Presence and location of TP53 mutation determines pattern of CDKN2A/ARF pathway inactivation in bladder cancer.

Transformation and immortalization require the inactivation of key cell cycle regulatory genes. We examined 19 bladder cancer cell lines derived from 17 patients for alterations in TP53, RB1, CDKN2A, and ARF. Twelve cell lines had a mutation in exons 5-11 of TP53 and, with only one exception, a concomitant loss of RB1 protein expression. Another group of seven cell lines had a wild-type TP53 gene or a mutation in exons 1-4 of TP53 and concomitant alterations in both CDKN2A and ARF in every case. This demonstrates the requirement, in all but one line, for inactivation of both the CDKN2A/RB1 and ARF/TP53 pathways in bladder cancer cell lines and provides the first evidence for potential differences in the penetrance of mutations in the transactivation and DNA-binding domains of TP53.

Global and gene-specific epigenetic patterns in human bladder cancer genomes are relatively stable in vivo and in vitro over time.

We used a methylation-sensitive arbitrarily primed PCR technique to analyze, in a nonselective manner, methylation alterations at GC-rich regions of the genome in metachronous tumors and their derived cell lines from two patients with transitional cell carcinoma of the bladder. The methylation status of the majority of evaluable sequences (83%) remained unchanged in the tumors from both patients relative to a panel of normal urothelium samples obtained from individuals free of bladder disease, in which we measured <1% interindividual variation. The 17% of methylation alterations represents sequences altered in either a cancer-specific (3%), tumor-specific (1%), or patient-specific (13%) manner. The proportion of the altered sequences analyzed that were CpG islands corresponds to approximately 7000 CpG islands altered in the genome. Surprisingly, few additional changes in methylation patterns were observed in cell lines derived from the tumors; however, all of the cell lines showed altered methylation in a common set of 3% of evaluable sequences. Three genes known to be aberrantly methylated in bladder cancer (p16, p15, and PAX6) were studied in detail by methylation-sensitive single nucleotide primer extension and showed increased methylation in culture at preexisting methylated sites for all of the exons but no de novo methylation in culture for the promoters in any cell line. Therefore, our investigation provides the first serial as well as parallel quantitation of the global epigenetic stability in two independent bladder cancer genomes over the course of progression and in culture. In addition, our investigation also provides the first direct comparison of the epigenetic and genetic patterns on the global scale, showing the epigenetic pattern to be relatively stable in vivo and in vitro over time within an individual.

Bladder cancer genotype stability during clinical progression.

Genomic instability is manifested by the accumulation of large numbers of mutations. The rate at which mutations accumulate has been difficult to estimate because serial comparisons are required. For further insight into how quickly mutations accumulate during clinical progression, cell lines sequentially isolated 6 or 11 months apart from two patients with metastatic bladder cancer were compared for loss of heterozygosity (LOH). The genomes were scanned at approximately 200 polymorphic microsatellite loci to increase the resolution and sensitivity for losses. The cell lines from both patients had evidence of genomic instability, with aneuploidy, chromosomal instability, and fractional allelic losses of 0.15 and 0. 48, respectively. However, additional changes were relatively infrequent, with more than 90% identity between the initial and recurrent cell lines. Allelic losses were not randomly scattered, but clustered on specific chromosomes. Therefore, the numbers of loci with further LOH during the clinical progression of some bladder cancers are small relative to the total number of loci with LOH, suggesting that most allelic losses accumulate before clinical presentation.

PAX6 methylation and ectopic expression in human tumor cells.

The mechanisms underlying the de novo methylation of CpG islands in human cancer remain almost completely unknown. We used a methylation-sensitive arbitrarily primed polymerase chain reaction (Ms AP-PCR) technique to scan genomic DNA for differential methylation patterns and identified a 550 bp band that was hypermethylated in a majority of colon and bladder cancer cells. This band corresponded to a CpG-rich region in exon 5 of PAX6, which is a highly conserved transcription regulatory factor involved in embryogenesis. Interestingly, exon 5 was very frequently methylated in solid tumors compared with adjacent normal tissues, 17 out of 27 (63%) in bladder, and colon, but the promoter remained unmethylated in all these cases. This methylation was not effective in blocking transcription since ectopic expression of PAX6 was seen in several tumors and cell lines with extensive exon 5 methylation. De novo methylation of the promoter was only seen in tumor cell lines and was associated with gene silencing since treatment with 5-aza-2'-deoxycytidine restored expression to the cells and resulted in a less methylated promoter. Thus, ectopic expression and hypermethylation of exon 5 of PAX6 demonstrate that methylation within a transcribed region, as opposed to promoter methylation, does not block gene expression.

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.