Mutations and Cancer

The ability of cancer cells to continually mutate may be central to how tumors evolve. In the case of solid tumors, there is approximately a 20-year interval from the exposure of an individual to a carcinogen until the clinical detection of a tumor. During this interval, cancer cells acquire properties that allow them to flourish in a changing environment. By the time a tumor is detectable, the cancer cells are able to divide where normal cells do not, to invade adjacent cellular architectures, to metastasize and eventually to kill the host. In addition, tumors have the capacity to rapidly develop resistance to a variety of chemotherapeutic agents. Each of these phenotypes can be mimicked by mutations in normal cells.

Chromosomal Alterations in Human Cancers

Mutations can be defined as a change in the nucleotide sequence in DNA. Mutations in cancer cells cover a wide spectrum from chromosomal alterations that encompasses millions of nucleotides to point mutations that involve only a few nucleotide substitutions in single genes. Multiple somatic chromosomal alterations are diagnostically associated with cancer cells and involve translocations, deletions, amplifications, and  aneuploidy (a change in the number of chromosomes in individual cells). Unique chromosomal changes occur at high frequencies in certain tumors and are of diagnostic significance. However, there is a striking heterogeneity of chromosomal alterations in cancer cells within individual tumors. In general, there is a positive correlation between the number of chromosomal changes within a tumor and the malignant potential of that tumor. As molecular techniques are becoming more sensitive, more and more chromosomal abnormalities are being reported in different tumors. In some tumors there is evidence for a sequential order in chromosomal mutations during  tumor progressions. Measurements of the number of copies of segments of the genome in tumor cells (DNA copy number) and the loss of pieces of DNA ( Loss of Heterozygosity) have established that many tumors harbor as many as 40 chromosomal alterations. It should be emphasized that these methodologies only score a very small fraction of the genome and as a result may greatly underestimate the number of small chromosomal rearrangements within the genome.

Point Mutations in Human Cancer

Only recently has evidence accumulated indicating that cancer cells not only contain multiple chromosomal alterations, but also contain thousands of smaller changes in nucleotide sequence. These studies have provided strong support of the mutator phenotype hypothesis involving a reduction in the fidelity of DNA replication and/or a decrease in the efficiency  repair of DNA. An early clue to the large numbers of mutations in cancer cells was the observations that the resistance of cancer cell lines to divergent chemotherapies was mediated by gene  amplification. The first direct evidence in support of a mutator phenotype in cancer was provided by the demonstration that cells from patients with  HNPCC (Hereditary Nonpolyposis Colon Cancer) exhibit  microsatellite instability in association with mutations in DNA repair genes. Microsatellites are short repetitive sequences of nucleotides in DNA that are subjected to slippage during copying by DNA polymerases. In normal individuals, mutations in microsatellites are corrected by the DNA  mismatch repair system. In HNPCC tumors, the deficits in mismatch repair result in expansions and contractions in the length of repetitive nucleotide sequences. Based on the enormous number of microsatellites in the human genome, it has been calculated that each tumor could harbor more than 100,000 mutations in these sequences alone. Repetitive sequences located within genes are also mutated at high frequencies in these tumors. Changes in the lengths of repetitive sequences have also been reported in a variety of tumors that are not known to be associated with mutations or deficiencies in mismatch repair, but may be associated with mutations in other genetic stability genes. Thus, repetitive sequences in DNA may be a “hot-spot” for mutagenesis and serve as a sentinel for the detection of a mutator phenotype in cancer.

Rarity of Spontaneous Mutations in Normal Cells

In normal cells, DNA replication and chromosomal segregation are accurate processes. Measurements of mutagenesis of cells grown in culture yield values of ∼2.0 × 10^–7 mutations/haploid gene/cell division. Taking into account this very low frequency of mutations, it seems improbable that the spontaneous mutation rate is sufficient to generate the large numbers of genetic alterations that are observed in cancer cells. If one assumes that  stem cells, which give rise to a cancer, have a similarly low rate of mutagenesis, then it can be calculated that the average stem cell would accumulate only one or two mutations during tumorigenesis. A few stem cells could accumulate as many as 12 mutations and thus account for the inactivation of  tumor suppressor genes in  retinoblastomas and other tumors. However, the normal spontaneous mutation rate is inadequate to account for thousands of mutations observed in most tumors.

Historical Perspective

The hypothesis that cancer cells exhibit a mutator phenotype was proposed more than 25 years ago and was set forth on the postulate that there is a decrease in the accuracy of DNA synthesis during tumorigenesis. These random mutations could arise by mutations in DNA polymerases that render them error-prone during DNA synthesis or by mutations in DNA repair enzymes making them less efficient in correcting DNA damage or mistakes in incorporation by DNA polymerases. The production of more errors and/or their lack of repair could result in an increase in point mutations and, indirectly, in chromosomal aberrations in cancer cells. Further mutations in these genes would result in cascading numbers of mutations as tumor cells multiplied. Independently, Nowell postulated that cancer cells accumulate multiple mutations by successive rounds of clonal selection. Analysis was based on chromosomal changes in the evolution of human tumors. Leukemias with minimal chromosome changes were considered to be early in clonal evolution, while highly  aneuploid solid tumors were considered to have undergone multiple rounds of clonal selection. Increases in both mutation rate and clonal selection could contribute to an increase in the number of mutations in tumors. Studies in bacteria suggest that these mechanisms may reinforce one another. Sequential rounds of selection for different mutants yielded populations of bacteria that invariably contained mutations in genes that normally function to maintain genetic stability. The conclusion is that selection for mutant clones simultaneously enriches for mutations in genes that can produce the mutant clones. With successive rounds of selection there is increasing enrichment for cells that express a mutator phenotype.

Tumor Evolution

The ability of cancer cells to express a mutator phenotype provides a mechanism for tumor cells to circumvent the host’s mechanisms that determine when cells divide and their position within tissues. As tumors expand, they encounter a sequence of restrictive blockages that curtail further expansion. As indicated in Fig. 1, impediments to expansion include the architecture of surrounding tissues, reduced nutrition, decreased oxygen, need for growth factors, and inadequate blood supply. Each of these impediments can be overcome by the selective clonal expansion of cancer cells with mutations in genes that impart the required phenotypes. Some of these mutations could be selected from a population of cells harboring multiple mutations. Others could result from new mutations induced by additional mutations that render genetic stability genes error prone. Thus, with each round of selection there would be a “piggy-backing” of mutations in genes that increase the overall mutation frequency in cancer cells. In order for a mutator phenotype to account for the many mutations in a cancer cell, it would have to be present early during the course of tumor progression. The continued expression of a mutator phenotype might not be compatible with the enhanced growth of cancer and thus may not be exhibited by cancer cells by the time a tumor is clinically apparent. Nevertheless, the footprints of its presence, multiple mutations, would persist in all cell descendents throughout the course of tumorigenesis ( Multistep Development).

Implications of a Mutator Phenotype

The multiplicity of mutations in cancer cells resulting from a mutator phenotype has implications with respect to pathogenesis, diagnosis, treatment, and prevention of human cancer. The presence of a large number of mutations in cancer cells implies that the malignant process is irreversible. Once a mutation occurs in a gene it is highly unlikely that a second mutation would occur at the same position to change the mutation back to the wild-type sequence. Also unlikely is the possibility that a second mutation would occur at a different locus and suppress the effects of the first mutation, particularly if the first mutation occurs at a tumor suppressor gene and inactivates that gene. Thus the concept of a mutator phenotype is not compatible with the hypotheses that cancer cells have the potential to differentiate into normal cells. The reports that certain tumors spontaneously regress does not necessarily indicate that these tumor cells have been transformed into normal cells. It is more likely to reflect the selective killing of tumor cells by normal host mechanisms that reject foreign cells.

The heterogeneity of mutations within a tumor provides a population of cancer cells harboring mutants that are resistant to many chemotherapeutic agents. In addition, the mutator phenotype exhibited by many cancer cells allows them to continuously generate new mutant clones that are resistant to and proliferate in the presence of chemotherapy, immunotherapy, and even radiation therapy. While such a scenario allows cancer cells to thwart many therapeutic protocols, it might offer new approaches to cancer therapy. It might be feasible to develop drugs that target the genetic instability of tumor cells. For example, nucleotide analogs can be designed that are preferentially incorporated into DNA by mutant error-prone DNA polymerases.

The large number of mutations in microsatellite sequences in cancer cells may facilitate the early detection of certain human malignancies. Rare tumor cells circulate, can be detected in the blood, and could account for the hematological spread of tumor cells to distant sites. In addition, cancer cells in tumors frequently undergo cell death (apoptosis) and shed their DNA into the blood. Thus, methods can be developed for the detection of altered microsatellite sequences in DNA and cells in blood for the early detection of human malignancy and for calibrating the potential of tumors for metastatic dissemination.

If a mutator phenotype is rate-limiting for tumor progression, it is important to identify the agents and genes involved in its expression. Sources for a mutator phenotype such as mutant DNA polymerases, which are error-prone, or mutations in DNA repair enzymes are not easily corrected. Other sources might be more easily attenuated. Recent evidence indicates that cells contain a number of error-prone DNA polymerases that are induced by DNA damage, and it might be feasible to develop drugs to selectively inhibit these enzymes. More immediately open to experimental analysis is increased DNA damage by normal cellular metabolites, such as oxygen-free radicals. These radicals can be scavenged by the administration of specific drugs and vitamins.

If a mutator phenotype constitutes a rate-limiting step for the development of a cancer, it might be possible to prevent cancers by delay. Even a twofold reduction in the rate of accumulation of mutations in cancer cells might have profound effects on the age at which patients succumb to certain adult cancers. Consider primary hepatoma resulting from hepatitis virus B infection that occurs in early infancy and persists chronically ( Hepatitis Virus Associated Hepatocellular Carcinoma). The persistence of chronic hepatitis in individuals increases the risk of subsequent hepatoma by more than 200-fold. It usually takes 40 additional years for the tumor to be clinically manifested. A reduction in the rate of mutation accumulation by only twofold could delay the clinical appearance of the tumor from age 50 to 90. It has been postulated that the persistence of hepatitis results in an inflammatory reaction with generation of oxygen-free radicals. Thus, drugs that scavenge  reactive oxygen species (ROS) might have a role in the prevention of primary hepatoma. It should be emphasized that prevention by delay does not affect the rate of initiation of the malignant process, but rather is directed at slowing down the rate of its progression. This approach may offer new direction to reducing the number of cancer-associated deaths.

 Microsatellite Instability