Oncogene (2012) 31, 2580–2592 & 2012 Macmillan Publishers Limited All rights reserved 0950-9232/12 www.nature.com/onc ORIGINAL ARTICLE NF-jB inducing kinase (NIK) modulates melanoma tumorigenesis by regulating expression of pro-survival factors through the b-catenin pathway

YM Thu1,6,YSu1, J Yang1, R Splittgerber1,SNa2, A Boyd3, C Mosse4, C Simons4 and A Richmond1,5

1Department of Cancer Biology, Vanderbilt University School of Medicine, Nashville, TN, USA; 2Oncology Research, Eli Lily and Co., Indianapolis, IN, USA; 3Division of Dermatology, Vanderbilt University School of Medicine, Nashville, TN, USA; 4Department of Pathology, Microbiology and Immunology, Vanderbilt University School of Medicine, Nashville, TN, USA and 5Department of Veterans Affairs Medical Center, Nashville, TN, USA

Nuclear factor-jB(NF-jB) inducing kinase (NIK) is a Introduction MAP3K that regulates the activation of NF-jB. NIK is often highly expressed in tumor cells, including melanoma, Melanoma is a deadly disease and number of efficacious but the significance of this in melanoma progression has therapies to treat malignant melanoma is limited. Thus, been unclear. Tissue microarray analysis of NIK expression finding new targets and understanding molecular reveals that dysplastic nevi (n ¼ 22), primary (n ¼ 15) and differences in melanoma is essential for the development metastatic melanoma (n ¼ 13) lesions showed a statistically of novel cancer therapeutics. Similar to other cancers, significant elevation in NIK expression when compared with malignant melanoma results from genetic and molecular benign nevi (n ¼ 30). Moreover, when short hairpin RNA changes, such as gain-of-function mutation in BRAF techniques were used to knock-down NIK, the resultant (BRAFV600E), N-RAS (N-RASQ61KorR), loss of p16INK4a, NIK-depleted melanoma cell lines exhibited decreased or activation of c-MET (Chin et al., 1998; Davies et al., proliferation, increased apoptosis, delayed cell cycle pro- 2002). Likewise, many major signaling pathways such as gression and reduced tumor growth in a mouse xenograft MAPK, b-catenin, Akt/PTEN and nuclear factor-kB model. As expected, when NIK was depleted there was (NF-kB) pathways are mis-regulated in melanoma. decreased activation of the non-canonical NF-jB pathway, However, until recently, use of this information for whereas canonical NF-jB activation remained intact. NIK targeted therapy in melanoma has met with only limited depletion also resulted in reduced expression of genes that success (Sullivan and Atkins, 2009). The remarkable contribute to tumor growth, including CXCR4, c-MYC and 50% response rate currently observed in patients c-MET, and pro-survival factors such as BCL2 and with metastatic melanoma with the BRAFV600E mutation survivin. These changes in gene expression are not fully (Puzanov and Flaherty, 2010; Chapman et al., 2011) explained by the attenuation of the non-canonical NF-jB was possible due to new advances in our understanding pathway. Shown here for the first time is the demonstration of molecular pathways crucial for melanoma tumori- that NIK modulates b-catenin-mediated transcription to genesis. promote expression of survivin. NIK-depleted melanoma Ongoing research has shown that various combina- cells exhibited downregulation of survivin as well as other b- tions of gene mutations, deletions and amplifications catenin regulated genes including c-MYC, c-MET and result in the formation of several subtypes of melanoma CCND2. These data indicate that NIK mediates both b- lesions, with different responses to therapy. Identi- catenin and NF-jB regulated transcription to modulate fication of the molecular pathways associated with each melanoma survival and growth. Thus, NIK may be a individual’s melanoma tumor will provide key insights promising therapeutic target for melanoma. for better classification of melanoma and more appro- Oncogene (2012) 31, 2580–2592; doi:10.1038/onc.2011.427; priate, personalized therapeutic intervention. It is published online 3 October 2011 increasingly apparent that NF-kB activation is one of the molecular changes important for melanoma growth Keywords: NIK; b-catenin; melanoma; tumorigenesis; (Devalaraja et al., 1999; Yang and Richmond, 2001; survival factors Dhawan and Richmond, 2002; Lev et al., 2003; Yang et al., 2010). Consequently, targeting NF-kBisan effective tool for inhibiting the growth of melanoma Correspondence: Dr A Richmond, Department of Cancer Biology, tumors with constitutive NF-kB activation (Yang et al., Vanderbilt University School of Medicine, 771 PRB, 2220 Pierce 2006, 2007, 2010). Avenue, Nashville, TN 37232, USA. The term NF-kB is used to denote a family of E-mail: [email protected] transcription factors which have crucial roles in the 6 This work is in partial fulfillment of the Ph.D. dissertation of Yee development of cancer due to its regulatory function in Mon Thu. Received 31 May 2011; revised 13 August 2011; accepted 20 August inflammation, proliferation, apoptosis and survival. 2011; published online 3 October 2011 Mammalian NF-kB members include p65 (RelA), NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2581 p105/p50, RelB, p100/p52 and c-Rel. The regulation of melanoma. These data suggest NIK may be a promising NF-kB activity is achieved by modulation of two therapeutic target for melanoma. distinct pathways: the canonical and the non-canonical pathways. The canonical pathway of NF-kB is regulated by the IkB kinase (IKK) complex, containing IKKa, Results IKKb and IKKg. The IKK complex phosphorylates inhibitor of kBs (IkBs), which usually sequester NF-kB To determine whether NIK levels are elevated in mela- members in the cytoplasm, thereby inhibiting the noma, we examined NIK expression in tissue micro- nuclear translocation of the transcription factors. Upon array of nevi (n ¼ 30), dysplastic nevi (n ¼ 22), primary phosphorylation, IkBs are degraded by the proteasome, melanoma (n ¼ 15) and metastatic melanoma (n ¼ 13) releasing NF-kB complexes to function as transcription lesions by immunofluorescence staining. Intriguingly, factors. NIK expression was higher in dysplastic nevi and in The non-canonical pathway of NF-kB is regulated primary and metastatic melanoma tissues compared by IKKa and NF-kB inducing kinase (NIK). Though with benign nevi (Po0.05) (Figures 1a and b). Likewise, NIK activates both IKKa and IKKb, it preferentially western blot analysis on lysates of 10 human melanoma phosphorylates IKKa (Ling et al., 1998). NIK, also patient tumors and four melanoma cell lines (Hs294T, known as MAP3K14 located on the human chromo- SKMel5, SKMel28 and WM115) showed that NIK some 17q21, was first identified as a kinase that activates protein expression was elevated in some melanoma NF-kB upon TNF and IL-1 receptor ligation, leading tissues (M2, M5, M6 and M7) and all melanoma cell to activation of the canonical pathway (Malinin lines compared with normal human epidermal melano- et al., 1997). However, later studies suggest that cytes (Figures 1c and d). Nevertheless, NIK mRNA NIK is dispensable for activation of the canonical expression in melanoma cells was not significantly up- NF-kB pathway (Smith et al., 2001; Yin et al., 2001). regulated in melanoma (Supplementary Figure S1A), NIK and IKKa phosphorylate p100, an IkB for this suggesting that NIK over-expression may be due to pathway. Upon phosphorylation, p100 undergoes par- protein stability. tial proteolysis to generate a 52kD protein (p52), which To understand the importance of NIK over-expres- can dimerize with RelB, and translocate to the nucleus sion in melanoma tumorigenesis, NIK was knocked to activate transcription of RelB/p52 responsive genes. down in two melanoma cell lines, Hs294T (derived Previous research has shown that alterations in the from a metastatic melanoma) and WM115 (derived NIK gene, NIK activity or NIK protein expression from a primary melanoma) with NIK targeting shRNA. occur in hematological cancers and that targeting NIK Western blot and quantitative RT–PCR (qRT–PCR) reduces tumor cell survival (Annunziata et al., 2007; analyses showed that NIK expression was efficiently Keats et al., 2007; Saitoh et al., 2008; Conze et al., 2010). reduced in Hs294T and WM115 (Figures 1e and f). Although the importance of NIK in tumorigenesis of However, NIK knock-down was gradually lost within solid tumors has been reported (Neely et al., 2010; the polyclonal population of Hs294T and WM115 Saitoh et al., 2010), its role in solid tumors is less cells selected to stably express the shRNA (Figure 1g). well-understood and detailed mechanisms of how NIK modulates tumor growth are not fully explored. Generation of clonal and polyclonal NIK knock-down cells We have previously shown that a higher amount of To prevent the growth selection against NIK-depleted NIK co-immunoprecipitates with the IKK complex in cells in polyclonal cultures, a clonal population of NIK- melanoma compared with melanocytes (Dhawan and deficient WM115 cells was selected by in-cell western Richmond, 2002). Constitutive activation of the IKK assay. The efficiency of the NIK silencing in WM115 complex has been identified in melanoma (Yang and clones was confirmed by western blot after treatment Richmond, 2001), explaining the enhanced expression of with the proteasome inhibitor (MG132), which inhibits NF-kB-regulated chemokines linked to melanoma the degradation of NIK (Supplementary Figure S1B). growth (Balentien et al., 1991; Devalaraja et al., 1999; This NIK-depleted WM115 clone (shNIK1(c)) was used Wang et al., 2000; Lev et al., 2003). As NIK can activate to examine the functional significance of NIK in the IKK complex, we asked whether NIK may also melanoma cells. In some experiments, polyclonal NIK contribute to the tumorigenesis of melanoma. In the knock-down WM115 cells (shNIK1(p)) and NIK knock- study described herein we establish the functional down Hs294T cells were included to further confirm the consequence of NIK upregulation in melanoma by effects of NIK depletion. showing that NIK contributes to melanoma growth and modulates the expression of pro-survival factors, such as BCL2 and survivin. Interestingly, although deficiency NIK deficiency in melanoma cells decreased proliferation, of NIK decreases activation of the non-canonical increased apoptosis and decreased cell cycle progression NF-kB pathway, it does not attenuate canonical NF- Consistent with phenotype of compromised growth, kB activity. However, NIK depletion reduces occupancy NIK-deficient Hs294T and WM115 cells exhibited a of b-catenin on the survivin promoter, resulting in significant lag in growth compared with non-silencing reduced survivin expression and enhanced tumor cell control cells (Figure 2a). Both Ki67 and BrdU staining apoptosis. NIK depletion also results in decreased suggest that the percentage of cells undergoing proli- proliferation and delayed cell cycle progression of feration was lower in knock-down populations than

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2582 in the non-silencing controls (Figures 2b and c). This pheno- we reasoned that the clonal knock-down cell line had type was confirmed in WM115 with another shRNA already developed a mechanism to maintain equili- targeting NIK (shNIK2) (Supplementary Figure S1C) brium between survival and death. At a basal level, the and in another melanoma cell line, Hs294T (Figure 2b). percentage of apoptotic cells was two-fold higher in the To determine whether NIK deficiency affects apop- knock-down population than in the control (Figure 2d, tosis, Annexin V and propidium iodide staining were white bars). Interestingly, knocking down IKKa in performed. For these experiments, only polyclonal NIK WM115 did not induce apoptosis (Supplementary knock-down cells in early passages were used, as Figure S1D), suggesting that NIK-mediated apoptosis

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2583 is not likely to result from decreased IKKa activity. (CXCR4, c-MYC, c-MET, CCND2 and LIN28B)(Pas- It has been documented that NF-kB activation and torino et al., 2003; Scala et al., 2005, 2006; Puri et al., NIK activity protect cells from TNFa-induced cell 2007; Viswanathan et al., 2009) were markedly reduced in death (Wang et al., 1996; Malinin et al., 1997; Van NIK knock-down cells (Table 1). NIK knock-down cell Antwerp et al., 1998). Consistent with previous studies, also exhibited upregulated expression of genes such as we observed that TNFa treatment markedly increased XIAP-associated factor 1 (XAF1), which has a negative the percentage of apoptotic cells in NIK-deficient popu- role in cancer cell survival (Plenchette et al., 2007). Similar lation (Figure 2d, dark bars). In addition, NIK knock- trends of gene expression were observed with control down WM115 cells exhibited a delay in cell cycle and NIK-depleted melanoma tumor samples (Table 1). progression compared with the control, most prominent Follow-up analysis of selected genes by qRT–PCR at 12 h after the release from thymidine block. At 12 h revealed statistically significant major changes in expres- after release, non-silencing cells had completed the cell sion for several genes identified by microarray analyses cycle, whereas a significant percentage of NIK-depleted (fold changes relative to the non-silencing control: c- cells were still in S and G2/M phases (Figure 2e). MYC(À3.39), c-MET(À4.30), CXCR4(À620.35), CCND2 (À58.08), XAF1(5.78) and LIN28B(À3.2 Â 107)with P NIK depletion in WM115 cells reduced xenograft tumor o0.05 for all genes) (Table 1). Taken together, our growth in nude mice data demonstrate that NIK depletion altered the expres- In agreement with in vitro results, depletion of NIK signi- sion of several genes crucial for tumor growth. ficantly decreased melanoma tumor growth and burden in nude mice (Po0.05) (Figures 3a and b). NIK deficiency in NIK depletion in WM115 did not reduce the canonical tumor samples was confirmed by both western blot and NF-kB activation qRT–PCR analyses (Figures 3c and d). Immunostaining Interestingly, both IKK kinase activity and nuclear of Ki67 or cleaved-caspase 3 was lower in NIK knock- phospho-p65 (Ser536) suggest that the canonical NF-kB down tumor sections than non-silencing control tumor activity was not diminished by NIK knock-down sections (Po0.05) (Figures 3e and f). These data suggest under basal or TNFa-stimulated conditions (Figures that tumors from NIK knock-down WM115 grew and 4a and b). Similarly, the basal level of some NF-kB- turned over slower than control tumors. In summary, NIK regulated chemokines and cytokines (IL-6, IL-8/ deficiency substantially attenuated melanoma tumor CXCL8, CCL2 and CXCL1) in the conditioned media growth of WM115 cells both in vitro and in vivo. from either WM115 clonal or polyclonal knock-down cells were not consistently reduced between these two NIK depletion altered expression of genes important cell lines as compared with the control (Figures 4c and for melanoma tumor growth d). The above data suggest that NIK depletion did not significantly diminish the canonical NF-kB activation in To elucidate potential mechanisms by which depletion of WM115 melanoma cells. NIK contributed to reduced tumorigenicity of melanoma cells, the global gene expression profile of polyclonal and clonal NIK knock-down WM115 cells as compared with NIK depletion decreased the non-canonical NF-kB the non-silencing control was examined by gene expres- activation sion microarray analyses. Although results of these In contrast to the canonical NF-kB activation, depletion experiments suggested clear differences in the expression of NIK in WM115 cells reduced the basal processing of of many genes, the differences were not statistically p100 to p52 and nuclear localization of p52 (Figures 4e significant when stringent analyses (Po0.05) were ap- and f). NIK-deficient cells still responded to the ligation plied, possibly owing to the variations between polyclonal through LTbR although to the less extent than the and clonal knock-down populations. However, using less non-silencing control (Figure 4f). Knock-down of NIK stringent criteria for analysis (Po0.15), several genes that in another melanoma cell line, Hs294T, resulted in a have a crucial role in tumor growth or progression similar phenotype (Figures 4e and f).

Figure 1 NIK expression is upregulated in melanoma tissues and cells. (a) Left to right: first column: Immunofluorescent analysis of NIK expression in benign (n ¼ 30), dysplastic (n ¼ 22) nevi, primary (n ¼ 15) and metastatic (n ¼ 13) melanoma tissue samples from tissue microarray (TMA). NIK-positive cells are green, based upon Alexa 488 anti-rabbit antibody (Molecular Probes, Eugene, OR, USA) recognition of NIK antibody. Nuclei are blue based upon Hoechst stain. Second column: enlargement of area denoted by box in the adjacent left panel showing fine cellular details of the histology of the lesions. Third column: H&E stain of the tissues shown in the first column. Fourth column: enlargement of area denoted by box in the adjacent left panel showing fine cellular details of the histology of the lesions (scale bar ¼ 100 mm). (b) Quantitation of NIK staining from TMA, based upon scans of images from duplicate or triplicate core samples of each lesion using the Ariol SL-50 imaging system (Genetix, San Jose, CA, USA) calculated as described in Supplementary Materials and Methods: (area of NIK staining/area of Hoechst) x100. Statistical significance was determined by analysis of variance, where Po0.05 indicates significance. (c, d) Western blot analysis of NIK expression in melanoma tissues (c) and cell lines (Hs294T, SKMel5, SKMel28 and WM115) (d) compared with normal human epidermal melanocytes (NHEM). Densitometric scans from triplicate assays were quantitated, normalized to the loading control and calculated as fold difference from NHEM cells. (e) Western blot and (f). qRT–PCR analysis of NIK expression in NIK knock-down cells compared with the non- silencing (ns) control in Hs294T and WM115. Ns shRNA is a mock shRNA with the same vector backbone as the NIK shRNA. (g) Western blot analysis of NIK in different passages of Hs294T_shNIK1 and WM115_shNIK1 cells (P ¼ passage). Densitometric scans from triplicate assays were quantitated, normalized to the loading control and calculated as fold difference from ns.

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2584

Figure 2 Depletion of NIK decreases growth and survival of melanoma cells. (a) Growth curves of NIK knock down and non- silencing Hs294T and WM115 cells. Error bars are ±s.e.m. In Hs294T graph, s.e.m.o0.02. (n ¼ 9). (b) Quantitation of Ki67 staining in NIK-deficient Hs294T or WM115 cells compared with the non-silencing control. Error bars are ±s.e.m. (n ¼ 10). (c) BrdU incorporation analysis of NIK knock-down WM115 cells compared with the non-silencing control. Error bars are ±s.e.m. (n ¼ 7). (d) Apoptosis analysis of basal and TNFa-treated NIK knock-down WM115 compared with the control. Percentage of apoptosis is defined by % of cells that are Annexin V þ and Annexin V þ PI þ . Error bars are ±s.e.m. (n ¼ 9). In figures c and d insets are western blot showing the efficiency of NIK knock down. (e) Cell cycle progression of non-silencing shRNA control and NIK knock-down WM115 cells. Left panel: Percentage of unsynchronized and synchronized (thymidine block, 16 h) cells in each phase of cell cycle. Error bars are ±s.e.m. (n ¼ 3). Right panel: Time course graphs showing the progression of cell cycle in ns and NIK depleted WM115 populations. Times indicated (0 h, 4 h, 8 h and 12 h) were hours after the release from the thymidine block of 16 h. Scales for number of non-silencing control WM115 cells: unsynchronized: 0–4500, 0 h: 0–1600, 4 h: 0–1000, 8 h: 0–2500 and 12 h: 0–5000. Scales for number of clonal shNIK1(c) WM115 cells: unsynchronized: 0–4000, 0 h: 0–2200, 4 h: 0–1200, 8 h: 0–600 and 12 h: 0–1500.

To determine whether NIK deficiency in melanoma to the gene expression pattern from lymphoma, where cells exhibited expected changes in the gene expression cIAP2-MALT1 fusion protein induces NIK activation profile associated with NIK activity, we compared the (Rosebeck et al., 2011). In agreement with their obser- gene expression from NIK knock-down WM115 cells vations, genes (such as TLR4, IL1b and PTX3) upregu-

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2585 1600 ) 3 1400 ) 3 800 ns shNIK1 (c) 1200 ** 1000 600 **p<0.05 ** 800 400 ** 600 ** p=0.01 400 200

Average Tumor Size (mm Tumor Average 200

Average Tumor Size (mm Size Tumor Average 0 0 ns Day0 Day3 Day6 Day9 Day12Day15Day18Day21Day24 shNIK1(c)

ns shNIK1(c) 1.5

T1 T2 T3 T4 T1 T2 T3 T4 T5 1.0 p=0.03 104kD NIK Fold 42kD Actin 0.5 T= tumor

(Normalized to ns) 0.0 ns

shNIK1(c)

ns shNIK1 (c) ns shNIK1 (c)

15 100 80 10 p<0.05 60 40 p<0.05

per field 5 20 %of Ki67 positive 0 Number of caspase3 0 positive cells (per area) positive ns ns (c)

shNIK1(c) shNIK1 Figure 3 NIK depletion significantly reduces tumor burden in a melanoma xenograft model of NIK knock-down WM115 (shNIK1(c)) and non-silencing WM115. (a) Tumor growth over 24 days. (b) size of tumors at the 24 day end point. Error bars are ±s.e.m. (ns n ¼ 21 and shNIK1(c) n ¼ 22). (c) Western blot and (d) qRT–PCR analyses of NIK in tumor samples. Fold changes in figure d were calculated by normalizing to the loading control and comparing it with ns. (e) Ki67 (proliferation marker) and (f) cleaved caspase-3 (apoptosis marker) staining of non-silencing control WM115 tumors compared to WM115 with NIK knock-down. Error bars are ±s.e.m. (n ¼ 5). Statistical significance was determined by Mann–Whitney test, where Po0.05 indicates significance.

Table 1 Microarray data from NIK-depleted cells and tumors confirmed by qRT–PCR (normalized to the non-silencing control) to examine gene expression changes, which are indicative of less tumorigenecity compared with the control. Data are presented as fold changes from the control Microarray (cells) P-value Microarray (tumors) P-value qRT–PCR P-value c-Myc À3.90 0.14 À2.20 0.06 À3.39 o0.05 c-Met À3.72 0.12 À6.80 0.002 À4.30 o0.05 CXCR4 À14.50 0.08 À5.80 0.01 À620.35 o0.05 CCND2 À9.60 0.13 À1.04 0.92 À58.08 o0.05 Xaf1 3.78 0.17 3.65 0.01 5.78 o0.05 CD36 7.80 0.07 1.70 0.003 5.11 o0.05 TSC22D3 À10.40 0.04 À7.30 0.02 À33.05 o0.05 TRIM22 5.30 0.15 1.03 0.73 5.26 o0.05 ZNF711 À11.20 0.06 À7.10 0.003 À28.57 o0.05 LIN28B À18.02 0.09 À5.20 0.02 À32436417 o0.05

Abbreviation: NIK, NF-kB inducing kinase.

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2586

Figure 4 Knocking-down NIK in WM115 melanoma reduces the non-canonical but not the canonical NF-kB activation. (a) Basal IKK activity in NIK knock-down WM115 compared with WM115 non-silencing control in an in vitro kinase assay. Densitometric scans from duplicate assays were quantitated and normalized to the total immunoprecipitated IKKs. (b) Nuclear p-p65 in NIK knock- down WM115 compared with the non-silencing control (c ¼ cytoplasm, n ¼ nucleus). Densitometric scans from four assays were quantitated and normalized to the loading control. TNFa-treatment was used as a positive control in both a and b.(c, d) Cytokine array data from conditioned media of NIK knock-down cells (clonal and polyclonal) compared with the non-silencing control. Signal intensity for each cytokine was normalized to the internal positive control. Error bars are ±s.e.m. (n ¼ 4). Statistical significance was determined by Mann–Whitney test, where Po0.05 indicates significance. (e) Nuclear and cytoplasmic localization of p52 in non- silencing and NIK knock-down WM115 and Hs294T (c ¼ cytoplasm, n ¼ nucleus). Densitometric scans from triplicate assays were quantitated, normalized to the loading control and calculated as fold difference from ns. (f) The proteolytic processing of p100 to p52 in non-silencing and NIK knock-down WM115 and Hs294T. Cells were treated with LTbR agonist antibody or IgG (2 mg/ml) for 16 h. Densitometric scans from triplicate assays were quantitated, normalized to the loading control and calculated as fold difference from ns. (g) qRT–PCR analysis of gene expression in NIK-deficient WM115 cells normalized to the loading control, calculated as fold difference from ns and compared with the study by Rosebeck et al., 2011.

lated in cIAP2-MALT1-negative tumors (NIK low) lated in cIAP2-MALT1-positive tumors (NIK high) were increased in NIK-deficient melanoma, whereas were decreased in NIK-deficient cells compared with the genes (such as HMOX1, PIM2 and CXCR4) upregu- control (Po0.05 except PTX3) (Figure 4g). Based on

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2587 suggest that NIK regulates pro-survival proteins BCL2 and survivin at the transcriptional level. A similar decrease in protein expression and mRNA levels of pro- survival factors was observed in NikÀ/À MEFs, suggest- ing that NIK regulates the expression of these genes and the regulation is not melanoma specific (Figures 5c and d).

NIK modulated melanoma tumorigenesis by regulating b-catenin activity Although NIK knock-down WM115 cells exhibited decreased non-canonical NF-kB activity, this pathway has not been directly linked to the expression of pro- survival proteins such as survivin. Therefore, we postu- lated that NIK modulates the expression of survivin through an alternative pathway. Firestein et al. have identified NIK (MAP3K14) as one of the kinases that regulates b-catenin activation in colorectal cancer cells (Firestein et al., 2008). To explore the possibility that NIK may modulate b-catenin-mediated transcription, we ex- amined the mRNA expression of AXIN2 and TCF7,two genes typically targeted by b-catenin,inbothNIKknock- down melanoma cells and NikÀ/À MEFs by qRT–PCR. Figure 5 NIK modulates the expression of pro-survival factors. Data show that deficiency of NIK reduced expression of (a) Protein expression of survival factors such as BCL2, cIAP1 and survivin in NIK knock-down WM115 (both clonal and polyclonal) these two b-catenin regulated genes (Figures 6a and b). compared with the non-silencing control. Similar to WM115, In addition, microarray analysis confirmed by qRT–PCR Hs294T cells with NIK knock down exhibited decreased protein of NIK knock-down melanoma cells shows significant expression of survivin. Densitometric scans from triplicate assays reduction of b-catenin targets including c-MYC, c-MET were quantitated, normalized to the loading control and calculated as fold difference from ns. (b) mRNA levels (qRT–PCR) of BCL2, and CCND2, further supporting the concept that NIK survivin and cIAP1 in NIK-depleted and non-silencing WM115 can modulate b-catenin activity (Table 1) (He et al.,1998; cells. Fold changes were calculated by normalizing to the loading Boon et al., 2002; Cole et al., 2010). control and comparing it with ns. (c) Western blot and (d) qRT– In addition, the expression of pro-survival factors / / PCR of BCL2 and survivin in Nik þ þ and NikÀ À MEFs. In figure c such as survivin can be transcriptionally regulated by densitometric scans from triplicate assays were quantitated, normalized to the loading control and calculated as fold difference b-catenin (Zhang et al.,2001).Thus,wedetermined from WT. In figure d fold changes were calculated by normalizing whether NIK modulates survivin expression through to the loading control and comparing it with WT. affecting b-catenin transcriptional activity using chroma- tin immunoprecipitation analysis to monitor the presence of b-catenin at the survivin promoter. Both NIK knock- À/À above data, we conclude that knocking-down NIK down melanoma cells and Nik MEFs exhibited a reduced the non-canonical NF-kB activation in mela- decreased b-catenin occupancy at the promoter of the noma cells. survivin gene (Figures 6c and d), supporting the notion that NIK regulates survivin expression, and possibly other b-catenin targets such as c-MYC, c-MET and NIK knock-down attenuated the expression CCND2,throughb-catenin activity. of pro-survival factors To understand how NIK may affect b-catenin acti- One of the mechanisms by which tumor cells escape vity, nuclear and cytoplasmic extraction and immuno- apoptosis is by upregulating the expression of survival fluorescence staining of b-catenin was performed in factors. To understand the apoptotic phenotype of NIK knock-down melanoma cells and NikÀ/À MEFs. NIK-depleted cells, protein expression of a panel of The ratio of nuclear to cytoplasmic pool of b-catenin is pro-survival factors was examined. Some anti-apoptotic less in NIK-deficient cells as shown by both analyses proteins such as BCL-XL and cFLIP expression (Figures 6e and f), implying that NIK regulates sub- remained unchanged (Figure 5a). However, pro-survival cellular localization of b-catenin. proteins such as cIAP1, survivin and BCL2 were markedly decreased in NIK knock-down WM115 cells (Figure 5a). Depletion of NIK in Hs294T cells also Discussion diminished survivin protein levels (Figure 5a). qRT– PCR demonstrates that there is no difference in cIAP1 In this study, we demonstrate for the first time that NIK mRNA in NIK-depleted melanoma cells, though both regulates melanoma tumor cell growth and survival. We BCL2 and survivin mRNA levels were significantly also demonstrate that the mechanism for this effect on diminished (Figure 5b). Although NIK depletion may tumor growth involves NIK regulation of expression affect cIAP1 at a post-translational level, the above data of pro-survival genes in part through modulation

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2588

Figure 6 NIK regulates survivin expression through b-catenin activity. (a, b) Expression of traditional b-catenin transcriptional targets such as AXIN2 and TCF7 in NIK knock-down WM115 (a) and NikÀ/À MEFs. (b) Fold changes were calculated by normalizing to the loading control and comparing it with ns or WT. (c, d) chromatin immunoprecipitation (ChIP) analysis using b-catenin antibody and primers flanking the survivin promoter exhibited decreased b-catenin occupancy at the survivin promoter in NIK knock-down WM115 (c) and NikÀ/À MEFs. (d) Amount of DNA bound to IgG, b-catenin or histone was calculated as percentage of the 2% input. (e, f) Immunofluorescent staining and western blot analyses showing sub-cellular localization of b-catenin in NIK knock-down WM115 (e) and NikÀ/À MEFs (f). Images for immunofluorescence staining were taken using a  63 oil lens and pseudo-colored. Densitometric scans of western blot from triplicate assays were quantitated and normalized to the loading control.

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2589 of b-catenin transcriptional activity. NIK depletion in melanoma (Ng et al., 2004), but is increased at the decreases proliferation and increases apoptosis in mRNA level in NIK-depleted WM115 cells. Remarkably, melanoma cells in vitro and sensitizes melanoma cells a potent tumor promoter, LIN28B, is downregulated in to TNFa-induced apoptosis. NIK knock-down cells also NIK knock-down WM115 cells. LIN28B de-represses exhibit a defect in cell cycle progression. Although the oncogenes such as K-RAS and c-MYC by regulating the canonical NF-kB pathway is not decreased in NIK let-7 miRNA family (Viswanathan et al., 2009). Given the knock-down cells, pro-survival proteins such as important role of the RAS/RAF/MAPK pathway in BCL2, cIAP1 and survivin are diminished, suggesting melanoma, suppression of LIN28B in NIK knock-down that the canonical NF-kB pathway lacks the cyto- tumor cells is likely biologically relevant. As changes in protective function in response to TNFa when pro- CXCR4, c-MET, c-MYC, LIN28B and XAF1 observed in survival effectors are absent. Other studies have also cultured NIK knock-down melanoma cells were retained reported that cell death can be induced by TNFa in the in NIK knock-down tumors, our data indicate that these presence of NF-kB activation when other survival genes contribute to tumorigenicity. Based on its ability to signals such as those of cIAPs are disrupted in cancer regulate the expression of genes important for melanoma cells (Petersen et al., 2007; Varfolomeev et al., 2007; progression, we propose that NIK is a promising target Vince et al., 2007). for melanoma therapy. Consistent with in vitro data, the melanoma tumor Upregulation of survival factors is one of the mechan- xenograft data show that NIK silencing markedly isms by which cancer cells resist apoptosis (Baldwin, decreases tumor cell proliferation (based upon Ki67 2001). Previous studies suggest that pro-survival proteins staining) and melanoma tumor growth. Despite the such as BCL2, cIAP1 and survivin contribute to tumori- significant decrease in tumor volume, NIK-deficient genesis of melanoma. As a result, functional disruption tumors showed diminished cleaved caspase 3 staining, of these proteins impairs melanoma tumor cell growth suggesting the following possibilities: (1) cancer cells or survival (Grossman et al., 2001; Hilmi et al., 2008; undergoing apoptosis were already eliminated by the Lecis et al., 2010). Consistent with these findings, BCL2, time tumors were harvested; or (2) the turn-over rate of cIAP1 and survivin levels were markedly decreased melanoma cells in the rapidly growing control tumors in NIK-deficient melanoma cells. Yet, the canonical was faster due to the limited nutrients in the micro- NF-kB transcriptional activity, which typically regulates environment. The delay in cell cycle observed in NIK- the expression of survival proteins (Baldwin, 2001), is not depleted cells also likely contributes to the decreased diminished in NIK-deficient melanoma cells, in agreement tumor growth. Overall, NIK depletion markedly with the genetic data (Yin et al., 2001). The activity of the decreases the net melanoma tumor growth. non-canonical pathway is downregulated in NIK-deficient Similarly, studies of multiple myeloma, T-cell leuke- melanoma cells, possibly contributing to downstream mia and lymphoma suggest that NIK mediates prolif- effects of NIK on melanoma growth. However, the non- eration and survival and as a result, depleting NIK in canonical NF-kB pathway has not been directly linked to these cancers decreased tumor growth (Annunziata the expression of genes such as survivin. Interestingly, et al., 2007; Keats et al., 2007; Saitoh et al., 2008; NikÀ/À MEFs that also do not exhibit a defect in the Conze et al., 2010; Demchenko et al., 2010). Consistent canonical NF-kB activation (Yin et al., 2001), have with the previous literature, our data clearly demon- decreased survivin expression. Thus, we propose that strate that NIK contributes to melanoma tumor growth. decreased pro-survival factor expression in NIK-depleted However, which downstream effectors of NIK regulate cells is attributed to a non-NF-kB function of NIK. Thus, tumorigenesis has remained unclear. Here, we demon- NIK like IKKa and IKKb, appears to have NF-kB- strate that NIK regulates expression of genes that are independent functions (Chariot, 2009). pro-tumorigenic and pro-survival. Here our data suggest that NIK regulates b-catenin Gene expression data from NIK-deficient melanoma activity, a novel mechanism by which NIK mediates cells suggest that NIK modulates the expression of cancer progression. Downregulation of b-catenin tran- CXCR4, c-MET, c-MYC, LIN28B and XAF1, all of scriptional targets (c-MYC, CCND2, c-MET, AXIN2, which are important for melanoma and/or tumor malig- TCF7 and survivin) in NIK-deficient cells supports the nancy. For instance, high CXCR4 expression predicts notion that NIK mediates b-catenin transcriptional poor prognosis in malignant melanoma (Scala et al., activity. Specifically, NIK depletion reduces b-catenin 2005). The c-MET tyrosine kinase receptor and the occupancy at the promoter of survivin, diminishing the c-MYC transcription factor stimulate survival and expression of survivin. b-catenin is possibly recruited proliferation (Lai et al., 2009; Ruggero, 2009). It was to the survivin promoter by TCF4 (Kim et al., 2003). reported that both of these genes are amplified in meta- Interestingly, several studies have reported that b-catenin static melanoma patient tumors (Moore et al., 2008). activity contributes to melanoma tumor growth (Rubinfeld Specifically, both c-MET and c-MYC are proposed to et al., 1997; Delmas et al., 2007; Sinnberg et al., 2010). be targets for melanoma therapy (Pastorino et al., 2003; Although NIK knock-down cells exhibited a reduc- Puri et al., 2007). Depletion of NIK significantly dimi- tion in b-catenin-mediated transcription of genes impor- nishes the expression of CXCR4, c-MET and c-MYC tant for melanoma tumorigenesis, it is important to mRNA. In addition XAF1, a negative regulator of cell note that NIK depletion does not affect all of the survival that counteracts the function of the survival transcriptional targets of b-catenin. For example, Mitf, factor XIAP (Liston et al., 2001), is significantly decreased a transcriptional target of b-catenin, has an important

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2590 role in melanoma development by regulating genes such obtained from Vanderbilt Skin Research Core and cultured as BCL2 (McGill et al., 2002; Cheli et al., 2010). in Media 254 containing supplements for melanocytes (S-002- However, a difference in Mitf gene expression was not 5, Cascade Biologics, Carlsbad, CA, USA) under the same detected in our microarray analysis, suggesting that conditions. Melanoma cell lines were obtained from ATCC decreased BCL2 expression detected in NIK-deficient and were tested for mycoplasma monthly. cells may be regulated by transcription factors other than Mitf. Generating NIK knock-down cell lines The specific mechanisms by which NIK activates Melanoma cells (Hs294T or WM115) were infected with lentivirus b-catenin are yet to be determined. One possible mecha- carrying NIK short hairpin RNA (shRNA), (Sigma Mission shRNA pLKO.1_733 or_731, Sigma, St Louis, MO, USA) or nism is that NIK modulates the phosphorylation of IKKa shRNA (Sigma Mission shRNA pLKO.1_508) and each regulators of b-catenin, such as GSK3b. Alternatively, shRNA vector also encoded a puromycin resistance marker. Non- NIK may directly or indirectly regulate the nuclear silencing shRNA in pLKO vector containing shRNA insert that transport of b-catenin, based upon the finding that does not target human genes was used as a mock shRNA control. NIK-depleted cells exhibited reduced nuclear to cyto- A total of 48 h after infection, cells were cultured in media plasmic ratio of b-catenin. Although we did not examine containing puromycin (Sigma P9620) at a concentration of in detail the plasma membrane localization of b-catenin 2 ug/ml to select for cells with shRNA constructs. in this study, it is possible that the localization is altered in NIK-deficient cells and this change may contribute to Cytokine array NIK-mediated tumorigenesis. The membrane cadherins/ Non-silencing or knock-down cells were cultured for 24 h in b-catenin complex has a crucial role in invasiveness of equal volumes of serum free DMEM/F12 to obtain condi- melanoma by mediating homotypic or heterotypic cell tioned media. Cytokine array (AAH-CYT-G2000-8, RayBio- adhesions (McGary et al., 2002). Tech, Norcross, GA, USA) was performed according to the user manual using 50ml of collected conditioned media. Serum- NIK may also affect other signaling pathways that free media was used as a control to determine the background mediate melanoma gene expression and growth. Our level of cytokines/chemokines. group has demonstrated that ERK phosphorylation was reduced by over-expression of dominant-negative NIK in BrdU incorporation melanoma cells (Dhawan and Richmond, 2002) suggest- BrdU incorporation was performed using the BrdU staining kit ing that this MAP3K (NIK) can alter the MAPK (559619, BD Biosciences, San Jose, CA, USA). Briefly, cells were pathway, which is essential for melanoma growth and cultured to 70–80% confluency, and incubated with BrdU for survival. Therefore, NIK may modulate melanoma 4 h in serum-containing media. Following BrdU incorporation, progression by influencing other signaling pathways in cells were fixed, permeabilized, stained with FITC-conjugated addition to the ones mentioned above. anti-BrdU antibody, according to the manufacturer’s protocol, Recently, significant progress has been made in mela- and analyzed by flow cytometry (Beckman Coulter, CA, USA). noma therapy for those patients that exhibit BRAF mutation (Dhomen and Marais, 2009). However, it is Annexin V staining imperative to also increase our understanding of other Non-silencing and knock-down cells were plated at equal density pathways that are involved in melanoma growth so that and treated with TNFa (1000 ng/ml) (300-01A, PeproTech, therapies for patients that cannot be effectively treated Rock Hill, NJ, USA) where indicated. After 8 h of TNFa with BRAFV600E inhibitors can be developed. Owing to treatment, the Annexin V staining assay was performed as described in the user manual (TACS Annexin V 4830-01-K its non-essential role in the canonical NF-kB activation R&D Systems, Minneapolis, MN, USA) then samples were and its ability to regulate survival factors, NIK is an subjected to flow cytometry analysis (Beckman Coulter). attractive target to induce melanoma cell death without substantially reducing NF-kB activity, which is impor- Orthotopic tumor implant studies in mice tant for normal cellular processes. Moreover, as both Non-silencing control or NIK knock-down WM115 cells NIK and b-catenin are involved in other cancers such (3 Â 106 per mouse) were subcutaneously injected into athy- as multiple myeloma and colon cancer, it may be a mic-Foxn1nu/nu nude mice (Harlan Laboratories, Indianapolis, good therapeutic target for a number of malignancies IN, USA) and allowed to grow until palpable tumors (Malbon, 2005; Annunziata et al., 2007; Keats et al., developed (about 2–3 weeks). Tumors were measured every 2 2007). However, as NIK-depleted melanoma cells may days with a micro-caliper. Three weeks from the day of initial eventually escape apoptosis as observed in our studies, tumor detection (or when tumors reached 1.5 cm in one combination of NIK inhibitors with other therapies dimension), mice were killed and final tumor volumes were should be carefully considered when NIK is used as a measured by HBSS volume displacement. All experiments performed with mice were according to the guidelines targeted therapy in cancer patients. approved by Vanderbilt University School of Medicine Animal Care and Use Committee.

Materials and methods Statistical analysis Non-parametric statistical analyses (Mann–Whitney or Cell culture analysis of variance) were performed using the Graphpad Melanoma cells were cultured in DMEM/F12 containing Prism software (GraphPad Software, Inc, La Jolla, CA, 10% FBS at 37 1C with 95% air/5% CO2 in a water-jacketed USA). Statistically significant differences are assumed if the incubator. Normal human epidermal melanocytes were P-value p0.05.

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2591 Supplemental methods Functional Genomic Shared Resource Core, VA Flow Cyto- Methods for western blot, nuclear and cytoplasmic extraction, metry Core and Vanderbilt Immunohistochemistry Core for IKK kinase assay, qRT–PCR, chromatin immunoprecipita- technical help. We thank Dr Joseph T Roland (Epithelial tion, cell cycle analysis, immunofluorescence staining of tissue Biology Center) for assistance with the ariol scanning and microarray, immunohistochemistry and immunofluorescence quantitation of TMAs. We thank Dr Mark Kelly and staining are described in Supplementary Materials. Vanderbilt melanoma group for providing melanoma tissues and Dr Sara Kantrow for reviewing some of the TMA slides. Conflict of interest We thank Dr Fiona Yull and Dr Punita Dhawan for thoughtful discussion and Vanderbilt Editor Club for careful reading of the The authors declare no conflict of interest. manuscript. Grant Support: This work is supported by NIH, CA 098807-01 and the Department of Veterans Affairs, Career Scientist Award and Merit Award (to AR). Acknowledgements Author contributions: Designed experiments and wrote paper: YMT and AR. Performed experiments: YMT, YS, JY We would like to thank Dr Robert Schreiber (Washington and RS. Analyzed data: YMT, JY and RS. Shared reagents: SN. University) for generously sharing NikÀ/À MEFs, and Vanderbilt Contributed human tissue samples for TMA: AB, CS and CM.

References

Annunziata CM, Davis RE, Demchenko Y, Bellamy W, Gabrea A, Dhawan P, Richmond A. (2002). A novel NF-kappa B-inducing Zhan F et al. (2007). Frequent engagement of the classical and kinase-MAPK signaling pathway up-regulates NF-kappa B activity alternative NF-kappaB pathways by diverse genetic abnormalities in melanoma cells. J Biol Chem 277: 7920–7928. in multiple myeloma. Cancer Cell 12: 115–130. Dhomen N, Marais R. (2009). BRAF signaling and targeted Baldwin AS. (2001). Control of oncogenesis and cancer therapy therapies in melanoma. Hematol Oncol Clin North Am 23: resistance by the transcription factor NF-kappaB. J Clin Invest 107: 529–545, ix. 241–246. Firestein R, Bass AJ, Kim SY, Dunn IF, Silver SJ, Guney I et al. Balentien E, Mufson BE, Shattuck RL, Derynck R, Richmond A. (2008). CDK8 is a colorectal cancer oncogene that regulates beta- (1991). Effects of MGSA/GRO alpha on melanocyte transforma- catenin activity. Nature 455: 547–551. tion. Oncogene 6: 1115–1124. Grossman D, Kim PJ, Schechner JS, Altieri DC. (2001). Inhibition of Boon EM, van der Neut R, van de Wetering M, Clevers H, Pals ST. melanoma tumor growth in vivo by survivin targeting. Proc Natl (2002). Wnt signaling regulates expression of the receptor tyrosine Acad Sci U S A 98: 635–640. kinase met in colorectal cancer. Cancer Res 62: 5126–5128. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Costa LT Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin et al. (1998). Identification of c-MYC as a target of the APC J et al. (2011). Improved survival with vemurafenib in melanoma pathway. Science 281: 1509–1512. with BRAF V600E mutation. N Engl J Med 364: 2507–2516. Hilmi C, Larribere L, Giuliano S, Bille K, Ortonne JP, Ballotti R et al. Chariot A. (2009). The NF-kappaB-independent functions of IKK (2008). IGF1 promotes resistance to apoptosis in melanoma cells subunits in immunity and cancer. Trends Cell Biol 19: 404–413. through an increased expression of BCL2, BCL-X(L), and survivin. Cheli Y, Ohanna M, Ballotti R, Bertolotto C. (2010). Fifteen-year J Invest Dermatol 128: 1499–1505. quest for microphthalmia-associated transcription factor target Keats JJ, Fonseca R, Chesi M, Schop R, Baker A, Chng WJ genes. Pigment Cell Melanoma Res 23: 27–40. et al. (2007). Promiscuous mutations activate the noncanonical Chin L, Merlino G, DePinho RA. (1998). Malignant melanoma: NF-kappaB pathway in multiple myeloma. Cancer Cell 12: modern black plague and genetic black box. Genes Dev 12: 131–144. 3467–3481. Kim PJ, Plescia J, Clevers H, Fearon ER, Altieri DC. (2003). Survivin Cole AM, Myant K, Reed KR, Ridgway RA, Athineos D, Van den and molecular pathogenesis of colorectal cancer. Lancet 362: Brink GR et al. (2010). Cyclin D2-cyclin-dependent kinase 4/6 is 205–209. required for efficient proliferation and tumorigenesis following Apc Lai AZ, Abella JV, Park M. (2009). Crosstalk in Met receptor loss. Cancer Res 70: 8149–8158. oncogenesis. Trends Cell Biol 19: 542–551. Conze DB, Zhao Y, Ashwell JD. (2010). Non-canonical NF-kappaB Lecis D, Drago C, Manzoni L, Seneci P, Scolastico C, Mastrangelo E activation and abnormal B cell accumulation in mice expressing et al. (2010). Novel SMAC-mimetics synergistically stimulate ubiquitin protein ligase-inactive c-IAP2. PLoS Biol 8: melanoma cell death in combination with TRAIL and Bortezomib. e1000518. Br J Cancer 102: 1707–1716. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S et al. Lev DC, Ruiz M, Mills L, McGary EC, Price JE, Bar-Eli M. (2003). (2002). Mutations of the BRAF gene in human cancer. Nature 417: Dacarbazine causes transcriptional up-regulation of interleukin 8 949–954. and vascular endothelial growth factor in melanoma cells: a possible Delmas V, Beermann F, Martinozzi S, Carreira S, Ackermann J, escape mechanism from chemotherapy. Mol Cancer Ther 2: Kumasaka M et al. (2007). Beta-catenin induces immortalization 753–763. of melanocytes by suppressing p16INK4a expression and Ling L, Cao Z, Goeddel DV. (1998). NF-kappaB-inducing kinase cooperates with N-Ras in melanoma development. Genes Dev 21: activates IKK-alpha by phosphorylation of Ser-176. Proc Natl Acad 2923–2935. Sci U S A 95: 3792–3797. Demchenko YN, Glebov OK, Zingone A, Keats JJ, Bergsagel PL, Liston P, Fong WG, Kelly NL, Toji S, Miyazaki T, Conte D et al. Kuehl WM. (2010). Classical and/or alternative NF-kappaB (2001). Identification of XAF1 as an antagonist of XIAP anti- pathway activation in multiple myeloma. Blood 115: 3541–3552. Caspase activity. Nat Cell Biol 3: 128–133. Devalaraja MN, Wang DZ, Ballard DW, Richmond A. (1999). Malbon CC. (2005). Beta-catenin, cancer, and G proteins: not just Elevated constitutive IkappaB kinase activity and IkappaB- for frizzleds anymore. Sci STKE 2005: pe35. alpha phosphorylation in Hs294T melanoma cells lead to Malinin NL, Boldin MP, Kovalenko AV, Wallach D. (1997). MAP3K- increased basal MGSA/GRO-alpha transcription. Cancer Res 59: related kinase involved in NF-kappaB induction by TNF, CD95 1372–1377. and IL-1. Nature 385: 540–544.

Oncogene NF-jB inducing kinase (NIK) in melanoma YM Thu et al 2592 McGary EC, Lev DC, Bar-Eli M. (2002). Cellular adhesion pathways Scala S, Ottaiano A, Ascierto PA, Cavalli M, Simeone E, Giuliano P and metastatic potential of human melanoma. Cancer Biol Ther et al. (2005). Expression of CXCR4 predicts poor prognosis in 1: 459–465. patients with malignant melanoma. Clin Cancer Res 11: 1835–1841. McGill GG, Horstmann M, Widlund HR, Du J, Motyckova G, Sinnberg T, Menzel M, Kaesler S, Biedermann T, Sauer B, Nahnsen S Nishimura EK et al. (2002). Bcl2 regulation by the melanocyte et al. (2010). Suppression of casein kinase 1alpha in melanoma cells master regulator Mitf modulates lineage survival and melanoma cell induces a switch in beta-catenin signaling to promote metastasis. viability. Cell 109: 707–718. Cancer Res 70: 6999–7009. Moore SR, Persons DL, Sosman JA, Bobadilla D, Bedell V, Smith DD Smith C, Andreakos E, Crawley JB, Brennan FM, Feldmann M, et al. (2008). Detection of copy number alterations in metastatic Foxwell BM. (2001). NF-kappaB-inducing kinase is dispensable for melanoma by a DNA fluorescence in situ hybridization probe activation of NF-kappaB in inflammatory settings but essential for panel and array comparative genomic hybridization: a southwest lymphotoxin beta receptor activation of NF-kappaB in primary oncology group study (S9431). Clin Cancer Res 14: 2927–2935. human fibroblasts. J Immunol 167: 5895–5903. Neely RJ, Brose MS, Gray CM, McCorkell KA, Leibowitz JM, Sullivan RJ, Atkins MB. (2009). Molecular-targeted therapy in Ma C et al. (2010). The RET/PTC3 oncogene activates classical malignant melanoma. Expert rev anticancer ther 9: 567–581. NF-kappaB by stabilizing NIK. Oncogene 30: 87–96. Van Antwerp DJ, Martin SJ, Verma IM, Green DR. (1998). Inhibition Ng KC, Campos EI, Martinka M, Li G. (2004). XAF1 expression is of TNF-induced apoptosis by NF-kappa B. Trends Cell Biol 8: significantly reduced in human melanoma. J Invest Dermatol 123: 107–111. 1127–1134. Varfolomeev E, Blankenship JW, Wayson SM, Fedorova AV, Pastorino F, Brignole C, Marimpietri D, Pagnan G, Morando A, Kayagaki N, Garg P et al. (2007). IAP antagonists induce auto- Ribatti D et al. (2003). Targeted liposomal c-myc antisense oligo- ubiquitination of c-IAPs, NF-kappaB activation, and TNFalpha- deoxynucleotides induce apoptosis and inhibit tumor growth and dependent apoptosis. Cell 131: 669–681. metastases in human melanoma models. Clin Cancer Res 9: Vince JE, Wong WW, Khan N, Feltham R, Chau D, Ahmed AU et al. 4595–4605. (2007). IAP antagonists target cIAP1 to induce TNFalpha- Petersen SL, Wang L, Yalcin-Chin A, Li L, Peyton M, Minna J et al. dependent apoptosis. Cell 131: 682–693. (2007). Autocrine TNFalpha signaling renders human cancer cells Viswanathan SR, Powers JT, Einhorn W, Hoshida Y, Ng TL, susceptible to Smac-mimetic-induced apoptosis. Cancer Cell 12: Toffanin S et al. (2009). Lin28 promotes transformation and 445–456. is associated with advanced human malignancies. Nat Genet 41: Plenchette S, Cheung HH, Fong WG, LaCasse EC, Korneluk RG. 843–848. (2007). The role of XAF1 in cancer. Curr Opin Investig Drugs 8: Wang CY, Mayo MW, Baldwin Jr AS. (1996). TNF- and cancer 469–476. therapy-induced apoptosis: potentiation by inhibition of NF- Puri N, Ahmed S, Janamanchi V, Tretiakova M, Zumba O, Krausz T kappaB. Science 274: 784–787. et al. (2007). c-Met is a potentially new therapeutic target for Wang D, Yang W, Du J, Devalaraja MN, Liang P, Matsumoto K treatment of human melanoma. Clin Cancer Res 13: 2246–2253. et al. (2000). MGSA/GRO-mediated melanocyte transformation Puzanov I, Flaherty KT. (2010). Targeted molecular therapy in involves induction of Ras expression. Oncogene 19: 4647–4659. melanoma. Semin Cutan Med Surg 29: 196–201. Yang J, Amiri KI, Burke JR, Schmid JA, Richmond A. (2006). BMS- Rosebeck S, Madden L, Jin X, Gu S, Apel IJ, Appert A et al. (2011). 345541 targets inhibitor of kappaB kinase and induces apoptosis in Cleavage of NIK by the API2-MALT1 fusion oncoprotein leads to melanoma: involvement of nuclear factor kappaB and mitochondria noncanonical NF-kappaB activation. Science 331: 468–472. pathways. Clin Cancer Res 12: 950–960. Rubinfeld B, Robbins P, El-Gamil M, Albert I, Porfiri E, Polakis P. Yang J, Pan WH, Clawson GA, Richmond A. (2007). Systemic (1997). Stabilization of beta-catenin by genetic defects in melanoma targeting inhibitor of kappaB kinase inhibits melanoma tumor cell lines. Science 275: 1790–1792. growth. Cancer Res 67: 3127–3134. Ruggero D. (2009). The role of Myc-induced protein synthesis in Yang J, Richmond A. (2001). Constitutive IkappaB kinase activity cancer. Cancer Res 69: 8839–8843. correlates with nuclear factor-kappaB activation in human mela- Saitoh Y, Martinez Bruyn VJ, Uota S, Hasegawa A, Yamamoto N, noma cells. Cancer Res 61: 4901–4909. Imoto I et al. (2010). Overexpression of NF-kappaB inducing kinase Yang J, Splittgerber R, Yull FE, Kantrow S, Ayers GD, Karin M et al. underlies constitutive NF-kappaB activation in lung cancer cells. (2010). Conditional ablation of Ikkb inhibits melanoma tumor Lung Cancer 70: 263–270. development in mice. J Clin Invest 120: 2563–2574. Saitoh Y, Yamamoto N, Dewan MZ, Sugimoto H, Martinez BV, Yin L, Wu L, Wesche H, Arthur CD, White JM, Goeddel DV et al. Iwasaki Y et al. (2008). Overexpressed NF-\{kappa\}B inducing (2001). Defective lymphotoxin-beta receptor-induced NF-kappaB kinase contributes to the tumorigenesis of adult T-cell leukemia and transcriptional activity in NIK-deficient mice. Science 291: 2162–2165. Ho. Blood 111: 5118–5129. Zhang T, Otevrel T, Gao Z, Gao Z, Ehrlich SM, Fields JZ et al. (2001). Scala S, Giuliano P, Ascierto PA, Ierano C, Franco R, Napolitano M Evidence that APC regulates survivin expression: a possible et al. (2006). Human melanoma metastases express functional mechanism contributing to the stem cell origin of colon cancer. CXCR4. Clin Cancer Res 12: 2427–2433. Cancer Res 61: 8664–8667.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene