Abstract
Background: Several studies have demonstrated that different genetic profiles contribute to melanoma development and progression. Materials and Methods: To evaluate the existence of different molecular aberration patterns in melanoma associated with v-raf murine sarcoma viral oncogene homolog B1 (BRAF) or 9p21 locus alterations, eleven patient-derived melanoma cell lines were characterized. Multiplex ligation probe amplification (MLPA) was used to detect chromosomal alterations. Single- strand conformation analysis and sequencing were performed to study BRAF, neuroblastoma RAS viral (v-ras) oncogene homolog (NRAS), v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog (c-KIT), melanocortin 1 receptor (alpha melanocyte stimulating hormone receptor) (MC1R), cyclin-dependent kinase inhibitor 2A (CDKN2A) and cyclin-dependent kinase 4 (CDK4) genes. Results: BRAFV600E mutation was detected in 54% of cell lines. NRAS was mutated in one cell line also carrying multiple copies of NRAS. All cell lines with MC1R variants harboured BRAFV600E. Concurrent loss of MUTYH (1p33), gains of c-MYC (8q24) and of CDK6 (7q21) were found to be significantly associated in cell lines (45%) that harboured biallelic 9p21 deletions including CDKN2B-CDKN2A-MTAP. Conclusion: These data suggest the existence of a specific pattern of somatic alterations in genes that are involved in DNA repair (MUTYH) and in cell cycle regulation (c-MYC, CDK6, CDKN2A and CDKN2B). Interestingly, all MC1R variants were associated with BRAFV600E and all cell lines from visceral metastases harboured BRAFV600E.
Melanoma is a complex and heterogeneous disease. Several studies have identified loci with established importance (1). The 9p21 region has been widely studied since deletions affecting this region have been reported in approximately 50% of tumours, being higher in cultured cell lines (2, 3). The 9p21 locus contains a well known tumour- suppressor gene (TSG), cyclin-dependent kinase inhibitor 2A (CDKN2A), which encodes two distinct cell cycle regulatory proteins: p16INK4A and p14ARF. Comparative microarray analysis in melanoma cell lines with and without homozygous deletion of CDKN2A showed several classes of genes involved in different pathways (4). Loss of heterozygosity (LOH) studies in melanoma tumours suggest the existence of other TSGs located at 9p21 (5).
Greshock et al. (6) have shown that DNA copy number aberrations associated with v-raf murine sarcoma viral oncogene homolog B1 (BRAF)-mutated melanomas are different from those found in neuroblastoma RAS viral (v- ras) oncogene homolog (NRAS)-mutated and BRAF/NRAS wild-type tumours. Previously, two expression studies revealed a BRAF mutation-associated expression signature, finding a number of significant differentially expressed genes comparing melanoma cell lines with and without BRAF mutations (7, 8). However, another study concluded that there was no specific gene expression profile associated with BRAF mutation status (9).
The main objective of this study was to investigate the existence of different molecular aberration patterns associated with BRAF- activating mutations and 9p21 alterations in a set of eleven human cutaneous melanoma-derived cell lines by Multiplex ligation probe amplification (MLPA) approach. Furthermore, BRAF, CDKN2A, cyclin-dependent kinase 4 (CDK4), v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog (c-KIT), melanocortin 1 receptor (alpha melanocyte stimulating hormone receptor) (MC1R) and NRAS genes were also studied by mutational analysis.
Materials and Methods
Cell lines and culture conditions. Eleven human malignant melanoma cell lines were obtained from Caucasian patients with sporadic melanoma (10). Melanoma cell lines were grown in Dulbecco's modified Eagle's medium (DMEM) (Gibco, Barcelona, Spain), supplemented with 10% foetal calf serum (FCS) (Gibco) and antibiotics (penicillin/streptomycin; Sigma/Aldrich, Madrid, Spain). All the cells were incubated at 37°C in an atmosphere with 5%-CO2.
DNA extraction. The PUREGENE DNA Isolation Kit (Gentra Systems, Minneapolis, MN, USA) was used to isolate genomic DNA from cell lines according to the manufacturer's instructions.
PCR amplification. Promoter (-34G>T variant), intronic (IVS2-105) and coding regions of the CDKN2A gene (exons 1α, 2 and 3 of p16INK4A and exon 1β of p14ARF) and exon 2 of CDK4, MC1R and c-KIT exons 9, 11, 13, 17, 18 were amplified by PCR using primers and conditions previously described (11-14). For BRAF exons 11 and 15 and NRAS exons 1 and 2, primers were designed to amplify the exons where the most common mutations are detected. All PCRs were carried out using the PCR Master Mix (Promega Co., Madison, WI, USA) following the manufacturer's instructions. PCR conditions were: initial denaturizing step at 95°C for 5 min, followed by 35 cycles (95°C for 1 min, Tm (c-KIT 56°C, BRAF 56°C, NRAS 57°C and MC1R 55°C) for 1 min, 72°C for 1 min), and a final extension at 72°C for 10 min and maintaining at 4°C until single- strand conformational polymorphism (SSCP) or sequencing studies were carried out. Primers for BRAF exon 11F: TTTCTTTTTCTGTTTGGCTTG, 11R: TGTGGTGACATT GTGA CAAGT, exon 15F: TGCTTGCTCTGATAGGAAAA, and exon 15R: TGAGGCTATTTTTCCACTGA; for NRAS exon 1F: CGCCAATTAACCCTGATTAC, exon 1R: GCTGACCTGATCCT GTCTCT, and exon 2F: CCCCTTACCCTCCACACC, exon 2R: TCTGAAAGGATGATCTTTGTGTT.
Mutational analysis. Mutation screening for BRAF, CDK4, CDKN2A, c-KIT and NRAS loci was performed by SSCP (15). Samples with abnormal migration products were sequenced using the BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) in an ABI3100 automatic sequencer (Applied Biosystems). MC1R was directly sequenced. Specific internal MC1R primers were designed to analyze the entire coding sequence (INT-F: TACATCTCCATCTTCTACGC and INT-R: GTGCTGAAGACGACACTG).
MLPA. Copy number variations were analyzed by MLPA (P024B CDKN2A/2B, P172 and P027) (MRC-Holland, Amsterdam, Netherlands). All kits were used according to the manufacturer's instructions. Amplified samples were analysed on automated sequencer (Applied Biosystems). MLPA results were evaluated using a custom MLPA analysis programme (SeqPilot- JSI Medisys, Kippenheim, Germany). For each fragment, the peak area was calculated and normalized against the mean peak area of control samples (consisting of six human DNA from normal tissue with normal gene dosage). A difference was considered significant if the ratio was less than 0.5 (loss) or higher than 1.5 (gain). A ratio close to 1.0 indicates two copies present (i.e. wild-type, wt); 0.00, both copies deleted (i.e. homozygous deletion, HD); 0.5, one copy deleted (i.e. loss of heterozygosity, LOH), 1.5 or higher, one copy duplicated or multiple copies. Experiments were carried out in duplicate. For each cell line, results from the MLPA kits employed were combined and arranged in figures to represent chromosome order.
Statistical analysis. Fisher's exact test (two-sided) was used to evaluate correlations between copy number alterations. P-values <0.05 were considered significant.
Results
Analysis of BRAF, NRAS, c-KIT, MC1R and CDK4 mutational status. Eleven cultured melanoma cell lines were analysed for mutations in exons 11 and 15 of the BRAF gene, and exons 1 and 2 of the NRAS gene, using SSCP and sequencing. Six out of eleven cell lines carried activating mutations in BRAF (BRAFV600E). The mutation involving the hotspot codon 600 in exon 15 was c.1799T>A in all cell lines except in one (CPM) carried c.1798-99GT>AA tandem mutation causing, in all cases, a valine-to-lysine amino acid change at residue 600 (Table I). Interestingly, all cell lines derived from visceral metastases belonged to the BRAFV600E carrying cell lines, while all cell lines from cutaneous metastases except one belonged to the BRAF wild-type (BRAFwt) (Table I).
Somatic mutation affecting NRAS was detected in one cell line (TPR), which carried a missense mutation changing a glutamine to histidine at hotspot codon 61 (p.Q61H, c.183A>C) (Table I). No mutations were detected in c-KIT. Three c-KIT variants were found in 45% of the cell lines, all of them with BRAF or NRAS mutated (Table I). M36 carried p.I798I (c.2394C>T) in exon 17. The variant p.L862L (c.2586G>C) in exon 18 was detected in cell lines TPR, DB, CPM and M16, which also carried IVS16-77G>A in intron 16. All the variants have previously been described except IVS16-77G>A.
DNA sequence analysis of the MC1R gene revealed that four out of sixof the BRAFV600E cell lines carried concomitant MC1R allelic variants (Table I). The M3 cell line had variants p.R142H (c.425G>A) and p.V60L (c.178 G>T). Cell line M16 harbored variant p.R163Q (c.488G>A), and p.T314T (c.942A>G) and p.V92M (c.274 G>A) variants were present in the DB cell line. Cell line JC carried a p.T272M (c.815C>T) change. All the variants have previously been described. In contrast, MC1R variants were not detected in the BRAFwt cell lines.
SSCP analysis of exon 2 of the CDK4 gene, and DNA sequencing revealed no mutations in any of the cell lines.
Analysis of copy number alterations. The presence of deletions affecting the 9p21 region was evaluated by MLPA assay. Overall, almost all the cell lines (10/11) carried deletions within the CDKN2A-CDKN2B-MTAP region (Table II). However, retention of cyclin-dependent kinase inhibitor 2B (CDKN2B) and of methylthioadenosine phosphorylase (MTAP) was found in two and five cases, respectively. Homozygous deletions were present in 72% of cell lines (8/11), while loss of heterozygosity (LOH) was observed in 18% (2/11). Only the DB cell line retained all loci. Furthermore, three samples harboured hemizygotic deletions. In cell line M36, the affected loci were CDKN2A-CDKN2B. In cell line M16 LOH involved tyrosine kinase, endothelial (TEK), embryonic lethal, abnormal vision, Drosophila-like 2 (ELAVL2) and also part of MTAP to the interferon, beta 1, fibroblast (IFNB1) gene. TPR presented hemizygotic deletions of practically the whole 9p21 region. A complex pattern was observed in the M28 cell line including homozygous deletions CDKN2A-CDKN2B and partial deletions of MTAP, as well as of the IFNB1 gene. Only the CPM cell line was affected by an amplification encompassing the region from kelch-like 9 (Drosophila) (KLHL9) to the myeloid/lymphoid or mixed-lineage leukemia (trithorax homolog, Drosophila); translocated to, 3 (MLLT3) gene. Homozygous deletion in melanoma cell lines was confirmed by PCR amplification of CDKN2A. In only three cell lines CDKN2A was amplified completely: M36, TPR and DB. The absence of amplification correlates with the homozygotic deletions detected with MLPA. Sequencing of PCR products revealed the presence of CDKN2A point mutations in two cases. Whereas the TPR cell line presented a proline-to-leucine substitution at codon 114 (c.341CC>TT) in the retained allele, the DB cell line carried nonsense mutation p.R58X (c.172 CC>TT) in homozygosis.
To further characterize the cell lines, additional MLPA probes were used to enable the analysis of DNA copy number losses and gains in chromosomal regions known to be relevant in melanoma pathogenesis. We identified a significant correlation between losses of mutY homolog (E. coli) (MUTYH) (1p33), gains of v-myc myelocytomatosis viral oncogene homolog (avian) (c-MYC) (8q24) and gains of cyclin-dependent kinase 6 (CDK6) (7q21.3) (p<0.05) in cell lines (45%) that harboured homozygous deletions in 9p21 restricted to CDKN2A and CDKN2B or MTAP but not affecting contiguous genes in the 9p21 region. 9p21 deletions in the CDKN2A region found by MLPA were confirmed by mutational analysis of CDKN2A (Table II).
No other chromosomal alteration was found to be associated with either BRAF or the 9p21 locus.
Discussion
In the present study eleven human cutaneous melanoma cell lines were genetically characterized for the presence of genetic aberration patterns.
The MAPK pathway is a key regulator of melanoma cell proliferation, as dysregulation of this pathway has been identified by gain of function mutations in NRAS or BRAF in approximately 13% and 60% of melanoma cell lines (2). In our study, 54% of melanoma cell lines (6/11) carried BRAF mutations. NRAS mutation was detected in one cell line with BRAFwt (p.Q61H mutation). As previously reported, we found a strong association between MC1R variants and BRAF mutations (16). It is believed that increased generation of reactive oxygen species in carriers of MC1R variants induces the A>T transversion characteristic of the common p.V600E BRAF mutation (16). We did not identify a genomic profile related to BRAF mutations.
A specific pattern of copy number variation losses of MUTYH and gains of c-MYC were significantly associated with gains of CDK6.
Frequent LOH on chromosome 1p indicates the existence of putative TSGs in this region that may predispose to tumor development or contribute to tumor progression.
The MUTYH protein is a base excision repair enzyme involved in repair of DNA damage caused by oxidative stress. Defects in this repair mechanism leaded to accumulation of mutations (17). Recently, Santonocito et al., has shown that MUTYH germline variants Y165C, G382D and V479F, known to be associated with adenomatous polyposis and colorectal cancer, are not associated with melanoma risk (18, 19). This finding is in contrast with ours. One explanation for these discrepant results is that MUTYH variants are located in exons 7, 13 and 15 whereas the MLPA probe for MUTYH gene hybridized to a region of exon 5. Furthermore, two putative melanoma susceptibility loci on chromosome 1p (1p22, 1p36) have been identified (20-23). To date, no candidate genes have been identified at either 1p22 or 1p36. Although not previously associated with melanoma, due its function and location, MUTYH could be a candidate melanoma gene.
c-MYC is a cell-cycle regulator that influences the activity of CDK4/CDK6 complex, which is critical for G1 progression and G1/S transition (24). Overexpression of c-MYC has been found in 40% of melanomas (25). Zhuang et al. (26) found that one of the major functions of c-MYC overexpression during melanoma progression is to continuously suppress senescence induced by mutated BRAF or NRAS. Regarding CDK6, large genomic amplifications of 7q21 in 59% of melanoma cell lines have been reported (27).
Losses of MUTYH and gains of c-MYC and CDK6 genes were found among cell lines carrying biallelic deletions of the 9p21 region including CDKN2A and CDKN2B or MTAP genes. Deletions associated with the loss of TSGs CDKN2A and CDKN2B have been frequently found in melanoma cell lines (4) and impaired the production of p15INK4B, p16INK4A and p14ARF proteins. Biallelic deletions in CDKN2A are associated with reduced overall survival (28). Given that, the MUTYH-MYC-CDK6 pattern is expected to be involved in tumor progression rather than initiation. MTAP gene is located approximately 100 kb telomeric of CDKN2A and therefore may also serve as a TSG. The loss of MTAP expression has been detected in melanoma cell lines and had an effect on tumour progression (29).
Other alterations in the 9p21 locus involving point mutations, large deletions and duplications were observed in cell lines without loss of MUTYH and gains of c-MYC (Table II). CDKN2A point mutations were detected in two cases. The TPR cell line showed hemizygotic deletions of the whole 9p21 region and carried a missense mutation (p.P114L) in the retained allele. The p16INK4A protein with this missense mutation loses the ability to bind CDK4/6 complexes (30). The DB cell line presented a normal DNA copy number 9p21 locus and a premature CDKN2A stop codon (p.R58X) in homozygosis, leading to production of non-functional p14ARF and p16INK4A. The possible mechanism leading to this homozygotic mutation would be loss of the chromosome harbouring the wild-type allele followed by chromosomal duplication or gene conversion on the other allele. Both mutations detected in CDKN2A are UV signature mutations characterized by C-T and CC-TT transitions at the dipyrimidine site, suggesting the probable UV aetiology of the mutations (31).
There is evidence that there is more than one TSG situated in the 9p21 region (5). Hemizygous deletions of tumour-suppressor IFNB1 and IFNW1 indicated that defects in the IFN gene cluster could act as a selective event for growth advantage. Surprisingly, in the CPM cell line, we detected duplication, including KIAA1354, IFNW1, IFNB1 and MLLT3 genes, to the homozygously deleted region. This amplification could be due to chromosomal copy number alterations
This study provides a comprehensive characterization of eleven melanoma cell lines using mutational analyses and MLPA. This approach offers a considerable advantage over other established techniques due to its low cost, the small quantities of DNA required and the possibility to test multiple loci in a single reaction. Limitations of this study include the small sample size. Despite the low number of samples, it was possible to confirm known mutations and their frequencies in human melanoma, such as BRAFV600E and NRAS.
Conclusion
In conclusion, a specific pattern of somatic alterations was described in genes that are involved in DNA repair (MUTYH) and in cell cycle regulation (c-MYC, CDK6, CDKN2A and CDKN2B). MC1R variants were associated with the BRAFV600E carrying group of melanoma cell lines. Among known genes implicated in melanoma, our study suggests the involvement of MUTYH in melanoma development. Further studies with larger numbers of samples are imperative as the role of MUTYH in melanoma needs to be clarified. Interestingly all the melanoma cell lines derived from visceral metastases belong to the BRAFV600E carrying group, while all cell lines from cutaneous metastases except one belong to the BRAFwt group.
Acknowledgements
This GenoMEL research has been supported by the European Commission under the 6th Framework Programme, Contract nr: LSHC-CT-2006-018702, by the grants 03/0019 and 06/0265 from the Fondo de Investigaciones Sanitarias. Zighereda Ogbah was supported by financial support provided by Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR) and Federico Simonetta by GenoMEL Exchange Programe fund by the European Commission under the 6th Framework Programme.
We also thank the principal investigators of the GenoMEL founding grants David Elder, Julia Newton Bishop and Nelleke Gruis for their support and to the leader of the exchange Programme Giovanna Bianchi-Scarrá and to Paola Ghiorzo for reviewing the manuscript. Thanks are also due to Irene Madrigal for MLPA technical support.
This work was developed in the Melanoma Unit of Hospital Clinic I Provincial de Barcelona, IDIBAPS.
Footnotes
-
This article is freely accessible online.
-
Conflicts of Interest
All authors declare no conflicts of interest or financial disclosures.
- Received February 8, 2012.
- Revision received March 12, 2012.
- Accepted March 13, 2012.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved