Cytogenetic and loss of heterozygosity (LOH) studies have long indicated the presence of a tumor suppressor gene (TSG) on 9p involved in the development of melanoma. Although LOH at 9p has been reported in approximately 60% of melanoma tumors, only 5–10% of these tumors have been shown to carry CDKN2A mutations, raising the possibility that another TSG involved in melanoma maps to chromosome 9p. To investigate this possibility, a panel of 37 melanomas derived from 35 individuals was analyzed for CDKN2A mutations by single-strand conformation polymorphism analysis and sequencing. The melanoma samples were then typed for 15 markers that map to 9p13–24 to investigate LOH trends in this region. In those tumors demonstrating retention of heterozygosity at markers flanking CDKN2A and LOH on one or both sides of the gene,multiplex microsatellite PCR was performed to rule out homozygous deletion of the region encompassing CDKN2A. CDKN2Amutations were found in tumors from 5 patients [5 (14%) of 35], 4 of which demonstrated LOH across the entire region examined. The remaining tumor with no observed LOH carried two point mutations, one on each allele. Although LOH was identified at one or more markers in 22 (59%)of 37 melanoma tumors corresponding to 20 (57%) of 35 individuals,only 11 tumors from 9 individuals [9 (26%) of 35] demonstrated LOH at D9S942 and D9S1748, the markers closest to CDKN2A. Of the remaining 11 tumors with LOH, 9 demonstrated LOH at two or more contiguous markers either centromeric and/or telomeric to CDKN2A while retaining heterozygosity at several markers adjacent to CDKN2A. Multiplex PCR revealed one tumor carried a homozygous deletion extending from D9S1748 to the IFN-αlocus. In the remaining eight tumors, multiplex PCR demonstrated that the observed heterozygosity was not attributable to homozygous deletion and stromal contamination at D9S1748, D9S942, or D9S974, as measured by comparative amplification strengths, which indicates that retention of heterozygosity with flanking LOH does not always indicate a homozygous deletion. This report supports the conclusions of previous studies that at least two TSGs involved in melanoma development in addition to CDKN2A may reside on chromosome 9p.

Cutaneous malignant melanoma is an often fatal form of skin cancer the incidence of which is rising worldwide. In particular,Queensland has the highest incidence of melanoma with a lifetime incidence in men and women of 1/12 and 1/17, respectively(1). The chromosomal region 9p21–22 was thought to harbor a TSG2involved in melanoma pathogenesis because of the high frequency of cytogenetic alterations and LOH observed in uncultured tumors. The CDKN2A gene mapping to chromosome band 9p21 was cloned as a candidate TSG involved in both melanoma predisposition and the genesis of many sporadic tumor types including melanoma. The presence of germ-line mutations in approximately one-half of those melanoma families that demonstrate linkage to this region confirms CDKN2A as a major gene involved in melanoma predisposition. HDs and mutations in CDKN2A have been reported in cell lines derived from a wide variety of tumor types (reviewed in Refs. 2, 3, 4, 5); however, the frequency of aberrations detected in uncultured tumors is much lower, even in those tumors displaying LOH on chromosome 9p21–22 (5, 6, 7, 8). Because p16 has been shown to play a role in replicative senescence (reviewed in Ref.9), it is likely that a proportion of aberrations are late-stage events that are selected by cell culture.

Although the high rate of CDKN2A mutations in immortalized cell lines is probably attributable to the acquisition of mutations during culture and/or the preferential establishment of cell lines from tumors carrying CDKN2A mutations, this still leaves the paradoxically low frequency of CDKN2A mutations in those tumors demonstrating 9p21 LOH. A review of the relevant literature reveals that although LOH on chromosome 9p is observed in 30–70% of all uncultured melanomas, mutations in CDKN2A have been observed only in 5–10% of samples. Several explanations for this paradox have been put forward including the following: failure to identify 100% of the coding region mutations because of technical limitations and/or their presence outside the coding region;transcriptional silencing of CDKN2A by methylation of the CpG island spanning the promoter and exon 1α of the gene; the inactivation of CDKN2A by HD; and the presence of another TSG(s) within this region that accounts for 9p21 LOH in a subset of tumors. It is the difficulty in assessing the presence of HDs in tumors presenting with LOH on chromosome 9p21 that lies at the foundation of some of the debate surrounding the possibility of another TSG mapping to this region.

Before the cloning of the CDKN2A gene, Cairns et al.(10) reported the detection of HDs at IFNA on 9p21 by the ROH at this marker flanked by regions of LOH. Comparative multiplex PCR was used to demonstrate that this apparent ROH was attributable to an amplification signal originating from a small number of stromal cells present in those tumor samples. This evidence was taken to represent HD of this locus. With the isolation of new polymorphic markers mapping either side of the CDKN2A locus, Cairns et al. reexamined a panel of primary bladder tumors in which a low frequency of CDKN2Amutations had been reported previously (8). HDs were reported in a total of 126 (71%) of 178 primary bladder tumors demonstrating 9p21 LOH (11), and the authors concluded that this argued against the presence of another TSG mapping to chromosome 9p21. They scored the tumors as carrying HDs when one or more markers near CDKN2A demonstrated apparent ROH and flanking markers exhibited LOH. However, in contrast to their original paper, in this study, the authors did not report the use of multiplex PCR, without which one cannot conclusively distinguish between apparent ROH attributable to a HD and real ROH. Moreover, examination of the figure (Fig. 1) supplied in this article, casts some doubt on their conclusions. In Fig. 1b there is very little residual signal observed from the lost allele (markers IFNA and D9S171) indicating very little stromal contamination in this tumor; however, at those markers demonstrating ROH (PKY2 and PKY3) the signals obtained from the tumor show no reduction in intensity compared with the normal lane. This is surprising if the same amount of DNA and the same number of cycles were used to amplify all markers. Unwittingly, this article(11) is at the center of the controversy regarding additional TSG loci mapping to chromosome 9p, because numerous subsequent studies have reported ROH with flanking LOH at markers surrounding CDKN2A and have interpreted this to represent the presence of HDs, without performing either multiplex PCR or Southern blotting to verify the conclusions.

Several previous LOH studies in melanoma tumors have indicated that there may be at least two additional TSG loci mapping to chromosome 9p,one telomeric to IFNA and one centromeric to D9S171(12, 13, 14, 15). The low frequency of CDKN2A mutations in other tumor types with a high rate of LOH on chromosome 9p, e.g., bladder cancer and NSCLC, has also prompted a renewal of LOH studies in a wide variety of tumors to examine whether CDKN2A is the major target of LOH on chromosome 9p in these cancers (16, 17, 18, 19, 20, 21, 22, 23).

We have analyzed 37 melanoma tumors from 35 patients for mutations in the CDKN2A gene and for LOH at 9p13–24. Multiplex amplification of two and/or three microsatellite markers was performed to ascertain HDs at those markers flanking CDKN2Ademonstrating apparent ROH where markers on one or both sides of the gene demonstrated LOH. This report provides further evidence for two TSG loci, in addition to CDKN2A, mapping to chromosome 9p21–22 that may be involved in the progression of melanoma (14, 15).

Patients and Tumor Samples.

We have studied 37 histologically confirmed melanoma samples from 35 unrelated cutaneous malignant melanoma patients. DNA was extracted as described in Walker et al.(24). Lymphoblastoid cell lines were established from each individual as a source of constitutional DNA. Of the 37 samples, 34 were derived from melanoma metastasis, and the remaining 3 samples were obtained from primary tumors. Two metastases were obtained from patients 40867–01 and 40438–01.

SSCP Analysis of CDKN2A.

SSCP analysis was used to screen for CDKN2A mutations in all of the samples using primer sequences taken from Hussussian et al.(25). PCR reactions included 75 ng of template, 15 pmol of each primer, 0.15 units Taq, 10% DMSO, 1.5 mm final MgCl2concentration, and 1 μCi of [α-32P]dCTP in a 15-μl total volume. Amplification was carried out using a touchdown PCR protocol in which the annealing temperature was dropped 1° every 2 cycles from 65°C to 56°C, followed by an additional 15 cycles with an annealing temperature of 55°C. For all cycles, 60 s annealing and extension times and a 45 s denaturation time were used. Resultant PCR products were heated to 95°C for 5 min with two volumes of stop dye, snap-cooled on ice, and then separated on a 0.5× MDE (FMC Inc.) matrix in 0.6×Tris/borate/EDTA. Gels were routinely run at 9 W for 6–8 h or 5 W overnight. Multiple DNA samples with known mutations(26) were included in every SSCP experiment as positive controls.

Sequencing.

CDKN2A amplification for sequencing was performed under the same conditions, and the PCR products were purified using a gel extraction kit (Qiagen, Hilden, Germany). Applied Biosystems Incorporated (ABI) dye terminator sequencing kits were used according to the manufacturer’s specifications and reactions were run on an ABI 377 automated sequencer (Applied Biosystems). Sequences generated from each tumor were aligned with Sequencher (Gene Codes Corporation) for analysis, in addition to being scanned manually for heterozygous peaks not detected by the sequencing analysis software.

Microsatellite Analysis.

A panel of 11 microsatellites mapping from 9p13 to 9p24 was typed in all of the samples to detect LOH (Table 3). Four additional microsatellite markers (D9S168, D9S736, D9S1604, and D9S15) were amplified in those samples demonstrating LOH, to further delineate potential borders of loss. All of the primer sequences were obtained from the Genome Database.3 PCR reactions, cycling conditions, and electrophoresis were carried out as described previously (26) with the exception of D9S1748, which was amplified using the conditions described for D9S974. For all of the equivocal results, PCRs were repeated, the samples were rerun several times, and the autoradiographs independently scored by at least two co-authors. In addition,various exposures of all of the autoradiographs were taken to ensure that overexposure of a given lane did not obscure interpretation of the signal. Multiplex amplification of two and/or three microsatellite markers was performed to ascertain for HDs at markers flanking CDKN2A in eight samples. In these cases, D9S974, D9S942, and D9S1748 were multiplexed with another marker mapping to chromosome 9p21 and demonstrating LOH in that sample, so that the level of stromal contamination could be assessed. Markers D9S942 and D9S1748 were multiplexed with D9S171 and/or D9S169, and D9S974 was multiplexed with D9S162. LOH can result from either deletion of chromosomal regions or mitotic recombination. In the latter case, LOH manifests as both a decrease in signal from the lost allele and an increase in signal from the retained allele (allelic imbalance). This adds complexity to the interpretation of data generated from duplex PCRs, especially when one marker demonstrates LOH and the other marker may or may not be homozygously deleted. For this reason, triplex amplification with an additional marker mapping to chromosome 10q (either D10S539 or D10S221)was used as an additional control for equal loading. Because the 10q LOH status of these tumors was known, attempts were made to choose a control marker at which the tumor was heterozygous; however, this was not always possible, and, in some cases, the tumor was known to demonstrate LOH at the control marker. In duplex and triplex PCRs, only 27 cycles were used.

Generally, PCR reactions consisted of 100 ng template DNA; 5–20 pmol each primer; 200 nm dATP, dGTP, dTTP; 2 nmdCTP; and 1 μCi of [α-32P]dCTP in a final volume of 10 μl with 0.5 units of Amplitaq Gold using 1.5 mm MgCl2. PCR conditions consisted of an initial denaturation step at 94°C for 10 min followed by 30 cycles at 94°C for 45 s, 55°C for 45 s, and 72°C for 45 s, with a final 5 min extension at 72°C.

CDKN2A Mutations.

SSCP analysis revealed aberrant banding patterns in 7 of 37 tumors,corresponding to 5 of 35 individuals (Fig. 1). These samples were sequenced to reveal six different CDKN2A mutations in 5 of 35 (14%) cases (Fig. 2; Table 1). All of the six mutations result in amino acid substitutions or premature stop codons within CDKN2A. In patient 40861, a CC→TT mutation was detected in both of the metastases taken from the left and right axillary nodes, and in patient 40438, a CC→TT mutation was identified in metastases removed from the submandibular and preauricular nodes, which indicated that in these two patients, the CDKN2A mutations were probably present in the primary tumors. The most common mechanism of inactivating both copies of CDKN2A is via HD, or hemizygous deletion with a corresponding mutation on the remaining allele. Surprisingly, patient 41070 was found to carry two mutations within exon 2 of CDKN2A. No LOH was detectable in this tumor (results not shown) and sequencing of CDKN2A in the germ line of this individual revealed no mutation (results not shown). Subcloning and sequencing confirmed that this tumor had acquired two CDKN2Asomatic mutations, one on each allele (results not shown).

It is notable that the mutations found in this panel of melanoma tumors further support the importance of UV radiation in the genesis of melanoma, because five of six mutations were CC→TT tandem substitutions or C→T transitions at the 3′ pyrimidine of a dipyrimidine pair. In all of the tumors carrying CDKN2Amutations, the original site of the primary lesion was a sun-exposed site, e.g., forehead, back of neck, back, leg, or arm.

Microsatellite Analysis to Detect LOH.

An example of LOH for each marker analyzed is presented in Fig. 3, and the results for those tumors demonstrating LOH are summarized in Table 2. LOH was identified at one or more markers in 22 (57%) of 37 of the melanomas; however, only 10 (24%) of 37 tumors demonstrated LOH at D9S942 and/or D9S1748, the markers located closest to the CDKN2A gene. Of the remaining 12 tumors with LOH: 1 tumor was noninformative at D9S942 and D9S1748 but demonstrated LOH at every informative marker analyzed; 7 tumors demonstrated LOH at one or more markers either centromeric and/or telomeric to CDKN2A but demonstrated apparent ROH at several markers adjacent to CDKN2A; 2 tumors demonstrated ROH at D9S942 with flanking LOH at several centromeric markers; and the remaining 2 tumors demonstrated LOH at one or more noncontiguous markers. Of the tumors carrying CDKN2A mutations, six of seven demonstrated LOH across the whole region examined, whereas the seventh had no LOH but carried two point mutations (detailed above). Replication errors were observed in tumors from four patients, although replication error at multiple markers was not observed in any individual. The apparent ROH of markers at or near CDKN2A with LOH detected on either one or both sides of the gene indicated that either (a)these tumors are homozygously deleted for this region or (b)the ROH surrounding CDKN2A is real. In the latter case this would suggest that the LOH observed targeted other regions of 9p, in turn suggesting the presence of one or more additional TSGs. We,therefore, used multiplex amplification of microsatellite markers to distinguish between these possibilities.

Multiplex PCR to Detect HD at Markers Flanking CDKN2A.

For each tumor sample demonstrating apparent ROH at D9S1748, D9S942,and/or D9S974 and LOH at centromeric and/or telomeric markers, at least one marker flanking CDKN2A was multiplexed with another marker in either of the other 9p deleted regions so that the level of stromal contamination could be assessed. In addition, triplex amplification with an additional marker mapping to chromosome 10 was also performed. Using this technique, we demonstrated that the apparent ROH observed in sample 40977–01 was indeed attributable to HD of these markers (Fig. 4). This deletion extended from IFNA to D9S1748 (results not shown). However the remaining seven tumors analyzed were not homozygously deleted for any of the markers mapping within or near CDKN2A (D9S1748, D9S974, or D9S942).

CDKN2A Aberrations.

We detected a relatively low proportion of CDKN2A mutations in uncultured melanomas (7 of 37 melanoma tumors corresponding to 5(14%) of 35 individuals). Table 3 summarizes the results of all of the CDKN2A mutation analyses on uncultured melanomas reported to date. The total frequency of CDKN2A mutations detected in all of the melanomas tested is ∼8%. This is likely to be an underestimate because not every study screened the entire gene for mutations. The overall frequency of CDKN2A mutations is slightly higher in primary tumors compared with metastases (10 versus 6%, respectively). It has been hypothesized that the close proximity of the alternate exon 1β of CDKN2A means that, if both p16 and p14ARF are inactivated by HD, it could result in an additional growth advantage, which may explain the slightly higher frequency of point mutations in primary tumors (27). Although it is likely that aberration of CDKN2A is likely to occur by a combination of mutation, HD, or transcriptional silencing, in those studies that have combined these analyses with 9p21 LOH, these mechanisms of inactivation together still account only for just under one-half of all melanomas demonstrating LOH on chromosome 9p21.

LOH on Chromosome 9p21: Evidence for Additional TSG Loci Mapping to Chromosome 9p21–24 Involved in Melanoma.

This study supports the data presented by Ruiz et al.(15) and several earlier studies that indicated the presence of possibly two or more additional TSGs on 9p21–22 involved in melanoma development. Fig. 5 summarizes the data obtained from the present and previous studies, which indicate two regions of loss in addition to CDKN2A. In many studies, a SRO that specifically targets CDKN2A and excludes these additional loci is not evident, because many tumors demonstrate LOH over a large portion of 9p21. Nevertheless, there is remarkable concordance among studies regarding the location of the additional TSG loci. Holland et al.(12) identified a very small additional SRO telomeric to CDKN2A, mapping between D9S157 and IFNA; however, subsequent studies (Refs. 14, 15, and the present study) have failed to confirm this distal border. Given that multiplex PCR was not performed in the former study,D9S157 cannot be confidently assumed to mark the distal border of this region (Fig. 5, Region 1). The current study used D9S168, by far the most telomeric marker analyzed in any melanoma study to date. We still observe LOH at this marker, which suggests the possibility that the candidate region to which a putative TSG maps may extend into 9p23–24, far beyond D9S157. Further localization of the putative TSG telomeric to CDKN2A requires analysis of additional markers mapping to 9p23–24 in the various tumor panels. The SRO centromeric to CDKN2A (Fig. 5, Region 2) has been localized to between D9S265 and D9S161 by Ruiz et al.(15). Although our results cannot further refine this SRO, they overlap with this region and support the location of an additional TSG centromeric to CDKN2A. It should be noted that the original HDs identified on chromosome 9p21 in melanoma cell lines by Fountain et al.(28) encompassed the D9S126 locus and the polymorphic RFLP marker S3, both of which map within region 2 (Fig. 5). Small HDs affecting CDKN2A cannot be ruled out in the five tumors in this study that demonstrate LOH at all of the 9p markers but that do not carry detectable CDKN2A mutations. Nevertheless, it is tempting to speculate that these tumors may also support the presence of additional TSG loci mapping to this region.

Evidence for Additional TSG Loci Mapping to Chromosome 9p21–24 Involved in Other Cancers.

A review of the literature reveals that LOH studies in a wide range of other cancers also suggest that multiple TSGs may map to chromosome 9p. These studies are summarized in Fig. 6. If a TSG is involved in multiple tissue types, comparing LOH data from different tumor types can help to further localize the candidate region. Conversely, given the myriad of genes potentially involved in the complex processes of tumor initiation, evasion of apoptosis, and metastasis, it is likely that multiple TSGs involved in more than one tumor type will reside on any one chromosome arm.

Nevertheless, a comparison of the candidate 9p TSG loci identified in other tumor types with the candidate TSG loci involved in melanoma reveals several interesting points. The potential melanoma TSG, mapping to Region 2 (Fig. 5) overlaps with a candidate TSG locus,region D (Fig. 6), which is common to NSCLC, SCLC, bladder cancer, HNSCC, and pituitary adenomas, which indicates that this region may harbor a TSG involved in a wide range of tissue types. The presence of HDs at D9S126 with heterozygosity at flanking markers in NSCLC has potentially delineated this SRO to a 1 Mb region between D9S265 and D9S259 (16). The Hel-N1(human elav-like neuronal protein 1) gene encodes a neural-specific RNA-binding protein that is expressed in SCLC, maps to within 100 kb of the D9S126 marker (29), and has been investigated as a candidate TSG. Mutation analyses have been performed in a panel of lung tumors and cell lines, as well as a panel of childhood acute lymphocytic leukemia samples; however, no somatic mutations have been found (29, 30), which makes it unlikely that Hel-N1 is the target of the 9p LOH centromeric to CDKN2A.

In contrast, the SRO identified in melanomas as Region 1,spans 3 smaller delineated regions, one of which is common to NSCLC,SCLC, HNSCC, pituitary adenomas, and oral carcinomas, the other two regions having only been identified in breast carcinomas. It should be noted that the small SRO defined by Holland et al.(12) is identical to the SRO identified by Farrell et al.(18) in pituitary adenomas and overlaps with Region C (Fig. 6). It remains to be seen whether an analysis of additional markers mapping to chromosome 9p21–24 will reveal more than one candidate region in melanoma. If so,this may explain the more distal LOH observed in this study. The only detailed LOH study using markers mapping to 9p23–24 has been performed in breast carcinoma samples, and because this region is relatively underinvestigated, it is unknown at present whether similar analyses in a wide range of tumor types will support the presence of additional TSGs in this region.

Functional Evidence for Additional TSG Loci Mapping to 9p21.

The reintroduction of CDKN2A into a variety of cell types results in growth suppression. To investigate chromosome 9p for the presence of additional TSGs, Parris et al.(31)have used microcell-mediated chromosome transfer, harboring different microdeletions that remove CDKN2A, into the cell line UACC-903. The first variant, designated chromosome 9a, carried a deletion encompassing only CDKN2A, CDKN2B, and p14ARF. The second variant, designated chromosome 9b, harbored a much larger deletion extending from IFNA to D9S171. The hybrids constructed with chromosome 9a demonstrated an increased ability to suppress growth in soft agar and suppressed both tumor formation and metastasis in nude mice to a greater extent than the hybrids constructed with chromosome 9b. These data suggest the presence of a TSG in addition to CDKN2A that is inactivated by the microdeletion extending from IFNA to D9S171. These data contrast with the candidate regions delineated by LOH studies (Fig. 5), which map just outside this region; however, one possibility is that the deletion disrupts sequences important for the regulation of one of the putative TSGs. Nevertheless, hybrids containing chromosome 9b did result in suppression of growth in soft agar and a statistically significant reduction in tumor volume compared with the control when injected into nude mice. Together, these data provide functional evidence for one or more TSG loci mapping to chromosome 9p in addition to CDKN2A. Moreover, the presence of hybrids, which have lost their ability to suppress tumor formation because of the acquisition of small regions of HD, may provide a resource with which to map the location of these novel TSGs.

In conclusion, the LOH data in this study support previous reports that additional TSG loci involved in melanoma progression map both centromeric and telomeric to CDKN2A. Furthermore, they indicate the importance of multiplex PCR to distinguish possible HDs manifesting as apparent ROH from real ROH.

Fig. 1.

Examples of aberrant shifts detected in SSCP analysis. SSCP analysis was performed on three overlapping subfragments of CDKN2A exon 2 (2A, 2B, 2C). N, normal; T,tumor.

Fig. 1.

Examples of aberrant shifts detected in SSCP analysis. SSCP analysis was performed on three overlapping subfragments of CDKN2A exon 2 (2A, 2B, 2C). N, normal; T,tumor.

Close modal
Fig. 2.

Sequence chromatograms demonstrating mutations in melanoma tumors. The upper panel in each case, the tumor; the lower panel, the wild-type sequence; arrows, the position of mutated bases. The sequence traces correspond to the following tumors: a, 41125; b, 40438; c,40861; d, 40597; e and f,41070.

Fig. 2.

Sequence chromatograms demonstrating mutations in melanoma tumors. The upper panel in each case, the tumor; the lower panel, the wild-type sequence; arrows, the position of mutated bases. The sequence traces correspond to the following tumors: a, 41125; b, 40438; c,40861; d, 40597; e and f,41070.

Close modal
Fig. 3.

Representative autoradiographs demonstrating LOH for each of the microsatellite markers analyzed on chromosome 9. Top of each panel, patient number; bottom of each panel, microsatellite marker; N, normal; T, tumor; arrows, the alleles that were scored as lost in each case.

Fig. 3.

Representative autoradiographs demonstrating LOH for each of the microsatellite markers analyzed on chromosome 9. Top of each panel, patient number; bottom of each panel, microsatellite marker; N, normal; T, tumor; arrows, the alleles that were scored as lost in each case.

Close modal
Fig. 4.

Representative autoradiograph demonstrating example of multiplex PCR to assess HD at D9S1748. At the top,patient numbers; N, normal; T, tumor. Triplex PCR with D9S171 (•), D9S1748 (▪), and D10S539 (○). Arrows, either LOH or, in sample 40977–01 at marker D9S1748, HD.

Fig. 4.

Representative autoradiograph demonstrating example of multiplex PCR to assess HD at D9S1748. At the top,patient numbers; N, normal; T, tumor. Triplex PCR with D9S171 (•), D9S1748 (▪), and D10S539 (○). Arrows, either LOH or, in sample 40977–01 at marker D9S1748, HD.

Close modal
Fig. 5.

Summary of LOH studies detecting multiple regions of loss on chromosome 9p in uncultured melanomas. Gray shading,the SRO for each study. Dark gray shading, the actual markers used in the study demonstrating LOH (+/−). Light gray shading, the markers not analyzed in that particular study but which fall within the region delineated by the flanking markers used to define the SRO. +/+, the marker that places a proximal or distal border on the SRO. Black bars to the right, the two SROs and the position of CDKN2A. The order of the microsatellite markers has been deduced from the maps provided in Refs. (in parentheses at top and in Fig. 6).

Fig. 5.

Summary of LOH studies detecting multiple regions of loss on chromosome 9p in uncultured melanomas. Gray shading,the SRO for each study. Dark gray shading, the actual markers used in the study demonstrating LOH (+/−). Light gray shading, the markers not analyzed in that particular study but which fall within the region delineated by the flanking markers used to define the SRO. +/+, the marker that places a proximal or distal border on the SRO. Black bars to the right, the two SROs and the position of CDKN2A. The order of the microsatellite markers has been deduced from the maps provided in Refs. (in parentheses at top and in Fig. 6).

Close modal
Fig. 6.

Summary of SROs on chromosome 9p in multiple tumor types. Gray shading, the SRO for each study. Dark gray shading, the actual markers used in the study demonstrating LOH (+/−). Light gray shading, the markers not analyzed in that particular study, but which fall within the region delineated by the flanking markers used to define the SRO.+/+, the marker that places a proximal or distal border on the SRO. Black bars to the right, the SROs and the position of CDKN2A. The order of the microsatellite markers has been deduced from the maps provided in Refs. (in parentheses at top and in Fig. 5). In the study by Weist et al.(16) the SRO represents HD present at D9S126.

Fig. 6.

Summary of SROs on chromosome 9p in multiple tumor types. Gray shading, the SRO for each study. Dark gray shading, the actual markers used in the study demonstrating LOH (+/−). Light gray shading, the markers not analyzed in that particular study, but which fall within the region delineated by the flanking markers used to define the SRO.+/+, the marker that places a proximal or distal border on the SRO. Black bars to the right, the SROs and the position of CDKN2A. The order of the microsatellite markers has been deduced from the maps provided in Refs. (in parentheses at top and in Fig. 5). In the study by Weist et al.(16) the SRO represents HD present at D9S126.

Close modal

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.

2

The abbreviations used are: TSG, tumor suppressor gene; HD, homozygous deletion; LOH, loss of heterozygosity;IFNA, IFN α locus; ROH, retention of heterozygosity; SCLC, small cell lung cancer; NSCLC, non-SCLC; SSCP, single-strand conformation polymorphism; smallest region of overlap; HNSCC, head and neck squamous cell carcinoma.

3

Internet address: http://gdbwww.gdb.org/.

Table 1

Summary of CDKN2A mutations detected in uncultured melanomas

Patient no.BaseaCodonSequencebBase changePredicted aac change on p16Predicted aac change on p14ARF
41125-01 238 80 caccCgacc C→T Arg Silent 
41070-01 204 69 cgcgGagcc G→T Glu→stop Gly→Val 
 299/300 100 gggCCgggg CC→TT Ala→Val Arg→Trp 
40597-01 330/331 110/111 ctcgGGccg GG→AA Trp→stop  
     Gly→Ser Gly→Lys 
40861-01 341/342 114 ctgcCCgtg CC→TT Pro→Leu Arg→Cys 
40438-01 341 114 ctgcCcgtg C→T Pro→Leu Silent 
Patient no.BaseaCodonSequencebBase changePredicted aac change on p16Predicted aac change on p14ARF
41125-01 238 80 caccCgacc C→T Arg Silent 
41070-01 204 69 cgcgGagcc G→T Glu→stop Gly→Val 
 299/300 100 gggCCgggg CC→TT Ala→Val Arg→Trp 
40597-01 330/331 110/111 ctcgGGccg GG→AA Trp→stop  
     Gly→Ser Gly→Lys 
40861-01 341/342 114 ctgcCCgtg CC→TT Pro→Leu Arg→Cys 
40438-01 341 114 ctgcCcgtg C→T Pro→Leu Silent 
a

Nucleotide numbering starts at the A of the initiation codon.

b

The sequence is written 5′→3′ for the coding strand. Boldface letters, bases that were mutated.

c

aa, amino acid.

Table 2

Summary of LOH analysis in a panel of 37 melanoma tumorsa

Summary of LOH analysis in a panel of 37 melanoma tumorsa
Summary of LOH analysis in a panel of 37 melanoma tumorsa
a

Gray shading, minimum regions of LOH.

Table 3

Summary of literature reporting mutation, methylation, and homozygous deletion of CDKN2A in uncultured cutaneous melanomasa

ReferenceFrequency of CDKN2A aberration
9p21 LOHMutationMethylationHomozygous deletionb
(6)    2/4 (50%) 1° tumors  0/13 1° tumors   
(32)    5/34 (15%) 1° tumors   
(33)    1/61 (2%) metastases   7/61 (11%) metastases (Southern blotting) 
(34)   12/40 (30%) metastases  0/3 1° tumors 2/21 (5%) metastases   9/40 (22%) metastases (Southern blotting) 
(14)   40/64 (63%) metastases  2/66 (3%) metastases   8/66 (12%) metastases (multiplex PCR) 
(35)   14/26 (54%) 1° tumors  1/26 (4%) 1° tumors   
(36)    3/25 (12%) metastases   0/25 metastases (multiplex PCR) 
(27)    4/12 (33%) 1° tumors 0/9 metastasesc   
(37)    0/6 1° tumors  1/6 (16%) metastases    
(38)   15/27 (55%)  3/21 (14%) metastases 0/1° tumors 3/29 metastases0/1 1° tumors  
(39)    6/31 (19%) 1° tumorsd   
(15)   21/48 (44%)  0/24 (0%) 1° tumors 1/24 (4%) metastases 0/2 1° tumors0/5 metastasis  0/8 1° tumors 0/8 metastases (multiplex PCR/Southern blotting) 
(40)   10/10 (100%)e  1/10 (10%) 1° tumors   
(41)    1/30 (3%) 1° tumorsf 0/12 1° tumors  6/39 (15%) 1° tumors (immunohistochemistry) 
(42)   32/45 (71%)  2/14g (14%) 1° tumors   2/45 (4%) 1° tumors 
Present study  21/37 (57%)  1/3 (66%) 1° tumors 4/32 (9%) metastases   1/8 metastases (multiplex PCR)h 
Total 157/291i (54%) 19/193 (10%) 1° tumors17/265 (6%) metastases 3/49 (6%) 33/300 (11%) 
ReferenceFrequency of CDKN2A aberration
9p21 LOHMutationMethylationHomozygous deletionb
(6)    2/4 (50%) 1° tumors  0/13 1° tumors   
(32)    5/34 (15%) 1° tumors   
(33)    1/61 (2%) metastases   7/61 (11%) metastases (Southern blotting) 
(34)   12/40 (30%) metastases  0/3 1° tumors 2/21 (5%) metastases   9/40 (22%) metastases (Southern blotting) 
(14)   40/64 (63%) metastases  2/66 (3%) metastases   8/66 (12%) metastases (multiplex PCR) 
(35)   14/26 (54%) 1° tumors  1/26 (4%) 1° tumors   
(36)    3/25 (12%) metastases   0/25 metastases (multiplex PCR) 
(27)    4/12 (33%) 1° tumors 0/9 metastasesc   
(37)    0/6 1° tumors  1/6 (16%) metastases    
(38)   15/27 (55%)  3/21 (14%) metastases 0/1° tumors 3/29 metastases0/1 1° tumors  
(39)    6/31 (19%) 1° tumorsd   
(15)   21/48 (44%)  0/24 (0%) 1° tumors 1/24 (4%) metastases 0/2 1° tumors0/5 metastasis  0/8 1° tumors 0/8 metastases (multiplex PCR/Southern blotting) 
(40)   10/10 (100%)e  1/10 (10%) 1° tumors   
(41)    1/30 (3%) 1° tumorsf 0/12 1° tumors  6/39 (15%) 1° tumors (immunohistochemistry) 
(42)   32/45 (71%)  2/14g (14%) 1° tumors   2/45 (4%) 1° tumors 
Present study  21/37 (57%)  1/3 (66%) 1° tumors 4/32 (9%) metastases   1/8 metastases (multiplex PCR)h 
Total 157/291i (54%) 19/193 (10%) 1° tumors17/265 (6%) metastases 3/49 (6%) 33/300 (11%) 
a

Blank, analysis not performed in that study; 1°, primary.

b

The method of determining homozygous deletions is provided in parentheses.

c

Metastases examined corresponded to primary tumors, including those in which CDKN2A mutations were detected.

d

Only mutations that lead to a change in coding sequences are included.

e

These tumors were selected for mutation screening based on LOH at 9p.

f

Exon 1 and 2 were sequenced in 11 tumors; however, only exon 2 was sequenced in 19 tumors.

g

CDKN2A mutations in remaining 31 samples were reported previously (39).

h

These eight tumors demonstrated retention of heterozygosity flanked on one or both sides by LOH, and multiplex amplification of D9S942, D9S1748, and D9S974 was used to assess homozygous deletions.

i

Tumors preselected for LOH were not included in the final total.

1
MacLennan R., Green A. C., McLeod G. R., Martin N. G. Increasing incidence of cutaneous melanoma in Queensland, Australia..
J. Natl. Cancer Inst.
,
84
:
1427
-1432,  
1992
.
2
Smith-Sorensen B., Hovig E. CDKN2A (p16INK4A) somatic and germline mutations..
Hum Mutat.
,
7
:
294
-303,  
1996
.
3
Pollock P. M., Pearson J. V., Hayward N. K. Compilation of somatic mutations of the CDKN2 gene in human cancers: non-random distribution of base substitutions..
Genes Chromosomes Cancer
,
15
:
77
-88,  
1996
.
4
Foulkes W. D., Flanders T. Y., Pollock P. M., Hayward N. K. The CDKN2A (p16) gene and human cancer..
Mol. Med.
,
3
:
5
-20,  
1997
.
5
Ruas M., Peters G. The p16INK4a/CDKN2A tumor suppressor and its relatives..
Biochim Biophys Acta
,
1378
:
F115
-F177,  
1998
.
6
Ohta M., Nagai H., Shimizu M., Rasio D., Berd D., Mastrangelo M., Singh A. D., Shields J. A., Shields C. L., Croce C. M., Huebner K. Rarity of somatic and germline mutations of the cyclin-dependent kinase 4 inhibitor gene, CDK4I, in melanoma..
Cancer Res.
,
54
:
5269
-5272,  
1994
.
7
Spruck C. H. , III, Gonzalez-Zulueta, M., Shibata, A., Simoneau, A..
R.
,
Lin,M.F.,Gonzales,F.,Tsai,Y.C.,andJones,P.A.p16geneinunculturedtumours.Nature(Lond.),370
:
183
-184,  
1994
.
8
Cairns P., Mao L., Merlo A., Lee D. J., Schwab D., Eby Y., Tokino K., van der Riet P., Blaugrund J. E., Sidransky D. Rates of p16 (MTS1) mutations in primary tumors with 9p loss..
Science (Washington DC)
,
265
:
415
-417,  
1994
.
9
Huschtscha L. I., Reddel R. R. p16(INK4a) and the control of cellular proliferative life span..
Carcinogenesis (Lond.)
,
20
:
921
-926,  
1999
.
10
Cairns P., Tokino K., Eby Y., Sidransky D. Homozygous deletions of 9p21 in primary human bladder tumors detected by comparative multiplex polymerase chain reaction..
Cancer Res.
,
54
:
1422
-1424,  
1994
.
11
Cairns P., Polascik T. J., Eby Y., Tokino K., Califano J., Merlo A., Mao L., Herath J., Jenkins R., Westra W., Rutter J. L., Buckler A., Gabrielson E., Tockman M., Cho K. R., Hedrick L., Bova G. S., Isaacs W., Koch W., Schwab D., Sidransky D. Frequency of homozygous deletion at p16/CDKN2 in primary human tumours..
Nat Genet.
,
11
:
210
-212,  
1995
.
12
Holland E. A., Beaton S. C., Edwards B. G., Kefford R. F., Mann G. J. Loss of heterozygosity and homozygous deletions on 9p21–22 in melanoma..
Oncogene
,
9
:
1361
-1365,  
1994
.
13
Puig S., Ruiz A., Lazaro C., Castel T., Lynch M., Palou J., Vilalta A., Weissenbach J., Mascaro J. M., Estivill X. Chromosome 9p deletions in cutaneous malignant melanoma tumors: the minimal deleted region involves markers outside the p16 (CDKN2) gene..
Am. J. Hum. Genet.
,
57
:
395
-402,  
1995
.
14
Ohta M., Berd D., Shimizu M., Nagai H., Cotticelli M. G., Mastrangelo M., Shields J. A., Shields C. L., Croce C. M., Huebner K. Deletion mapping of chromosome region 9p21–p22 surrounding the CDKN2 locus in melanoma..
Int. J. Cancer
,
65
:
762
-767,  
1996
.
15
Ruiz A., Puig S., Lynch M., Castel T., Estivill X. Retention of the CDKN2A locus and low frequency of point mutations in primary and metastatic cutaneous malignant melanoma..
Int. J. Cancer
,
76
:
312
-316,  
1998
.
16
Wiest J. S., Franklin W. A., Otstot J. T., Forbey K., Varella-Garcia M., Rao K., Drabkin H., Gemmill R., Ahrent S., Sidransky D., Saccomanno G., Fountain J. W., Anderson M. W. 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..
Cancer Res.
,
57
:
1
-6,  
1997
.
17
Mead L. J., Gillespie M. T., Hung J. Y., Rane U. S., Rayeroux K. C., Irving L. B., Campbell L. J. Frequent loss of heterozygosity in early non-small cell lung cancers at chromosome 9p21 proximal to the CDKN2a gene..
Int. J. Cancer
,
71
:
213
-217,  
1997
.
18
Farrell W. E., Simpson D. J., Bicknell J. E., Talbot A. J., Bates A. S., Clayton R. N. Chromosome 9p deletions in invasive and noninvasive nonfunctional pituitary adenomas: the deleted region involves markers outside of the MTS1 and MTS2 genes..
Cancer Res.
,
57
:
2703
-2709,  
1997
.
19
Waber P., Dlugosz S., Cheng Q. C., Truelson J., Nisen P. D. Genetic alterations of chromosome band 9p21–22 in head and neck cancer are not restricted to p16INK4a..
Oncogene.
,
15
:
1699
-1704,  
1997
.
20
Kim S. K., Ro J. Y., Kemp B. L., Lee J. S., Kwon T. J., Fong K. M., Sekido Y., Minna J. D., Hong W. K., Mao L. Identification of three distinct tumor suppressor loci on the short arm of chromosome 9 in small cell lung cancer..
Cancer Res.
,
57
:
400
-403,  
1997
.
21
Baud E., Catilina P., Boiteux J. P., Bignon Y. J. Human bladder cancers and normal bladder mucosa present the same hot spot of heterozygous chromosome-9 deletion..
Int. J. Cancer
,
77
:
821
-824,  
1998
.
22
An H. X., Claas A., Savelyeva L., Seitz S., Schlag P., Scherneck S., Schwab M. Two regions of deletion in 9p23–24 in sporadic breast cancer..
Cancer Res.
,
59
:
3941
-3943,  
1999
.
23
Nakanishi H., Wang X. L., Imai F. L., Kato J., Shiiba M., Miya T., Imai Y., Tanzawa H. Localization of a novel tumor suppressor gene loci on chromosome 9p21- 22 in oral cancer..
Anticancer Res.
,
19
:
29
-34,  
1999
.
24
Walker G. J., Palmer J. M., Walters M. K., Nancarrow D. J., Parsons P. G., Hayward N. K. Refined localization of the melanoma (MLM) gene on chromosome 9p by analysis of allelic deletions..
Oncogene
,
9
:
819
-824,  
1994
.
25
Hussussian C. J., Struewing J. P., Goldstein A. M., Higgins P. A., Ally D. S., Sheahan M. D., Clark W. H., Jr., Tucker M. A., Dracopoli N. C. Germline p16 mutations in familial melanoma.
Nat. Genet.
,
8
:
15
-21,  
1994
.
26
Castellano M., Pollock P. M., Walters M. K., Sparrow L. E., Down L. M., Gabrielli B. G., Parsons P. G., Hayward N. K. CDKN2A/p16 is inactivated in most melanoma cell lines..
Cancer Res.
,
57
:
4868
-4875,  
1997
.
27
Piccinin S., Doglioni C., Maestro R., Vukosavljevic T., Gasparotto D., D’Orazi C., Boiocchi M. p16/CDKN2, and CDK4 gene mutations in sporadic melanoma development and progression..
Int. J. Cancer
,
74
:
26
-30,  
1997
.
28
Fountain J. W., Karayiorgou M., Ernstoff M. S., Kirkwood J. M., Vlock D. R., Titus-Ernstoff L., Bouchard B., Vijayasaradhi S., Houghton A. N., Lahti J., Kidd V. J., Housman D. E., Dracopoli N. C. Homozygous deletions within human chromosome band 9p21 in melanoma..
Proc. Natl. Acad. Sci. USA
,
89
:
10557
-10561,  
1992
.
29
Cairns P., Okami K., King P., Bonacum J., Ahrendt S., Wu L., Mao L., Jen J., Sidransky D. Genomic organization and mutation analysis of Hel-N1 in lung cancers with chromosome 9p21 deletions..
Cancer Res.
,
57
:
5356
-5359,  
1997
.
30
Takeuchi C., Takeuchi S., Ikezoe T., Bartram C. R., Koeffler H. P., Taguchi H. Mutational analysis of the Hel-N1 gene in childhood acute lymphoblastic leukemia with LOH at 9p21..
Leukemia
,
13
:
650
-651,  
1999
.
31
Parris C. N., Harris J. D., Griffin D. K., Cuthbert A. P., Silver A. J., Newbold R. F. Functional evidence of novel tumor suppressor genes for cutaneous malignant melanoma..
Cancer Res.
,
59
:
516
-520,  
1999
.
32
Gruis N. A., Weaver-Feldhaus J., Liu Q., Frye C., Eeles R., Orlow I., Lacombe L., Ponce-Castaneda V., Lianes P., Latres E., Skolnick M., Cordon-Carlo C., Kamb A. Genetic evidence in melanoma and bladder cancers that p16 and p53 function in separate pathways of tumor suppression..
Am. J. Pathol.
,
146
:
1199
-1206,  
1995
.
33
Maelandsmo G. M., Florenes V. A., Hovig E., Oyjord T., Engebraaten O., Holm R., Borresen A. L., Fodstad O. Involvement of the pRb/p16/cdk4/cyclin D1 pathway in the tumorigenesis of sporadic malignant melanomas..
Br. J. Cancer
,
73
:
909
-916,  
1996
.
34
Flores J. F., Walker G. J., Glendening J. M., Haluska F. G., Castresana J. S., Rubio M. P., Pastorfide G. C., Boyer L. A., Kao W. H., Bulyk M. L., Barnhill R. L., Hayward N. K., Housman D. E., Fountain J. W. Loss of the p16INK4a and p15INK4b genes, as well as neighboring 9p21 markers, in sporadic melanoma..
Cancer Res.
,
56
:
5023
-5032,  
1996
.
35
Healy E., Sikkink S., Rees J. L. Infrequent mutation of p16INK4 in sporadic melanoma..
J. Investig. Dermatol.
,
107
:
318
-321,  
1996
.
36
Platz A., Ringborg U., Lagerlof B., Lundqvist E., Sevigny P., Inganas M. Mutational analysis of the CDKN2 gene in metastases from patients with cutaneous malignant melanoma..
Br. J. Cancer
,
73
:
344
-348,  
1996
.
37
Herbst R. A., Gutzmer R., Matiaske F., Mommert S., Kapp A., Weiss J., Arden K. C., Cavenee W. K. Further evidence for ultraviolet light induction of CDKN2 (p16INK4) mutations in sporadic melanoma in vivo..
J. Investig. Dermatol.
,
108
:
950
1997
.
38
Gonzalgo M. L., Bender C. M., You E. H., Glendening J. M., Flores J. F., Walker G. J., Hayward N. K., Jones P. A., Fountain J. W. Low frequency of p16/CDKN2A methylation in sporadic melanoma: comparative approaches for methylation analysis of primary tumors..
Cancer Res.
,
57
:
5336
-5347,  
1997
.
39
Kumar R. , Lundh Rozell, B., Louhelainen, J., and Hemminki, K..
Mutations in the CDKN2A (p16INK4a) gene in microdissected sporadic primary melanomas. Int. J. Cancer
,
75
:
193
-198,  
1998
.
40
Peris K., Chimenti S., Fargnoli M. C., Valeri P., Kerl H., Wolf P. UV fingerprint CDKN2a but no p14ARF mutations in sporadic melanomas..
J. Investig. Dermatol.
,
112
:
825
-826,  
1999
.
41
Fujimoto A., Morita R., Hatta N., Takehara K., Takata M. p16INK4a inactivation is not frequent in uncultured sporadic primary cutaneous melanoma..
Oncogene
,
18
:
2527
-2532,  
1999
.
42
Kumar R., Smeds J. , Lundh Rozell, B., and Hemminki, K..
Loss of heterozygosity at chromosome 9p21 (INK4–p14ARF locus): homozygous deletions and mutations in the p16 and p14ARF genes in sporadic primary melanomas. Melanoma Res.
,
9
:
138
-147,  
1999
.