A mutator phenotype in human cancers: Origin and consequences. Free

CS07-01

We have previously proposed that human cancers exhibit a mutator phenotype during their evolution . We hypothesized that the large numbers of somatic mutations observed in human cancers could not be generated by the extremely low rate of mutation in normal cells . Clonal mutations have been documented in a variety of human cancers. An extensive number of mutated genes have been identified in human tumors by brute force DNA sequencing . However, DNA sequencing is unable to detect random mutations within a tumor, and it is these mutations that would be associated with the heterogeneity of cancers cells within a tumor. The quantitation of random mutations in human cells has been precluded by lack of an assay to measure mutations at the level of single DNA molecules . We have established a method to detect and identify rare random mutations in human cells, at a frequency of 1 per 10⁸ base pairs . The assay is based on gene capture, by hybridization with a uracil-containing probe, followed by magnetic separation. Mutations that render the mutational target sequence non-cleavable by a restriction enzyme are quantified by dilution to single molecules and real-time quantitative PCR amplification. Our target sequence was TCGA in intron 6 of p53. The background was 10^-8 mutations/base pair in normal human cells, and mutations in this sequence were genetically neutral. Using this new mutation assay, we report here the results of a blinded experiment comparing the frequency of random mutations in five neoplasms with that in the corresponding normal tissues obtained from untreated patients. The mutation frequency in normal tissues is less than one mutation in 10⁸ base-pairs. In contrast to the paucity of mutations in normal tissues, all five tumors exhibited large numbers of mutations, the mean being 205 × 10^-8 mutations per base-pair. Frequencies ranged from 65 x 10^-8 for a peri-renal liposarcoma to 475 x 10^-8 mutations per base pair for a colon adenocarcinoma. Sequence analysis indicated all were single base substitutions and that the majority were not clonally expanded. The difference between the median mutation frequencies for the normal and tumor samples is highly significant [P = 0.009; Wilcoxon Rank-Sum (Mann-Whitney test)]. These results provide strong evidence for a mutator phenotype in carcinogenesis and indicate that it is an ongoing process. The large numbers of random mutations present within tumors has important biological implications. First, random mutations might be the basis for the morphologic heterogeneity of cancer cells within a tumor. Second, even though many of these mutations might be progenitors of clonal mutations they would not be detected by the proposed projects to establish an atlas of tumor cancer genes. Third, random mutations might be predictive for the emergence of drug resistance; tumors with large number of random mutations are more likely to contain maligant cells harboring enhancing mutations in drug resistant genes. This concept provides a molecular basis for the observed clinical efficacy of combination therapy, since any single cell would be unlikely to contain mutations that confer resistance to agents with different mechanisms of cytotoxicity. Fourth, random mutations may provide an independent criterion for grading tumors; tumors with large numbers of mutations may be further along the pathway to invasion and metastasis. Last, our results suggest that enhanced mutagenesis is an ongoing process in at least some tumors and thus carries the hopeful implication that inhibition of this process may impede tumor progression. References 1. Loeb, L. A., Springgate, C. F., and Battula, N. Errors in DNA replication as a basis of malignant change. Cancer Res., 34: 2311-2321, 1974. 2. Loeb, L. A., Loeb, K. R., and Anderson, J. P. Multiple mutations and cancer. Proc. Natl. Acad. Sci. USA, 100: 776-781, 2003. 3. Wang, T. L., Rago, C., Silliman, N., Ptak, J., Markowitz, S., Willson, J. K. V., Parmigiani, G., Kinzler, K. W., Vogelstein, B., and Velculescu, V. E. Prevalence of somatic alterations in the colorectal cancer cell genome. Proc. Natl. Acad. Sci. USA, 99: 3076-3080, 2002. 4. Futreal, P. A., Coin, L., Marshall, M., Down, T., Hubbard, T., Wooster, R., Rahman, N., and Stratton, M. R. A census of human cancer genes. Nat. Rev. Cancer, 4: 117-183, 2004. 5. Bielas, J. H. and Loeb, L. A. Quantification of random genomic mutations. Nature Meth., 2: 285-290, 2005.