RASSF1A gene, found at the 3p21.3 locus, is a tumor suppressor gene frequently hypermethylated in human cancers. In this study, we report that compared with melanocytes in normal choroid, RASSF1A is downregulated in uveal melanoma samples and in uveal melanoma cell lines. LOH at 3p21.3 was detected in 50% of uveal melanoma. Moreover, methylation of the RASSF1A promoter was detected in 35 of 42 tumors (83%) and RASSF1A was also weakly expressed at the mRNA level. These data indicate that LOH at the RASSF1A locus or RASSF1A promoter methylation may partly account for the suppression of RASSF1A expression observed in uveal melanoma. Furthermore, following ectopic expression in three RASSF1A-deficient melanoma cell lines (OCM-1, Mel270, and 92.1), RASSF1A weakly reduces cell proliferation and anchorage-independent growth of uveal melanoma cells without effect on ERK1/2 activation, cyclin D1 and p27Kip1 expression. This study explored biological functions and underlying mechanisms of RASSF1A in the ERK1/2 pathway in normal uveal melanocytes. We showed that siRNA-mediated depletion of RASSF1A increased ERK1/2 activation, cyclin D1 expression, and also decreased p27Kip1 expression in normal uveal melanocytes. Moreover, that the depletion of RASSF1A induced senescence-associated β-galactosidase activity and increased p21Cip1 expression suggests that RASSF1A plays a role in the escape of cellular senescence in normal uveal melanocytes. Interestingly, we found that RASSF1A was epigenetically inactivated in long-term culture of uveal melanocytes. Taken together, these data show that depletion of RASSF1A could be an early event observed during senescence of normal uveal melanocytes and that additional alterations are acquired during malignant transformation to uveal melanoma. Mol Cancer Res; 9(9); 1187–98. ©2011 AACR.

Two of the best-studied Ras effectors, which interact with their Ras-binding domains (RBD), are Raf and phosphatidylinositol 3-kinase (PI3K). Raf controls the MEK/ERK pathway and PI3K activity is required for activation of the protein kinase B, Akt. There is another group of Ras effectors, which also share a Ras Association (RA) domain (i.e., Ras-association domain family (RASSF; ref. 1). The biological role of each members of the RASSF group still remains unclear. Unlike classic Ras effectors such as Raf and PI3K, which are oncoproteins, there is evidence suggesting that members of the RASSF serve as tumor suppressor genes by modulating some of the growth inhibitory responses mediated by Ras (1–3). Tumor suppressor RASSF1A (RAS association domain family 1 isoform A) gene is located on chromosome 3 locus p21.3. Loss of RASSF1A gene expression is a frequent event in human cancer. More precisely, aberrant promoter methylation has been reported in at least 40 types of sporadic cancers, including those in the lung, head, and neck (1, 4). It has been shown that RASSF1A plays an important role in cell-cycle regulation, apoptosis, and microtubule stability (3, 5, 6). Indeed, ectopic expression of RASSF1A reduced the growth rate of human cancer cells, supporting a role for RASSF1A as a tumor suppressor gene (4, 7, 8). Furthermore, the knockout mouse model for RASSF1A shows an increased tendency to develop tumors (9). The mechanisms by which RASSF1A exerts its tumor suppression activities or the involved pathways are not yet fully understood.

In skin melanoma, it has been clearly indicated that BRAF and NRAS mutations are involved in the activation of the Raf/MEK/ERK pathway (10). Therefore, epigenetic inactivation of RASSF1A has been found in skin melanoma with or without NRAS or BRAF mutation and this, in turn, suggests that RASSF1A inactivation does not act as an alternative mechanism to oncogenic BRAF or RAS mutation (10).

Despite a common embryologic origin and similar histologic features, cutaneous and uveal melanoma cells differ in epidemiologic and cytogenetic aspects. Loss of 1 copy of chromosome 3 (monosomy 3) is strongly associated with poor survival and metastatic disease in uveal melanoma, suggesting the presence of tumor suppressor gene on chromosome 3 (11, 12). It remains unclear which genetic or environmental predisposing factors and cellular events are involved in the malignant transformation of normal uveal melanocytes to form uveal melanoma. The constitutive activation of ERK1/2, the downstream kinase of the mitogen-activated protein kinase pathway, has been observed in both primary uveal melanoma tumors and uveal melanoma cell lines without either RAS or BRAF mutations (13–15). Our previous findings showed the key role of the WTB-Raf/MEK/ERK pathway in the control of proliferation of uveal melanoma cells (16). To date, a somatic mutation in the heterotrimeric G protein α-subunit, GNAQ, is described in ERK1/2 activation, this being found in 46% uveal melanoma patients (17). Recently, a study reported other somatic mutation activating the ERK1/2 pathway in a closely related gene, GNA11 in 32% of GNAQWT uveal melanoma (18). Hypermethylation of the RASSF1A promoter reported in uveal melanoma has been suggested to be common epigenetic event in uveal melanoma development (19, 20). However, these studies did not investigate the relationship between loss of RASSF1A expression, monosomy 3, and RASSF1A promoter methylation. In addition, the exact implication of RASSF1A in uveal melanoma tumorigenesis remains to be characterized.

Bypassing cellular senescence and becoming immortal is a prerequisite step in the tumorigenesis (19). Cutaneous melanocytes expressing somatic mutation in BRAF and various primary cell types, which show constitutive activation of the ERK1/2 pathway, display classical hallmarks of senescence, suggesting that oncogene-induced senescence is a tumor-suppressive mechanism (22–24). Therefore, there must be additional mutations and/or epigenetic alterations required for immortalization to occur. Recent evidence suggests that early-stage cancer cells carry epigenetic modifications in growth- and differentiation-associated genes; these may predispose cells to the accumulation of additional changes in oncogenes and tumor suppressor genes and, ultimately, lead to the development of cancer (25).

In this study, we report the relationship between RASSF1A promoter hypermethylation, level of RASSF1A expression and chromosome 3 status by using frozen primary uveal melanoma tumors. We have investigated the consequences of ectopic expression of RASSF1A on the activation of ERK1/2 pathway and the proliferation of RASSF1A-deficient uveal melanoma cell lines. We show that one hallmark of transformation, anchorage-independent growth in soft agar, is independent of the ectopic expression of RASSF1A in uveal melanoma cell lines. In addition, we established pure culture of normal uveal melanocytes for studying the involvement of RASSF1A in the absence of other genetic alterations acquired during the malignant transformation to uveal melanoma. For the first time, by using siRNA-mediated depletion of endogenous RASSF1A expression in normal uveal melanocytes, we showed the functional implications of RASSF1A in uveal melanoma. We showed that both constitutive activation of ERK1/2 and cell-cycle proteins expression are linked to the presence of RASSF1A protein in normal uveal melanocytes.

Tumor samples and DNA extraction

The LOH analysis was done on DNA isolated from 9 uveal melanoma tumors for which matched adjacent normal scleral/choroidal tissue was available. Genomic DNA of 42 uveal melanoma tumors was isolated by using an easy-spin genomic DNA extraction kit according to the manufacturer's instructions (Macherey-Nagel).

Cell cultures and treatment

Normal uveal melanocytes were isolated, as previously described (26), from human enucleated eyes provided by the Faculty of Medicine of Caen, France. Normal uveal melanocytes were cultured in FIC medium (F-12 Ham's medium supplemented with 10% fetal calf serum (FCS), 2.5 μg/mL fungizone/amphotericin B, 2 mmol/L l-glutamine (Invitrogen), 10 ng/mL cholera toxin (Sigma-Aldrich), 0.1 mmol/L isobutylmethylxanthine (Sigma-Aldrich), and 5 ng/mL FGF2 (Sigma-Aldrich). Seven cell lines derived from primary human uveal melanoma were also cultured as previously described (16). These were 92.1 and Mel270 (both provided by Dr. M. Jager, University of Leiden, the Netherlands), OCM-1 (provided by Dr. F. Malecaze, CHU Toulouse, France), Mum2B and TP17 (provided by Dr. S. Guerin, CUO/LOEX université Laval Quebec, Canada), μ2 (human uveal melanoma created at the IBCP, Lyon, France), and μ2F, which when implanted in mice, causes hepatic metastases (provided by Dr.L. Baggetto, Lyon, France). The OMM1.3 metastasis cell line was established from the same patient as Mel270 cell line (27). These cells were grown in RPMI 1640 medium supplemented with 5% FCS, 2.5 μg/mL fungizone/amphotericin B, 50 μg/mL gentamycin, and 2 mmol/L l-glutamine (Gibco/BRL; Invitrogen). Cells were seeded in triplicate in 6-well plates at a density of 20,000 cells per well. The plates were incubated for 3 days and then treated with 5-aza-2′-deoxycytidine (5-aza-CdR, 15 μmol/L; Sigma) for 3 and 6 days. Cell proliferation was assessed by counting the cells and by using the MTT colorimetric assay.

FISH analysis

To detect copy number of chromosome 3 in cell lines, we conducted FISH experiments on metaphasic cell lines by using chromosome 3 classical satellite probe (alpha Satellite Red direct labeled; MB Biomedical). Hybridization with probe was done according to the manufacturer's instructions.

LOH analysis

LOH analysis was carried out by using DNA sequencing in uveal melanoma and adjacent normal scleral/choroidal tissue with polymorphic microsatellite markers located in the 3p21.3 region (D3S1478, D3S1578 obtained from the genome Database: www.ncbi.nlm.nih.gov.gate2.inist.fr). PCR amplifications using 5′-fluorescent dye–labeled (dye D3; Beckman) primers for microsatellite loci (oligo synthesis service; Promega) with tumor and normal template DNA were done with the HOT Start Taq DNA polymerase kit (Qiagen). Three μl of the DNA sample was combined with 30 μl of CEQ Sample Loading Solution (Beckman Coulter), and 0.20 μl of a 400-bp CEQ DNA Size Standard (Beckman Coulter). The samples were separated via capillary electrophoresis on an automated 8-capillary CEQ 8000 DNA sequencer (Beckman Coulter) after a denaturation for 2 minutes at 95°C. LOH was analyzed by using GeneMarker (Softgenetics). LOH was considered present when the peak allele signal from tumor DNA was less than 50% of that of the normal tissue counterpart.

Methylation-specific PCR

One microgram of genomic DNA was modified by using an EpiTect bisulfite kit (Qiagen). Modified DNA was amplified by using primers specific for the unmethylated (forward 5′-GGTTTTGTGAGAGTGTGTTTAG-3′, reverse 5′-CACTAACAAACACAAACCAAAC-3′) and methylated sequences (forward 5′-GGGTTTTGCGAGAGCGCG-3′, reverse 5′-GCTAACAAACGCGAACCG-3′). PCR amplifications were done by using the HOT Start Taq kit (Qiagen). For control, DNA from normal uveal melanocytes and lymphocytes, either treated or untreated in vitro with SssI methylase (Biolabs), were used. PCR products (size 169 bp) was separated on 2% agarose gels, stained with ethidium bromide, and visualized under UV illumination.

Mutation screening

We analyzed exon 15 of BRAF, exons 1 and 2 of NRAS by using specific PCR, as previously described (16). Exons 4 and 5 of GNAQ and GNA11 gene were analyzed as previously described (17, 18).

Quantitative real-time PCR

Total RNA was isolated from uveal melanoma cells lines and from uveal biopsies with Nucleospin RNA II kit according to the manufacturer's protocol (Macherey-Nagel). The quality of isolated RNA was analyzed on an electrophoresis agarose gel. One microgram of RNA from sample was reverse transcribed by using the SuperScript First-Strand Synthesis System for qPCR kit (Invitrogen). The PCR was conducted according to the manufacturer's protocol by using the Absolute QPCR SYBR Green Mixes (ABgene). Assays were run in triplicate on the iCycler iQ real-time PCR detection system (Biorad). The following primers were used: RASSF1A, forward (5′-CCTCTGTGGCGACTTCATCTG-3′) and reverse (5′-CAACAGTCCAGGCAGACGAG-3′; 108 bp) and β-actin, forward (5′-ACCTCATGAAGATCCTCACCGA-3′) and reverse (5′-CTTAATGTCACGCACGATTTCCC-3′; 77 bp) was used as an internal control. The amount of target was given by the formula, 2−(CtGene- Ctβ-actine) × 1,000, in which Ct is the threshold cycle value.

Ectopic RASSF1A expression and RASSF1A gene silencing

pcDNA-FLAG RASSF1A and pcDNA3.1 vectors were a generous gift from Professor G.J. Clark (J.G. Brown Cancer Center, Louisville; ref. 28). Cells were seeded at a density of 6 × 104 cells per well in a 6-well plate in RPMI culture medium supplemented with 5% FCS. After 24 hours, cells were transfected with 300 ng of pcDNA-FLAG RASSF1A, or pcDNA3.1 empty vectors, in the presence of 2 μg/mL Lipofectamine 2000 reagent (Invitrogen) for 4 hours. The culture medium was then replaced and cells were cultured in fresh growth medium supplemented with 600 ng/mL G418 for 15 days. The culture medium was replaced every 3 days. The double-stranded siRNA oligonucleotide targeting RASSF1A, published previously, was synthesized by Qiagen (29). As a control, a nonspecific siRNA duplex containing the same nucleotides but in irregular sequence (scrambled) was used. Transfection with siRNA was previously described (30).

Western blot analysis

Proteins were obtained from cells and were separated by SDS-PAGE and were transferred to polyvinylidene difluoride membrane (PerkinElmer-NEN), as previously described (16). Membranes were probed with monoclonal antibody directed against RASSF1A (Abcam), polyclonal antibodies against ERK1/2, phospho-ERK1/2 (T202/Y204), cyclin D1, p27Kip1 (1:1,000; Cell Signaling Technology), and goat antibody directed against actin (0.8 μg/mL; Sigma) to control for equal loading. The primary antibodies were tagged with specific peroxidase–conjugated secondary antibodies. Antibody complexes were detected by enhanced chemiluminescence (PerkinElmer-NEN).

Cell cycle progression analysis

We analyzed cell cycle progression by determining the DNA content of the cells with propidium iodide (PI). Cells were treated according to the manufacturer's instructions (Roche Applied Science) and were analyzed by flow cytometry (Epics XL; Beckman Coulter).

Senescence-associated β-galactosidase activity

Senescence was investigated with the senescence-associated β-galactosidase (SA β-gal) staining kit (Cell Signaling), according to the manufacturer's instructions.

Statistics

Values are given as mean ± SD of data from 3 independent experiments conducted in triplicate. The Student's t test was used for statistical analysis. The value of P < 0.05 was considered significant.

Downexpression of RASSF1A is frequent in primary uveal melanoma

We examined 12 uveal melanoma tumors for which the chromosome 3 status was analyzed by FISH and multiplex ligation–dependent probe amplification (12): 6 had chromosome 3 monosomy and 6 had disomy. To address whether the genomic imbalance of the RASSF1A locus could also contribute to the decrease of RASSF1A expression, LOH was investigated in 9 pairs of uveal melanoma tumors for which adjacent normal scleral/choroidal tissues were available and with known chromosome 3 status was determined. Among the 9 uveal melanoma tumors analyzed, LOH using 2 polymorphic microsatellite markers (D3S1478 and D3S1578) was identified in 4 tumors, and 1 uveal melanoma showed microsatellite instability (Table 1).

Table 1.

Correlation between expression of RASSF1A and its promoter methylation status in uveal melanoma tumors

PatientsChromosome 3 statusDNA methylation RASSF1A promotor statusRelative RASSF1A mRNA expression
UMU/M
Healthy Disomy +   1 
Disomy  +  0 
Disomy  +  0 
Disomy   + 0.4 
Disomy  +  0.3 
Disomy  +  0 
Disomy  +  0 
Monosomy  +  0 
Monosomy  +  0 
Monosomy  +  0 
10 Monosomy  +  0 
11 Monosomy  +  0 
12 Monosomy  +  0 
13 LOH−   + 0.2 
14 LOH−   + 0.9 
15 LOH−   + 0.6 
16 LOH−   + 0.3 
17 LOH + (D3S1478)  +  0.3 
18 LOH +(D3S1478)  +  0.1 
19 LOH + (D3S1478)  +  0 
20 LOH + (D3S1478) +   0.6 
21 Microsatellite alterations +   0 
PatientsChromosome 3 statusDNA methylation RASSF1A promotor statusRelative RASSF1A mRNA expression
UMU/M
Healthy Disomy +   1 
Disomy  +  0 
Disomy  +  0 
Disomy   + 0.4 
Disomy  +  0.3 
Disomy  +  0 
Disomy  +  0 
Monosomy  +  0 
Monosomy  +  0 
Monosomy  +  0 
10 Monosomy  +  0 
11 Monosomy  +  0 
12 Monosomy  +  0 
13 LOH−   + 0.2 
14 LOH−   + 0.9 
15 LOH−   + 0.6 
16 LOH−   + 0.3 
17 LOH + (D3S1478)  +  0.3 
18 LOH +(D3S1478)  +  0.1 
19 LOH + (D3S1478)  +  0 
20 LOH + (D3S1478) +   0.6 
21 Microsatellite alterations +   0 

NOTE: Chromosome 3 status was determined by FISH in healthy tissue from human enucleated eyes and in tumors of patients 1 to 12. The microsatellite markers, D3S1478 and D3S1578, located on chromosome 3p21.3 and flanking the RASSF1A locus, were used to analyze the LOH on 9 uveal melanoma tumors (13–21). The methylation of RASSF1A promoter was analyzed by MSP on PCR products with specific primers from bisulfite-treated genomic DNA. The expression level of RASSF1A was analyzed by qPCR on total RNA available.

Abbreviations: LOH−, no deleted locus; LOH+, deleted 3p21.3 locus; M, Methylated; U, Unmethylated.

To quantify the DNA methylation status of RASSF1A promoter, we conducted methylation-specific PCR (MSP) on bisulfite-treated genomic DNA of 21 uveal melanoma tumors for which the cytogenetic analysis had been previously reported. A methylated RASSF1A promoter was detected in 19 of 21 (90%) of uveal melanoma (Table 1). Also examined were 21 other primary uveal melanoma for which the chromosome 3 status was not unknown: a methylated RASSF1A promoter could be detected in 16 of 21 (76%) of uveal melanoma analyzed. In total, 42 primary uveal melanoma were examined and of these 83% (35/42) showed a methylated RASSF1A promoter (data not shown).

The mRNA level of RASSF1A in primary uveal melanoma tumors was determined by quantitative real-time PCR. RASSF1A mRNA was weakly expressed in frozen primary uveal melanoma tumors compared with healthy tissue from human enucleated eyes (Table 1). Such a decrease in RASSF1A expression in uveal melanoma can be explained either by the loss of one copy of RASSF1A (as a result of monosomy 3 or LOH at the RASSF1A locus) or by the promoter methylation on the other allele.

Downexpression of RASSF1A in uveal melanoma cell lines

To evaluate the transcriptional expression of RASSF1A in uveal melanoma cell lines, RASSF1A mRNA levels were analyzed by quantitative PCR (qPCR) in normal uveal melanocytes, primary uveal melanoma cell lines (OCM-1, Mel270, μ2, Mum2B, 92.1, TP17, and μ2F) and metastatic uveal melanoma cell line (OMM1.3). We first confirmed by FISH that loss of one copy of chromosome 3 was not detected in any cell lines (data not shown). RASSF1A expression was reduced in 4 of 7 primary and in metastatic uveal melanoma cell line when compared with normal uveal melanocytes (Fig. 1A). We also determined RASSF1A promoter methylation status in uveal melanoma cell lines and normal uveal melanocytes. The normal uveal melanocytes showed an unmethylated RASSF1A promoter (Fig. 1B). In contrast, 4 of 7 of primary uveal melanoma cell lines (OCM-1, Mel270, Mum2B, and 92.1) and metastatic uveal melanoma cell lines (OMM1.3) had a methylated RASSF1A promoter (Fig. 1B). The RASSF1A promoter was methylated in all cell lines expressing a low level of RASSF1A (Fig. 1A and B). Interestingly, the TP17, in which the RASSF1A promoter is unmethylated, expressed strong levels of RASSF1A when compared with methylated RASSF1A promoter cell lines (Fig. 1A and B). Moreover, the metastatic uveal melanoma cell line OMM1.3, showed methylated RASSF1A promoters, similar to its corresponding primary cell line (Mel270). Because a decrease in RASSF1A transcript levels was observed in both primary and metastatic tumors, it indicates that this epigenetic modification is not restricted to metastatic disease.

Figure 1.

Correlation between RASSF1A expression, RASSF1A promoter methylation status, and presence of mutation in NRAS, BRAF, and GNAQ genes in uveal melanoma cell lines. A, relative expression of RASSF1A between normal uveal melanocytes, primary (OCM-1, 92.1, Mel270, Mum2B, μ2, TP17, and μ2F), and metastatic uveal melanoma cells lines (OMM1.3), columns, mean of triplicate experiments; bars, SD. B, DNA methylation status of RASSF1A promoter was assessed in cell lines by MSP. U: unmethylated promoter and M: methylated promoter. DNA treated with SssI methylase was used as a positive control (C+) and DNA from lymphocytes was used as negative control (C−). NUM, normal uveal melanocytes; UM, uveal melanoma.

Figure 1.

Correlation between RASSF1A expression, RASSF1A promoter methylation status, and presence of mutation in NRAS, BRAF, and GNAQ genes in uveal melanoma cell lines. A, relative expression of RASSF1A between normal uveal melanocytes, primary (OCM-1, 92.1, Mel270, Mum2B, μ2, TP17, and μ2F), and metastatic uveal melanoma cells lines (OMM1.3), columns, mean of triplicate experiments; bars, SD. B, DNA methylation status of RASSF1A promoter was assessed in cell lines by MSP. U: unmethylated promoter and M: methylated promoter. DNA treated with SssI methylase was used as a positive control (C+) and DNA from lymphocytes was used as negative control (C−). NUM, normal uveal melanocytes; UM, uveal melanoma.

Close modal

Analysis of the mutational status of NRAS and BRAF did not show any correlation between methylation of the RASSF1A promoter and an activating mutation in NRAS and BRAF (Table 2). We detected a single-base substitution in exon 5 of the heterotrimeric G protein α-subunit (GNAQ) of Mel270, μ2, and OMM1.3 cell lines (Table 2). The site R183 in exon 4 of GNAQ was not mutated in uveal melanoma cell lines. Analysis of the exons 4 and 5 of GNA11 gene showed no mutation in uveal melanoma cell lines (Table 2). Surprisingly, the Q209P mutation is not found in μ2F, the cell line that causes hepatic metastases when implanted in mice, suggesting that this mutation is not essential for metastasis induction. Moreover, we could not show a relationship between methylation of the RASSF1A promoter and the presence of an activating mutation in GNAQ in uveal melanoma cells (Fig. 1B and Table 2).

Table 2.

Sequences of exon 15 of BRAF, exons 1 and 2 of NRAS, exons 4 and 5 of GNAQ, and GNA11 were analyzed in normal uveal melanocytes and in primary and metastatic uveal melanoma cell lines after PCR amplification and sequencing of genomic DNA

Primary uveal melanoma cell linesMetastatic uveal melanoma cell line
Normal uveal melanocytesOCM-192.1Mel270Mum2Bμ2TP17μ2F0MM1.3
BRAF wt V600E wt wt wt wt wt wt wt 
NRAS wt wt wt wt wt wt wt wt wt 
GNAQ          
Q209 wt wt wt Q209P wt Q209P wt wt Q209P 
R183 wt wt wt wt wt wt wt wt wt 
GNA11          
Q209 wt wt wt wt wt wt wt wt wt 
R183 wt wt wt wt wt wt wt wt wt 
Primary uveal melanoma cell linesMetastatic uveal melanoma cell line
Normal uveal melanocytesOCM-192.1Mel270Mum2Bμ2TP17μ2F0MM1.3
BRAF wt V600E wt wt wt wt wt wt wt 
NRAS wt wt wt wt wt wt wt wt wt 
GNAQ          
Q209 wt wt wt Q209P wt Q209P wt wt Q209P 
R183 wt wt wt wt wt wt wt wt wt 
GNA11          
Q209 wt wt wt wt wt wt wt wt wt 
R183 wt wt wt wt wt wt wt wt wt 

Abbreviation: Wt, wild type.

RASSF1A expression is regulated by methylation of its promoter

To evaluate whether the genomic DNA demethylation can contribute to RASSF1A expression, we treated OCM-1, 92.1, and Mel270 cell lines with DNA methyltransferase inhibiting drug 5-Aza-CdR. This agent restored RASSF1A expression 3 and 6 days posttreatment (Fig. 2A). To avoid a potential nonspecific effect on RASSF1A expression restoration, we then examined the methylation status of RASSF1A promoter when the uveal melanoma cells were treated with this drug (Fig. 2B). The restored RASSF1A expression after 5-Aza-CdR treatment inversely correlates with RASSF1A promoter methylation status (Fig. 2B). These data showed that methylation of RASSF1A promoter may play an important role in the silencing of RASSF1A transcription in uveal melanoma.

Figure 2.

Expression of RASSF1A by treatment with 5-Aza-CdR in uveal melanoma cell lines. OCM-1, 92.1, and Mel270 cell lines were treated for 3 and 6 days with 5-Aza-CdR (10 μmol/L). A, RNA was extracted and the transcriptional induction of RASSF1A in relation to a housekeeping gene was analyzed by qPCR (mean values ± SD of values from 3 independent experiments). B, PCR products from MSP assay with specific primers showing DNA methylation status of RASSF1A promoter at 3 days of treatment with 5-Aza-CdR (10 μmol/L). U: unmethylated promoter and M: methylated promoter.

Figure 2.

Expression of RASSF1A by treatment with 5-Aza-CdR in uveal melanoma cell lines. OCM-1, 92.1, and Mel270 cell lines were treated for 3 and 6 days with 5-Aza-CdR (10 μmol/L). A, RNA was extracted and the transcriptional induction of RASSF1A in relation to a housekeeping gene was analyzed by qPCR (mean values ± SD of values from 3 independent experiments). B, PCR products from MSP assay with specific primers showing DNA methylation status of RASSF1A promoter at 3 days of treatment with 5-Aza-CdR (10 μmol/L). U: unmethylated promoter and M: methylated promoter.

Close modal

RASSF1A is not sufficient to reduce the ERK1/2 activation in uveal melanoma cell lines

We then wanted to characterize the contribution of RASSF1A expression on the constitutive activation of the ERK1/2 signaling pathway and on the control of cyclin D1 and p27Kip1 cell-cycle proteins expression in uveal melanoma. By gene transfer technology, ectopic RASSF1A expression was obtained in OCM-1, 92.1, and Mel270. Figure 3A shows that overexpression of RASSF1A did not affect the level of ERK1/2 activation in uveal melanoma cells lines studied (Fig. 3A). Consequently, expression levels of cyclin D1 and p27Kip1 proteins were not modified (Fig. 3A). To determine whether theses processes occurred at the transcriptional level, we showed by using qPCR that RASSF1A overexpression was not influencing the mRNA expression levels of the cell-cycle regulators, cyclin D1, p27Kip1, and p21 (Fig. 3B). Following this, we evaluated the contribution of RASSF1A in uveal melanoma cell proliferation. Overexpression of RASSF1A reduced cell proliferation by only 18%, 19%, and 28% in OCM-1, 92.1, and Mel270, respectively (P < 0.05; Fig. 3C). Moreover, RASSF1A expression did not modify the distribution of the cells in the different phases of the cell cycle and did not induce apoptosis in OCM-1, 92.1, and Mel270 uveal melanoma cell lines (data not shown). Furthermore, a clonogenic assay showed that the overexpression of RASSF1A decreased foci number by only 26% in Mel270 cell lines (Fig. 3D). Subsequent to this, the role of RASSF1A in the cell transformation of uveal melanoma cell lines (OCM-1, 92.1, and Mel270) was analyzed by studying the ability of cells to grow under anchorage-independent conditions. The ability of cells to form colonies in soft agar, a specificity closely associated with the malignant phenotype, was used to assess cell transformation. The 3 RASSF1A-overexpressing cell lines formed numerous and large colonies in soft agar similar to the control cells (Fig. 3E). These data showed that RASSF1A expression is not sufficient to decrease the transformation of uveal melanoma.

Figure 3.

Ectopic expression of RASSF1A did not alter ERK1/2 activation and the regulation of cell-cycle regulators, cyclin D1, p27, and p21. OCM-1, 92.1, and Mel270 were transfected with Flag-tagged RASSF1A. A, 72 hours posttransfection cell extracts were used for immunoblotting with the indicated primary antibody. The immunoblot photographs were representative of 3 independent experiments. B, relative mRNA expression of indicated protein was quantified by qPCR in stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) OCM-1, 92.1 and Mel270 cell lines (*, P < 0.001 compared with control Flag-pcDNA). C, effects of RASSF1A expression on cell proliferation. The MTT colorimetric assay was used to assess melanoma cell proliferation after a 6 days culture period and the percentage of viable cells was obtained (*, P < 0.05 compared with Flag-pcDNA). D, quantification of the number of foci in stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) Mel270 cell line. Data are presented as the mean ± SD of values from 3 independent experiments (*, P < 0.05). Foci were photographed on day 12. E, stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) Mel270 cell line were resuspended in complete medium containing 0.3% agar. Plates were incubated at 37°C for 2 weeks. Macroscopic colonies in 3 randomly chosen 9-cm2 areas were photographed on day 14. Similar results were obtained in the OCM-1 and 92.1 cell lines.

Figure 3.

Ectopic expression of RASSF1A did not alter ERK1/2 activation and the regulation of cell-cycle regulators, cyclin D1, p27, and p21. OCM-1, 92.1, and Mel270 were transfected with Flag-tagged RASSF1A. A, 72 hours posttransfection cell extracts were used for immunoblotting with the indicated primary antibody. The immunoblot photographs were representative of 3 independent experiments. B, relative mRNA expression of indicated protein was quantified by qPCR in stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) OCM-1, 92.1 and Mel270 cell lines (*, P < 0.001 compared with control Flag-pcDNA). C, effects of RASSF1A expression on cell proliferation. The MTT colorimetric assay was used to assess melanoma cell proliferation after a 6 days culture period and the percentage of viable cells was obtained (*, P < 0.05 compared with Flag-pcDNA). D, quantification of the number of foci in stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) Mel270 cell line. Data are presented as the mean ± SD of values from 3 independent experiments (*, P < 0.05). Foci were photographed on day 12. E, stable RASSF1A (Flag-RASSF1A) and control (Flag-pcDNA) Mel270 cell line were resuspended in complete medium containing 0.3% agar. Plates were incubated at 37°C for 2 weeks. Macroscopic colonies in 3 randomly chosen 9-cm2 areas were photographed on day 14. Similar results were obtained in the OCM-1 and 92.1 cell lines.

Close modal

RASSF1A is involved in cell cycle control in normal uveal melanocytes

Several studies have examined the effects of exogenous expression of RASSF1A on the diminution of cell proliferation and on the tumorigenic phenotype on immortalized cells (3, 4, 7). To investigate the function of RASSF1A in the nonimmortalized cell division, we first examined the effect of RASSF1A depletion on cell proliferation and the cell-cycle status in freshly isolated normal uveal melanocytes. siRNA-mediated depletion of RASSF1A resulted in a significant decrease in RASSF1A expression, 3 and 5 days posttransfection (Fig. 4A). Normal uveal melanocytes transfected with control siRNA grew slowly in the exponential phase, with a doubling time of approximately 72 hours, following stimulation with 10 ng/mL basic fibroblast growth factor (bFGF) and with substances, such as isobutylmethylxanthine and cholera toxin, as previously described (Fig. 4B; refs. 26, 31–33). RASSF1A downregulated normal uveal melanocytes grew more quickly, with a doubling time of approximately 36 hours at 3 days. Interestingly, cells cease proliferating after 3 days (Fig. 4B). Moreover, cell-cycle analysis with flow cytometry revealed that siRNA-mediated depletion of RASSF1A led to a 14% decrease of cells in the G1 phase, 37% and 45% increase of cells in the S and G2-M phases, respectively, as compared with controls at 3 days (P < 0.05; Fig. 4C). Five days posttransfection, no difference of cell-cycle distribution was observed when comparing control normal uveal melanocytes and RASSF1A downregulated normal uveal melanocytes, confirming the proliferative arrest observed in Figure 4B. Until day 5, RASSF1A inactivation did not induce a sub-G1 phase, indicating the absence of apoptosis (Fig. 4C).

Figure 4.

RASSF1A is involved in cell-cycle control and in ERK1/2 activation. Normal uveal melanocytes were transfected with siRNA specifically targeting the RASSF1A transcript and with siRNA scramble (negative control) for 5 days. A, relative amount of transcript as determined by qPCR in control cells (Ct) and after 3 and 5 days of incubation (mean values ± SD of 3 independent experiments *, P < 0.05 compared with control siRNA). B, after the addition of FIC medium, cells were counted at 3 and 5 days (mean values ± SD of 4 independent experiments). C, 3 days posttransfection, cells in sub-G1, G1, S, and G2-M phases of the cell cycle were quantified by flow cytometry to detect cellular DNA content with propidium iodide (PI) (mean values ± SD of 3 independent experiments *, P < 0.05; **, P = 0.05 compared with control siRNA). D, effects of RASSF1A siRNA on ERK1/2 activation and cyclin D1 and p27Kip1 expression in stimulated normal uveal melanocytes were analyzed by Western blotting at days 3 and 5 by using the indicated primary antibody.

Figure 4.

RASSF1A is involved in cell-cycle control and in ERK1/2 activation. Normal uveal melanocytes were transfected with siRNA specifically targeting the RASSF1A transcript and with siRNA scramble (negative control) for 5 days. A, relative amount of transcript as determined by qPCR in control cells (Ct) and after 3 and 5 days of incubation (mean values ± SD of 3 independent experiments *, P < 0.05 compared with control siRNA). B, after the addition of FIC medium, cells were counted at 3 and 5 days (mean values ± SD of 4 independent experiments). C, 3 days posttransfection, cells in sub-G1, G1, S, and G2-M phases of the cell cycle were quantified by flow cytometry to detect cellular DNA content with propidium iodide (PI) (mean values ± SD of 3 independent experiments *, P < 0.05; **, P = 0.05 compared with control siRNA). D, effects of RASSF1A siRNA on ERK1/2 activation and cyclin D1 and p27Kip1 expression in stimulated normal uveal melanocytes were analyzed by Western blotting at days 3 and 5 by using the indicated primary antibody.

Close modal

Molecular mechanisms, by which RASSF1A impacts on cell proliferation, was then investigated. We showed that downregulation of RASSF1A led to an increase in ERK1/2 activation after 3 and 5 days of culture (Fig. 4D). Consequently, the relative amounts of protein expression showed that RASSF1A inactivation led to an increase in the levels of cyclin D1 after 3 and 5 days of culture (Fig. 4D). In contrast, inactivation of RASSF1A leads to a decrease in the levels of p27Kip1, a negative regulator of the G1-to S-phase transition (Fig. 4D). Taken together, these data showed the role of endogenous RASSF1A in the control of ERK1/2 signaling pathway and in cell-cycle control in normal uveal melanocytes.

Inhibition of endogenous RASSF1A induces premature senescence in normal uveal melanocytes

Observations of RASSF1A downregulated normal uveal melanocytes under phase-contrast microscope showed dramatic morphologic changes. RASSF1A inactivation caused loss of the dendritic shape and, most remarkably, the enlargement and flattening of the cells, as described in senescent cells (Fig. 5A; ref. 34). Furthermore, nuclear condensation, cell detachment, and nuclear fragmentation were no longer observed in RASSF1A downregulated normal uveal melanocytes (Fig. 5A). Interestingly, RASSF1A downregulated normal uveal melanocytes accumulated senescence-associated β-galactosidase activity (SA β-gal; Fig. 5A). To determine whether RASSF1A inactivation could be associated with senescence in normal uveal melanocytes, we investigated the level of RASSF1A expression in senescent normal uveal melanocytes. For this, fresh isolated normal uveal melanocytes were grown in culture and have been passaged for 23 generations over a period of 9 months and were maintained in a nonproliferative state defined as replicative senescence. We called these cells senescent uveal melanocytes (senescent uveal melanocytes) and they showed a high level of SA-β-galactosidase activity (Fig. 5A). By qPCR, we observed significantly decreased expression of RASSF1A in senescent uveal melanocytes (Fig. 5B and C). Moreover, the RASSF1A promoter was epigenetically inactivated by methylation in senescent uveal melanocytes (Fig. 5B). A 2-fold decrease in cyclin D1 mRNA levels was observed in senescent uveal melanocytes when compared with normal uveal melanocytes (Fig. 5C). In contrast, the level of p27kip1 mRNA expression remained unchanged in these cells (Fig. 5C). We also showed that RASSF1A inactivation led to a 3-fold increase in cyclin D1 mRNA expression in normal uveal melanocytes (Fig. 5C). Interestingly, RASSF1A inactivation seemed to be responsible for the increased expression of the cyclin-dependent kinase inhibitor p21Cip1, one of the major contributors of senescence, with a similar level as in senescent uveal melanocytes (Fig. 5C). Taken together, these findings suggest that RASSF1A could participate to the process of cellular senescence escape in uveal melanoma.

Figure 5.

Inactivation of endogenous RASSF1A activates premature senescence. A, normal uveal melanocytes were transfected with siRNA scramble (control siRNA) or with siRNA targeted RASSF1A for 5 days. Representative phase-contrast photomicrographs (left) of transfected normal uveal melanocytes and senescent uveal melanocytes (late-passage uveal melanocytes undergone replicative senescence). The transfected normal uveal melanocytes and senescent uveal melanocytes were stained for SA–β-galactosidase activity (right). Percentage of SA–β-gal activity intensity was calculated with the ImageJ software. Sixty cells from each cell population were scored; (mean values ± SD of 3 separate experiments are shown. *, P < 0.05 compared with normal uveal melanocytes control siRNA). B, analyse of RASSF1A expression in senescent uveal melanocytes compared with normal uveal melanocytes. qPCR and Western blotting (WB) were done to examine the level of RASSF1A and actin mRNA and protein. The DNA methylation status of RASSF1A promoter was assessed by MSP in normal uveal melanocytes and senescent uveal melanocytes. DNA treated with SssI Methylase was used as a positive control (C+) and DNA from lymphocytes was used as negative control (C-). U, unmethylated promoter and M, methylated promoter. C, in transfected normal uveal melanocytes and in senescent uveal melanocytes the relative mRNA expression of indicated protein was quantified by qPCR (mean values ± SD of 3 independent experiments *, P < 0.05, **, P = 0.06 compared with normal uveal melanocytes control siRNA). NUM, normal uveal melanocytes; SUM, senescent uveal melanocytes.

Figure 5.

Inactivation of endogenous RASSF1A activates premature senescence. A, normal uveal melanocytes were transfected with siRNA scramble (control siRNA) or with siRNA targeted RASSF1A for 5 days. Representative phase-contrast photomicrographs (left) of transfected normal uveal melanocytes and senescent uveal melanocytes (late-passage uveal melanocytes undergone replicative senescence). The transfected normal uveal melanocytes and senescent uveal melanocytes were stained for SA–β-galactosidase activity (right). Percentage of SA–β-gal activity intensity was calculated with the ImageJ software. Sixty cells from each cell population were scored; (mean values ± SD of 3 separate experiments are shown. *, P < 0.05 compared with normal uveal melanocytes control siRNA). B, analyse of RASSF1A expression in senescent uveal melanocytes compared with normal uveal melanocytes. qPCR and Western blotting (WB) were done to examine the level of RASSF1A and actin mRNA and protein. The DNA methylation status of RASSF1A promoter was assessed by MSP in normal uveal melanocytes and senescent uveal melanocytes. DNA treated with SssI Methylase was used as a positive control (C+) and DNA from lymphocytes was used as negative control (C-). U, unmethylated promoter and M, methylated promoter. C, in transfected normal uveal melanocytes and in senescent uveal melanocytes the relative mRNA expression of indicated protein was quantified by qPCR (mean values ± SD of 3 independent experiments *, P < 0.05, **, P = 0.06 compared with normal uveal melanocytes control siRNA). NUM, normal uveal melanocytes; SUM, senescent uveal melanocytes.

Close modal

This study shows for the first time that RASSF1A downregulation in uveal melanoma results from biallelic inactivation of the RASSF1A gene by promoter hypermethylation and/or LOH at chromosome 3p21.3 or monosomy 3. Although we could not investigate the status of chromosome 3 or LOH in the locus of RASSF1A gene in all uveal melanoma tumors, we were able to show RASSF1A promoter hypermethylation in 83% of primary tumors studied. This frequency of RASSF1A promoter hypermethylation is only slightly higher than what has been previously reported in uveal melanoma at 50% to 70% (19, 20). A further finding in our study is the demonstration of at least one allele with a methylated RASSF1A promoter in 4 of 7 primary and metastatic uveal melanoma cell lines. This is in agreement with data previously published (19, 20). A decrease in RASSF1A transcript levels was detected in both primary and metastatic cell lines, indicating that this modification is not restricted to metastatic disease. We also showed that RASSF1A promoter methylation status directly correlates with the transcriptional expression of the gene in uveal melanoma cell lines treated with a demethylating agent.

In contrast to cutaneous melanoma, uveal melanoma exhibit a constitutive activation of ERK1/2 in the absence of any mutation in NRAS and BRAF (10, 16). We found the Q209P mutation in the heterotrimeric G protein α-subunit, GNAQ in 58% (11/19) of uveal melanoma tumors. We detected the Q209L mutation in the related gene, GNA11 in the 50% (4/8) of the GNAQWT uveal melanoma tumors (unpublished data). These mutation have been shown to be an alternative route to ERK1/2 activation in uveal melanoma and may be an early event in uveal melanoma development (17, 18, 34). Our data indicated no relationship between alteration of GNAQ and GNA11 and methylation of RASSF1A promoter in uveal melanoma cell lines and tumors. Moreover, the normal uveal melanocytes and senescent uveal melanocytes did not show the GNAQ and GNA11 mutations, suggesting that these mutations may be additional deregulations occurring during uveal melanoma progression and may appear after the downregulation of RASSF1A. These data are consistent with the suggestion that early-stage cancer cells carry epigenetic modifications of growth-associated genes, which may predispose the cells to an accumulation of additional changes in oncogenes and tumor suppressor genes, ultimately leading to cancer (35). RASSF1A overexpression did not result in any difference in the level of ERK1/2 activation or in expression levels of cyclin D1 and p27 in the cell lines OCM-1 (which expresses BRAFV600E), Mel270 (which expresses GNAQQ209P), or 92.1 (wild type for BRAF and GNAQ). In wild-type GNAQ and GNA11 uveal melanoma cells, the deregulation leading to constitutive activation of ERK1/2 remains to be determined.

In lung and breast tumor–derived epithelial cells, overexpression of RASSF1A leads to cessation of cell progression at the G1 phase by inhibiting the accumulation of cyclin D1 without any changes in cyclin D1 mRNA levels (36). In these cells, RASSF1A inactivation results in an increase in cyclin D1 protein level (36, 37). To our knowledge, our study presents novel data showing that in RASSF1A-inactivated normal uveal melanocytes, the level cyclin D1 protein is increased with a modulation of cyclin D1 mRNA levels. It has been reported that the presence of cyclin D1 protein is associated with a more aggressive clinical course and a histologically unfavorable prognosis in uveal melanoma (38). Interestingly, endogenous RASSF1A inactivation led to a decrease in the cell-cycle–negative regulator p27Kip1 at the protein level but not at the mRNA level, suggesting that RASSF1A is involved in the stability of p27Kip1. These findings are consistent with the reported reverse correlation between the number of mitotic figures and p27Kip1 expression in primary uveal melanoma (39). Our current results, showing that the activation of ERK1/2 is induced via RASSF1A inactivation, suggest that (a) RASSF1A is a safeguard to ERK overactivation in normal uveal melanocytes; and (b) that RASSF1A may be specialized to transmit inhibitory growth signals via ERK signaling in normal uveal melanocytes. Moreover, these data indicate that loss of RASSF1A may provide a permissive environment for acquiring additional alterations required during uveal melanoma progression.

One of the critical steps in human tumorigenesis is cellular immortalization, a process in which cells must escape senescence and acquire an infinite lifespan. Our results presented here show that endogenous RASSF1A inactivation can produce 2 precisely opposite outcomes in nonimmortalized uveal melanocytes: (a) mitogenesis with ERK1/2 activation, and (b) senescence (i.e., SA-β-Gal activity and upregulation of p21Cip1 expression; ref. 21). We previously showed that stimulation by mitogens and growth factors, such as bFGF, is required for growing normal uveal melanocytes, and this was mediated by activation of ERK1/2 signaling pathway (26, 30, 31). In this study, the RASSF1A inactivation has an additional effect on ERK1/2 overactivation and provides an initial proliferative advantage but also accelerates senescence in cultured normal uveal melanocytes. Indeed, it has been shown that activated ERK1/2 pathway in nonimmortalized intestinal epithelial cells led to and induced expression of p21Cip1 and to senescence (40). Moreover, in cutaneous naevi, BRAFV600E induces senescence as a physiologic mechanism in humans, thus limiting the progression of premalignant lesions, and this oncogene requires additional cooperating events for tumor development (22). In agreement with phenomenon described in these studies (22), our data showed that inhibition of endogenous RASSF1A in nonimmortalized normal uveal melanocytes promotes premature senescence through overactivation of the ERK1/2 pathway.

The senescent uveal melanocytes obtained during late passages of freshly isolated human normal uveal melanocytes were distinguished from immortal-, quiescent-, or terminally differentiated uveal melanocytes by morphologic changes, induction of SA-β-Gal activity, and upregulation of p21Cip1 expression. Inhibition of endogenous RASSF1A in normal uveal melanocytes induces the same phenotype (morphologic changes, increased SA-β-Gal activity and p21Cip1 expression). A plausible role for RASSF1A can be inferred from the regulated expression of p21Cip1. This is in agreement with previous studies, which have shown immunoexpression of uveal melanoma cells for p21Cip1 (38, 39).

It has recently been shown that oncogenic RAS requires immortalizing changes to promote oncogenic transformation (21). Although a cell might undergo malignant transformation, it may not indefinitely proliferate in the absence of immortalization. Tumor-suppressive genes—such as pRb, p53, p16—may initially exert a protective effect on cells by limiting their growth potential, but the loss of these genes may lead to increased proliferation, and, eventually, to cancer development (23). Our findings show that RASSF1A expression is downregulated by an epigenetic mechanism in senescent uveal melanocytes.

In conclusion, models of normal and senescent uveal melanocytes established here support the notion that normal uveal melanocytes possess fail-safe mechanisms that limit the consequences of overactivation of the ERK1/2 mitogenic pathway. Our data show that the high frequency of hypermethylation of the RASSF1A promoter, associated with initiation of cellular senescence, may be an early event in uveal melanoma progression. Overexpression of RASSF1A alone in immortal uveal melanoma cell lines does not seem to be sufficient to induce decrease in the overactivation of ERK1/2 and oncogenic transformation; additional alterations are required during the malignant transformation to uveal melanoma.

No potential conflicts of interest were disclosed.

We thank Dr. Galateau for receiving the authors in her laboratory, J. Cheret for technical assistance with quantitative qPCR, G. Zalcman for scientific exchange, and L. Poulain and M. Duval for the scientific and technical assistance with flow cytometry facility, Grecan, Caen.

This study was supported by grants from Association pour la Recherche contre le Cancer (ARC) and Ligue nationale contre le cancer and the Conseil Regional de Basse-Normandie.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
van der Weyden
L
,
Adams
DJ
. 
The Ras-association domain family (RASSF) members and their role in human tumourigenesis
.
Biochim Biophys Acta
2007
;
1776
:
58
85
.
2.
Vos
MD
,
Clark
GJ
. 
RASSF family proteins and ras transformation
.
Methods Enzymol
2006
;
407
:
311
22
.
3.
Agathanggelou
A
,
Cooper
WN
,
Latif
F
. 
Role of the ras-association domain family 1 tumor suppressor gene in human cancers
.
Cancer Res
2005
;
65
:
3497
508
.
4.
Dammann
R
,
Li
C
,
Yoon
JH
,
Chin
PL
,
Bates
S
,
Pfeifer
GP
. 
Epigenetic inactivation of a ras association domain family protein from the lung tumour suppressor locus 3p21.3
.
Nat Genet
2000
;
25
:
315
9
.
5.
Spugnardi
M
,
Tommasi
S
,
Dammann
R
,
Pfeifer
GP
,
Hoon
DSB
. 
Epigenetic inactivation of RAS association domain family protein 1 (RASSF1A) in malignant cutaneous melanoma
.
Cancer Res
2003
;
63
:
1639
43
.
6.
Liu
L
,
Guo
C
,
Dammann
R
,
Tommasi
S
,
Pfeifer
GP
. 
RASSF1A interacts with and activates the mitotic kinase aurora-A
.
Oncogene
2008
;
27
:
6175
86
.
7.
Kuzmin
I
,
Gillespie
JW
,
Protopopov
A
,
Geil
L
,
Dreijerink
K
,
Yang
Y
, et al
The RASSF1A tumor suppressor gene is inactivated in prostate tumors and suppresses growth of prostate carcinoma cells
.
Cancer Res
2002
;
62
:
3498
502
.
8.
Dreijerink
K
,
Braga
E
,
Kuzmin
I
,
Geil
L
,
Duh
FM
,
Angeloni
D
, et al
The candidate tumor suppressor gene, RASSF1A, from human chromosome 3p21.3 is involved in kidney tumorigenesis
.
Proc Natl Acad Sci U S A
2001
;
98
:
7504
9
.
9.
Tommasi
S
,
Dammann
R
,
Zhang
Z
,
Wang
Y
,
Liu
L
,
Tsark
WM
, et al
Tumor susceptibility of rassf1a knockout mice
.
Cancer Res
2005
;
65
:
92
8
.
10.
Reifenberger
J
,
Knobbe
CB
,
Sterzinger
AA
,
Blaschke
B
,
Schulte
KW
,
Ruzicka
T
, et al
Frequent alterations of ras signaling pathway genes in sporadic malignant melanomas
.
Int J Cancer
2004
;
109
:
377
84
.
11.
Tschentscher
F
,
Prescher
G
,
Horsman
DE
,
White
VA
,
Rieder
H
,
Anastassiou
G
, et al
Partial deletions of the long and short arm of chromosome 3 point to two tumor suppressor genes in uveal melanoma
.
Cancer Res
2001
;
61
:
3439
42
.
12.
Damato
B
,
Dopierala
J
,
Klaasen
A
,
van Dijk
M
,
Sibbring
J
,
Coupland
SE
. 
Multiplex ligation-dependent probe amplification of uveal melanoma: correlation with metastatic death
.
Invest Ophthalmol Vis Sci
2009
;
50
:
3048
55
.
13.
Rimoldi
D
,
Salvi
S
,
Liénard
D
,
Lejeune
FJ
,
Speiser
D
,
Zografos
L
, et al
Lack of braf mutations in uveal melanoma
.
Cancer Res
2003
;
63
:
5712
5
.
14.
Malaponte
G
,
Libra
M
,
Gangemi
P
,
Bevelacqua
V
,
Mangano
K
,
D'Amico
F
, et al
Detection of braf gene mutation in primary choroidal melanoma tissue
.
Cancer Biol Ther
2006
;
5
:
225
7
.
15.
Gear
H
,
Williams
H
,
Kemp
EG
,
Roberts
F
. 
Braf mutations in conjunctival melanoma
.
Invest Ophthalmol Vis Sci
2004
;
45
:
2484
8
.
16.
Calipel
A
,
Mouriaux
F
,
Glotin
A
,
Malecaze
F
,
Faussat
A
,
Mascarelli
F
. 
Extracellular signal-regulated kinase-dependent proliferation is mediated through the protein kinase A/B-Raf pathway in human uveal melanoma cells
.
J Biol Chem
2006
;
281
:
9238
50
.
17.
Van Raamsdonk
CD
,
Bezrookove
V
,
Green
G
,
Bauer
J
,
Gaugler
L
,
O'Brien
JM
, et al
Frequent somatic mutations of GNAQ in uveal melanoma and blue naevi
.
Nature
2009
;
457
:
599
602
.
18.
Van Raamsdonk
CD
,
Griewank
KG
,
Crosby
MB
,
Garrido
MC
,
Vemula
S
,
Wiesner
T
, et al
Mutations in GNA11 in uveal melanoma
.
N Engl J Med
2010
;
363
:
2191
9
.
19.
Merhavi
E
,
Cohen
Y
,
Avraham
BCR
,
Frenkel
S
,
Chowers
I
,
Pe'er
J
, et al
Promoter methylation status of multiple genes in uveal melanoma
.
Invest Ophthalmol Vis Sci
2007
;
48
:
4403
6
.
20.
Maat
W
,
van der Velden
PA
,
Out-Luiting
C
,
Plug
M
,
Dirks-Mulder
A
,
Jager
MJ
, et al
Epigenetic inactivation of RASSF1a in uveal melanoma
.
Invest Ophthalmol Vis Sci
2007
;
48
:
486
90
.
21.
Fridman
AL
,
Tainsky
MA
. 
Critical pathways in cellular senescence and immortalization revealed by gene expression profiling
.
Oncogene
2008
;
27
:
5975
87
.
22.
Michaloglou
C
,
Vredeveld
LCW
,
Soengas
MS
,
Denoyelle
C
,
Kuilman
T
,
van der Horst
CMAM
, et al
Brafe600-associated senescence-like cell cycle arrest of human naevi
.
Nature
2005
;
436
:
720
4
.
23.
Lin
AW
,
Barradas
M
,
Stone
JC
,
van Aelst
L
,
Serrano
M
,
Lowe
SW
. 
Premature senescence involving p53 and p16 is activated in response to constitutive MEK/MAPK mitogenic signaling
.
Genes Dev
1998
;
12
:
3008
19
.
24.
Mooi
WJ
,
Peeper
DS
. 
Oncogene-induced cell senescence–halting on the road to cancer
.
N Engl J Med
2006
;
355
:
1037
46
.
25.
Grønbaek
K
,
Treppendahl
M
,
Asmar
F
,
Guldberg
P
. 
Epigenetic changes in cancer as potential targets for prophylaxis and maintenance therapy
.
Basic Clin Pharmacol Toxicol
2008
;
103
:
389
96
.
26.
Mouriaux
F
,
Casagrande
F
,
Pillaire
MJ
,
Manenti
S
,
Malecaze
F
,
Darbon
JM
. 
Differential expression of g1 cyclins and cyclin-dependent kinase inhibitors in normal and transformed melanocytes
.
Invest Ophthalmol Vis Sci
1998
;
39
:
876
84
.
27.
Folberg
R
,
Kadkol
SS
,
Frenkel
S
,
Valyi-Nagy
K
,
Jager
MJ
,
Pe'er
J
, et al
Authenticating cell lines in ophthalmic research laboratories
.
Invest Ophthalmol Vis Sci
2008
;
49
:
4697
701
.
28.
Vos
MD
,
Ellis
CA
,
Bell
A
,
Birrer
MJ
,
Clark
GJ
. 
Ras uses the novel tumor suppressor RASSF1 as an effector to mediate apoptosis
.
J Biol Chem
2000
;
275
:
35669
72
.
29.
Ahmed-Choudhury
J
,
Agathanggelou
A
,
Fenton
SL
,
Ricketts
C
,
Clark
GJ
,
Maher
ER
, et al
Transcriptional regulation of cyclin A2 by RASSF1A through the enhanced binding of p120E4F to the cyclin A2 promoter
.
Cancer Res
2005
;
65
:
2690
7
.
30.
Calipel
A
,
Lefevre
G
,
Pouponnot
C
,
Mouriaux
F
,
Eychène
A
,
Mascarelli
F
. 
Mutation of B-Raf in human choroidal melanoma cells mediates cell proliferation and transformation through the MEK/ERK pathway
.
J Biol Chem
2003
;
278
:
42409
18
.
31.
Lefèvre
G
,
Babchia
N
,
Calipel
A
,
Mouriaux
F
,
Faussat
A
,
Mrzyk
S
, et al
Activation of the FGF2/FGFR1 autocrine loop for cell proliferation and survival in uveal melanoma cells
.
Invest Ophthalmol Vis Sci
2009
;
50
:
1047
57
.
32.
Hu
DN
,
McCormick
SA
,
Ritch
R
. 
Studies of human uveal melanocytes in vitro: growth regulation of cultured human uveal melanocytes
.
Invest Ophthalmol Vis Sci
1993
;
34
:
2220
7
.
33.
Lefevre
G
,
Calipel
A
,
Mouriaux
F
,
Hecquet
C
,
Malecaze
F
,
Mascarelli
F
. 
Opposite long-term regulation of c-Myc and p27Kip1 through overactivation of Raf-1 and the MEK/ERK module in proliferating human choroidal melanoma cells
.
Oncogene
2003
;
22
:
8813
22
.
34.
Dimri
GP
,
Lee
X
,
Basile
G
,
Acosta
M
,
Scott
G
,
Roskelley
C
, et al
A biomarker that identifies senescent human cells in culture and in aging skin in vivo
.
Proc Natl Acad Sci U S A
1995
;
92
:
9363
7
.
35.
Onken
MD
,
Worley
LA
,
Long
MD
,
Duan
S
,
Council
ML
,
Bowcock
AM
, et al
Oncogenic mutations in GNAQ occur early in uveal melanoma
.
Invest Ophthalmol Vis Sci
2008
;
49
:
5230
4
.
36.
Shivakumar
L
,
Minna
J
,
Sakamaki
T
,
Pestell
R
,
White
MA
. 
The RASSF1A tumor suppressor blocks cell cycle progression and inhibits cyclin D1 accumulation
.
Mol Cell Biol
2002
;
22
:
4309
18
.
37.
Whang
YM
,
Kim
YH
,
Kim
JS
,
Yoo
YD
. 
RASSF1A suppresses the c-Jun-NH2-kinase pathway and inhibits cell cycle progression
.
Cancer Res
2005
;
65
:
3682
90
.
38.
Coupland
SE
,
Bechrakis
N
,
Schüler
A
,
Anagnostopoulos
I
,
Hummel
M
,
Bornfeld
N
, et al
Expression patterns of cyclin D1 and related proteins regulating G1-S phase transition in uveal melanoma and retinoblastoma
.
Br J Ophthalmol
1998
;
82
:
961
70
.
39.
Mouriaux
F
,
Maurage
CA
,
Labalette
P
,
Sablonnière
B
,
Malecaze
F
,
Darbon
JM
. 
Cyclin-dependent kinase inhibitory protein expression in human choroidal melanoma tumors
.
Invest Ophthalmol Vis Sci
2000
;
41
:
2837
43
.
40.
Boucher
M
,
Jean
D
,
Vézina
A
,
Rivard
N
. 
Dual role of MEK/ERK signaling in senescence and transformation of intestinal epithelial cells
.
Am J Physiol Gastrointest Liver Physiol
2004
;
286
:
G736
46
.