. Shimada Y , Saito A , Suzuki M , Takahashi E , Horie M . Cloning of a novel gene (ING1L) homologous to ING1, a candidate tumor suppressor. Cytogenet Cell Genet 1998, 83: 232-235. Shiomi T, Okada Y . M T 1 - M M P and M M P - 7 in invasion and metastasis of human cancers. Cancer Metastasis Rev 2003, 22: 145-152. Shutoh M , Oue N , Aung PP, Noguchi T, Kuraoka K , Nakayama H , Kawahara K , Yasui W. D N A methylation of genes linked with retinoid signaling in gastric carcinoma: expression of the retinoid acid receptor beta, cellular retinol-binding protein 1, and tazarotene-induced gene 1 genes is associated with D N A methylation. Cancer 2005,104: 1609-1619. - 7 4 -Siddik Z H . Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 2003, 22: 7265-7279. Sironi E, Cerri A , Tomasini D , Sirchia S M , Porta G , Rossella F, Grati F R , Simoni G Loss of heterozygosity on chromosome 4q32-35 in sporadic basal cell carcinomas: evidence for the involvement of p 3 3 I N G 2 / I N G l L and SAP30 genes. J Cutan Pathol 2004, 31: 318-322. Steck P A , Pershouse M A , Jasser S A , Yung W K , L i n H , Ligon A H , Langford L A , Baumgard M L , Hattier T, Davis T, Frye C, H u R, Swedlund B , Teng D H , Tavtigian S V . Identification of a candidate tumor suppressor gene, M M A C 1 , at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet 1997,15: 356-362. Stein S, Thomas E K , Herzog B , Westfall M D , Rocheleau J V , Jackson RS 2nd, Wang M , Liang P. N D R G 1 is necessary for p53-dependent apoptosis. J Biol Chem. 2004, 279: 48930-48940. Takahashi M , Seki N , Ozaki T, Kato M , Kuno T, Nakagawa T, Watanabe K , Miyazaki K , Ohira M , Hayashi S, Hosoda M , Tokita H , Mizuguchi H , Hayakawa T, Todo S, Nakagawara A . Identification of the p33(INGl)-regulated genes that include cyclin B l and proto-oncogene D E K by using c D N A microarray in a mouse mammary epithelial cell line N M u M G . Cancer Res 2002, 62: 2203-2209. Tallen G , Riabowol K , Wol f f J E A . Expression of p33INGl m R N A and chemosensitivity in brain tumor cells. Anticancer Research 2003, 23:1631-1636. -75 -Thoma B S , Vasquez K M . Critical D N A damage recognition functions of X P C - h H R 2 3 B and X P A - R P A in nucleotide excision repair. Mol Carcinog 2003, 38:1-13. Thompson JF, Scolyer R A , Kefford R F . Cutaneous melanoma. Lancet 2005, 365: 687-701 Toyama T, Iwase H , Watson P, Muz ik H , Saettler E , Magliocco A , DiFrancesco L , Forsyth P, Garkavtsev I, Kobayashi S, Riabowol K . Suppression of ING1 expression in sporadic breast cancer. Oncogene 1999, 18: 5187-5193. Tsang F .C. , Po L .S . , Leung K . M . , Lau A . , Siu W. Y . , Poon R. Y . I N G l b decreases cell proliferation through p53 dependent and independent mechanisms. FEBS Lett 2003, 553: 277-285. Urso C. Are growth phases exclusive to cutaneous melanoma? J Cutan Patho 2004, 57: 560 Valery C, Grob JJ, Verrando P Identification by c D N A microarray technology of genes modulated by artificial ultraviolet radiation in normal human melanocytes: relation to melanocarcinogenesis. J Invest Dermatol 2001, 117: 1471-1482. Vieyra D, Toyama T, Hara Y , Boland D , Johnston R, Riabowol K . ING1 isoforms differentially affect apoptosis in a cell age-dependent maimer. Cancer Res 2002, 62:4 445-4452. - 7 6 -Vieyra D , Loewith R, Scott M , Bonnefin P, Boisvert F M , Cheema P, Pastyryeva S, Meijer M , Johnston R N , Bazett-Jones DP , McMahon S, Cole .D, Young D , Riabowol Kar l . Human ING1 proteins differentially regulate histone acetylation, The Journal of Biological Chemistry 2002, 277:29832-29839. Wagner M J , Gogela-Spehar M , Skirrow R C , Johnston R N , Riabowol K , Helbing C C . Expression of novel I N G variants is regulated by thyroid hormone in the Xenopus laevis tadpole. J Biol Chem 2001, 276: 47013-47020. Wang J, Chin M Y , L i G . The novel tumor suppressor p33ING2 enhances nucleotide excision repair via inducement of histone H4 acetylation and chromatin relaxation. Cancer Res. 2006, 66:1906-1911 Wang Y , L i G . ING3 promotes UV-induced apoptosis via Fas/caspase-8 pathway in melanoma cells. JBiol Chem. 2006, 281:11887-11893. Wang Y , Wang J, L i G . Leucine zipper-like domain is required for tumor suppressor ING2-mediated nucleotide excision repair and apoptosis. FEBS Lett 2006, 580: 3787-3793. Yamashita T, Kuwahara T, Gonzalez S, Takahashi M Non-invasive visualization of melanin and melanocytes by reflectance-mode confocal microscopy. J Invest Dermatol 2005, 124: 235-240. - 7 7 -Yang G , Zhang G , Pittelkow M R , Ramoni M , Tsao H . Expression Profiling of U V B Response in Melanocytes Identifies a Set of p53-Target Genes. J Invest Dermatol. 2006. Y o u M J , Castrillon D H , Bastian B C , O'Hagan R C , Bosenberg M W , Parsons R, Chin L , DePinho R A . Genetic analysis of Pten and Ink4a/Arf interactions in the suppression of tumorigenesis in mice. Proc Natl Acad Sci USA 2002 ,99: 1455-1460. Zhang J, L i u L , Pfeifer G P . Methylation of the retinoid response gene TIG1 in prostate cancer correlates with methylation o f the retinoic acid receptor beta gene. Oncogene 2004, 23: 2241-2249. Zuo L , Weger J, Yang Q , Goldstein A M , tucker M A , Walker G J , Hayward N , Dracopoli N C . Germline mutations in the p l 6 I N K 4 a binding domain of C D K 4 in familial melanoma. Nat Genet 1996, 12: 97-99. - 7 8 -


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items