Prognostic Significance of Nuclear Factor-KB p105/p50 in Human Melanoma and Its Role in Cell Migration

Transcriptional factor nuclear factor-kappaB (NF-kappaB) family has been shown to play an important role in tumor pathogenesis and serve as a potential target in cancer therapy. However, it is necessary to clarify the specific functions of NF-kappaB members, which would provide the basis for the selective blockade and reduction of therapeutic side effects resulting from unspecific inhibition of NF-kappaB members. In this study, we explored the role of NF-kappaB p105/p50 in melanoma pathogenesis in vitro and in vivo. We found that the expression of NF-kappaB p105/p50 significantly increased in dysplastic nevi, primary melanoma, and metastatic melanoma compared with normal nevi (P = 0.0004, chi(2) test). Furthermore, NF-kappaB p105/p50 nuclear staining increased with melanoma progression and strong NF-kappaB p105/p50 nuclear staining was inversely correlated with disease-specific 5-year survival of patients with tumor thickness >2.0 mm (P = 0.014, log-rank test). Multivariate Cox regression analysis revealed that nuclear expression of NF-kappaB p105/p50 is an independent prognostic factor in this subgroup. Moreover, we found that up-regulation of NF-kappaB p50 enhanced melanoma cell migration, whereas small interfering RNA knockdown inhibited cell migration. In addition, overexpression of NF-kappaB p50 induced RhoA activity and Rock-mediated formation of stress fiber in melanoma cells. Taken together, our data indicate that NF-kappaB p105/p50 may be an important marker for human melanoma progression and prognosis as well as a potentially selective therapeutic target.


Introduction
Cutaneous malignant melanoma is the most aggressive form of skin cancer, and its incidence is rising rapidly among the Caucasian population worldwide (1).In 2005, there were 59,580 estimated new melanoma patients and f7,770 patients died from this disease in the United States (2).Although early localized melanoma is curable with surgical excision (3), the median survival time for patients with metastatic melanoma is 6 to 10 months (4).Treatment for the advanced disease is ineffective despite combined therapeutic regimes of surgery, immunotherapy, radiotherapy, and chemotherapy (5)(6)(7)(8).
Nuclear factor-nB (NF-nB) has been shown to play an important role in melanoma proliferation, apoptosis resistance, invasion, and metastasis (9).Cell cycle-related factors, such as cyclin D1 and cyclin-dependent kinase 2, as well as antiapoptotic factors, such as melanoma inhibitor of apoptosis, which are all downstream effectors of NF-nB, are shown to be overexpressed in melanoma (10,11).NF-nB also regulates the expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and metalloproteinases that facilitate the invasion and metastasis of melanoma (12)(13)(14)(15)(16). Recently, inhibition of NF-nB was shown to be a potential means to sensitize melanoma cells to anticancer drugs (17).
Despite close association between NF-nB expression and tumor progression, complete inhibition of NF-nB activity may result in severe side effects because NF-nB regulates the expression of >150 genes, some of which are involved with normal cellular functions, such as immune response and cell proliferation (18)(19)(20).Lack of selective blockade of NF-nB activity is due to the poor understanding of specific functions of NF-nB family members in different cell types.The NF-nB family is composed of five mammalian members: RelA/p65, RelB, c-Rel, NF-nB1 p105/p50, and NF-nB2 p100/p52.Active forms of NF-nB subunit p50 and p52 are produced by ubiquitin-dependent proteolytic process of the COOH-terminal domains of NF-nB1 p105 and NF-nB2 p100, respectively.In most cells, NF-nB subunit forms homodimers or heterodimers and sequestered by InB in cytoplasm.The activation of NF-nB dimers in response to stimuli results in nuclear translocation of the dimers, which then regulate a wide variety of gene expression (20).Thus far, only a few reports showed specific roles of individual NF-nB members under certain circumstances.For example, overexpression of RelA inhibited tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis, whereas overexpression of c-Rel enhanced this process (21).Depletion of NF-nB subunit RelA led to hepatic cell death and embryonic lethality of mice (22).In human keratinocytes, overexpression of RelA caused cell growth arrest (23).The p50 homodimer is also known to be responsible for lipopolysaccharide tolerance (24).To define the role of specific NF-nB member in melanoma pathogenesis, here, we show that nuclear expression of NF-nB p105/p50 is increased with melanoma progression and inversely correlated with disease-specific 5-year survival of patients with tumor thickness >2.0 mm.In addition, we show that NF-nB p105/p50 enhances melanoma cell migration, RhoA activity, as well as stress fiber formation.

Materials and Methods
Tissue microarray construction.The construction and composition of the melanoma tissue microarray (TMA) were described previously (25).Briefly, formalin-fixed, paraffin-embedded tissue blocks containing 16 normal nevi, 66 dysplastic nevi, 204 primary melanomas, and 58 metastatic melanomas were retrieved from the 1990 to 1998 archives of the Department of Pathology, Vancouver General Hospital (Vancouver, British Columbia, Canada).The use of human skin tissues in this study was approved by the medical ethics committee of the University of British Columbia and was done in accordance with the Declaration of Helsinki Guidelines.For each case, the most representative tumor area was selected and marked on H&E-stained slides.The TMAs were assembled using a tissue array instrument (Beecher Instruments, Silver Spring, MD).Duplicate 0.6-mm-thick tissue cores were taken from each biopsy specimen.Three composite high-density TMA blocks containing 107, 126, and 111 cases from a total of 344 patients were designed.Multiple 4-Am sections were cut with a Leica microtome (Leica Microsystems, Inc., Bannockburn, IL) and transferred to adhesive-coated slides.One section from each TMA was routinely stained with H&E.The remaining sections were stored at room temperature for immunohistochemical staining.
Immunohistochemistry. TMA slides were deparaffinized by heating at 55jC for 30 minutes followed by three washes with xylene.Tissues were rehydrated in a series of ethanol washes and rinsed with PBS.Antigen retrieval was carried out by microwaving the slides at high power in 10 mmol/L citrate buffer (pH 6.0) for 4 minutes.Endogenous peroxidase activity was quenched with 0.3% hydrogen peroxide in PBS for 20 minutes.Nonspecific binding was blocked with normal goat serum for 30 minutes.NF-nB subunit p105/p50 immunoreactivity was studied using the polyclonal rabbit anti-p50 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) at a dilution of 1:500.The antibody was applied for 1 hour at room temperature.The slides were washed with PBS thrice and then incubated with a biotinylated anti-rabbit antibody (Santa Cruz Biotechnology) for 30 minutes.After washing with PBS, the slides were incubated with horseradish peroxidase (HRP)-conjugated streptavidin (Santa Cruz Biotechnology) for 45 minutes.Following the wash with PBS, the signals were developed with 3,3 ¶-diaminobenzidine substrate (Vector Laboratories, Burlington, Ontario, Canada) for 5 minutes and counterstained with hematoxylin.The slides were dehydrated and sealed with coverslips.Negative control was done by omitting the primary p105/p50 antibody.
Evaluation of immunostaining.Due to loss of biopsy cores or insufficient tumor cells present in the cores, 14 cases of normal nevi, 51 cases of dysplastic nevi, 185 cases of primary melanomas, and 55 cases of melanoma metastases could be evaluated for p105/p50 staining.The p105/p50 staining in TMAs was examined blinded by three independent observers (including one dermatopathologist) simultaneously, and a consensus score was reached for each core.The positive reaction of p105/p50 was scored into four grades according to the intensity of the staining: 0, 1+, 2+, and 3+.The percentages of p50/p105 staining were also scored into four categories: 1 (0-25%), 2 (26-50%), 3 (51-75%), and 4 (76-100%).For cases in which multiple biopsy cores were available, 80% of the biopsies had uniform staining.In the cases with a discrepancy between duplicated cores, the average score from the two tissue cores was taken as the final score.The sum of the intensity and percentage scores is used as the final staining score (25).The staining pattern of the biopsies was defined as follows: 1, negative; 2 to 3, weak; 4 to 5, moderate; and 6 to 7, strong.In addition, the percentage of cells showing positive staining in the nucleus was assessed.We classified the nuclear staining as strongly positive if >50% of cells contained NF-nB p105/p50 in the nuclear compartment and weakly positive if V50% of cells show NF-nB p105/p50 in the nucleus.
Statistical analysis of TMA.Statistical analysis was done with the Statistical Package for the Social Sciences version 11.5 software (SPSS, Chicago, IL).The m 2 test was used to compare the quantitative differences of p105/p50 staining in various stages of melanoma progression.The Kaplan-Meier method and log-rank test were used to evaluate the correlations between p105/p50 expression and patient survival.Cox regression model was used for multivariate analysis.A P < 0.05 was considered significant.
Cell culture and transfection.MMRU and Sk-mel-3 human melanoma cell lines were cultured in DMEM supplemented with 10% fetal bovine serum (FBS; Invitrogen, Burlington, Ontario, Canada).All cells were maintained in 5% CO 2 atmosphere at 37jC.Cells were grown to 70% to 80% confluency in six-well plates before plasmids and small interfering RNA (siRNA) transfection.
Cells were transfected with expression vector pCMV or pCMV-hp50 encoding human NF-nB p50 (26) by LipofectAMINE reagent (Qiagen, Mississauga, Ontario, Canada) according to the manufacturer's instructions.Twenty hours after transfection, the medium containing transfection reagents was removed.The cells were rinsed twice with PBS and incubated in fresh medium.For siRNA transfection, cells were incubated with nonspecific control siRNA or NF-nB p105 siRNA (Qiagen) in serum-free medium for 20 hours followed by incubation in the complete medium supplemented with 10% FBS.The cells were harvested at different time points and lysed for Western blot assay.
Wound healing assay.After MMRU and Sk-mel-3 melanoma cells were transfected with plasmids or siRNA, cells were cultured in fresh medium for 24 hours and treated with 10 Ag/mL mitomycin C (Sigma) for 2 hours.After washing with PBS, a standard 200-AL pipette tip was drawn across the center of each well to produce a wound of f0.5-mm in width.The wounded monolayers were washed twice to remove nonadherent cells, and fresh medium was added.The photographs were taken at the same position of the wounds at various time intervals.The starting wound edges were defined in each photo by white lines according to the scratch at 0-hour time point.The numbers of migrating cells across these white lines in the photos were counted to quantitate the rates of migration (27).
RhoA pull-down assay.MMRU melanoma cells were seeded into 100mm plates and cultured to 80% confluency.The cells were transfected with vector pCMV or pCMV-hp50 for 20 hours or without transfection as control.Then, all cells were serum starved for 24 hours.Transfected cells were incubated with medium containing 10% FBS for 3 minutes at 37jC.Cells were washed with 10 mL of ice-cold TBS [50 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl] and lysed in 300 AL lysis buffer [50 mmol/L Tris-HCl (pH 7.2), 150 mmol/L NaCl, 10 mmol/L MgCl 2 , 1% (v/v) Triton X-100, 0.5% (w/v) sodium deoxycholate, 0.1% (w/v) SDS, 10 Ag/mL each of aprotinin and leupeptin, 1 mmol/L phenylmethylsulfonyl fluoride].The cell lysates were cleared by centrifugation at 12,000 Â g for 15 minutes at 4jC. Cell lysates (0.5 mg protein) in 500 AL lysis buffer were mixed with 20 Ag glutathione S-transferase fusion protein immobilized to glutathione-Sepharose 4B beads, and the mixture was rotated for 1 hour at 4jC.The samples were then washed with lysis buffer thrice.Bound proteins were fractionated in 10% SDS-PAGE and immunoblotted with monoclonal mouse anti-RhoA antibody (1:200; Santa Cruz Biotechnology).The total cell lysates were also separated on 10% SDS-PAGE and blotted with anti-RhoA antibody to detect total RhoA expression.
Immunofluorescence.MMRU cells were transfected with vector pCMV or pCMV-hp50 and subcultured onto coverslips in six-well plates.After 5 hours, the cells were serum starved overnight.Then, cells were collected after stimulation with complete medium containing 10% FBS for a desired period.The cells were fixed with 2 mL of fixation solution (2% paraformaldehyde and 0.5% Triton X-100 in PBS) for 30 minutes at 4jC.After washing with PBS, the cells were incubated with normal goat serum for 1 hour followed by anti-p50 primary antibody for 1 hour (Santa Cruz Biotechnology) and Cy2-conjugated goat anti-rabbit secondary antibody (Jackson ImmunoResearch, West Grove, PA) for 1 hour.The cells were then stained with phalloidin-rhodamine (1 unit/coverslip; Invitrogen) for 30 minutes.Finally, the coverslips were incubated with 1:3,000 dilution of stock Hoechst 33258 (20 mmol/L) for 10 minutes and the cells were visualized under a fluorescent microscope.Photos were taken with a cooled mono 12-bit Retiga-Ex camera equipped with Northern Eclipse imaging software.
To quantify the intensity of F-actin, 10 images per slide were taken under a fluorescent microscope.Each image contained an average of three to five cells.Images were analyzed using ImageJ software (NIH, Bethesda, MD), and the mean of relative cellular fluorescent intensity was measured (28).Data were presented from three independent experiments.

Results
Clinicopathologic features of TMA.For 185 cases of primary melanoma that NF-nB p105/p50 staining was available, there were 106 male and 79 female.Of 55 cases of metastatic melanoma that were informative, there were 38 male and 17 female.The mean average age of melanoma patients was 57 years, ranging from ages 21 to 92.Breslow thickness, which is the distance measured from the top layer of epidermis to the deepest point of tumor penetration, was used as the criteria for melanoma staging (29).Among 185 cases of primary melanoma, 55 were V1 mm, 81 were 1.1 to 2.0 mm, 23 were 2.1 to 4.0 mm, and 26 were >4.0 mm.Superficial spreading melanoma accounted for 79 cases, lentigo maligna melanoma accounted for 25 cases, and the remaining 81 cases consisted of nodular melanoma, acrolentigous melanoma, and desmoplastic melanoma.The majority of the primary melanomas were isolated from sun-protected sites (146 cases; trunk, arm, leg, and feet), and 59 cases were obtained from sunexposed sites (head and neck).Tumor ulceration was observed in 28 cases (Supplementary Table S1).
Strong nuclear expression of NF-KB p105/p50 is inversely correlated with 5-year survival of patients with tumor thickness >2 mm.Because tumor thickness is an important prognostic marker for melanoma, we divided the patients with primary melanoma into two subgroups: (a) low-risk group, thickness V2.0 mm and (b) high-risk group, thickness >2.0 mm.To evaluate whether nuclear expression of NF-nB p105/p50 in human melanoma is correlated with a worse prognosis, Kaplan-Meier survival curves were plotted.We found that whereas nuclear Figure 2. Expression of NF-nB p105/p50 and its nuclear localization in normal nevi, dysplastic nevi, primary melanoma, and metastatic melanoma.A, overall NF-nB p105/p50 expression is significantly increased in dysplastic nevi (DN ), primary melanoma (PM ), and metastatic melanoma (MM) compared with normal nevi (NN ).B, significant differences of nuclear staining of NF-nB p105/p50 were observed between normal nevi and dysplastic nevi (P = 0.035, m 2 test), between dysplastic nevi and primary melanoma (P = 0.023, m 2 test), and between primary melanoma and metastatic melanoma (P = 0.008, m 2 test).
NF-nB p105/p50 expression did not correlate with 5-year survival of patients with tumor thickness V2.0 mm (P = 0.82, log-rank test; Fig. 3A), strong NF-nB p105/p50 nuclear staining is significantly correlated with a poorer disease-specific 5-year survival of patients with melanomas >2.0 mm (P = 0.014, log-rank test; Fig. 3B).To further examine whether strong NF-nB p105/p50 nuclear staining is an independent prognostic marker for this subgroup, we did a multivariate analysis, including p105/p50 nuclear staining, age, sex, tumor thickness, location, and ulceration.Our result indicated that NF-nB p105/p50 nuclear staining reached a remarkable significance for predicting the patient outcome independent of other clinicopathologic variables for 5-year disease-specific survival (Fig. 3D).We also evaluated the prognostic value of nuclear NF-nB expression in metastatic melanomas but did not find a significant correlation (P = 0.97, log-rank test; Fig. 3C).
NF-KB p50 promotes melanoma cell migration.Because cell migration is one of the central steps of cancer metastasis that greatly shortens the survival of melanoma patients, we investigated the involvement of NF-nB p50 in melanoma cell migration.We first transiently transfected MMRU and Sk-mel-3 melanoma cells with pCMV-hp-p50 expression vector or control pCMV vector.Cells were harvested at 24, 48, and 72 hours after transfection, and extracts were analyzed by Western blot.The maximal level of NF-nB p50 expression was observed at 24 hours after transfection, and the increased level of NF-nB p50 lasted up to 72 hours in MMRU melanoma cells and 48 hours in Sk-mel-3 melanoma cells (Fig. 4A), respectively.Wound healing assay was done after the transfected cells were incubated with 10 Ag/mL mitomycin C for 2 hours, which inhibits cell division, so that the difference in cell motility was not due to the differences in cell proliferation.MMRU and Sk-mel-3 melanoma cells transfected with NF-nB p50 almost healed the wound 12 or 24 hours after scratch, whereas the cells transfected with vector were unable to heal the wound at the same time point (Fig. 4C).Overexpression of NF-nB p50 resulted in 2-fold increase of migrated cells into wound area compared with vector controls (Fig. 4D).
Figure 3. Prognostic value of NF-nB p105/p50 nuclear expression in subgroups of melanoma patients.A, disease-specific 5-year survival of patients with melanoma thickness V2.0 mm (P = 0.82, log-rank test).B, disease-specific 5-year survival of patients with melanoma thickness >2.0 mm (P = 0.014, log-rank test).C, disease-specific 5-year survival of patients with metastatic melanomas (P = 0.97, log-rank test).D, multivariate Cox regression analysis of NF-nB p105/p50 nuclear expression in patients with melanoma thickness >2.0 mm.Coding of variables: NF-nB p105/p50 nuclear staining was coded as 1 (weak expression) and 2 (strong expression).Thickness was coded as 1 (>4.0 mm) and 2 (V4.0 mm).Ulceration was coded as 1 (absent) and 2 (present).Location was coded as 1 (extremities and trunk) and 2 (head and neck).Age was coded as 1 (z57 years) and 2 (<57 years).Sex was coded as 1 (male) and 2 (female).A, MMRU and Sk-mel-3 melanoma cells were transfected with NF-nB p50 or control vector.Extracts of the cells were prepared at 24, 48, and 72 hours after transfection and analyzed for the expression of NF-nB p50 by Western blot analysis.B, wound healing assay was done on monolayers of MMRU (top ) and Sk-mel-3 (bottom ) melanoma cells after 24 hours of transfection.The photographs of MMRU and Sk-mel-3 melanoma cells were taken at 12 and 24 hours after wounds were made.C, quantitation of (B).Columns, mean of triplicates; bars, SD.Similar results were observed in three independent experiments.Magnification, Â100.
Silencing of NF-KB p50 inhibits melanoma cell migration.To test whether a physiologic level of NF-nB p50 plays a role in melanoma cell migration, we used siRNA to knockdown NF-nB p105, the precursor of NF-nB p50 in MMRU melanoma cells.Figure 5A depicts the efficiency of the siRNA transfection as measured by Western blot analysis.Accompanied with >90% downregulation of NF-nB p105 expression level, the level of NF-nB p50 also gradually decreased from 24 to 72 hours after transfection.Wound healing assay was done 24 hours after transfection.Migration ability of cells transfected with NF-nB p105 siRNA was significantly reduced, whereas cells transfected with control siRNA almost closed the wound 20 hours after the wounds were introduced (Fig. 5B).NF-nB p105 siRNA reduced the number of migrated cells by 2-fold compared with the control siRNA (Fig. 5C).
NF-KB p50 enhances RhoA activity and stress fiber formation via Rock.It is well established that RhoA-Rock pathway plays an important role in cell migration through reorganization of stress fiber (30).To address whether changes of NF-nB p50-transfected cell migration were related to in vivo level of active RhoA, RhoA pull-down assay was done.Cells were transfected with pCMV-hp50 or pCMV vector or without transfection as control.Then, all the cells were serum starved for 24 hours and RhoA activity was induced by serum stimulation of transfected cells for 3 minutes.Figure 6A shows that RhoA activity is undetectable or very low in untransfected or pCMV vectortransfected control cells, whereas the RhoA activity is greatly induced in cells transfected with NF-nB p50.Moreover, seruminduced formation of stress fiber is more excessive in MMRU cells transfected with NF-nB p50 than those with vector control after 30 minutes of serum stimulation, whereas Rock inhibitor Y27632 effectively abrogated the regulatory function of NF-nB p50 in stress Figure 5. of NF-nB p50 inhibits melanoma cell migration.A, MMRU melanoma cells were transfected with NF-nB p105 siRNA or control siRNA.Western blot assay was done to measure the expression level of NF-nB p105/p50 and actin at 24, 48, and 72 hours after transfection.B, the monolayer of MMRU melanoma cells was scratched 24 hours after transfection and photographed 20 hours later.Magnification, Â100.C, quantitation of (B).B, NF-nB p50 enhances stress fiber formation mediated by Rock.Cells were transfected with pCMV vector or pCMV-hp50, respectively, followed by serum starvation overnight and serum stimulation for 30 minutes.For Rock inhibitor treatment, cells were pretreated with serum-free medium containing 10 Amol/L Y27632 for 2 hours after transfection with pCMV vector or pCMV-hp50 and serum starvation overnight and then incubated with complete medium containing 10% FBS and 10 Amol/L Y27632 for 30 minutes.Magnification, Â400.C, quantitation of (B ).
fiber formation (Fig. 6B).The relative fluorescence intensity of F-actin in cells transfected with NF-nB p50 was 2-fold higher than that in cells transfected with vector, whereas there is no significant difference between cells transfected with NF-nB p50 or vector after treatment with Rock inhibitor Y27632 (Fig. 6C).

Discussion
In the present study, we sought to determine the role of NF-nB p105/p50 in melanoma pathogenesis.Initially, we used TMA technology and immunohistochemistry to investigate the expression level of NF-nB p105/p50 in various stages of melanocytic lesions.Expression level of NF-nB subunit RelA has been well studied in melanoma.NF-nB subunit RelA and Ser 529 -phosphorylated RelA are overexpressed in melanoma biopsies (31,32).However, little is known of the expression level of NF-nB p105/p50 in melanoma.Our data showed that the expression of NF-nB p105/p50 was significantly increased in dysplastic nevi, primary melanoma, and metastatic melanoma comparing with normal nevi (Fig. 2A).In addition, we assessed the nuclear staining of NF-nB p105/p50 in melanoma biopsies because nuclear translocation of NF-nB subunits is one of critical steps for transcriptional activation of NF-nB.We have, for the first time, found that strong nuclear staining of NF-nB p105/p50 correlated with the melanoma progression (Fig. 2B).
The progression of melanoma requires multiple genetic alterations.For example, we previously found that phosphorylated Akt expression is increased in melanoma biopsies (25).In addition, BRAF is mutated in f60% to 70% of superficial spreading melanomas (33).The expression of the small GTPase RhoC is related to metastatic phenotype of melanomas (34).In the present study, we found that nuclear staining of NF-nB p105/p50 is correlated with disease-specific 5-year survival of patients with tumor thickness >2.0 mm, and nuclear expression of NF-nB p105/ p50 is an independent prognostic factor for patients in this subgroup (Fig. 3B and D).Our data suggest that increased nuclear expression of NF-nB p105/p50 is an important event during melanoma progression, possibly involved in the vertical growth phase.
Recent reports showed that NF-nB subunit p50 might be more important than other NF-nB subunits in the pathogenesis of classic Hodgkin's lymphoma (35), nasopharyngeal carcinoma (36), and cervical carcinoma (37).In addition, the expression level of InBa, which retains NF-nB subunits in the cytoplasm and inhibits their nuclear was higher in nevi than in melanoma biopsies (31).Therefore, we speculate that increased expression of NF-nB RelA and p105/p50 occurs during melanoma development, but their transcriptional potential is inhibited by InBa.During melanoma progression, the expression of InBa is decreased, resulting in nuclear accumulation of NF-nB dimers and enhanced transcriptional activity.
Cell migration is one of the critical steps for tumor invasion and progression (38).Our in vitro data by overexpressing NF-nB p50 (Fig. 4) and treatment with p105 siRNA (Fig. 5) clearly showed that this subunit is crucial for the motility of melanoma cells.The process of cell migration is related to the network of various types of extracellular growth factors, chemokines, transmembrane receptors, and intracellular factors (39)(40)(41).Because nuclear NF-nB can lead to enhanced gene transcription, its role in melanoma cell migration is most likely associated with the activation of its downstream targets.It is well known that Rho GTPase family proteins are essential elements to regulate cell shape, polarity, and locomotion (42).Rho proteins are activated when bound to GTP and inactivated on hydrolysis of GTP to GDP (43).Three main families regulate their activity: guanine nucleotide exchange factors that catalyze exchange of GDP to GTP, GTPaseactivating proteins that stimulate hydrolyze GTP to GDP, and guanine nucleotide dissociation inhibitors that seem to block the spontaneous activation (44)(45)(46).RhoA, Rac1, and Cdc42 are three well-studied members of Rho family, all of which are pivotal regulators of reorganization of actin during cell migration (47,48).Therefore, it is possible that NF-nB p50 may activate RhoA to induce cell migration through regulation of actin organization.The results from RhoA pull-down assay indeed indicate that RhoA activity is dramatically enhanced in cells transfected with NF-nB p50 compared with vector control (Fig. 6A).We also investigated the effect of NF-nB p50 on the activity of Rac1 and Cdc42, but no significant difference in Rac1 or Cdc42 activity was found between melanoma cells transfected with NF-nB p50 and vector control (data no shown).Furthermore, our findings provided evidence that overexpression of NF-nB p50 resulted in enhanced stress fiber formation, whereas Rock inhibitor Y27632 can inhibit this effect (Fig. 6B and C), which confirmed the regulatory role of NF-nB p50 in RhoA-Rock pathway.
On the other hand, previous study suggested that NF-nB could be activated by RhoA, Cdc42, and Rac1 (49).These Rho proteins induced phosphorylation of InB and translocation of p50/p50 and p50/p65 dimers to the nucleus.Therefore, we speculate that NF-nB p50 serves as a positive feedback factor to RhoA activity.
In summary, we showed that constitutive activity of NF-nB p105/ p50 plays an important role in human melanoma pathogenesis.Elevated activity of NF-nB p105/p50 may facilitate tumor progression by enhancing cell migration and RhoA activity as well as stress fiber formation via Rock.Our data indicate that NF-nB p105/p50 may serve as an important prognostic marker and a promising therapeutic target for melanoma.

Figure 4 .
Figure 4. Overexpression of NF-nB p50 promotes melanoma cell migration.A, MMRU and Sk-mel-3 melanoma cells were transfected with NF-nB p50 or control vector.Extracts of the cells were prepared at 24, 48, and 72 hours after transfection and analyzed for the expression of NF-nB p50 by Western blot analysis.B, wound healing assay was done on monolayers of MMRU (top ) and Sk-mel-3 (bottom ) melanoma cells after 24 hours of transfection.The photographs of MMRU and Sk-mel-3 melanoma cells were taken at 12 and 24 hours after wounds were made.C, quantitation of (B).Columns, mean of triplicates; bars, SD.Similar results were observed in three independent experiments.Magnification, Â100.

Figure 6 .
Figure 6.NF-nB p50 enhances RhoA activity and stress fiber formation via Rock.A, NF-nB p50 enhances RhoA activity determined by RhoA pull-down assay.Lane 1, MMRU cells without transfection were serum starved for 24 hours; lanes 2 and 3, MMRU cells were transfected with pCMV vector or pCMV-hp50, respectively, followed by serum starvation for 24 hours and serum stimulation for 3 minutes.MMRU melanoma cells transfected with NF-nB p50 display enhanced activity of RhoA comparing with cells transfected with vector.B, NF-nB p50 enhances stress fiber formation mediated by Rock.Cells were transfected with pCMV vector or pCMV-hp50, respectively, followed by serum starvation overnight and serum stimulation for 30 minutes.For Rock inhibitor treatment, cells were pretreated with serum-free medium containing 10 Amol/L Y27632 for 2 hours after transfection with pCMV vector or pCMV-hp50 and serum starvation overnight and then incubated with complete medium containing 10% FBS and 10 Amol/L Y27632 for 30 minutes.Magnification, Â400.C, quantitation of (B ).