PTEN inactivation is rare in melanoma tumours but occurs frequently in melanoma cell lines.

Deletions detected in cytogenetic and loss of heterozygosity (LOH) studies indicate that at least one tumour suppressor gene maps to the long arm of chromosome 10. Previous deletion mapping studies have observed LOH on 10q in about 30% of melanomas analysed. The PTEN gene, mapping to chromosome band 10q23.3, encodes a protein with both lipid and protein phosphatase activity. Somatic mutations and deletions in have been detected in a variety of cell lines and tumours, including melanoma samples. We performed mutation analyses and extensive allelic loss studies to investigate the role this gene plays in melanoma pathogenesis. We found that a total of 34 out of 57 (60%) melanoma cell lines carried hemizygous deletions of chromosome 10q encompassing the PTEN locus. A further three cell lines carried smaller deletions excluding PTEN. Inactivation of both PTEN alleles by exon-specific homozygous deletion or mutation was observed in 13 out of 57 (23%) melanoma cell lines. The mutation spectrum observed does not indicate an important role for ultraviolet radiation in the genesis of these mutations, and evidence from three cell lines supports the acquisition of PTEN aberrations in culture. Ten out of 49 (20%) matched melanoma tumour/normal samples harboured hemizygous deletions of either the whole chromosome or most of the long arm. Mutations within were detected in only one of the 10 tumours demonstrating LOH at 10q23 that were analysed. These results suggest that PTEN inactivation may be important for the propagation of melanoma cells in culture, and that another chromosome 10 tumour suppressor gene may be important for melanoma pathogenesis.

Localization of multiple melanoma tumor-suppressor genes on chromosome 11 by use of homozygosity mapping-of-deletions analysis.

Loss-of-heterozygosity (LOH) studies have implicated one or more chromosome 11 tumor-suppressor gene(s) in the development of cutaneous melanoma as well as a variety of other forms of human cancer. In the present study, we have identified multiple independent critical regions on this chromosome by use of homozygosity mapping of deletions (HOMOD) analysis. This method of analysis involved the use of highly polymorphic microsatellite markers and statistics to identify regions of hemizygous deletion in unmatched melanoma cell line DNAs. Regions of loss were defined by the presence of an extended region of homozygosity (ERH) at > or =5 adjacent markers and having a statistical probability of < or =.001. Significant ERHs were similar in nature to deletions identified by LOH analyses performed on uncultured melanomas, although a higher frequency of loss (24 [60%] of 40 vs. 16 [34%] of 47) was observed in the cell lines. Overall, six small regions of overlapping deletions (SROs) were identified on chromosome 11 flanked by the markers D11S1338/D11S907 (11p13-15.5 [SRO1]), D11S1344/D11S11385 (11p11.2 [SRO2]), D11S917/D11S1886 (11q21-22.3 [SRO3]), D11S927/D11S4094 (11q23 [SRO4]), AFM210ve3/D11S990 (11q24 [SRO5]), and D11S1351/D11S4123 (11q24-25 [SRO6]). We propose that HOMOD analysis can be used as an adjunct to LOH analysis in the localization of tumor-suppressor genes.

Functional localization of a melanoma tumor suppressor gene to a small (< or = 2 Mb) region on 11q23.

We have previously demonstrated the existence of a melanoma tumor suppressor gene(s) on the long arm of chromosome 11 through suppression of tumorigenicity assays. Although loss of heterozygosity studies also support this finding, only a large critical region (44 cM) has been identified to date on 11q22-25. To further localize a tumor suppressor gene(s) within this region, we have now generated and characterized nine melanoma microcell hybrids, each retaining an introduced fragment of 11q. Of the nine hybrids, four were suppressed for tumor formation in nude mice, while five formed tumors at the same rate as the parental melanoma cell line (UACC 903). Molecular analysis of the hybrids with 118 microsatellite markers narrowed the location of a putative suppressor gene to a small (< or =2 Mb) candidate region on 11q23 between the markers D11S1786 and D11S2077 and within the larger region frequently deleted in melanoma tumors and cell lines. While multiple tumor suppressor genes are likely to reside on 11q22-25, the presence of this region in all four suppressed hybrids supports the simplest model that a single locus is responsible for the suppressed phenotype observed in UACC 903.

In vitro loss of heterozygosity targets the PTEN/MMAC1 gene in melanoma.

Gross genetic lesions of chromosome 10 occur in 30-50% of sporadic human melanomas. To test the functional significance of this observation, we have developed an in vitro loss of heterozygosity approach in which a wild-type chromosome 10 was transferred into melanoma cells, where there was selection for its breakage and regional deletion to relieve its growth suppressive effects. The overlap of these events was at band 10q23, the site of the recently isolated PTEN/MMAC1 tumor suppressor gene, suggesting it as a potential target. Although the gene was expressed in the parental cells, both of its chromosomal alleles contained truncating mutations. In vitro loss of heterozygosity resulted in loss of the chromosomally introduced wild-type PTEN/MMAC1, and ectopic expression of the gene caused cell growth suppression. Thus, this approach identified PTEN/MMAC1 as a target in malignant melanoma and may provide an alternative means to localizing tumor suppressor genes.

Virtually 100% of melanoma cell lines harbor alterations at the DNA level within CDKN2A, CDKN2B, or one of their downstream targets.

The cyclin-dependent kinase inhibitor 2A (CDKN2A), or p16INK4a, gene on 9p21 is important in the genesis of both familial and sporadic melanoma. Homozygous deletions and intragenic mutations of this gene have been identified in both melanoma cell lines and uncultured tumors, although the frequency of these alterations is higher in the cell lines. A proportion of melanoma cell lines and tumors without deletion/mutation of CDKN2A have also been determined to harbor transcriptionally inactive CDKN2A alleles or carry alterations in other components of the pathway through which p16INK4a acts on pRb to mediate cell cycle arrest. We sought to determine the frequency of these alternative events (in relationship to those that specifically inactivate CDKN2A) in a panel of 45 melanoma cell lines. Surprisingly, at the DNA level alone, 96% (43/45) of melanoma cell lines examined were found to be deleted/mutated/methylated for CDKN2A (34/45), homozygously deleted for CDKN2A's neighbor and homolog CDKN2B (6/45), and/or mutated/amplified for CDK4 (5/45). In two of these 43 cases, homozygous deletions of CDKN2A were detected along with a CDK4 mutation or amplification of the cyclin D1 (CCND1) gene. The latter discoveries were made in two of three cell lines which harbored extremely large (3-6 Mb) homozygous deletions on 9p21; all other homozygous deletions in similarly affected cell lines (N = 23) were confined to a region immediately surrounding the CDKN2A/CDKN2B loci. These results suggest that (1) only melanoma cells with alterations in this pathway can be propagated in culture, and (2) the homozygous deletions on 9p21 in the cell lines, which are also mutated/amplified for CDK4 or CCND1, could serve to target tumor suppressor genes other than CDKN2A.

Low frequency of p16/CDKN2A methylation in sporadic melanoma: comparative approaches for methylation analysis of primary tumors.

Methylation of the 5' CpG island of the p16 tumor suppressor gene represents one possible mechanism for inactivation of this cell cycle regulatory gene that is also a melanoma predisposition locus. We have investigated the potential contribution of somatic silencing of the p16 gene by DNA methylation in 30 cases of sporadic cutaneous melanoma. The methylation status of the 5' CpG island of p16 was initially determined by Southern analysis and then reevaluated (in a blinded manner) using methylation-specific PCR, methylation-sensitive single nucleotide primer extension, and bisulfite genomic sequencing. All methodologies yielded concordant results, and significant levels of methylation were observed in 3 of the 30 (10%) melanoma DNAs analyzed. Of the three tumors found to be methylated, two were also positive for LOH on 9p21 (where the p16 gene resides), implying that both p16 alleles were inactivated, one via deletion and the other via methylation-associated transcriptional silencing. The association between methylation and transcriptional silencing of p16 was also further supported by inducing p16 expression with a DNA demethylating agent (5-aza-2'-deoxycytidine) in a melanoma cell line known to harbor a methylated p16 allele. Although methylation-associated gene silencing does not represent a common mechanism for p16 inactivation in sporadic melanoma, our findings provide support that PCR-based techniques, such as methylation-specific PCR and methylation-sensitive single nucleotide primer extension, can be reliably used for the accurate detection and quantitation of aberrant levels of DNA methylation in tumor specimens.

Identification of a novel region of homozygous deletion on chromosome 9p in squamous cell carcinoma of the lung: the location of a putative tumor suppressor gene.

Cytogenetic and molecular studies have implied the presence of tumor suppressor genes (TSGs) on chromosome 9p that are critical in the development of lung and other cancers. The p16/CDKN2 gene, a cyclin dependent kinase inhibitor, is a well-defined TSG on 9p21. Although the frequency of mutations in the p16/CDKN2 gene has been detected in approximately 30% of non-small cell lung cancer, loss of heterozygosity on 9p has been observed in greater than 70% of non-small cell lung cancers. These and other deletion mapping studies have suggested the existence of additional TSGs on 9p. This study examined chromosome 9p for TSG loci by analyzing 23 squamous cell carcinomas of the lung with 21 microsatellite markers. Loss of heterozygosity was detected in all of the tumors, and homozygous deletions of the p16/ CDKN2 locus were observed in 6 of the 23 tumors (26%). In addition, a novel region of homozygous deletion was detected in six tumors (26%) at D9S126, approximately 2.5 cM proximal to p16/CDKN2. A single tumor contained a homozygous deletion at both the p16/CDKN2 locus and the D9S126 locus. The possibility of homozygous loss was confirmed by multiplex PCR using both the D9S126 marker and a chromosome 9p control marker. Fluorescence in situ hybridization analysis with P1 and cosmid probes containing D9S126 also confirmed these data. The minimum region of homozygous deletion was determined by testing markers immediately proximal and distal to the D9S126 region. The data identify a homozygous loss on the short arm of chromosome 9 suggesting the presence of a novel TSG locus, proximal to p16/CDKN2 and located between D9S265 and D9S259.

Analysis of the CDKN2A, CDKN2B and CDK4 genes in 48 Australian melanoma kindreds.

Germline mutations within the cyclin-dependent kinase inhibitor 2A (CDKN2A) gene and one of its targets, the cyclin dependent kinase 4 (CDK4) gene, have been identified in a proportion of melanoma kindreds. In the case of CDK4, only one specific mutation, resulting in the substitution of a cysteine for an arginine at codon 24 (R24C), has been found to be associated with melanoma. We have previously reported the identification of germline CDKN2A mutations in 7/18 Australian melanoma kindreds and the absence of the R24C CDK4 mutation in 21 families lacking evidence of a CDKN2A mutation. The current study represents an expansion of these efforts and includes a total of 48 melanoma families from Australia. All of these families have now been screened for mutations within CDKN2A and CDK4, as well as for mutations within the CDKN2A homolog and 9p21 neighbor, the CDKN2B gene, and the alternative exon 1 (E1beta) of CDKN2A. Families lacking CDKN2A mutations, but positive for a polymorphism(s) within this gene, were further evaluated to determine if their disease was associated with transcriptional silencing of one CDKN2A allele. Overall, CDKN2A mutations were detected in 3/30 (10%) of the new kindreds. Two of these mutations have been observed previously: a 24 bp duplication at the 5' end of the gene and a G to C transversion in exon 2 resulting in an M531 substitution. A novel G to A transition in exon 2, resulting in a D108N substitution was also detected. Combined with our previous findings, we have now detected germline CDKN2A mutations in 10/48 (21%) of our melanoma kindreds. In none of the 'CDKN2A-negative' families was melanoma found to segregate with either an untranscribed CDKN2A allele, an R24C CDK4 mutation, a CDKN2B mutation, or an E1beta mutation. The last three observations suggest that these other cell cycle control genes (or alternative gene products) are either not involved at all, or to any great extent, in melanoma predisposition.

Loss of the p16INK4a and p15INK4b genes, as well as neighboring 9p21 markers, in sporadic melanoma.

Although homozygous deletions of the cyclin-dependent kinase inhibitor 2 gene p16INK4a on 9p21 have been reported frequently in metastatic melanoma cell lines, and intragenic mutations within the p16INK4a gene have been detected in familial melanoma kindreds, specific targeting of this gene in the development of sporadic melanoma in vivo remains controversial. Southern analyses were performed in this study to initially assess the frequency of hemi- or homozygous losses of p16INK4a, as well as its neighboring family member, p15INK4b, and other candidate regions within 9p21, in sporadic melanoma. Overall, 22 of 40 (55%) uncultured sporadic melanoma DNAs were determined to harbor deletions of 1-11 markers/genes located on 9p21. This included 10 tumors (25%; 10 of 40) with homozygous deletions limited to either the p16INK4a gene only (20%; 2 of 10), both the p16INK4a and p15INK4b genes (10%; 1 of 10), another novel 9p21 gene, FB19 (10%; 1 of 10), or all three of these genes plus surrounding markers (60%; 6 of 10). In subsequent single-strand conformation polymorphism and sequencing analyses, intragenic mutations in the p16INK4a gene were also revealed in two (10%; 2 of 21) melanoma DNAs that retained one copy of this locus. By comparison, the frequency of pl6INK4a and p15INK4b homozygous deletions, as well as p16INK4a mutations, in melanoma cell lines (analyzed in parallel) was 2-3-fold higher at 61 (23 of 38) and 24% (9 of 38), respectively. These findings indicate that (a) p16INK4a is inactivated in vivo in over one-fourth (27.5%; 11 of 40) of sporadic melanomas; (b) mutation/deletion of p16INK4a may confer a selective growth advantage in vitro; and (c) other 9p21 tumor suppressor genes could be targeted during the development of melanoma.

Homozygous loss of the p15INK4B gene (and not the p16INK4 gene) during tumor progression in a sporadic melanoma patient.

Homozygous deletions of 9p21, including the cyclin-dependent kinase inhibitor genes p16INK4 and p15INK4B, have been reported frequently in melanoma (as well as other tumor) cell lines. Germline mutations within the p16INK4 gene have also been described in a proportion of familial melanoma kindreds, suggesting that p16INK4 is the 9p21 "melanoma" gene. We have previously concluded that deletion of this chromosomal region can occur early (before metastasis) and in vivo in sporadic melanoma due to the identification of identical hemizygous losses on 9p21 in six autologous melanoma cell lines established from an individual patient (DX). These related cell lines have now been used to evaluate the timing of deletion/mutation of the p16INK4 and p15INK4B genes during tumor progression in melanoma. Surprisingly, homozygous deletions of a < or = 200-kb region surrounding p15INK4B, but not p16INK4, were detected in all six cell lines. Furthermore, single strand conformation polymorphism and sequencing analysis of the remaining p16INK4 allele in each case revealed only one intragenic mutation (in DX-6), whereas Western analysis provided evidence that p16INK4 protein was expressed in all six instances. These findings, taken together with those generated on other unrelated melanoma tumors and cell lines, suggest that hemizygous loss (or haplo-insufficiency) of the p16INK4 gene may be enough to place a melanocyte on a tumor pathway, and/or that the p16INK4 gene is not the sole 9p21 locus targeted in sporadic melanoma.

Mutations of the CDKN2/p16INK4 gene in Australian melanoma kindreds.

The cyclin dependent kinase inhibitor 2 (CDKN2) gene on chromosome 9p21 is potentially involved in the genesis of many cancers and is currently under intense investigation as a possible melanoma susceptibility locus. We have analyzed 18 Australian melanoma kindreds for mutations within the coding and neighboring splice junction portions of the CDKN2 gene. In seven kindreds (including our six largest), CDKN2 mutations were found to segregate with the putative melanoma chromosome previously assigned by 9p haplotype analysis. These changes included the duplication of a 24 bp repeat, a deleted C residue resulting in the introduction of a premature stop codon, and four single basepair changes causing amino acid substitutions. Mutations segregated to 46 of 51 affected individuals in these seven kindreds, with three apparent sporadic cases in one family and one in each of another two families. Penetrance was variable (55-100%) among the different mutations. These data provide additional strong support that the CDKN2 gene is the chromosome 9p21 familial melanoma locus.

Loss of expression of the p16/cyclin-dependent kinase inhibitor 2 tumor suppressor gene in melanocytic lesions correlates with invasive stage of tumor progression.

Sporadic and familial malignant melanoma susceptibility has been linked to defects in the chromosomal region 9p21. Recently, a putative 9p21 tumor suppressor gene, the cyclin dependent kinase inhibitor 2 (CDKN2) or p16 gene, has been shown to be deleted, mutated, or rearranged in a high percentage of sporadic melanoma cell lines, as well as mutated in the germline of a proportion of familial melanoma patients. CDKN2 encodes a M(r) 16,000 protein (p16) that plays a key role in cell cycle control by binding to the cyclin-dependent kinase 4 enzyme and inhibiting its ability to phosphorylate critical substrates necessary for transition past the G1 phase of the cell cycle. Thus, mutations or deletions of the CDKN2 gene could result in abnormal proliferation via defective cell cycle control. The correlation of 9p21 cytogenetic and molecular alterations with the clinical stages of melanoma progression suggests that dysfunction of a gene within this chromosomal region is critical to the evolution of melanoma. However, it remains unclear whether this gene is the CDKN2 gene. If so, then loss of p16 is potentially an initiating or early event in melanoma progression. To address the issues of what is the potential involvement of the CDKN2 gene in sporadic melanoma and precisely when during the clinically evident stages of melanoma progression defects in CDKN2 occur, we have evaluated by immunohistochemistry the expression of p16 protein in 103 melanocytic lesions representing all stages in the progression of melanoma. Our results suggest that loss of p16 protein expression is (a) not necessary for tumor initiation in malignant melanoma because all melanomas in situ and the majority of primary invasive melanomas retain expression of this protein; and (b) potentially more related to invasiveness or the ability to metastasize, because 52% of primary invasive tumors and 72% of metastatic lesions show partial or complete loss of expression of p16.

Distinct deletions of chromosome 9p associated with melanoma versus glioma, lung cancer, and leukemia.

Deletions of DNA on chromosome 9p21-22 are frequently observed in cells derived from melanomas, gliomas, non-small cell lung cancers, and acute lymphoblastic leukemia. The minimal deletion shared by the latter three cancers extends from the interferon-alpha locus towards the centromere; its centromeric end is flanked by the gene encoding methylthioadenosine phosphorylase. We have determined that the telomeric end of the minimal homozygous deletion shared by two melanoma cell lines does not include the methylthioadenosine phosphorylase locus. Thus, a distinct region of DNA is lost in melanoma. The physical size of this region remains to be defined precisely, but it may extend over several million base pairs.

Molecular definition of a chromosome 9p21 germ-line deletion in a woman with multiple melanomas and a plexiform neurofibroma: implications for 9p tumor-suppressor gene(s).

Cutaneous malignant melanoma (CMM) is often familial, but the mode of inheritance and the chromosomal location of melanoma susceptibility locus are controversial. Identification of a 34-year-old woman with eight primary malignant melanomas, multiple atypical moles, and a de novo constitutional cytogenetic rearrangement involving chromosomes 5p and 9p suggested the presence of a melanoma predisposition gene at one of these locations. A high-resolution karyotype showed a partial deletion of a dark-staining Giemsa band, either 5p14 or 9p21. The patient was heterozygous for five 5p14 RFLPs. In situ hybridization with D9S3 indicated that this 9p21 marker was deleted. Gene dosage studies demonstrated the deletion of two more distal 9p21 markers, D9S126 and IFNA. In addition, she was hemizygous for the more proximal 9p21 short tandem-repeat polymorphism at D9S104. D9S18, D9S19, and D9S33 were retained, localizing the deletion to 9p21 between D9S19 on the proximal side and D9S33 on the distal side. Pulsed-field gel electrophoresis with D9S19 and D9S33 did not reveal any junction fragments in the patient's DNA. This germ-line deletion suggests that mutations in a 9p21 gene may initiate melanoma tumorigenesis.

Mutations in the neurofibromatosis 1 gene in sporadic malignant melanoma cell lines.

Neurofibromatosis type 1 (NF1) is a common autosomal dominant disorder characterized by progressive and variable involvement of tissues predominantly derived from the neural crest and a predisposition toward malignancies. The NF1 gene encodes neurofibromin, a GTPase-activating protein containing a GAP-related domain (NF1-GRD) that is capable of down-regulating ras by stimulating its intrinsic GTPase activity. We report a homozygous deletion of most of NF1 in one of eight malignant melanoma cell lines leading to loss of detectable mRNA and protein, as well as the apparent absence of protein and mRNA in another melanoma. This data suggests that NF1 can function as a tumour suppressor gene in the development or progression of malignant melanoma.

Molecular genetics of human malignant melanoma.

Due to a variety of known and unknown control mechanisms, the human genome is remarkably stable when compared to most other species. The long latency periods of most solid tumors, during which the cell undergoes malignant transformation, are presumably due to this stability. The molecular basis responsible for the induction of genetic instability and the resultant biological characteristics manifest in tumor populations is not well understood. The discovery of both oncogenes and tumor suppressor genes, however, has placed the phenomenon of human genome stability on a more solid conceptual footing. These types of genes clearly place multiple barriers to oncogenic transformation, and traversing these barriers apparently requires both time and the accumulation of genetic defects that cannot be corrected. The evolution of neoplasias can, therefore, be predicted to be due to: (1) consistent and progressive loss of tumor suppressor genes; (2) gene amplification, resulting in the over-expression of proteins that aid in tumor progression; (3) gene mutation, which alters the orderly biochemistry of the normal cell; (4) genes that allow a cell like the melanocyte to escape the confining nature of the epidermis and to invade through the dermis into the circulatory and lymphatic systems in order to disseminate itself to other organs (e.g., proteolytic enzymes, enzyme inhibitors, integrins, metastases genes, chemotactic factors etc.); (5) factors, perhaps such as TGF beta 2, that may impact negatively on MHC antigens and confuse host defense mechanisms; and (6) S.O.S.-type genes, which may be expressed as a direct response to the accumulating damage in an attempt to correct the damage, but that may then become part of the problem instead of the solution. The extraordinary plasticity and instability of the genome of a melanoma cell suggests an inordinate amount of genetic flux. In addition to activating and inactivating various genes, this constant shuffling and rearranging of the genome in neoplasms such as MM may be constantly altering gene dose. Cytogenetic and molecular biological studies have been the Rosetta stone for understanding the etiological relevant genetic events in human cancers. Genetic alterations fundamental to the pathology of MM have begun to be defined. Studies designed to understand these perturbations at the biochemical and organismic level are underway.(ABSTRACT TRUNCATED AT 400 WORDS)