Download Significance of multiple mutations in cancer

Carcinogenesis vol.21 no.3 pp.379–385, 2000
Significance of multiple mutations in cancer
Keith R.Loeb and Lawrence A.Loeb1
Department of Pathology, University of Washington School of Medicine,
Seattle, WA 98195 and the Fred Hutchinson Cancer Research Center,
Seattle, WA 98109, USA
1To
whom correspondence should be addressed
Email: [email protected]
There is increasing evidence that in eukaryotic cells, DNA
undergoes continuous damage, repair and resynthesis. A
homeostatic equilibrium exists in which extensive DNA
damage is counterbalanced by multiple pathways for DNA
repair. In normal cells, most DNA damage is repaired
without error. However, in tumor cells this equilibrium
may be skewed, resulting in the accumulation of multiple
mutations. Among genes mutated are those that function
in guaranteeing the stability of the genome. Loss of this
stability results in a mutator phenotype. Evidence for a
mutator phenotype in human cancers includes the frequent
occurrence of gene amplification, microsatellite instability,
chromosomal aberrations and aneuploidy. Current experiments have centered on two mechanisms for the generation
of genomic instability, one focused on mutations in mismatch repair genes resulting in microsatellite instability,
and one focused on mutations in genes that are required
for chromosomal segregation resulting in chromosomal
aberrations. This dichotomy may reflect only the ease by
which these manifestations can be identified. Underlying
both pathways may be a more general phenomenon involving the selection for mutator genes during tumor progression. During carcinogenesis there is selection for cells
harboring mutations that can overcome adverse conditions
that limit tumor growth. These mutations are produced by
direct DNA damage as well as secondarily as a result of
mutations in genes that cause a mutator phenotype. Thus, as
tumor progression selects for cells with specific mutations, it
also selects for cancer cells harboring mutations in genes
that normally function in maintaining genetic instability.
Introduction
Mutations are not only a hallmark of cancer but may be central
to how cancers evolve. Cancer cells divide where normal cells
do not; they invade, metastasize and kill the host of origin.
The facts that cancer is inheritable at the cellular level and
that cancer cells contain multiple mutations, suggest that tumor
progression is driven by mutagenesis. Molecular techniques
are progressively making it easier to dissect the human genome,
from chromosomes down to genes and ultimately nucleotide
sequences. With each deeper level of exploration, more and
more mutations are being documented in cancer cells. The
emerging concept is that genomes of cancer cells are unstable,
and this instability results in a cascade of mutations some of
which enable cancer cells to bypass the host regulatory
processes that control cell location, division, expression,
© Oxford University Press
adaptation and death. Genetic instability is manifested by
extensive heterogeneity of cancer cells within each tumor. In
addition, tumors invariably develop resistance to chemotherapeutic agents. Each of the tumor phenotypes involves, or can
be mimicked by, specific mutations introduced in critical genes.
These mutations either arise from copying unrepaired DNA
damage or from errors committed during DNA synthesis. In
this review we will consider the generation of mutations in
human cells, the multiplicity of mutations present in human
tumors, and the relationship of multiple mutations to tumor
progression. Recent studies have suggested two pre-eminent
mechanisms for the generation of mutations in cancer cells,
one involving deficits in DNA repair and one involving deficits
in chromosomal partitioning during cell division. We will
consider the hypothesis that there are thousands of mutations
in cancer cells and that there are many mechanisms for the
generation of a mutator phenotype in cancer cells.
Mutations result from DNA damage
It has become increasingly recognized that, rather than being
inert, cellular DNA undergoes continuous damage and resynthesis. DNA is damaged by both environmental and cellular
(endogenous) sources. Many of the environmental agents
that damage DNA have been demonstrated to be mutagens.
Epidemiologic data indicate that many of these agents are also
human carcinogens. The association between environmental
chemicals and human cancers was most definitively established
in situations where small groups of individuals have been
exposed to an inordinately high concentration of a specific
chemical that elicited a rare tumor. DNA damage by chemicals
can be divided into two categories: (i) those that produce large
bulky adducts and are repaired by the nucleotide excision
pathway; and (ii) those that cause small alterations, such as
alkylating agents that add methyl and ethyl groups onto
nucleotides in DNA, and are repaired by the base excision
repair pathway. In addition to DNA damage by environmental
agents, Ames et al. (1) have emphasized that many foods
contain natural chemicals that also damage DNA and produce
similar alterations. Moreover, it has been argued that consumption of natural DNA-damaging chemicals from foods is much
greater than our exposure to DNA-damaging chemicals produced by industry (2). Cellular metabolic processes also
generate reactive chemical intermediates with the potential to
damage DNA and, as a result, might also be a source of
spontaneous mutations and cancers (3). Even water has the
potential to damage DNA, causing hydrolysis of the glycosylic
bond and the formation of mutagenic abasic sites in DNA.
Based on rates of depurination of DNA in vitro, it has been
estimated that each cell undergoes 10 000 depurinations per
day (4). DNA damage by reactive oxygen species has been
estimated to occur at an equally high frequency (5,6), and
many of the damaged bases, such as 8-oxo-deoxyguanosine,
miscode (7–9). Even though deamination of cytosine to thymidine is less frequent than oxidative damage to DNA (10),
379
K.R.Loeb and L.A.Loeb
Fig. 1. Factors governing mutation accumulation in cancer cells. A variety
of agents have the potential to damage DNA. Mutations are generated by
DNA damage that escapes the multiple mechanisms for DNA repair. Thus,
there is a dynamic equilibrium between DNA damage and DNA repair and
this equilibrium is perturbed in cancer cells.
the result is a change in nucleotide sequence in the DNA.
Mechanisms have evolved to repair each DNA lesion, but
considering the high frequency at which they occur and the
compact and inaccessible structure of human chromatin, it is
likely that a significant fraction of these lesions escape DNA
repair and produce mutations during replication of the unrepaired DNA by DNA polymerases (Figure 1). In general, DNA
polymerases copy past small alterations such as methyl or
ethyl groups with high efficiency, and thus are likely to produce
mutations depending on the miscoding potential of the altered
nucleotide (11). Bulky adducts are not easily bypassed by
normal cellular polymerases and mutagenesis is dependent
on the induction of damage-inducible error-prone pathways
frequently involving special error-prone DNA synthesis (12).
Taken together, these results suggest that the nucleotide
sequence of cellular DNA is maintained at a homeostatic
equilibrium, such that an increase in the production of DNA
damage or a reduction in DNA repair results in an increased
frequency of mutations (Figure 1).
(17). Blocks to accuracy could be minimized by replication
proteins that function in concert with DNA polymerase in DNA
replication and/or DNA repair. In addition, misincorporated
nucleotides are excised by the mismatch repair system (18)
and this system may interact with a variety of adducted
nucleotides in DNA (19). Thus, there are multiple systems
with overlapping specificities for the repair of DNA.
There is increasing evidence for the interchangeability of
DNA polymerases (20) and thus mutation rates might be
altered by varying the relative expression of different DNA
polymerases (21) as well as by mutations that render these
enzymes error-prone (22). The increase in DNA polymerase
β in certain tumors suggests the possibility that polymerase β
could substitute for a more accurate DNA polymerase, resulting
in increased mutagenesis (23). In Escherichia coli, DNA
damage results in the induction of an error-prone response
referred to as the SOS pathway (24). It has long been suspected
that this pathway facilitates the bypass of blocking lesions
during DNA replication through induction of proteins that
render the normal DNA polymerases error-prone. Recently,
one of the proteins involved in the SOS response, UmuD’2C,
has been identified as an error-prone DNA polymerase that
was designated as Pol V (25,26). Moreover, a similar errorprone DNA polymerase has been identified in yeast (27) and
found to be mutated in Xeroderma pigmentosum variant cells
(XP-V) (28,29). Conceivably, error-prone DNA polymerases
are elevated in certain cancers and contribute to the accumulation of DNA mutations in these tumors.
The aforementioned results lead one to conclude that mutation rates in cells are maintained at a homeostatic equilibrium
in which the extensive DNA damage is counterbalanced by
multiple mechanisms for DNA repair (Figure 1). In normal
cells, only infrequently do lesions in DNA escape the screen
provided for by DNA repair and direct misincorporation at the
time of DNA synthesis. In eukaryotic cells there is the emerging
concept of checkpoints in which cells sterilize the DNA
immediately prior to the onset of DNA replication (30).
Mutations in G1/S checkpoint genes allow DNA replication in
the presence of unrepaired lesions (31) and result in enhanced
mutagenesis.
Mutations arising from misincorporation by DNA polymerases
Multiple chromosomal aberrations in many human cancers
Mutations can also result from nucleotide misincorporation by
DNA polymerases in copying non-damaged DNA templates
during DNA replication or even during DNA repair synthesis.
Studies on the fidelity of DNA synthesis indicate that purified
DNA polymerases incorporate non-complementary nucleotides
at rates much greater than one would predict based on the rare
spontaneous mutations in human cells. The error rates of
purified mammalian DNA polymerases in vitro thus far documented (13) vary from one in 5000 for DNA polymerase β,
the enzyme responsible for the synthesis of short segments of
DNA during repair synthesis, to one in 10 000 000 for DNA
polymerases δ and ε, enzymes involved in DNA replication
(14,15). The latter polymerases contain a 3⬘→5⬘ exonucleolytic
activity that excises non-complementary nucleotides immediately after misincorporation by the polymerase. The frequencies
of misincorporation by any DNA polymerase can vary as
much as 100-fold as a result of the nucleotide sequence in
DNA (16). DNA polymerases pause at sites that have the
potential to form secondary structures and these pause sites
have been shown to be loci for enhanced misincorporation
One of the hallmarks of cancer cells is genetic instability. This
is manifested both at the single nucleotide level, resulting in
point mutations, or at the chromosomal level resulting in
translocations, deletions, amplifications and whole chromosome aneuploidy. Both of these forms of genetic instability
can lead to a mutator phenotype via altered protein expression,
function or gene dosage effects. There is a growing debate in
the literature whether single nucleotide changes or large
macromolecular chromosomal abnormalities are more prevalent in cancers or constitute a causative factor in cancer
formation (32). Aneuploidy, a change in chromosome number,
is a defining characteristic of many human cancers and is one
of the most frequent changes observed in cancer. Since each
chromosome contains thousands of genes, it is amazing that
changes in chromosome number are compatible with cellular
viability. Specific chromosomal exchanges occur at high frequencies and are diagnostic of certain tumors. In general, there
is a positive correlation between the number of chromosomal
alterations within a tumor and the malignant potential of
that cancer. This suggests that a progressive diminution in
380
Significance of multiple mutations in cancer
chromosome maintenance pathways leads to an increasing
number of errors during carcinogenesis, and promotes the
development of highly anaplastic tumors. Chromosomal
changes visible by light microscopy must involve deletions,
additions or translocations of DNA segments that span millions
of nucleotide bases. Conceivably, these visible chromosomal
aberrations are the tip of an iceberg: hidden amongst these
large rearrangements could be an even greater number of
smaller changes in the nucleotide sequence of DNA in tumor
cells.
Cancer cells exhibit a mutator phenotype
The large number of chromosomal aberrations and the striking
heterogeneity of tumors suggest that cancer cells are genetically
unstable. Two overlapping mechanisms were initially and
independently proposed to account for this instability. First, it
was proposed that cancer cells exhibit a mutator phenotype
based on the postulate that there was an increase in the rate
of errors in DNA synthesis during tumor cell division (33).
These errors could arise from mutations in DNA polymerases
that render them error-prone, or from mutations in DNA repair
proteins that render them defective. Mutations arising from
either process would occur throughout the genome. Some of
these mutations would be in additional genes normally required
for maintaining genetic stability. As a result, there would ensue
a cascade of mutations as tumor cells undergo successive cell
divisions. Second, in 1976, Nowell analyzed the sequence of
chromosomal aberration in tumors, and proposed a model for
mutation accumulation based on successive waves of clonal
selection (34). Recent studies suggest the possibility that these
models are not only tightly coupled but may be interdependent
(vide infra).
Evidence for a mutator phenotype in human cancers
The hypothesis that cancer cells exhibit a mutator phenotype
predicts that by the time cancers are detected clinically,
they already contain enormous numbers of mutations. Recent
evidence indeed indicates that human cancers harbor thousands
of mutations. Perhaps the earliest hint of the large number of
mutations in tumor cells were studies indicating that gene
amplification in tumor cells accounts for the rapid emergence
of their resistance to the drug PALA (35). Amplification of
resistant genes was undetectible (⬍10–9) in primary diploid
cells but was observed at frequencies as large as 10–3 to 10–4
in transformed cells derived from a variety of tumors. The
extensiveness of point mutations in tumor cells was initially
delineated by studies on microsatellite instability in colon
cancer. Perucho and colleagues, using arbitrary primers for
PCR amplification, detected multiple bands at different loci of
DNA from colon tumors but not from adjacent normal cells
(36,37). Based on the changes detected in the limited number
of nucleotide sequences that were sampled, these authors
calculated that the tumors they examined contained more than
100 000 mutations (38). The mutations these authors observed
were in microsatellites, repetitive nucleotide sequences located
between genes. However, it is becoming increasingly apparent
that repetitive sequences are also present within genes and
also undergo expansion or contraction specifically in cancer
cells at very high frequencies (39,40). Contraction or expansion
of repeats within genes provides an attractive mechanism
for inactivation of tumor suppressor genes during tumor
progression. High levels of microsatellite instability are especi-
ally prominent in hereditary non-polyposis colon cancer
(HNPCC). In this syndrome, mutation in one allele of the
genes involved in DNA mismatch repair is inherited (41) and
loss of the second wild-type allele occurs with high frequency.
In other cancers that exhibit expansion or deletion of repetitive
sequences, diminished DNA repair appears to be mediated by
methylation with reduced expression of a least one of the
genes involved in mismatch repair (42,43).
It has been assumed that the changes in the length of
repetitive sequences are mediated by slippage of DNA polymerases. We recently reported that damage to plasmids containing
repetitive sequences by reactive oxygen species results in
enhanced slippage specifically within the repetitive sequences
suggesting a role of oxygen free radicals in enhancing microsatellite instability (44). In this system the preference for the
induction of frameshifts in microsatellite sequences by reactive
oxygen species is 70-fold greater than in non-repetitive
sequences. It is now important to establish if microsatellite
instability in tumors results from DNA damage by reactive
oxygen species. If so, this instability might then be diminished
by drugs that scavenge reactive oxygen species; this approach
may be particularly relevant for tumor prevention in HNPCC
families.
Microsatellite instability in tumors may be a harbinger for
other types of mutations that are also present in tumors in
large numbers. Tumors that exhibit microsatellite instability
frequently contain alterations in the lengths of repetitive
sequences within a variety of cancer-associated genes
including: APC, IGF, TGF-β, hMSH3 and hMSH6 (39,40).
Tissue-culture cells defective in mismatch repair also exhibit
elevations in non-selectable genes such as hgprt and oubain
resistance (45). Nearly all breast (46) and ovarian tumors (47)
studied utilizing comparative genomic hybridization have been
shown to contain multiple changes in gene copy number. This
technique only scores for changes that involve thousands of
nucleotides. Studies using PCR-amplified gene fragments have
revealed as many as 56 regions that exhibit loss of heterozygosity in breast cancer (48). This technique scans only a
minute fraction of the genome and thus a much larger number
of alterations are likely to be present throughout the genome.
From these data we have calculated that a tumor with 50
regions of loss of heterozygosity detected by PCR-amplification
of a limited number of gene segments might contain as many
as 100 000 other alterations throughout the entire genome (49).
Cell cycle checkpoints, apoptosis and the mutator phenotype
Alterations in the timing of cell-cycle phases provide another
mechanism for the generation of a mutator phenotype. During
cell-cycle progression there are several regulatory pathways
that function as checkpoints and monitor the repair of damage
before proceeding to the next stage in the cell cycle (30). If
activated, these checkpoints function to transiently arrest cellcycle progression so that damage can be repaired or that the
proper assembly of cell components can be completed. The
elimination of cell-cycle checkpoints is a facile way to develop
a mutator phenotype. Mutations in p53, for example, result in
decreasing the stringency of the DNA damage checkpoint,
allowing cancer cells to replicate damaged DNA, and to
accumulate mutations in daughter cells that lead to genomic
instability (50). In the presence of unrepairable damage,
apoptosis is triggered as a final means to halting the spread of
mutations. Again, many cancers contain mutations that delay
or prevent the apopotic response and thus promote the survival
of genetically unstable malignant cells (51).
381
K.R.Loeb and L.A.Loeb
Fig. 2. Different pathways that could result in a mutator phenotype in tumor
cells.
Chromosomal instability
Aneuploidy, frequently present in many cancers, presents
another manifestation of genetic instability. Some argue that
aneuploidy is the defining characteristic of cancer and serves
as the predominant mode of genetic instability, since the
majority of human cancers fail to exhibit detectable microsatellite instability (38,52). Although aneuploidy is an easily
recognized feature of cancers, little is known about its development or the selective advantage it provides for the tumor.
Presumably, altered gene dosage enhances genetic instability
and thus represents another form of the mutator phenotype.
Aneuploidy can arise from chromosomal fragmentation, translocation, amplification or non-disjunction (resulting in the
mis-segregation of entire chromosomes). The generation of
aneuploidy can be an early or even an initial event in the
development of genetic instability. It is conceivable that
progressive increases in aneuploidy represent a predominant
pathway for the development of genetic instability and it is
independent of the accumulation of other types of mutations.
There is evidence suggesting that aneuploidy first develops
from a transient tetraploid intermediate that exhibits profound
genetic instability and rapidly accumulates additional chromosomal abnormalities (53). To achieve a tetraploid intermediate,
one requires defective mitosis or endoreduplication: two rounds
of DNA synthesis without an intervening mitosis. For this
reason and because of the presence of morphologically abnormal mitosis in tumors, many have predicted cancer cells to be
defective in mitotic pathways. Lengauer et al. (54) studied a set
of aneuploid colorectal carcinoma cell lines and demonstrated a
pronounced chromosome instability (CIN) phenotype that
encompassed gains or losses of chromosomes in excess of
10–2/cell/generation. Chromosomal instability was considered
as a causative factor leading to aneuploidy rather than a result
of aneuploidy, since diploid cell lines that were genetically
modified to produce an aneuploid population did not exhibit
chromosome instability. Fusion experiments between diploid
and aneuploid cell lines suggested that the molecular basis
underlying CIN is dominant. One mechanism for the development of chromosomal instability is a defective mitotic
checkpoint, the regulatory pathway that monitors the assembly
of the mitotic spindles. In support of this hypothesis, Cahill
et al. (55) identified heterozygous mutations in hBub1, a
protein that participates in the mitotic checkpoint, in 5%
382
of the aneuploid cell lines that exhibited CIN. Since the
heterozygous allele resulted in a chromosome instability phenotype, these investigators surmised that the mutation functioned
in a dominant-negative manner inactivating the mitotic checkpoint. This prediction was supported by additional studies in
which expression of mutant proteins was shown to alter the
mitotic checkpoint in diploid cell lines.
The precise manner by which mutations in hBub1 promote
aneuploidy has not been established. Bub1 was first identified
along with several other mitotic checkpoint genes [Bub1–3
(56) and Mad1–3 (57)] in genetic screens for yeast genes
required for pre-anaphase delay in response to spindle disruption. Additional studies suggest that Bub1 acts as a sensor that
binds unattached kinetochores and delays the onset of mitosis
until all chromosomes are properly attached to the mitotic
spindle (58). The presence of unrepairable mitotic spindle
damage normally activates a checkpoint-mediated cell-cycle
arrest, which leads to cell elimination via apoptosis. Cells
expressing the Bub1 mutant proteins have a compromised
mitotic checkpoint and escape the subsequent apoptosis normally induced by severe spindle damage (58). The role of
apoptosis in the elimination of cells arising from a defective
mitosis is supported by studies in which the expression of
Bcl-XL, an anti-apoptotic gene, is sufficient to induce polyploidy in a murine prolymphocytic cell line (59). That Bub1
is essential in higher eukaryotes is supported by the fact that
homozygous Bub1 mutations are embryonic lethal in
Drosophila (60). Cells from these larvae show severe mitotic
abnormalities including accelerated exit from metaphase,
chromosome missegregation and fragmentation.
Following a transient mitotic arrest in the presence of spindle
damage, some cancer cells are able to resume the cell cycle
by entering a second round of DNA synthesis leading to
endoreduplication (two rounds of DNA synthesis without an
intervening mitosis). Cells lacking functional gene products
involved in the G1/S checkpoint such as p53 (61,62), pRb
(62), p16INK4A (62) or p21Waf1/Cip1 (61,63), as well as cells
that over-express myc (64) have been shown to undergo
endoreduplication following a transient mitotic arrest in the
presence of spindle damage.
Whether a mitotic checkpoint defect really plays a role in
the development of aneuploidy has been questioned by two
recent studies. In one, no mutations were detected in five
additional mitotic checkpoint genes surveyed from the 19
aneuploid colorectal tumors described above (65). In the other,
no hBub1 mutations were detected amongst 31 head and
neck carcinomas (66). However, there are many proteins that
participate in the segregation of chromosomes; mutations in
any one of them can contribute to the formation of aneuploidy.
A recent study by Lee et al. (67) detected mutations in p53,
Bub1 or Mad3L (mitotic checkpoint) in four spontaneous
thymic lymphomas arising in mice homozygous for Brca2
truncation. These authors further showed that a dominantnegative Bub1 can reverse a growth defect and initiate transformation in mouse embryonic fibroblasts homozygous for
the Brca truncation. These combined results suggest that
deficiencies in the mitotic checkpoint promote chromosomal
aberrations, aneuploidy and transformation, but it may occur
only in cells harboring other mutations.
Types of genetic instability in cancer
It has been suggested that there are two major mechanisms
for the generation of genetic instability (Figure 2). One
Significance of multiple mutations in cancer
Fig. 3. Evolution of a tumor based on selection for mutator mutations. As cancers develop there is likely to be a series of barriers that prevent further growth.
Only rare mutant cells within the tumor can overcome each of these barriers. Some of these mutant cells would be generated by mutations in genes required
for the maintenance of genetic stability in normal cells. Thus, with each round of selection there would be a progressive enrichment for mutations in mutator
genes.
Fig. 4. Timing of mutations during tumor progression. The mutator
phenotype hypothesis proposes that genes required for the maintenance of
genetic stability are among the early targets of DNA damage by
carcinogens. Mutations in these genes would result in other mutations
throughout the genome. Mutations in cancer associated genes would occur
later.
mechanism involves mutations in DNA mismatch correction
genes and is manifested by microsatellite instability. The
other involves mutations in genes required for chromosomal
segregation and is manifested by fragmentation of chromosomes and/or duplication and deletions of whole chromosomes,
rather than by sequence changes involving only a few nucleotides. In the first mechanism, two allelic mutations are required
to generate microsatellite instability. In the second mechanism,
only one mutation may be required, since the dominant nature
of chromosome instability phenotype observed in cell fusion
(54) suggests that modification in only one allele can result in
chromosomal instability. Thus, it could be argued that a mutator
phenotype is more likely to be manifested by chromosomal
instability than by point mutations. However, this dichotomy
may be artificial and simply reflect the relative sensitivities of
both assays and the difficulties inherent in detection of random
point mutations other than changes in the lengths of microsatellites. Chromosomal alterations involve millions of nucleotides
and can be readily demonstrated by cytogenetic techniques.
There are hundreds of thousands of microsatellites in the
human genome and each target microsatellite contains many
nucleotides; slippage at any single position within a repeat
sequence should result in a detectable change in its length
after PCR amplification and gel electrophoresis. Thus, scoring
for mutations in only a few microsatellite sequences may not
be adequate for determining whether or not a specific tumor
exhibits microsatellite instability. Mutations generated by genetic instability are likely to be random events and only a few
would result in a sufficiently selective advantage to generate
a clonal population of cancer cells. Methods have been
developed to detect clonal mutations. However, we still lack
methods for quantitating random nucleotide changes in DNA
sequences in tumors, particularly when they involve only a
few nucleotide substitutions.
Coupling of increased mutagenesis and clonal selection
Initially, two mechanisms, increased mutations and clonal
selection, were invoked to account for a mutator phenotype in
cancer. Recent experiments in bacteria (68) and modeling
383
K.R.Loeb and L.A.Loeb
studies (69) provide evidence that both mechanisms are operative and, moreover, intradependent. Mao et al. (68) showed
that the spontaneous mutation rate in bacteria progressively
increases from 1/100 000 to 1/200 after four rounds of selection
for mutations in different genes. Moreover, exposure to a
mutagen followed by subsequent rounds of selection for
different mutations resulted in a population that was 100%
mutators. The inference is that with each round of selection,
one not only selects for a particular mutation, such as resistance
to rifamycin, but one also selects for mutations that increase
the mutation rate throughout the genome, i.e. mutators. The
progressive enrichment of mutator mutations with successive
rounds of clonal selection may be relevant to tumor progression.
As cancers expand, they encounter a series of restrictive
blockades that limits further growth. These limitations to
growth might include interference by surrounding tissues,
reduced nutrition, reduced oxygen levels, need for growth
factors, inadequate blood supply, etc. Each of these blockades
might be overcome by mutations that provide a growth
advantage and establish new clonal populations (Figure 3).
With each round of selection there would be a ‘piggy-backing’
of mutator mutants and increase in mutation frequency. At
some step in tumor progression the mutator phenotype might
be detrimental and be lost during clonal selection. However,
the multiple mutations created would be persistent and present
even in later-stage tumors.
Future directions
The findings that multiple mutations are present in human
tumors brings forth a series of important questions: what is
the origin of these mutations? If they result from random
events, how can we develop quantitative assays for their
assessment? Are they phenomena that parallel tumor progression or are they rate limiting for tumor progression? If they
are rate limiting, can we reduce the rate of accumulation and
thus prevent cancer by delay? Since it takes ~20 years from
the time of exposure to a carcinogen to the development of a
clinically detectable tumor, even a 2-fold decrease in the rate
of tumor progression would reduce cancer death in adults (70).
So far, only a limited number of mutator genes have been
identified based on their presence in inherited human diseases
that exhibit a high proclivity towards the development of
specific tumors, or based on the production of mutators in
bacteria or in cell culture. Considering the large number of
genes involved in DNA replication, recombination and DNA
repair, and those required for the correct partitioning of
chromosomes during cell replications, it seems likely that
many other mutator genes will soon be discovered, and that
some of these will have an important role in the pathogenesis
of specific cancers (Figure 4). Interestingly, the concept of a
mutator phenotype was based on mutations in DNA polymerases that render them error-prone (33); yet, we still have
only fragmentary evidence for the presence of mutations in
DNA polymerases in tumors (71). If a mutator phenotype is
important in the development of cancer it is likely to be an
early event. Mutations in many of the oncogenes found in
different tumors are likely to occur later. As a result, we would
argue that drugs directed against many cancer-associated genes
are unlikely to reverse the malignant phenotype.
DNA sequencing provides a powerful tool to establish
whether specific mutations are present in a tumor. However,
this and related technologies are inadequate for detecting
384
random mutations within a cell population. New techniques
are required to detect random mutations and to determine if
they accumulate during tumor progression. Stratification of
normal individuals based on accumulation of random mutations
may be predictive of the development of certain tumors.
Stratification of tumors on the basis of accumulated mutations
may be predictive of malignant potential.
Acknowledgements
This work was supported by grants from the National Cancer Institute (R35
CA 39903, CA 80993) and the National Institute of Aging (AG 01751).
K.R.Loeb is a research fellow of the College of American Pathologists.
References
1. Ames,B.N., Gold,L.S. and Willet,W.C. (1995) The causes and prevention
of cancer. Proc. Natl Acad. Sci. USA, 92, 5258–5265.
2. Ames,B.N., Profet,M. and Gold,L.S. (1990) Nature’s chemicals and
synthetic chemicals: comparative toxicology. Proc. Natl Acad. Sci. USA,
87, 7782–7786.
3. Loeb,L.A. (1989) Endogenous carcinogenesis: molecular oncology into
the twenty-first century—presidential address. Cancer Res., 49, 5489–5496.
4. Lindahl,T. and Nyberg,B. (1972) Rate of depurination of native
deoxyribonucleic acid. Biochemistry, 11, 3610–3618.
5. Cathcart,R., Scheiers,E., Saul,R.L. and Ames,B.N. (1984) Thymine glycol
and thymidine glycol in human and rat urine: A possible assay for
oxidative DNA damage. Proc. Natl Acad. Sci. USA, 81, 5633–5637.
6. Shigenaga,M.K., Cimeno,C.J. and Ames,B.N. (1989) Urinary
8-hydroxy-2⬘ deoxyguanosine as a biological marker of in vivo oxidative
DNA damage. Proc. Natl Acad. Sci. USA, 86, 9697–9701.
7. Basu,A.K.,
Wood,M.L.,
Niedernhofer,L.J.,
Ramos,L.A.
and
Essigmann,J.M. (1993) Mutagenic and genotoxic effects of three vinyl
chloride-induced DNA lesions: 1,N6-ethenoadenine, 3,N4-ethenocytosine
and 4-amino-5-(imidazol-2-yl)imidazole. Biochemistry, 32, 12793–12801.
8. Cheng,K.C., Cahill,D.S., Kasai,H., Nishimura,S. and Loeb,L.A. (1992)
8-hydroxyguanine, an abundant form of oxidative DNA damage, causes
G→T and A→C substitutions. J. Biol. Chem., 267, 166–172.
9. Shibutani,S., Takeshita,M. and Grollman,A.O. (1991) Insertion of specific
bases during DNA synthesis past the oxidation-damaged base 8-oxo-dG.
Nature, 349, 431–434.
10. Lindahl,T. and Nyberg,B. (1974) Heat-induced deamination of cytosine
residues in deoxyribonucleic acid. Biochemistry, 13, 3405–3410.
11. Loeb,L.A. and Cheng,K.C. (1990) Errors in DNA synthesis: a source of
spontaneous mutations. Mutat. Res., 238, 297–304.
12. Murli,S. and Walker,G.C. (1993) SOS mutagenesis. Curr. Opin. Genet.
Dev., 3, 719–725.
13. Umar,A. and Kunkel,T.A. (1996) DNA-replication fidelity, mismatch repair
and genomic instability in cancer cells. Eur. J. Biochem., 238, 297–307.
14. Hindges,R. and Hubscher,U. (1995) Production of active mouse DNA
polymerase delta in bacteria. Gene, 158, 241–246.
15. Syvaoja,J., Suomensaari,S., Nishida,C., Goldsmith,J.S., Chui,G.S.J., Jain,S.
and Linn,S. (1990) DNA polymerases α, δ and ε: three distinct enzymes
from HeLa cells. Proc. Natl Acad. Sci. USA, 87, 6664–6668.
16. Kunkel,T.A. and Alexander,P.S. (1986) The base substitution fidelity of
eucaryotic DNA polymerases. J. Biol. Chem., 261, 160–166.
17. Fry,M. and Loeb,L.A. (1992) A DNA polymerase α pause site is a hot
spot for nucleotide misinsertion. Proc. Natl Acad. Sci. USA, 89, 763–767.
18. Modrich,P. (1987) DNA mismatch correction. Annu. Rev. Biochem., 56,
435–466.
19. Wang,H., Lawrence,C.W., Li,G.-M. and Hays,J.B. (1999) Specific binding
of human MSH2-MSH6 mismatch-repair protein heterodimers to DNA
incorporating thymine- or uracil-containing UV light photoproducts
opposite mismatched bases. J. Biol. Chem., 274, 16894–16900.
20. Sweasy,J.B. and Loeb,L.A. (1992) Mammalian DNA polymerase β can
substitute for DNA polymerase I during DNA replication in Escherichia
coli. J. Biol. Chem., 267, 1407–1410.
21. Srivastava,D.K., Husain,I., Arteaga,C.L. and Wilson,S.H. (1999) DNA
polymerase beta expression differences in selected human tumors and cell
lines. Carcinogenesis, 20, 1049–1054.
22. Clairmont,C.A. and Sweasy,J.B. (1996) Dominant negative rat DNA
polymerase β mutants interfere with base excision repair in Saccharomyces
cerevisiae. J. Bacteriol., 178, 656–661.
Significance of multiple mutations in cancer
23. Canitrot,Y., Frechet,M., Servant,L., Cazaux,C. and Hoffmann,J.-S. (1999)
Overexpression of DNA polymerase β: a genomic instability enhancer
process. FASEB J., 13, 1107–1111.
24. Witkin,E.M. (1976) Ultraviolet mutagenesis and inducible DNA repair in
Escherichia coli. Bacteriol. Rev., 40, 869–907.
25. Tang,M., Bruck,I., Eritja,R., Turner,J., Frank,E.G., Woodgate,R.,
O’Donnell,M. and Goodman,M.F. (1998) Biochemical basis of SOSinduced mutagenesis in Escherichia coli: reconstitution of in vitro lesion
bypass dependent on the UmuD’2C mutagenic complex and RecA protein.
Proc. Natl Acad. Sci. USA, 95, 9755–9760.
26. Tang,M., Shen,X., Frank,E.G., O’Donnell,M., Woodgate,R. and
Goodman,M.F. (1999) UmuD’2C is an error-prone DNA polymerase,
Escherichia coli pol V. Proc. Natl Acad. Sci. USA, 96, 8919–8924.
27. Johnson,R.E., Prakash,S. and Prakash,L. (1999) Requirement of DNA
polymerase activity of yeast Rad30 protein for its biological function.
J. Biol. Chem., 274, 15975–15977.
28. Masutani,C., Araki,M., Yamada,A., Kusumoto,R., Nogimori,T.,
Maekawa,T., Iwai,S. and Hanaoka,F. (1999) Xeroderma pigmentosum
variant (XP-V) correcting protein from HeLa cells has a thymine dimer
bypass DNA polymerase activity. EMBO J., 18, 3491–3501.
29. Johnson,R.E., Kondratick,C.M., Prakash,S. and Prakash,L. (1999) hRAD30
mutations in the variant form of xeroderma pigmentosum. Science, 285,
263–265.
30. Hartwell,L.H. and Kastan,M.B. (1994) Cell cycle control and cancer.
Science, 266, 1821–1827.
31. Crook,T., Marston,N.J., Sara,E.A. and Vousden,K.H. (1994) Transcriptional
activation by p53 correlates with suppression of growth but not
transformation. Cell, 79, 814–827.
32. Orr-Weaver,T.L. and Weinberg,R.A. (1998) A checkpoint on the road to
cancer. Nature, 392, 223–224.
33. Loeb,L.A., Springgate,C.F. and Battula,N. (1974) Errors in DNA replication
as a basis of malignant change. Cancer Res., 34, 2311–2321.
34. Nowell,P.C. (1976) The clonal evolution of tumor cell populations. Science,
194, 23–28.
35. Tlsty,T.D., Margolin,B.H. and Lum,K. (1989) Differences in the rates of
gene amplification in nontumorigenic and tumorigenic cell lines as
measured by Luria–Delbruck fluctuation analysis. Proc. Natl Acad. Sci.
USA, 86, 9441–9445.
36. Peinado,M.A., Malkhosyan,S., Velazquez,A. and Perucho,M. (1992)
Isolation and characterization of allelic loss and gains in colorectal tumors
by arbitrarily primed polymerase chain reaction. Proc. Natl Acad. Sci.
USA, 89, 10065–10069.
37. Ionov,Y., Peinado,M.A., Malkhosyan,S., Shibata,S. and Perucho,M. (1993)
Ubiquitous somatic mutations in simple repeated sequences reveal a new
mechanism for colonic carcinogenesis. Nature, 363, 558–561.
38. Perucho,M. (1996) Cancer of the microsatellite mutator phenotype. Biol.
Chem., 377, 675–684.
39. Rampino,N., Yamamoto,H., Ionov,Y., Li,Y., Sawai,H., Reed,J.C. and
Perucho,M. (1997) Somatic frameshift mutations in the BAX gene in colon
cancers of the microsatellite mutator phenotype. Science, 275, 967–969.
40. Markowitz,S., Wang,J., Myeroff,L. et al. (1995) Inactivation of the type
II TGF-β receptor in colon cancer cells with microsatellite instability.
Science, 268, 1336–1338.
41. Fishel,R., Lescoe,M.K., Rao,M.R.S., Copeland,N.G., Jenkins,N.A.,
Garber,J., Kane,M. and Kolodner,R. (1993) The human mutator gene
homolog MSH2 and its association with hereditary nonpolyposis colon
cancer. Cell, 75, 1027–1038.
42. Herman,J.G., Umar,A., Polyak,K. et al. (1998) Incidence and functional
consequences of hMLH1 promoter hypermethylation in colorectal
carcinoma. Proc. Natl Acad. Sci. USA, 95, 6870–6875.
43. Leung,S.Y., Yuen,S.T., Chung,L.P., Chu,K.M., Chan,A.S.Y. and Ho,J.C.I.
(1999) hMLH1 promoter methylation and lack of hMLH expression in
sporadic gastric carcinomas with high-frequency microsatellite instability.
Cancer Res., 59, 159–164.
44. Jackson,A.L., Chen,R. and Loeb,L.A. (1998) Induction of microsatellite
instability by oxidative DNA damage. Proc. Natl Acad. Sci. USA, 95,
12468–12473.
45. Richards,B., Zhang,H., Phear,G. and Meuth,M. (1997) Conditional mutator
phenotypes in hMSH2 deficient tumor cell lines. Science, 277, 1523–1526.
46. Kallioniemi,A., Kallioniemi,O.-P., Piper,J., Tanner,M., Stokke,T., Chen,L.,
Smith,H.S., Pinkel,D., Gray,J.W. and Waldman,F.M. (1994) Detection and
mapping of amplified DNA sequences in breast cancer by comparative
genomic hybridization. Proc. Natl Acad. Sci. USA, 91, 2156–2160.
47. Iwabuchi,H., Sakamoto,M., Sakunaga,H. and Ma,Y.Y. (1995) Genetic
analysis of benign, low-grade and high-grade ovarian tumors. Cancer Res.,
55, 6172–6180.
48. Kerangueven,F., Noguchi,T., Coulier,F., Allione,F., Wargniez,V., SimonyLafontaine,J., Longy,M., Jacquemier,J., Sobol,H., Eisinger,F. and
Birnbaum,D. (1997) Genome-wide search for loss of heterozygosity shows
extensive genetic diversity of human breast carcinomas. Cancer Res., 57,
5469–5474.
49. Jackson,A.L. and Loeb,L.A. (1998) The mutation rate and cancer. Genetics,
148, 1483–1490.
50. Kuerbitz,S.J., Plunkett,B.S., Walsh,W.V. and Kastan,M.B. (1992) Wildtype p53 is a cell cycle checkpoint determinant following irradiation. Proc.
Natl Acad. Sci. USA, 89, 7491–7495.
51. LaCasse,E.C., Baird,S., Korneluk,R.G. and MacKenzie,A.E. (1998) The
inhibitors of apoptosis (IAPs) and their emerging role in cancer. Oncogene,
17, 3247–3259.
52. Duesberg,P., Rausch,C., Rasnick,D. and Hehlmann,R. (1998) Genetic
instability of cancer cells is proportional to their degree of aneuploidy.
Proc. Natl Acad. Sci. USA, 95, 13692–13697.
53. Andreassen,P.R., Martineau,S.N. and Margolis,R.L. (1996) Chemical
induction of mitotic checkpoint override in mammalian cells results in
aneuploidy following a transient tetraploid state. Mutat. Res., 372, 181–194.
54. Lengauer,C., Kinzler,K.W. and Vogelstein,B. (1997) Genetic instability in
colorectal cancers. Nature, 386, 623–627.
55. Cahill,D., Lengauer,C., Yu,J., Riggins,G.J., Willson,J.K., Markowitz,S.D.,
Kinzler,K.W. and Vogelstein,B. (1998) Mutations of mitotic checkpoint
genes in human cancers. Nature, 392, 300–303.
56. Hoyt,M.A., Totis,L. and Roberts,B.T. (1991) S. cerevisiae genes required
for cell cycle arrest in response to loss of microtubule function. Cell, 66,
507–517.
57. Liehr,J.G. (1997) Androgen-induced redox changes in prostate cancer
cells: what are causes and effects? J. Natl Cancer Inst., 89, 3–4.
58. Taylor,S.S. and McKeon,F. (1997) Kinetochore localization of murine
Bub1 is required for normal mitotic timing and checkpoint response to
spindle damage. Cell, 89, 727–735.
59. Minn,A.J., Boise,L.H. and Thompson,C.B. (1996) Expression of Bcl-XL
and loss of p53 can cooperate to overcome a cell cycle checkpoint induced
by mitotic spindle damage. Genes Dev., 10, 2621–2631.
60. Basu,J., Bousbaa,H., Logarinho,E., Li,Z., Williams,B.C., Lopes,C.,
Sunkel,C.E. and Goldberg,M.L. (1999) Mutations in the essential spindle
checkpoint gene bub1 cause chromosome missegregation and fail to block
apoptosis in Drosophila. J. Cell Biol., 146, 13–28.
61. Lanni,J.S. and Jacks,T. (1998) Characterization of the p53-dependent
postmitotic checkpoint following spindle disruption. Mol. Cell. Biol., 18,
1055–1064.
62. Khan,S.H. and Wahl,G.M. (1998) p53 and pRb prevent rereplication in
response to microtubule inhibitors by mediating a reversible G1 arrest.
Cancer Res., 58, 396–401.
63. Stewart,Z.A., Leach,S.D. and Pietenpol,J.A. (1999) p21(Waf1/Cip1)
inhibition of cyclin E/Cdk2 activity prevents endoreduplication after
mitotic spindle disruption. Mol. Cell Biol., 19, 205–215.
64. Li,X. and Nicklas,R.B. (1995) Mitotic forces control a cell-cycle
checkpoint. Nature, 373, 630–632.
65. Cahill,D.P., da Costa,L.T., Carson-Walter,E.B., Kinzler,K.W., Vogelstein,B.
and Lengauer,C. (1999) Characterization of MAD2B and other mitotic
spindle checkpoint genes. Genomics, 58, 181–187.
66. Yamaguchi,K., Okami,K., Hibi,K., Wehage,S.L., Jen,J. and Sidransky,D.
(1999) Mutation analysis of hBUB1 in aneuploid HNSCC and lung cancer
cell lines. Cancer Lett., 139, 183–187.
67. Lee,H., Trainer,A.H., Friedman,L.S., Thistlethwaite,F.C., Evans,M.J.,
Ponder,B.A. and Venkitaraman,A.R. (1999) Mitotic checkpoint inactivation
fosters transformation in cells lacking the breast cancer susceptibility gene,
Brca2. Mol. Cell, 4, 1–10.
68. Mao,E.F., Lane,L., Lee,J. and Miller,J.H. (1997) Proliferation of mutators
in a cell population. J. Bacteriol., 179, 417–422.
69. Taddei,F., Radman,M., Maynard-Smith,J., Toupance,B., Gouyon,P.H. and
Godelle,B. (1997) Role of mutator alleles in adaptive evolution. Nature,
387, 700–702.
70. Loeb,L.A. (1996) Genetic instability in cancer. In Lindahl,T. (ed.) Genetic
Instability in Cancer. Cold Spring Harbor Press, Plainview, NY, pp.
329–342.
71. Sekowski,J.W., Malkas,L.H., Schnaper,L., Bechtel,P.E., Long,B.J. and
Hickey,R.J. (1998) Human breast cancer cells contain an error-prone DNA
replication apparatus. Cancer Res., 58, 3259–3263.
Received September 8, 1999; accepted September 20, 1999
385