Abeloff's Clinical Oncology, 4th Edition

Part I – Science of Clinical Oncology

Section B – Genesis of Cancer

Chapter 10 – DNA Damage Response Pathways and Cancer

James M. Ford,Michael B. Kastan




DNA repair and the cellular response to DNA damage are critical for maintaining genomic stability.



Defects in DNA repair or the response to DNA damage encountered from endogenous or external sources results in an increased rate of genetic mutations, often leading to the development of cancer.



Inherited mutations in DNA damage response pathway genes often result in cancer susceptibility.



The major active pathways for DNA repair in humans are nucleotide excision repair, base excision repair, mismatch DNA repair, translesional DNA synthesis, and homologous recombination or nonhomologous end joining processes for double-strand break repair.



Defects in nucleotide excision repair lead to the skin cancer-prone syndrome xeroderma pigmentosum, as well as Cockayne syndrome and trichothiodystrophy.



Defects in base excision repair can result in enhanced colon adenomas and cancers.



Defects in mismatch repair result in hereditary nonpolyposis colorectal cancer syndrome.



Defects in DNA double-strand break repair and response pathways underlie a number of cancer-prone disorders, including ataxia-telangiectasia, Nijmegen breakage syndrome, Bloom syndrome, Werner syndrome, Rothmund-Thompson syndrome, and Fanconi anemia.



The highly cancer-prone Li-Fraumeni syndrome, due to inherited p53 mutations, and breast-ovarian cancer syndrome, due to inherited mutations of the BRCA1 and BRCA2 genes, exhibit defects in multiple DNA repair and DNA damage response pathways.


Cancer is a genetic disease that is caused by the accumulation over time of changes to the normal DNA sequence resulting in alterations, loss, or amplification of genes that are important for normal cellular functions and growth properties, including many proto-oncogenes and tumor suppressor genes. Nearly all cancers are clonal in origin; that is, they originate from a single progenitor cell rather than a group of cells. The development of cancer in a particular cell type or tissue is caused by a series of specific mutations, each of which could be caused by DNA replication errors or unrepaired endogenous or exogenous DNA damage or be the result of inherited mutations. For the most common cancers, multiple genetic events occur in many different genes during the process of carcinogenesis, suggesting that an early and perhaps necessary event in the cancer process is an underlying defect in mechanisms to maintain genomic stability.[1] In fact, alterations in the specific genes that are required for recognizing, processing, and responding to DNA damage may result in an enhanced rate of accumulation of additional mutations, recombinational events, chromosomal abnormalities, and gene amplification.[2]

In addition, cancer cells must be able to tolerate increased amounts of unrepaired DNA damage associated with genomic instability and therefore frequently inactivate DNA damage-inducible signaling and checkpoint pathways. Therefore, DNA repair and the DNA damage response are essential not only for the basic processes of transcription and replication required for cellular survival, but also for maintaining genomic stability and avoiding the development of malignancies. Numerous links have been identified between oncogenesis and acquired or inherited defects in genomic stability that cause a “mutator” phenotype, highlighting the key role of DNA protection systems in tumor prevention. This chapter reviews the major DNA repair mechanisms that are active in mammalian cells and our emerging understanding of DNA damage-signaling pathways that integrate with other cellular processes that regulate transcription, replication, cell division, and apoptosis in response to DNA damage. The relevance of these mechanisms to cancer is explored by focusing on several human cancer predisposition syndromes that are caused by underlying defects in DNA damage processing.

Advances in cancer genetics have defined three general groups of genes that are involved in the development of human cancers: oncogenes, tumor suppressor genes, and DNA damage repair and response genes. The latter set of genes is particularly important to hereditary cancer susceptibility, owing to their direct involvement in maintaining genomic stability. Much of what we know about cancer genes in sporadic tumors comes from the study of relatively rare inherited cancer syndromes caused by mutations passed along in the germline DNA of families and predisposing to the development of cancers, often at a very young age and at a high incidence, in affected carriers. Individuals who inherit a germline mutation in genes that are involved in or required for DNA repair usually are at increased risk for the development of cancer, owing to the enhanced frequency of mutations and increased genomic instability. Susceptibility to cancer may also be affected by environmental factors and multiple low-penetrance modifier genes.

Many converging lines of experimental evidence reveal the complexity of the cellular responses to DNA damage and their role in malignant transformation.[3] A number of interrelated biochemical pathways exist that influence the following actions: (1) the metabolism of potentially mutagenic or carcinogenic agents, (2) the efficiency and manner by which damaged DNA is recognized and repaired, (3) cell cycle progression and the coordination of DNA replication and cell division relative to the repair of lesions, and (4) the decision point determining survival or the active induction of programmed death of cells that carry different types and amounts of DNA damage. Many cellular pathways have evolved that require hundreds of gene products for the direct repair of DNA damage and involve excision of damaged DNA bases and joining of broken DNA strands ( Table 10-1 ). The central role of DNA damage responses in neoplastic transformation has been highlighted by the discovery that mutations in several classes of genes that are required for DNA repair and the maintenance of genomic integrity result in a predisposition to the development of certain malignancies.[4] In fact, a number of rare inherited disorders have been described that appear to be caused by defects in the repair of DNA lesions ( Table 10-2 ), and many of these are associated with an increased risk of developing certain cancers.[5]

Table 10-1   -- Human DNA Repair Pathways

DNA Repair Pathway

Type of DNA Damage

Approximate No. of Genes

Nucleotide excision repair

Bulky or helix-distorting DNA adducts, e.g., ultraviolet photoproducts, carcinogen adducts


Base excision repair

Oxidative DNA damage



Spontaneous depurination


Mismatch repair

Mispaired nucleotides



1–15 nucleotide insertion-deletion loops


Homologous recombination

Double-strand DNA breaks, DNA cross-links


Nonhomologous end joining

Double-strand DNA breaks




Table 10-2   -- Human Genetic Diseases Involving Defects in DNA Damage Response Pathways



Biologic Functions

Clinical Features


Xeroderma pigmentosum


Nucleotide excision repair

Sunlight hypersensitivity

UV, chemical carcinogens



Translesional DNA synthesis

Neurologic defects





Skin cancers


Cockayne syndrome


Transcription-coupled repair

Growth retardation

UV, chemical carcinogens




Mental retardation

Reactive oxygen species




Premature aging





Sunlight hypersensitivity




Nucleotide excision repair

Sulfur-deficient brittle hair





Dry, scaly skin





Mental and physical retardation





Sunlight sensitivity


Hereditary nonpolyposis Colorectal cancer (Lynch syndrome)


Mismatch repair

Colorectal, endometrial, gastric, bile duct cancers

6-Thioguanine and cisplatin resistance

Ataxia telangiectasia


DNA damage-responsive kinase

Cerebellar ataxia

Ionizing radiation
















Ataxia telangiectasia–like disease


Double-strand break repair

Similar to AT


Nijmegen breakage syndrome


Double-strand break repair


Ionizing radiation









Lymphomas, neuroblastoma







Bloom syndrome


DNA helicase

Sunlight hypersensitivity

UV, hydroxyurea



Homologous recombination at stalled replication forks?

Growth retardation





Leukemias, lymphomas





Breast and intestinal cancers


Werner syndrome


DNA helicase

Premature aging

4-NQO, camptothetin



Homologous recombination?





Translesional synthesis?

Soft tissue sarcomas





Melanoma, thyroid cancer


Rothmund-Thompson syndrome


DNA helicase

Growth deficiency





Sunlight sensitivity





Osteogenic sarcomas





Squamous cell carcinomas


Fanconi anemia


Interstrand crosslink repair

Growth retardation

Bifunctional alkylating agents



Homologous recombination

Anatomic defects

Ionizing radiation




Bone marrow failure





Myeloid leukemia





Squamous cell cancers


Li-Fraumeni syndrome



Breast cancer




Cell cycle checkpoints

Brain cancers




Nucleotide excision repair

Adrenocortical carcinoma










Bone and soft tissue sarcomas


Li-Fraumeni–like syndrome


DNA damage responsive kinase

Similar to Li-Fraumeni syndrome


Breast-ovarian cancer syndrome


Double-strand break repair

Breast cancer

Ionizing radiation?



Nucleotide excision repair

Ovarian cancer

UV, cisplatin

UV, ultraviolet.





DNA undergoes several types of spontaneous modifications, and it also can react with many physical and chemical agents, some of which are endogenous products of normal cellular metabolism (e.g., reactive oxygen species) whereas others, including ionizing radiation and ultraviolet light, are threats from the external environment ( Fig. 10-1 ). One pronounced example is exposure to genotoxic compounds in cigarette smoke, which contributes to the development of some of the most common cancers seen in Western countries. Most active chemotherapeutic agents function by damaging DNA through alkylation, cross-linking, and other means, and mechanisms to repair these lesions determine the sensitivity of a tumor to such treatments. Damage to DNA can cause genetic mutations, and these mutations can lead to the development of cancer. DNA damage also may result in cell death, which can have serious consequences for the organism of which the cell is a part, for example, loss of irreplaceable neurons in the brain. Accumulation of damaged DNA is thought to contribute to some of the features of aging. Therefore, it is not surprising that a complex set of cellular surveillance and repair mechanisms has evolved to reverse or limit potentially deleterious DNA damage. Some of these DNA repair systems are so important that life cannot be sustained without them. An increasing number of human hereditary diseases that are characterized by severe developmental problems or a predisposition to cancer have been found to be linked to deficiencies in DNA repair (see Table 10-2 ).


Figure 10-1  Cellular responses to DNA damage. Different types of DNA damage cause a variety of different types of lesions, and these, in turn, are dealt with by a variety of DNA repair mechanisms and signal various cellular response pathways. The outcome of DNA damage may be cell survival of a normal cell, cell death, or mutagenesis, possibly leading toward malignant transformation.




The results of DNA damage are diverse and frequently adverse. Acute effects arise from disturbed DNA metabolism, triggering cell cycle arrest or cell death. Long-term effects result from irreversible mutations contributing to oncogenesis and inherited genetic disorders. Many lesions block transcription, and this has elicited the development of a dedicated repair system, transcription-coupled repair (TCR), which displaces or removes the stalled RNA polymerase and assures preferential repair of lesions within the transcribed strand of expressed genes. [6] [7] [8] Transcriptional stress due to DNA lesions that block RNA polymerase and DNA strand breaks caused by DNA damage or stalled replication forks constitute two major signals for DNA damage-inducible responses, including apoptosis, [9] [10] [11]through both p53-dependent and independent mechanisms.[12]

Lesions also may interfere with DNA replication. Recently, a class of at least 10 specialized DNA polymerases was discovered that appear devoted to overcoming damage-induced replication stress. [13] [14] [15] These special polymerases take over temporarily from the stalled replicative DNA polymerases. Though translesion polymerases protect the genome, this solution to replication blocks comes at the expense of a higher replicative error rate, and mutations in some of these polymerases cause cancer susceptibility.[5] Therefore, detection of DNA lesions may occur by blocked transcription, replication, or specialized sensors. Although the precise molecular mechanisms by which the cell senses altered DNA remains obscure, such signals result in a complex cellular response that includes cell cycle checkpoints, DNA repair, and apoptosis.


DNA damage checkpoints initially were defined as regulatory pathways that control the ability of cells to arrest the cell cycle in response to DNA damage, allowing time for repair.[16] However, in addition to controlling cell cycle arrest, proteins that are involved in these pathways have been shown to control the activation of DNA repair pathways, [3] [17] [18] [19] [20] [21] the movement of DNA repair proteins to sites of DNA damage, [22] [23] [24] [25] [26] [27] and activation of transcriptional responses. [28] [29] [30] When damage is too significant or it benefits the tissue or organism as a whole, a cell may opt for the ultimate mode of rescue by initiating its own death via apoptosis. [31] [32] [33] As the DNA damage response pathway has been better defined at a molecular level, it has been seen as a network of interacting pathways that together execute the response.[34] Initial recognition of DNA damage occurs by a variety of damage-specific DNA binding proteins that either by themselves or together with complexes of associated proteins that are not directly involved in DNA repair may signal the DNA damage response.[35] Transduction and amplification of the DNA damage signal often is carried out by an overlapping set of conserved protein kinases, including the phosphoinositide-3-kinase-related proteins, which include ataxia-telangiectasia mutated (ATM) and ATM-Rad3-related (ATR) proteins, the checkpoint kinases Chk1 and Chk2, and others. [3] [36] [37] [38] [39] [40] [41] Many of these protein kinases are themselves targets for phosphorylation and activation; they then further target downstream genes that are critical to oncogenesis such as p53 and BRCA1. [34] [36] [42] [43] [44] The ultimate targets of this highly regulated DNA damage response include mechanisms for DNA repair, and although much of DNA repair is constitutive, a number of regulatory connections between the DNA damage response pathway and DNA repair have emerged.[3] In mammals, a large number of genes that are involved in DNA repair are transcriptionally induced in response to DNA damage, suggesting that many facets of repair are inducible, similar to the RecA-dependent SOS response in bacteria that enhances DNA repair and mutagenesis following DNA damage. [3] [21] [45] In fact, the p53 tumor suppressor gene is a central mediator of the DNA damage-inducible transcriptional response in humans, and p53 mutant mammalian cells are deficient in several aspects of DNA repair. [17] [18] [19] [20] [21] [46] Therefore, the mammalian DNA damage-inducible response pathway is highly regulated and fine-tuned to determine whether a particular cell type proceeds to a cell cycle checkpoint and DNA repair or to cell death following a significant damage insult. Defects at any level of these pathways can alter repair and result in carcinogenesis (see Fig. 10-1 ).


DNA repair may be defined as the cellular responses that are associated with the restoration of the normal nucleotide sequence following events that damage or alter the genome.[47] Given the wide variety of DNA damage that a cell encounters, it is not surprising that a large number of repair systems are available to handle these insults. Indeed, many of the repair systems are broadly overlapping and interacting, several sharing certain strategies and even specific gene products. Much of what is known about the basic mechanisms of many types of DNA repair comes from the study of lower organisms, such as bacteria and yeast, since many aspects of these pathways have been conserved through evolution. Inherited defects in any of the major DNA repair pathways in humans, in general, predisposes to malignancy, and several of these syndromes will be discussed in detail. In humans, a great deal has been learned about DNA repair from the often rare, autosomal recessive hereditary syndromes associated with defects in DNA repair genes.[5]

Nucleotide Excision Repair

The most versatile and ubiquitous mechanisms for DNA repair are those in which the damaged or incorrect part of a DNA strand is excised and then the resulting gap is filled by repair replication using the complementary strand as template. The redundancy of genetic information provided by the duplex DNA structure is essential to the maintenance of the genome by this “cut and patch” mode known asexcision repair. Each DNA strand can serve as a template for replication-based repair of the other strand. Excision repair was discovered in the early 1960s through basic studies on the effects of ultraviolet (UV) irradiation on DNA synthesis and repair replication in bacteria. [48] [49] [50] Nucleotide excision repair (NER) functions to remove many types of lesions, including bulky base adducts of chemical carcinogens, intrastrand cross-links, and UV-induced cyclobutane primidine dimers (CPDs) and 6-4 photoproducts. Such lesions may serve as structural blocks to transcription and replication owing to distortion of the helical conformation of DNA, and they also may result in mutations if translesional replication occurs or if they are not repaired correctly. The sequential steps for NER are (1) recognition of the damaged site, (2) incision of the damaged DNA strand near the site of the defect, (3) removal of a stretch of the affected strand containing the lesion, (4) repair replication to replace the excised region with a corresponding stretch of normal nucleotides using the complementary strand as a template, and (5) ligation to join the repair patch at its 3′ end to the contiguous parental DNA strand ( Fig. 10-2 ). [51] [52] This excision repair pathway can remove DNA damage from sites throughout the genome and is termed global genomic repair (GGR). The majority of human NER genes have been identified and cloned, and many have been shown to be mutated in hereditary NER-deficient, cancer-prone diseases. [21] [46] [53]


Figure 10-2  Mechanism for human nucleotide excision repair. Ultraviolet irradiation-induced adducts in genomic DNA are recognized by the XPE and XPC/hHR23B protein heteroduplexes that recruit the XPA/RPA complex and the larger TFIIH protein complex. Dimers that occur in the transcribed strand of an expressed gene result in a blocked RNA polymerase II molecule, which together with the CSA and CSB gene products serves to recruit the downstream repair machinery. The TFIIH complex contains helicases, including XPB and XPD, that unwind the DNA and allow the other repair proteins access for incision and excision of the damaged DNA oligonucleotide. After excision, repair replication based on the normal DNA template and ligation of the newly synthesized DNA sequence occurs. In total, more than 25 proteins participate in NER.



A unique problem arises if a bulky lesion is encountered by a translocating RNA polymerase making messenger RNA, before repair enzymes have removed the damage and restored intact DNA. The polymerase may be arrested at the site of the lesion and prevent access to the damage by repair enzymes. Furthermore, the arrest of transcription in human cells is a strong signal for p53 activation and can trigger apoptosis.[12] In this situation, a dedicated excision repair pathway known as transcription-coupled repair (TCR) comes to the rescue to displace the RNA polymerase and then efficiently repairs the blocking lesion so that transcription may resume—and so that the cell may survive.[45] The existence of a mechanism to facilitate the preferential repair of the transcribed strand of active genes in both eukaryotes and prokaryotes raises a number of questions as to its evolutionary role. Certainly, strand-specific repair of active genes should be important for maintaining genomic stability in multicellular organisms by helping to avoid transforming mutations in expressed proto-oncogenes and tumor suppressor genes. However, the lack of an increased incidence of malignancy in individuals with Cockayne syndrome (CS), a disease in which TCR has been selectively lost but GGR has been retained, argues against the idea that this NER pathway is critical in the process of transformation. The existence of TCR in unicellular and prokaryotic organisms suggests that its function might be more important to the basic processes of transcription and replication required for cellular survival than for avoidance of transforming mutations. Recent evidence suggests that cells from patients with CS, trichothiodystrophy (TTD), and xeroderma pigmentosum/CS share a defect in repair of oxidative DNA damage that might explain the overlapping progeroid features of these syndromes.[54]

Recently, it has become apparent that the GGR subpathway of NER is damage inducible and highly regulated by both transcriptional and post-translational mechanisms following DNA damage, in concert with damage inducible cell cycle checkpoints and apoptosis. [3] [46] In fact, the p53 gene, which is central to maintaining genomic stability in human cells, is required for efficient GGR of UV-light- and carcinogen-induced DNA damage and functions as a DNA damage-activated transcription factor that directly regulates the expression of several NER damage recognition genes. [19] [20] [21] Similarly, several other important cancer-related genes have been shown to transcriptionally regulate the DNA damage recognition NER genes XPC and DDB2, including BRCA1 and E2F1. [55] [56] Therefore, the GGR pathway of NER appears relevant to suppressing DNA damage-induced malignancy and highly regulated by genes involved in tumor suppression.

Further evidence for the importance of exquisite regulation of DNA damage recognition and repair activity to carcinogenesis come from a new appreciation for the role of the ubiquitin-proteasome system in maintenance of genomic stability.[57] Many complex intracellular signaling processes, including DNA repair, are controlled not only by the regulated expression of proteins, but also by their assembly and targeted degradation through post-translational ubiquitination. With regard to NER, the same DNA damage recognition proteins that are found to be transcriptionally induced following DNA damage (XPC and DDB2) are also rapidly ubiquitinated following DNA damage by an E3 ubiquitin ligase activity that contains the DDB2 binding partner DDB1, resulting in a higher order of regulation.[46] The ubiquitin-proteasome system has been found to regulate other repair pathways as well, including translesional synthesis, and the Fanconi anemia–associated homologous recombination.[57] Therefore, mammalian cells have evolved a proteolytic pathway to limit their repair capability through restricting certain types of DNA repair activities. Considering that most DNA damage recognition complexes identify a variety of DNA damages or metabolic conditions rather than binding to specific DNA sequences, it is plausible that a “checkpoint” mechanism is required to limit DNA damage binding from interfering with other cellular processes involving unconventional DNA structures and that following inducible expression of DNA repair genes, levels are actively reduced to avoid gratuitous DNA repair and associated mutagenesis mediated by error-prone polymerases.

Recently, a class of specialized error-prone DNA polymerases, termed ζ (zeta) to σ (sigma), were discovered that seem to be devoted specifically to overcoming damage-induced replicational stress. [13] [14] [15] [58] These special polymerases take over temporarily from the stalled replicative DNA polymerases (δ [delta] and ε [epsilon]). They have more flexible base-pairing properties permitting translesion DNA synthesis, with each polymerase probably designed for a specific category of injury. Though translesion polymerases protect the genome, this solution to replication blocks comes at the expense of a higher error rate. For instance, inherited defects in pol η (eta), encoded for by the XPV/POLH/RAD30 gene, which specializes in relatively error-free bypassing of UV-induced cyclobutane pyrimidine dimers, causes a variant form of the skin cancer–prone disorder xeroderma pigmentosum. [59] [60]

Human Nucleotide Excision Repair–Deficient Syndromes and Cancer

A direct correlation between unrepaired DNA damage and carcinogenesis in humans was first established when James Cleaver found that the cancer-prone hereditary disease xeroderma pigmentosum (XP) involved a defect in the repair of DNA lesions produced by UV light.[61] Since then, at least three syndromes have been attributed to inborn errors in NER: XP, CS, and TTD, all characterized by exquisite sun sensitivity.

XP is a rare, autosomal recessive disease in which homozygous individuals display several characteristics: (1) extreme sensitivity of the skin to sun exposure that is evident by 1 year of age, (2) pigmentation abnormalities and premalignant lesions in sun-exposed skin, (3) increases up to 4000-fold in the incidence of skin cancers (predominantly squamous and basal cell carcinomas but also melanomas) and ocular neoplasms, occurring three to five decades earlier than in the general population, and (4) a 10- to 20-fold increased incidence of internal cancers in non-sun-exposed sites. [5] [62] [63]Overall, the life span is reduced by approximately 30 years among patients with XP, and many die due to malignancies.[64] Approximately 20% of patients with XP also display progressive neurologic degeneration, characterized by peripheral neuropathy, sensorineural deafness, progressive mental retardation, and cerebellar and pyramidal tract involvement.[65] XP occurs worldwide, in all ethnic groups and with a frequency varying from one to ten patients per million.

The biochemical defect in cells from most XP individuals is in NER,[61] though in a small number of cases (termed XP-variants), excision repair appears normal, and a defect exists in bypass replication at unrepaired lesions due to a mutation in the pol h (eta) translesional synthesis gene (XPV).[66] Complementation analysis via fusion of cells from different patients has demonstrated genetic heterogeneity within XP and has provided evidence for the existence of at least seven excision-deficient complementation groups, termed XP-A to XP-G, in addition to XP-variant.[5]

CS is another autosomal recessive disease that is associated with defective TCR of UV-damaged and oxidative-damaged DNA. [67] [68] [69] It is characterized by cutaneous photosensitivity, cachectic dwarfism, skeletal abnormalities, retinal degeneration, cataracts, severe mental retardation, and neurologic degeneration characterized by primary demyelination. [65] [70] In contrast to patients with XP, those with CS are not at increased risk for developing skin cancers. The average life span of individuals with CS is only 12 years, most patients succumbing to infectious or renal complications rather than cancer.[71] CS is characterized by the existence of at least three complementation groups. Several patients have been described in XP groups B, D, and G who share the DNA repair defects and clinical features of CS together with the cutaneous manifestations of XP. [72] [73]

TTD is an autosomal recessive condition that shares many of the signs and symptoms of CS, with the additional hallmark of brittle hair and nails, due to reduced sulfur content in the component proteins. As with CS, XPB and XPD are among the responsible genes that are implicated, but there is a third complementation group, TTD-A, for which no gene has been identified. The favored model for TTD is that of a transcription deficiency with respect to the genes relevant to the phenotype, including sulfur-containing proteins.[74] It is also conceivable that TTD and CS could be diseases of “premature cell death” in which the transcription deficiency and deficiency in TCR could cause the apoptosis of certain classes of metabolically active cells that sustain significant endogenous oxidative damage (e.g., neurons).

Analysis of the specific abnormalities in NER displayed by the various genetic complementation groups of XP, CS, and TTD allow correlations to be drawn with their heterogeneous clinical features. Specifically, only those subgroups of patients who display a defect in GGR are at significantly increased risk for developing UV-induced malignancies. In contrast, the neurologic symptoms and developmental abnormalities that are associated with other complementation groups of XP and CS are found only in the groups that are defective in TCR. The fact that the TFIIH complex, containing the XPB and XPD proteins, is common to both core NER and transcriptional initiation, supports the suggestion that the clinical phenotype of patients with defects in TCR might actually be due to abnormalities in transcription rather than in repair itself.[74]

Although these observations might explain the molecular basis for many of the clinical characteristics of XP and CS, they present an apparent paradox with regard to these patients’ cancer risk. Many currently recognized oncogenes and tumor suppressor genes are known to possess important cellular functions and to be actively expressed in normal cells. Because CS cells are defective in the repair of actively expressed genes, it would be reasonable to expect that these patients would acquire mutations in genes leading to transformation more readily than normal patients. However, this is not supported by the clinical picture. It has been demonstrated that defects in TCR specifically activate DNA damage induced apoptosis, [9] [10] which may eliminate potentially mutagenic, premalignant cells.

Another puzzling aspect of the clinical phenotype of XP is why these patients do not appear to be at a greater risk for developing neoplasms other than skin cancers. Although a disproportionate number of relatively rare tumors such as brain sarcomas and extraglossal carcinomas of the oral cavity have been described in XP patients under 40 years of age,[62] individuals with XP do not appear to be at significantly increased risk for more common solid or hematologic malignancies. It might be that the early mortality that XP patients experience or a decreased exposure to non-UV environmental carcinogens during early life may partially explain these observations. However, modest alterations in NER activity caused by functional polymorphisms in XP genes may contribute to the risk of solid tumors. [75] [76]

Base Excision Repair

A major source of DNA damage to cellular genomes arises from normal metabolism in the cytoplasmic environment through hydrolysis and exposure to reactive metabolites that cause oxidation and alkylation of DNA. The repair system that is primarily involved in identifying and removing such lesions, as well as for dealing with the spontaneous loss of purines from DNA, is the base excision repair (BER) pathway. [77] [78] The essential nature of BER for viability is highlighted by the fact that although a number of BER proteins have been discovered, only recently has a single human hereditary disease been identified that appears to result from a mutation in a gene that is unique to this pathway. [79] [80] [81] [82] The enormous task that is required for BER is exemplified by the fact that a human being spontaneously loses on the order of a trillion guanines from the DNA in his or her body every hour and each of these must be replaced. Similarly, a large number of cytosines become deaminated spontaneously, and the resulting product, uracil, must be removed and replaced with cytosine to restore the correct nucleotide sequence. In most cases, BER is initiated by one of a set of lesion-specific glycosylases that recognize the altered or inappropriate base and cleaves it from its sugar moiety in the DNA ( Fig. 10-3 ). Different DNA glycosylases remove different kinds of damage, conferring specificity to the process. Once the base is removed, the apurinic/apyrimidinic (AP)-site is removed by an AP-endonuclease or an AP-lyase, which cleave the DNA strand 5′ or 3′ to the AP-site, respectively. The remaining deoxyribose phosphate residue is excised by a phosphodiesterase with the resulting gap filled by a DNA polymerase and the strand sealed by DNA ligase. The major oxidized purine lesion is 8-oxo-7,8-dihydroguanine (8-oxoG), which is abundant and has strong mutagenic properties. Oxidized pyrimidines include thymine glycol, 5-hydroxycytosine, and formamidopyrimidines. Oxidized bases, including both 8-oxoG and thymine glycol, share the property of blocking DNA replication and transcription and must be repaired efficiently to maintain genomic stability. [83] [84]


Figure 10-3  Excision repair pathways for DNA damage. The three main excision repair pathways in human cells—base excision repair (BER), nucleotide excision repair (NER), and mismatch repair (MMR)—proceed through similar steps to restore the normal DNA sequence. Following recognition of altered DNA bases, employing lesion-specific glycosylases for BER, XP proteins for NER, and MutS homologs for MMR, incision of DNA is achieved by endonucleases and displacement or degradation of single-stranded sections of DNA that contain the damage by enzymes with helicase and exonuclease activity. Repair replication of the resulting DNA gap and strand ligation results in the repaired double-stranded DNA molecule. The many repair enzymes that are involved in each specific step are tightly coupled and may be regulated or inducible by DNA damage response pathways.



In mammalian cells, the gene functions that are responsible for the strand incision steps of BER include the glycosylases hNTH1, which removes oxidized pyrimidines; hOGG1, which targets oxidized purines; and MYH, which removes adenines mispaired with an 8-oxoG, together with AP endonuclease 1 (APE1).[85] Attempts to engineer mice that are deficient in the core enzymes that are required for BER have typically resulted in early embryonic death, whereas knockout of individual glycosylases produce mice with no overt phenotype at all.[86] This attests to the importance of the repair of DNA lesions from endogenous causes during embryonic development as well as the likely redundancy between individual glycosylases. It is also consistent with the near absence of known human hereditary diseases characterized by defects in BER genes.

However, given the mutagenic and cytotoxic potential of the classes of DNA damage that are BER substrates, it seems likely that altered activity in these pathways would result in enhanced cancer risk. The most direct evidence for a role for BER in cancer comes from the discovery that germline mutations in the MYH gene, involved in processing 8-oxoG lesions, is associated with recessive inheritance of a predisposition to develop multiple colorectal adenomas (polyposis) and colon cancers. [79] [80] [81] Tumors from affected individuals exhibit excess transversions of a guanine-cytosine pair to a thymine-adenine pair in the APC gene, itself associated with colon carcinogenesis and causative of familial adenomatous polyposis. Therefore, biallelic inherited mutations in MYH result in a polyposis-like syndrome termed MYH-associated polyposis (MAP). Patients with MAP tend to develop tens to hundreds of polyps by the age of 40, and nearly 50% present with colon cancer.[82] Box 10-1 discusses the affect of genetic variations in other DNA repair genes.

Box 10-1 


Despite many decades of investigation, the exact cause of most cancers remains unknown, with a few important exceptions (e.g., certain cancers of the lung, skin, and cervix). Rather, cancer is associated with a broad and heterogeneous group of genetic and environmental influences, making the development of schemes for risk assessment and targeted prevention difficult. However, 30 years ago, prior to much of our current understanding of the specific molecular and genetic changes associated with cancer, Dr. Larry Loeb proposed that a common early event in the development of many cancers is the expression of a “mutator phenotype” resulting from functional mutations in genes that normally function to maintain genetic stability.[87] On the basis of calculations of the estimated fidelity of DNA replication and repair in normal human cells and the rarity of spontaneous mutations that occur in normal cells, Loeb noted the statistical unlikelihood that the large number of chromosomal aberrations and genetic mutations that are observed in human malignancies would occur by chance in a single cell. However, he speculated that if a cell exhibited unusual levels of genetic instability due to inherited or acquired mutations in the genes that regulate the processes of DNA replication and repair, the rate of additional mutations occurring in other genes important for carcinogenesis will be dramatically elevated. As is obvious from the many specific examples that are discussed in this chapter, Loeb's prediction has been borne out by many subsequent studies of multistep carcinogenesis and the identification of cancer susceptibility syndromes caused by inherited mutations in DNA repair genes.[88] However, these highly penetrant hereditary cancer syndromes (e.g., HNPCC) caused by inactivating mutations in tumor suppressor or DNA repair genes account for only a fraction of most common types of cancers (usually fewer than 10%). Therefore, an emerging hypothesis is that more common polymorphic genetic variation in DNA repair genes may result in variability in DNA repair capacity between individuals and result in altered cancer susceptibility.[76] Rapid advances in DNA sequencing technologies have allowed for large-scale genetic epidemiology studies to be performed to address this possibility. To date, the results have been rather inconsistent. A meta-analysis of associations between single-nucleotide polymorphisms (SNPs) in base excision repair genes and cancer risk found SNPs in the 8-oxoguanine DNA glycosylase (OGG1), apurinic/apyrimidinic endonuclease (APE1/APEX1) and XRCC genes affected lung cancer risk. [89] [90] Several other studies show similar results, though many do not find significant increased relative risk of cancer associated with DNA repair SNPs. This may be due to methodologic limitations, such as study size, false positives, and population heterogeneity, but an important possibility is that a single common sequence variant might not be detectable in population association studies. Rather, a combination of multiple variants in the same gene or genes in common pathways might be more important in carcinogenesis. Sir Walter Bodmer proposed just this by suggesting that multiple rare low-penetrance SNPs might together account for a substantial proportion of inherited cancer susceptibility, and recent results from his work in colon polyps and colon cancer support these ideas. [91] [92]

The mutator phenotype theory and individual genetic variation in DNA repair capacity also have major implications for the prevention of cancers. Whole-genome approaches to mapping genetic variation in DNA repair genes, potentially in combination with functional assays for individual DNA repair capacity,[93] might allow for targeted approaches for cancer prevention. Reducing the amount of DNA damage to which normal or premalignant cells are particularly vulnerable could slow the accumulation of additional mutations. Certainly, reducing exposure to known environmental carcinogens is important in this goal, but the majority of DNA damage likely occurs as the result of endogenous reactants of normal cellular metabolism, such as oxygen-reactive species, activated lipids, and metal cations. In fact, it has been estimated that oxidative radicals generate up to 10,000 DNA damage events per cell per day. Therefore, means to reduce the amount of oxidative DNA damage or enhance its repair, might slow the carcinogenic process sufficiently to prevent the clinical occurrence of some cancers. Indeed, evidence from several fields suggests that antioxidants might have chemopreventative properties. For example, epidemiologic evidence suggests that diets that are rich in the common trace element selenium might be associated with reduced cancer risk. However, clinical trials of antioxidant approaches to cancer prevention have not generally been successful. Given the complex genetic pathways that are involved, it is hoped that individualized risk assessment by using genetic approaches will allow for the identification of specific pharmacologic agents for cancer prevention. Certainly, our rapidly increasing understanding of processes for the prevention and repair of DNA damage provides many new targets for rational drug development.

Mismatch Repair

Mismatch repair (MMR) is another example of an excision repair mechanism that utilizes a similar strategy for genomic maintenance (see Fig. 10-3 ). MMR is a process that corrects mismatched nucleotides in the otherwise complementary paired DNA strands, arising from DNA replication errors and recombination, as well as from some types of base modifications. [94] [95] [96] This repair mode also can deal with small loops of single-stranded DNA at sites of insertions or deletions in the duplex DNA structure. The importance of this repair mechanism in maintaining genetic stability is illustrated by the observation that its absence results in a large increase in the frequency of spontaneously occurring mutations, particularly in microsatellite sequences of highly repetitive DNA.[97] Some of these spontaneous mutations arise from mistakes that are introduced during DNA replication, in spite of the operation of a “proofreading” system that also helps to ensure the high fidelity of replication. In humans, genetic defects in several mismatch repair genes have been linked to hereditary nonpolyposis colon cancer (HNPCC) as well as to sporadic cancers that exhibit instability in regions of DNA containing short repetitive sequences of nucleotides, a feature that is known as microsatellite instability (MSI).

As with other modes of excision repair, four principal steps are required for MMR: (1) mismatch recognition, (2) recruitment of additional MMR factors, (3) identification of the newly synthesized DNA strand containing the mismatched nucleotides, followed by their excision, and (4) resynthesis of the excised tract and ligation. The biochemical workings of this pathway are best understood in bacteria, but a similar set of events occurs in human cells. On the basis of functional homologies to their bacterial counterparts and sequence homology to corresponding yeast genes, a number of human genes have been cloned that participate in MMR, including those that are homologous to the bacterial MutS mismatch recognition protein (hMSH2, hMSH3, and hMSH6) and to the bacterial MutL gene (hMLH1 and hPMS2). [98] [99] [100] In humans, heterodimers of the MSH2/6 proteins recognize single-base-pair mismatches and short insertion-deletion loops, whereas MSH2/3 dimers recognize longer loops. Heterodimeric complexes of MLH1/PMS2 and MLH1/PMS1 interact with the MSH complexes and replication factors for strand discrimination and DNA excision. Similar to NER and BER, additional proteins are then recruited for repair replication on the basis of the original DNA template.

The MMR system also might interact with DNA damage due to certain alkylators and intercalating agents that assume structural alterations in DNA similar to those of mismatches and might actually result in erroneous or futile MMR cycles, resulting ultimately in apoptosis. Thus, intact MMR might confer chemosensitivity to these chemotherapeutic agents, and MMR-defective tumors may exhibit resistance to certain drugs.[101]

Human Mismatch Repair Deficiency and Cancer

HNPCC, also known as Lynch syndrome, is the most common hereditary colorectal cancer predisposition syndrome. [102] [103] HNPCC is an autosomal dominant inherited condition with an incidence ofone in 1000 in the general population. It accounts for approximately 5% of all colorectal cancers patients with HNPCC, who are also at elevated risk for cancers of the endometrium, ovary, stomach, small bowel, and other sites. Patients with HNPCC have an 80% lifetime risk for colorectal cancer and a 50% lifetime risk for endometrial cancer. The discovery of MSI, that is, the frequent alteration in the tract lengths of certain short repetitive nucleotide sequences, in some hereditary colorectal cancers provided the first indication that the etiology of these cancers might involve a problem in the MMR system.[104] The finding of germline MMR gene defects in patients with HNPCC established that these defects are the cause of the enhanced incidence of cancer. [98] [99] [100] Germline mutations in MLH1 and MSH2 together account for more than half of all cases of HNPCC. Defects in MSH6 cause a late-onset HNPCC phenotype. No strong genotype-phenotype correlations have been observed to date, but mutations in the MSH2 gene do appear to be associated with more extracolonic manifestations than are seen with mutations in the MLH1 gene.

MSI has been identified as a source of the genomic instability driving tumorigenesis in a number of sporadic tumor types in addition to those that arise in the context of inherited germline mutations of MMR genes. [104] [105] For example, up to 20% of sporadic colon cancers exhibit MSI, particularly when presenting in the ascending colon and in individuals younger than 50 years old, the majority due to epigenetic silencing of MLH1 gene expression by promoter hypermethylation. [106] [107] Whether through genetic or epigenetic inactivation, loss of MMR results in an elevated rate of mutations, particularly at microsatellite sequences, several of which occur in the coding sequences of other genes that are often found mutated in cancers, including TGF beta type II receptor, BAX, and the mismatch repair genes MSH3 and MSH6, themselves. Therefore, clear genetic evidence demonstrates that the phenotype of genetic instability associated with defects in MMR results in the genotype of tumors that arise owing to these defects.[108] Intriguingly, the survival of patients with MSI-associated colorectal cancer is better than those with more typical tumors exhibiting chromosomal instability. [109] [110] [111]These tumors also demonstrate different sources of genomic instability, resulting in distinct biologic and pathological characteristics.[112] Whether their more favorable outcome reflects differences in clinical behavior, responsiveness to therapy, or both remains to be fully determined.

Double-Strand Break Repair

Double-strand breaks (DSBs) in DNA that are induced by ionizing radiation, endogenously produced reactive oxygen radicals, chemicals, replication across single-strand breaks, and during repair of interstrand DNA cross-links are dealt with through either the recombination machinery or the relatively error-prone nonhomologous end-joining pathway ( Fig. 10-4 ). An unrepaired DSB is a highly lethal event, and even a single occurrence in the entire genome is thought to be sufficient to signal cell cycle checkpoints that prevent attempted DNA synthesis or cell division until repair has been completed or apoptosis if improperly repaired. DSBs also pose problems during mitosis, because intact chromosomes are a prerequisite for proper chromosome segregation during cell division. Thus, these lesions often induce various sorts of chromosomal aberrations, including aneuploidy, deletions (loss of heterozygosity), and chromosomal translocations, all of which are associated with carcinogenesis. Genetic recombination is the principal mechanism in cycling cells for dealing with DSBs that involve homologous stretches of nucleotides at the ends to be joined. If no such homology is present or if the cell is not cycling, then there is a system for nonhomologous end joining, which is more error-prone.


Figure 10-4  Mechanisms for DNA double-strand break repair. The repair of DNA DSBs is carried out by two mechanisms. Left, The rapid, but error-prone, nonhomologous end-joining that directly seals breaks but may result in the gain or loss of several nucleotides due to short areas of microhomologies used for annealing prior to ligation. Exposed DNA ends are recognized by the Ku70/80 heterodimer that recruits the DNA-dependent protein kinase catalytic subunit and other proteins assisting in strand alignment. The XRCC4-ligase IV heteroduplex joins the breaks. Right, The high-fidelity, homologous recombination of sister chromatids at sites of DSBs is the less prominent mode in mammalian cells. This DNA repair pathway is mediated by RAD51-associated proteins, including RAD52 that recognizes single-strand DNA ends and together with other proteins results in short nuclease-mediated resection. RAD51 then forms a nucleoprotein filament on the exposed strand and, probably with BRCA2 and other proteins, promotes strand invasion and displacement at homologous sequences. Thus, the undamaged sister molecule acts as a template for the resynthesis of the missing nucleotides.



Recent studies have identified a cascade of protein kinases that are involved in signaling cellular processes in response to DSBs. Many of these have been found to be defective in cancer-prone disorders that exhibit genomic instability, such as the ATM protein,[41] and the Chk2 protein kinase associated with a Li-Fraumeni–like cancer susceptibility syndrome. [113] [114] [115] A major target for these kinase activities is the p53 tumor suppressor gene. When activated, this protein is involved in inducing G1 arrest or apoptosis following ionizing radiation and other cellular stresses. [36] [37] [116] [117] [118] [119] [120]Germline mutation of p53 results in the Li-Fraumeni cancer susceptibility syndrome. Many other enzymes that are involved in processes required for effective DSB repair have been found in cancer-prone disorders, including MRE11 (AT-like disorder), NBS1 (Nijmegen breakage syndrome), BRCA1 and BRCA2 (breast-ovarian cancer syndrome), and the RecQ-like helicases (Werner, Bloom and Rothmund Thomson syndromes).[121]

Ataxia Telangiectasia

Ataxia telangiectasia (AT) was first identified as a disorder characterized by progressive neurodegeneration, immune deficiency, and cancer predisposition. A link to DNA damage responses arose when AT patients with lymphomas exhibited severe reactions to radiation therapy that was used to treat their tumors. A single gene, ATM, is responsible for the multiple and surprisingly diverse symptoms of this disease, including the predisposition to lymphoma and leukemia. It has been estimated that over 10% of AT patients develop cancer at an early age. AT is an autosomal recessive disease with an incidence of nearly one in 100,000 live births. Persons who are heterozygous for AT mutations, about 1% of the general population, may have an increased predisposition to cancer, in particular breast cancers, especially in individuals who express an ATM with missense mutations resulting in a dominant-negative effect on their wild-type ATM gene. [122] [123] [124] The ATM gene product is a central signaling protein in the DNA damage response, and cells lacking ATM fail to execute many critical cellular responses to DNA damage. For example, one hallmark of AT is what has been termed X-ray resistant DNA synthesis. It is now know that the ATM gene product is a key element in delaying the initiation of DNA replication following DNA damage resulting in strand breaks.

ATM is a protein kinase that is activated by introduction of DSBs into the genome and phosphorylates numerous substrates involved in controlling cellular responses to DNA damage, including p53, BRCA1, and CHK2.[34] In addition, ATM directly phosphorylates the NBS1 protein, which exists in a complex with the MRE11 and RAD50 proteins, a complex that is required both for nonhomologous end-joining and homologous recombination of DSBs.[125] Inherited germline mutations of the NBS1 and MRE11 genes, themselves, result in clinical variants of AT, termed Nijmegen breakage syndromeand AT-like disorder, respectively. [126] [127] In fact, it has recently been shown that cells from patients with NBS have defective ATR-dependent signaling and appear phenotypically similar to ATR-defective Seckel syndrome.[127] Therefore, ATM is central to a DNA damage response pathway that is critical for regulating cellular responses to stress, including recombination and repair following DSBs. Defects in many of the component proteins in the pathway result in genomic instability and a predisposition to cancer.


Diseases Involving Homologs of recQ

There are at least three cancer-prone diseases in humans in which the defect is in a homolog of the recQ gene that was originally discovered in bacteria.[128] The product of recQ is a helicase, which in E. coli is involved in processing the nascent DNA at arrested replication forks. Helicases are enzymes that separate the complementary strands of nuclei-acid duplexes using energy that is derived from ATP hydrolysis. In humans, recQ helicases are thought to function at the interface between DNA replication and recombination in dealing with damaged replication forks and interact with many other nuclear proteins that are required for DNA metabolism.[129] The biologic and clinical effects of the homozygous deficiency of these genes can be quite dramatic and profound. Bloom syndrome, a disorder that is caused by homozygous loss of the BLM helicase, is characterized by an extremely high frequency of genetic exchanges (so-called sister chromatid exchanges) that cause genomic instability and lymphoma, leukemias, and solid tumors of the GI tract and breast. [130] [131] Of interest, recent studies have identified an increased risk that individuals who are heterozygous for a BLM mutation will develop colorectal cancer, potentially due to haploinsufficiency. [132] [133] Werner syndrome results from a deficiency in another recQ homolog, WRN, and has features of profound premature aging as well as predisposition to sarcomas, melanoma, and cancer of the thyroid. [134] [135] [136] Yet another recQ homolog defect, Rothmund-Thompson syndrome, is characterized by growth deficiency and cancer predisposition, in particular to osteogenic sarcomas. [137] [138]

p53 Gene and Li-Fraumeni Syndrome

The discovery of the p53 tumor suppressor gene over 25 years ago inspired widespread investigations with the aim of understanding the basic biology behind its role in maintaining genomic stability and the cellular response to DNA damage. p53 is one of the most commonly mutated genes in human cancers[139] and its product is a multifunctional protein that regulates many physiologic processes, including cell cycle checkpoints, apoptosis, and DNA repair. [21] [116] [117] [140] [141] The primary role of p53 in tumor suppression has been attributed to its function as a transcription factor, regulating expression of several hundred different cellular genes,[142] although it appears to exhibit transcription-independent activities as well. Indeed, p53 appears to act as a central “node” that lies at the intersection of upstream signaling cascades induced by DNA damage and cellular stress responses and downstream DNA repair and DNA damage response pathways. In response to a variety of genotoxic stimuli, p53 protein is induced and stabilized. [34] [143] This activated p53 protein binds to DNA in a sequence-specific manner and regulates the transcription of downstream target genes that contain a consensus p53 response element in their promoter or intronic segments. These p53 target genes include those that are important for cell cycle checkpoints, such as p21;[144] apoptosis, such as BAX and PERP;[145] and DNA repair, such as the DDB2 and XPC genes that are required for NER. [19] [20] [21] [46] In addition, some evidence suggests that p53 protein might act in a transcription-independent manner to modulate BER through interactions with DNA polymerase beta and OGG1, and homologous recombination in conjunction with recQ proteins.[141]

Patients with the rare autosomal dominant Li-Fraumeni syndrome (LFS) are at increased risk for developing a number of common tumors at an early age due to an inherited germline defect in one allele of the p53 gene, including soft-tissue and osteosarcomas, breast cancer, brain tumors, lymphomas, leukemia, and adrenocortical carcinomas. Mutations in the p53 tumor suppressor gene account for 70% to 85% of classic LFS cases. [146] [147] [148] Although the heterozygote carriers of a defective p53 allele do not appear to have clinical problems or DNA repair defects, when the second allele has been mutated or lost, the absence of functional p53 results in severe problems for the cell. First of all, the p53-controlled pathway of apoptosis is disengaged, so severely damaged cells will survive and be at risk for carcinogenic transformation because of their genomic instability. That genomic instability derives from the fact that p53 is also an important regulator of cell cycle checkpoints. Thus, as with the situation in AT, the cells continue to progress through their growth cycle rather than pausing to allow time for DNA lesions to be repaired. Loss of p53 also leads to increased aneuploidy of cells, further contributing to genetic instability and the progression to malignancy or metastasis. Finally, p53 serves an important regulatory function in NER and perhaps in BER and recombination, and in its absence, some important mutagenic lesions are simply not repaired. That, of course, is a major contributor to the genomic instability and the consequent development of tumors. See Box 10-2 for a discussion of DNA repair and cancer treatment.

Box 10-2 


As was discussed in this chapter, many genes that are implicated in the development of cancer play roles in DNA repair. The mechanism of action for most cancer chemotherapeutic drugs, as well as radiation therapy, is thought to be through DNA damage. Therefore, it seems logical that cancers that acquired defects in DNA repair during the tumorigenic process would also be particularly susceptible to the cytotoxic effects of DNA-damaging therapeutic agents. However, for most common cancers, the clinical experience suggests otherwise. It is likely that the frequently concurrent inactivation of cell cycle checkpoints and apoptotic processes during tumorigenesis obscures the effect of DNA repair defects; therefore, the response of clinical tumors to various treatments remains very difficult to predict, even with genetic information. Nevertheless, several examples have emerged in which a detailed understanding of the DNA repair defects that are present in a particular tumor type might allow for the rational selection of certain treatment approaches that are likely to be more effective.

One example relates to the function of the BRCA1 gene, which is involved in several types of DNA repair, including double-strand break repair, nucleotide excision repair, and DNA crosslink repair.[46] [149] [150] Germline mutations in the BRCA1 gene predispose individuals to a very high risk for developing breast and ovarian cancers, and somatic inactivation of BRCA1 activity has also been observed in sporadic breast and ovarian cancers due to promoter methylation. Clinical and experimental data suggest that breast and ovarian tumors that are deficient in BRCA1 function are particularly sensitive to the chemotherapeutic drug cisplatin, which causes DNA damage to be repaired through the nucleotide excision and crosslink repair pathways, and ionizing radiation, which causes double-strand DNA breaks.[150] Another very exciting recent finding relates to the selective activity of PARP-1 inhibitors in tumors that are deficient for BRCA1/2. [151] [152] PARP-1 is the first described member of a large family of enzymes that can detect and bind to DNA nicks and strand breaks and is thought to play a key role in base excision repair.[153] Chemical inhibitors of PARP activity are thought to enhance killing of cells defective in double-strand break repair through synthetic lethality. The success of this approach in preclinical models has led to the rapid development of clinical trials of PARP inhibitors in breast cancers in women with known BRCA1/2 mutations, as well as in “triple-negative” breast cancers (ER/PR/Her-2 negative) that share phenotypic activity with BRCA1/2 mutant tumors. [154] [155]

A quite different example relates to the 15% to 20% of colorectal cancers that have inactivated the mismatch repair pathway and exhibit microsatellite instability (MSI). It has been appreciated for some time that individuals with colorectal cancer who express high levels of MSI have longer survival than do stage-matched patients with colorectal cancer without MSI.[109] However, whether these prognostic differences related to differences in tumor biology or sensitivity to chemotherapy was unclear. Recently, though, clinical studies suggest that patients with surgically resected stage II or III MSI-positive colorectal cancer do not benefit from 5-fluorouracil based adjuvant chemotherapy, as do patients with mismatch repair–intact colorectal cancers but nevertheless have better outcomes even without additional therapy. Experiments with mismatch repair–defective colon cancer cell lines suggest that they are also resistant to the cytotoxic effects of oxaliplatin (in fact, an intact mismatch repair pathway may be necessary to confer the apoptotic effects of platinum-induced DNA damage) but susceptible to the topoisomerase I inhibitor, CPT-11.[156] Since both these drugs are now being used for the treatment of colorectal cancer, concurrent diagnostic testing of tumors for MSI and mismatch repair activity may help guide their selection and use in individual patients. These examples and others suggest that our rapidly improving knowledge of the role of specific cancer genes in DNA repair pathways may significantly impact the treatment of human cancers. Efforts to obtain genotypic and phenotypic information from individual tumors will hopefully allow for tailored, rational selection of therapies for cancer treatment, an approach that is central to the emerging field of pharmacogenomics.

BRCA1, BRCA2, and Breast-Ovarian Cancer Susceptibility

Hereditary breast cancer includes a broad group of hereditary predisposition conditions in which breast cancer is a component tumor; these account for approximately 5% to 10% of all breast cancer cases. Hereditary syndromes of breast and ovarian cancer susceptibility have been particularly associated with germline mutations of two genes, BRCA1 and BRCA2, as well as rare cases due to mutations in the p53 gene in LFS, the PTEN gene in Cowden disease, and perhaps the ATR gene, Chk2 gene, and others. Recent experimental data suggest that both the BRCA genes might be involved in multiple DNA repair activities. [21] [55] [149] [150] [157] [158] [159]

The BRCA1 and BRCA2 genes are large and complex. Many hundreds of different germline mutations have been detected in each, but only rare sporadic breast or ovarian cancers have been found to harbor BRCA1 mutations. The exact biochemical functions of these proteins remain unknown, but increasing evidence suggests that they might be involved in various aspects of DNA repair and DNA damage response pathways. For example, BRCA1 is phosphorylated after exposure to DNA-damaging agents by ATM, ATR, and Chk2 and associates with a number of DNA repair proteins including MSH2, MSH6, ATM, RAD51; and the RAD50-MRE11-NBS1 protein complex following DNA damage and localizes to nuclear foci with these proteins after treatment with ionizing radiation and UV radiation. [24] [160] [161] The association of BRCA1 with RAD51, an enzyme that is involved in the coordination of recombination, suggests its involvement in DSB repair, and strong data exist that implicate BRCA1 in homologous recombination. [149] [162] Other studies suggest that BRCA1 might regulate cellular processes through transcriptional coactivation. BRCA1 has been shown to transcriptionally regulate the NER genes XPC and DDB2 and affect GGR of UV and cisplatin-induced DNA damage. [55] [150] [163] Chromosomal instability also is characteristic of breast tumors that harbor BRCA2 mutations, probably owing to defective recombination-mediated DSB repair. BRCA2 has been shown to bind to the RAD51 protein, an enzyme that is involved in the coordination of recombination, and together they colocalize to nuclear sites that contain DNA strand breaks caused by ionizing radiation. Structural studies of BRCA2 DNA-binding domains suggest that it might facilitate interactions of RAD51 with single-stranded DNA during recombination. [164] [165] The genomic instability that is associated with mutations of BRCA1 and 2 therefore might be due in part to the intact but error-prone nonhomologous end-joining repair pathway.[158]

MDC1 is another important regulator of ATM-dependent phosphorylation of BRCA1 and is required to activate Chk2 as part of the response of mammalian cells to DNA damage.[149] Therefore, a hypothesis is that BRCA1 lies at a critical intersection of the DSB response pathways. Post-translational regulation of BRCA1 by phosphorylation has established a link between BRCA1 and Chk2, BRCA2, ATR, and ATM and suggests mechanisms that promote genomic instability and increased susceptibility to cancer when mutant.

Fanconi Anemia, Cancer, and Interstrand Cross-Link Repair

The centrality and interwoven nature of DNA repair pathways for genomic stability have been highlighted by findings regarding Fanconi anemia, a rare, autosomal recessive disease that confers an increased risk of acute myeloid leukemia, squamous cell carcinomas of the head and neck and esophagus, gynecologic carcinomas, and liver tumors at a young age. [166] [167] At least 13 subtypes of Fanconi anemia have been determined by complementation analyses, and germline mutations in genes have been identified for most of them. [168] [169] The proteins that are encoded by these Fanconi anemia genes all cooperate in a common DNA repair pathway that is involved in interstrand crosslink repair. Germline homozygous inactivating mutations of the BRCA2 gene result in the D1 group.[170] Eight of these proteins are subunits of a nuclear E3 ligase, required for the monoubiquitination of the downstream D2 protein, which itself links Fanconi anemia proteins to BRCA1 in the response to DNA damage by colocalizing to nuclear sites occupied by RAD51 and BRCA2.[171] This process also is regulated by the ATM protein kinase, which phosphorylates FANCD2 in response to DNA damage.[172] It has long been appreciated that cells from Fanconi anemia patients are hypersensitive to DNA cross-linking agents, such as mitomycin C and cisplatin, in addition to being modestly sensitive to ionizing radiation. Although DNA crosslink repair in mammalian cells is poorly understood, it has been proposed that it utilizes components of both the excision repair and DSB repair systems to sequentially incise DNA near the site of a crosslink followed by homologous recombination or nonhomologous end-joining.[167] Therefore, the Fanconi anemia proteins appear to function at an interface between several DNA repair and DNA damage response pathways.


Recent molecular biology and genetic research has provided ample evidence to support the long-standing prediction that genomic instability is a major factor driving the onset and progression of carcinogenesis.[88] Overlapping and interacting mechanisms for DNA repair and the cellular response to DNA damage are critical components for the maintenance of genomic stability. Alterations in these pathways often are early events in the multistep acquisition of genetic mutations that lead to cancer development. Continued exploration of the DNA damage response will prove important for our improved understanding of cancer etiology, prevention, genetic susceptibility, diagnosis, and treatment.


  1. Sjoblom T, Jones S, Wood LD, et al: The consensus coding sequences of human breast and colorectal cancers.  Science2006; 314:268-274.
  2. Loeb LA, Loeb KR, Anderson JP: Multiple mutations and cancer.  Proc Natl Acad Sci USA2003; 100:776-781.
  3. Zhou BB, Elledge SJ: The DNA damage response: putting checkpoints in perspective.  Nature2000; 408:433-439.
  4. Hoeijmakers JH: Genome maintenance mechanisms for preventing cancer.  Nature2001; 411:366-374.
  5. Ford JM, Hanawalt PC: Role of DNA excision repair gene defects in the etiology of cancer.  Curr Top Microbiol Immunol1997; 221:47-70.
  6. Bohr VA, Smith CA, Okumoto DS, Hanawalt PC: DNA repair in an active gene: Removal of pyrimidine dimers from the DHFR gene of CHO cells is much more efficient than in the genome overall.  Cell1985; 40:359-369.
  7. Mellon I, Bohr VA, Smith CA, Hanawalt PC: Preferential DNA repair of an active gene in human cells.  Proc Natl Acad Sci USA1986; 83:8878-8882.
  8. Mellon I, Spivak G, Hanawalt PC: Selective removal of transcription-blocking DNA damage from the transcribed strand of the mammalian DHFR gene.  Cell1987; 51:241-249.
  9. Yamaizumi M, Sugano T: U.V.-induced nuclear accumulation of p53 is evoked through DNA damage of actively transcribed genes independent of the cell cycle.  Oncogene1994; 9:2775-2784.
  10. Ljungman M, Zhang F: Blockage of RNA polymerase as a possible trigger for UV light-induced apoptosis.  Oncogene1996; 13:823-831.
  11. Nelson WG, Kastan MB: DNA strand breaks: The DNA template alterations that trigger p53-dependent DNA damage response pathways.  Mol Cell Biol1994; 14:1815-1823.
  12. Ljungman M, Lane DP: Transcription: guarding the genome by sensing DNA damage.  Nat Rev Cancer2004; 4:727-737.
  13. Friedberg EC, Wagner R, Radman M: Specialized DNA polymerases, cellular survival, and the genesis of mutations.  Science2002; 296:1627-1630.
  14. Kunkel TA: Considering the cancer consequences of altered DNA polymerase function.  Cancer Cell2003; 3:105-110.
  15. Lehmann AR: New functions for Y family polymerases.  Mol Cell2006; 24:493-495.
  16. Weinert TA, Hartwell LH: The RAD9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae.  Science1988; 241:317-322.
  17. Ford JM, Hanawalt PC: Li-Fraumeni syndrome fibroblasts homozygous for p53 mutations are deficient in global DNA repair but exhibit normal transcription-coupled repair and enhanced UV resistance.  Proc Natl Acad Sci USA1995; 92:8876-8880.
  18. Ford JM, Hanawalt PC: Expression of wild-type p53 is required for efficient global genomic nucleotide excision repair in UV-irradiated human fibroblasts.  J Biol Chem1997; 272:28073-28080.
  19. Hwang BJ, Ford JM, Hanawalt PC, Chu G: Expression of the p48 xeroderma pigmentosum gene is p53-dependent and is involved in global genomic repair.  Proc Natl Acad Sci USA1999; 96:424-428.
  20. Adimoolam S, Ford JM: p53 and DNA damage-inducible expression of the xeroderma pigmentosum group C gene.  Proc Natl Acad Sci USA2002; 99:12985-12990.
  21. Adimoolam S, Ford JM: p53 and regulation of DNA damage recognition during nucleotide excision repair.  DNA Repair (Amst)2003; 2:947-954.
  22. Scully R, Chen J, Ochs RL, et al: Dynamic changes of BRCA1 subnuclear location and phosphorylation state are initiated by DNA damage.  Cell1997; 90:425-435.
  23. Scully R, Chen J, Plug A, et al: Association of BRCA1 with Rad51 in mitotic and meiotic cells.  Cell1997; 88:265-275.
  24. Cortez D, Wang Y, Qin J, Elledge SJ: Requirement of ATM-dependent phosphorylation of brca1 in the DNA damage response to double-strand breaks.  Science1999; 286:1162-1166.
  25. Wu X, Ranganathan V, Weisman DS, et al: ATM phosphorylation of Nijmegen breakage syndrome protein is required in a DNA damage response.  Nature2000; 405:477-482.
  26. Fitch ME, Cross IV, Ford JM: p53 responsive nucleotide excision repair gene products p48 and XPC, but not p53, localize to sites of UV-irradiation-induced DNA damage, in vivo.  Carcinogenesis2003; 24:843-850.
  27. Fitch ME, Nakajima S, Yasui A, Ford JM: In vivo recruitment of XPC to UV-induced cyclobutane pyrimidine dimers by the DDB2 gene product.  J Biol Chem2003; 278:46906-46910.
  28. Smith ML, Fornace Jr AJ: Mammalian DNA damage-inducible genes associated with growth arrest and apoptosis.  Mutat Res1996; 340:109-124.
  29. Fornace Jr AJ, Amundson SA, Bittner M, et al: The complexity of radiation stress responses: analysis by informatics and functional genomics approaches.  Gene Expr1999; 7:387-400.
  30. Zhao R, Gish K, Murphy M, et al: Analysis of p53-regulated gene expression patterns using oligonucleotide arrays.  Genes Dev2000; 14:981-993.
  31. Lowe SW, Schmitt EM, Smith SW, et al: p53 is required for radiation-induced apoptosis in mouse thymocytes.  Nature1993; 362:847-849.
  32. Lowe SW, Ruley HE, Jacks T, Housman DE: p53-dependent apoptosis modulates the cytotoxicity of anticancer agents.  Cell1993; 74:957-967.
  33. Clarke AR, Purdie CA, Harrison DJ, et al: Thymocyte apoptosis induced by p53-dependent and independent pathways.  Nature1993; 362:849-852.
  34. Kastan MB, Bartek J: Cell-cycle checkpoints and cancer.  Nature2004; 432:316-323.
  35. Cline SD, Hanawalt PC: Who's on first in the cellular response to DNA damage?.  Nat Rev Mol Cell Biol2003; 4:361-373.
  36. Canman CE, Lim DS, Cimprich KA, et al: Activation of the ATM kinase by ionizing radiation and phosphorylation of p53.  Science1998; 281:1677-1679.
  37. Kastan MB, Lim DS: The many substrates and functions of ATM.  Nat Rev Mol Cell Biol2000; 1:179-186.
  38. Shieh SY, Ahn J, Tamai K, et al: The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites.  Genes Dev2000; 14:289-300.
  39. Matsuoka S, Huang M, Elledge SJ: Linkage of ATM to cell cycle regulation by the Chk2 protein kinase.  Science1998; 282:1893-1897.
  40. Liu Q, Guntuku S, Cui XS, et al: Chk1 is an essential kinase that is regulated by Atr and required for the G(2)/M DNA damage checkpoint.  Genes Dev2000; 14:1448-1459.
  41. Shiloh Y: ATM and related protein kinases: safeguarding genome integrity.  Nat Rev Cancer2003; 3:155-168.
  42. Tibbetts RS, Brumbaugh KM, Williams JM, et al: A role for ATR in the DNA damage-induced phosphorylation of p53.  Genes Dev1999; 13:152-157.
  43. Xu B, O'Donnell AH, Kim ST, Kastan MB: Phosphorylation of serine 1387 in Brca1 is specifically required for the Atm-mediated S-phase checkpoint after ionizing irradiation.  Cancer Res2002; 62:4588-4591.
  44. Bakkenist CJ, Kastan MB: DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation.  Nature2003; 421:499-506.
  45. Hanawalt PC: Subpathways of nucleotide excision repair and their regulation.  Oncogene2002; 21:8949-8956.
  46. Ford JM: Regulation of DNA damage recognition and nucleotide excision repair: another role for p53.  Mutat Res2005; 577:195-202.
  47. Friedberg EC: DNA Repair,  New York, WH Freeman, 1985.
  48. Setlow RB, Carrier W: The disappearance of thymidine dimers from DNA: an error correcting mechanism.  Proc Natl Acad Sci USA1964; 51:226-231.
  49. Boyce R, Howard-Flanders P: Release of UV light-induced thymidine dimers from DNA in E. coli..  Proc Natl Acad Sci USA1964; 51:293-300.
  50. Pettijohn D, Hanawalt PC: Evidence for repair-replication of UV damage in bacteria.  J Mol Biol1964; 9:395-402.
  51. Wood RD: Nucleotide excision repair in mammalian cells.  J Biol Chem1997; 272:23465-23468.
  52. de Laat WL, Jaspers NG, Hoeijmakers JH: Molecular mechanism of nucleotide excision repair.  Genes Dev1999; 13:768-785.
  53. Wood RD, Mitchell M, Sgouros J, Lindahl T: Human DNA repair genes.  Science2001; 291:1284-1289.
  54. Andressoo JO, Mitchell JR, de Wit J, et al: An Xpd mouse model for the combined xeroderma pigmentosum/Cockayne syndrome exhibiting both cancer predisposition and segmental progeria.  Cancer Cell2006; 10:121-132.
  55. Hartman AR, Ford JM: BRCA1 induces DNA damage recognition factors and enhances nucleotide excision repair.  Nat Genet2002; 32:180-184.
  56. Lin PS, Sage J, Ford JM: The role of the Rb/E2F tumor suppressor pathway in nucleotide excision repair.  Proc Am Assoc Cancer Res2006; 47:818.
  57. Huang TT, D'Andrea AD: Regulation of DNA repair by ubiquitylation.  Nat Rev Mol Cell Biol2006; 7:323-334.
  58. Friedberg EC, Lehmann AR, Fuchs RP: Trading places: how do DNA polymerases switch during translesion DNA synthesis?.  Mol Cell2005; 18:499-505.
  59. Masutani C, Kusumoto R, Yamada A, et al: The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase eta.  Nature1999; 399:700-704.
  60. Johnson RE, Kondratick CM, Prakash S, Prakash L: hRAD30 mutations in the variant form of xeroderma pigmentosum.  Science1999; 285:263-265.
  61. Cleaver JE: Defective repair replication of DNA in xeroderma pigmentosum.  Nature1968; 218:652-656.
  62. Kraemer KH, Lee MM, Scotto J: DNA repair protects against cutaneous and internal neoplasia: evidence from xeroderma pigmentosum.  Carcinogenesis1984; 5:511-514.
  63. Bootsma D, Kraemer KH, Cleaver JE, Hoeijmakers JHJ: Nucleotide excision repair syndromes: xeroderma pigmentosum, Cockayne syndrome, and trichothiodystrophy.   In: Scriver CR, Beaudet AL, Sly WS, Valle D, ed. The Metabolic & Molecular Bases of Inherited Disease,  New York: McGraw-Hill; 2001:677-703.
  64. Kraemer KH, Slor H: Xeroderma pigmentosum.  Clin Dermatol1985; 3:33-69.
  65. Robbins JH: Xeroderma pigmentosum: defective DNA repair causes skin cancer and neurodegeneration.  JAMA1988; 260:384-388.
  66. Wang YC, Maher VM, Mitchell DL, McCormick JJ: Evidence from mutation spectra that the UV hypermutability of xeroderma pigmentosum variant cells reflects abnormal, error-prone replication on a template containing photoproducts.  Mol Cell Biol1993; 13:4276-4283.
  67. Schmickel RD, Chu EHY, Trosko JE: Cockayne syndrome: a cellular sensitivity to ultraviolet light.  Pediatrics1977; 60:135-139.
  68. Venema J, Mullenders LHF, Natarajan AT, et al: The genetic defect in Cockayne syndrome is associated with a defect in repair of UV-induced DNA damage in transcriptionally active DNA.  Proc Natl Acad Sci USA1990; 87:4707-4711.
  69. Cooper PK, Nouspikel T, Clarkson SG, Leadon SA: Defective transcription-coupled repair of oxidative base damage in Cockayne syndrome patients from XP group G.  Science1997; 275:990-993.
  70. Timme TL, Moses RE: Review: Diseases with DNA damage-processing defects.  Am J Med Sci1988; 295:40-48.
  71. Nance MA, Berry SA: Cockayne syndrome: review of 140 cases.  Am J Med Gen1992; 42:68-84.
  72. Robbins JH, Kraemer KH, Lutzner MA, et al: Xeroderma pigmentosum: an inherited disease with sun sensitivity, multiple cutaneous neoplasms and abnormal repair.  Ann Intern Med1974; 80:221-228.
  73. Vermeulen W, Scott RJ, Rodgers S, et al: Clinical heterogeneity within xeroderma pigmentosum associated with mutations in the DNA repair and trasncription gene ERCC3.  Am J Hum Genet1994; 54:191-200.
  74. Lehmann AR: The xeroderma pigmentosum group D (XPD) gene: one gene, two functions, three diseases.  Genes Dev2001; 15:15-23.
  75. Benhamou S, Sarasin A: ERCC2/XPD gene polymorphisms and cancer risk.  Mutagenesis2002; 17:463-469.
  76. Mohrenweiser HW, Wilson DM, Jones IM: Challenges and complexities in estimating both the functional impact and the disease risk associated with the extensive genetic variation in human DNA repair genes.  Mutat Res2003; 526:93-125.
  77. Demple B, Harrison L: Repair of oxidative damage to DNA: enzymology and biology.  Annu Rev Biochem1994; 63:915-948.
  78. Seeberg E, Eide L, Bjoras M: The base excision repair pathway.  Trends Biochem Sci1995; 20:391-397.
  79. Al-Tassan N, Chmiel NH, Maynard J, et al: Inherited variants of MYH associated with somatic G:C ➙ T:A mutations in colorectal tumors.  Nat Genet2002; 30:227-232.
  80. Jones S, Emmerson P, Maynard J, et al: Biallelic germline mutations in MYH predispose to multiple colorectal adenoma and somatic G:C ➙ T:A mutations.  Hum Mol Genet2002; 11:2961-2967.
  81. Halford SE, Rowan AJ, Lipton L, et al: Germline mutations but not somatic changes at the MYH locus contribute to the pathogenesis of unselected colorectal cancers.  Am J Pathol2003; 162:1545-1548.
  82. Cheadle JP, Sampson JR: MUTYH-associated polyposis: from defect in base excision repair to clinical genetic testing.  DNA Repair (Amst)2007; 6:274-279.
  83. Le Page F, Kwoh EE, Avrutskaya A, et al: Transcription-coupled repair of 8-oxoguanine: requirement for XPG, TFIIH, and CSB and implications for Cockayne syndrome.  Cell2000; 101:159-171.
  84. Bohr VA: Repair of oxidative DNA damage in nuclear and mitochondrial DNA, and some changes with aging in mammalian cells.  Free Radic Biol Med2002; 32:804-812.
  85. Morland I, Rolseth V, Luna L, et al: Human DNA glycosylases of the bacterial Fpg/MutM superfamily: an alternative pathway for the repair of 8-oxoguanine and other oxidation products in DNA.  Nucleic Acids Res2002; 30:4926-4936.
  86. Friedberg EC, Meira LB: Database of mouse strains carrying targeted mutations in genes affecting biological responses to DNA damage: version 5.  DNA Repair (Amst)2003; 2:501-530.
  87. Loeb LA, Springgate CF, Battula N: Errors in DNA replication as a basis of malignant changes.  Cancer Res1974; 34:2311-2321.
  88. Loeb LA: Mutator phenotype may be required for multistage carcinogenesis.  Cancer Res1991; 51:3075-3079.
  89. Hung RJ, Hall J, Brennan P, Boffetta P: Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review.  Am J Epidemiol2005; 162:925-942.
  90. Kiyohara C, Takayama K, Nakanishi Y: Association of genetic polymorphisms in the base excision repair pathway with lung cancer risk: a meta-analysis.  Lung Cancer2006; 54:267-283.
  91. Fearnhead NS, Wilding JL, Winney B, et al: Multiple rare variants in different genes account for multifactorial inherited susceptibility to colorectal adenomas.  Proc Natl Acad Sci USA2004; 101:15992-15997.
  92. Fearnhead NS, Winney B, Bodmer WF: Rare variant hypothesis for multifactorial inheritance: susceptibility to colorectal adenomas as a model.  Cell Cycle2005; 4:521-525.
  93. Spitz MR, Wei Q, Dong Q, et al: Genetic susceptibility to lung cancer: the role of DNA damage and repair.  Cancer Epidemiol Biomarkers Prev2003; 12:689-698.
  94. Kolodner R: Biochemistry and genetics of eukaryotic mismatch repair.  Genes Dev1996; 10:1433-1442.
  95. Kolodner RD, Marsischky GT: Eukaryotic DNA mismatch repair.  Curr Opin Genet Dev1999; 9:89-96.
  96. Jiricny J, Nystrom-Lahti M: Mismatch repair defects in cancer.  Curr Opin Genet Dev2000; 10:157-161.
  97. Peltomaki P: Role of DNA mismatch repair defects in the pathogenesis of human cancer.  J Clin Oncol2003; 21:1174-1179.
  98. Fishel R, Lescoe MK, Rao MRS, et al: The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer.  Cell1993; 75:1027-1038.
  99. Hemminki A, Peltomaki P, Mecklin JP, et al: Loss of the wild type MLH1 gene is a feature of hereditary nonpolyposis colorectal cancer.  Nat Genet1994; 8:405-410.
  100. Papadopoulos N, Nicolaides NC, Wei YF, et al: Mutation of a mutL homolog in hereditary colon cancer.  Science1994; 263:1625-1629.
  101. Karran P, Bignami M: DNA damage tolerance, mismatch repair and genome instability.  Bioessays1994; 16:833-839.
  102. Lynch HT, Smyrk T: Hereditary nonpolyposis colorectal cancer (Lynch syndrome): an updated review.  Cancer1996; 8:1149-1167.
  103. Lynch HT, de la Chapelle A: Hereditary colorectal cancer.  N Engl J Med2003; 348:919-932.
  104. Parsons R, Li GM, Longley MJ, et al: Hypermutability and mismatch repair deficiency in RER+ tumor cells.  Cell1993; 75:1227-1236.
  105. Thibodeau SN, Bren G, Schaid D: Microsatellite instability in cancer of the proximal colon.  Science1993; 260:816-819.
  106. Kuismanen SA, Holmberg MT, Salovaara R, et al: Epigenetic phenotypes distinguish microsatellite-stable and -unstable colorectal cancers.  Proc Natl Acad Sci USA1999; 96:12661-12666.
  107. Nakagawa H, Nuovo GJ, Zervos EE, et al: Age-related hypermethylation of the 5′ region of MLH1 in normal colonic mucosa is associated with microsatellite-unstable colorectal cancer development.  Cancer Res2001; 61:6991-6995.
  108. Ford JM, Whittemore AS: Predicting and preventing hereditary colorectal cancer.  JAMA2006; 296:1521-1523.
  109. Gryfe R, Kim H, Hsieh ET, et al: Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer.  N Engl J Med2000; 342:69-77.
  110. Hemminki A, Mecklin JP, Jarvinen H, et al: Microsatellite instability is a favorable prognostic indicator in patients with colorectal cancer receiving chemotherapy.  Gastroenterology2000; 119:921-928.
  111. Samowitz WS, Curtin K, Ma KN, et al: Microsatellite instability in sporadic colon cancer is associated with an improved prognosis at the population level.  Cancer Epidemiol Biomarkers Prev2001; 10:917-923.
  112. Ji H, Kumm J, Zhang M, et al: Molecular inversion probe analysis of gene copy alterations reveals distinct categories of colorectal carcinoma.  Cancer Res2006; 66:7910-7919.
  113. Bell DW, Varley JM, Szydlo TE, et al: Heterozygous germ line hCHK2 mutations in Li-Fraumeni syndrome.  Science1999; 286:2528-2531.
  114. Meijers-Heijboer H, van den Ouweland A, Klijn J, et al: Low-penetrance susceptibility to breast cancer due to CHEK2(*)1100delC in noncarriers of BRCA1 or BRCA2 mutations.  Nat Genet2002; 31:55-59.
  115. Vahteristo P, Bartkova J, Eerola H, et al: A CHEK2 genetic variant contributing to a substantial fraction of familial breast cancer.  Am J Hum Genet2002; 71:432-438.
  116. Kastan MB, Onyekwere O, Sidransky D, et al: Participation of p53 protein in the cellular response to DNA damage.  Cancer Res1991; 51:6304-6311.
  117. Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB: Wild-type p53 is a cell cycle checkpoint determinant following irradiation.  Proc Natl Acad Sci USA1992; 89:7491-7495.
  118. Giaccia AJ, Kastan MB: The complexity of p53 modulation: emerging patterns from divergent signals.  Genes Dev1998; 12:2973-2983.
  119. Khanna KK, Keating KE, Kozlov S, et al: ATM associates with and phosphorylates p53: mapping the region of interaction.  Nat Genet1998; 20:398-400.
  120. Banin S, Moyal L, Shieh S, et al: Enhanced phosphorylation of p53 by ATM in response to DNA damage.  Science1998; 281:1674-1677.
  121. O'Driscoll M, Jeggo PA: The role of double-strand break repair: insights from human genetics.  Nat Rev Genet2006; 7:45-54.
  122. Gatti RA, Tward A, Concannon P: Cancer risk in ATM heterozygotes: a model of phenotypic and mechanistic differences between missense and truncating mutations.  Mol Genet Metab1999; 68:419-423.
  123. Dork T, Bendix R, Bremer M, et al: Spectrum of ATM gene mutations in a hospital-based series of unselected breast cancer patients.  Cancer Res2001; 61:7608-7615.
  124. Scott SP, Bendix R, Chen P, et al: Missense mutations but not allelic variants alter the function of ATM by dominant interference in patients with breast cancer.  Proc Natl Acad Sci USA2002; 99:925-930.
  125. Carney JP, Maser RS, Olivares H, et al: The hMre11/hRad50 protein complex and Nijmegen breakage syndrome: linkage of double-strand break repair to the cellular DNA damage response.  Cell1998; 93:477-486.
  126. Stewart GS, Maser RS, Stankovic T, et al: The DNA double-strand break repair gene hMRE11 is mutated in individuals with an ataxia-telangiectasia-like disorder.  Cell1999; 99:577-587.
  127. O'Driscoll M, Ruiz-Perez VL, Woods CG, et al: A splicing mutation affecting expression of ataxia-telangiectasia and Rad3-related protein (ATR) results in Seckel syndrome.  Nat Genet2003; 33:497-501.
  128. Nakayama H, Nakayama K, Nakayama R, et al: Isolation and genetic characterization of a thymineless death-resistant mutant of Escherichia coli K12: identification of a new mutation (recQ1) that blocks the RecF recombination pathway.  Mol Gen Genet1984; 195:474-480.
  129. Hickson ID: RecQ helicases: Caretakers of the genome.  Nat Rev Cancer2003; 3:169-178.
  130. Ellis NA, Groden J, Ye TZ, et al: The Bloom's syndrome gene product is homologous to RecQ helicases.  Cell1995; 83:655-666.
  131. Ellis NA, German J: Molecular genetics of Bloom's syndrome.  Hum Mol Genet1996; 5:1457-1463.
  132. Gruber SB, Ellis NA, Scott KK, et al: BLM heterozygosity and the risk of colorectal cancer.  Science2002; 297:2013.
  133. Goss KH, Risinger MA, Kordich JJ, et al: Enhanced tumor formation in mice heterozygous for Blm mutation.  Science2002; 297:2051-2053.
  134. Yu CE, Oshima J, Fu YH, et al: Positional cloning of the Werner's syndrome gene.  Science1996; 272:258-262.
  135. Oshima J: The Werner syndrome protein: an update.  Bioessays2000; 22:894-901.
  136. Shen JC, Loeb LA: The Werner syndrome gene: the molecular basis of RecQ helicase-deficiency diseases.  Trends Genet2000; 16:213-220.
  137. Kitao S, Shimamoto A, Goto M, et al: Mutations in RECQL4 cause a subset of cases of Rothmund-Thomson syndrome.  Nat Genet1999; 22:82-84.
  138. Wang LL, Gannavarapu A, Kozinetz CA, et al: Association between osteosarcoma and deleterious mutations in the RECQL4 gene in Rothmund-Thomson syndrome.  J Natl Cancer Inst2003; 95:669-674.
  139. Hollstein M, Sidranksky D, Vogelstein B, Harris CC: p53 mutations in human cancers.  Science1991; 253:49-53.
  140. Levine AJ: p53, the cellular gatekeeper for growth and division.  Cell1997; 88:323-331.
  141. Sengupta S, Harris CC: p53: traffic cop at the crossroads of DNA repair and recombination.  Nat Rev Mol Cell Biol2005; 6:44-55.
  142. Wei CL, Wu Q, Vega VB, et al: A global map of p53 transcription-factor binding sites in the human genome.  Cell2006; 124:207-219.
  143. Oren M: Regulation of the p53 tumor suppressor protein.  J Biol Chem1999; 274:36031-36034.
  144. El-Deiry WS, Tokino T, Velculescu VE, et al: WAF1, a potential mediator of p53 tumor suppression.  Cell1993; 75:817-825.
  145. Vousden KH, Lu X: Live or let die: the cell's response to p53.  Nat Rev Cancer2002; 2:594-604.
  146. Malkin D, Li FP, Strong LC, et al: Germ line p53 mutations in a familial syndrome of breast cancer, sarcomas and other neoplasms.  Science1990; 250:1233-1238.
  147. Srivastava S, Zou ZQ, Pirollo K, et al: Germline transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome.  Nature1990; 348:747-749.
  148. Frebourg T, Barbier N, Yan YX, et al: Germline p53 mutations in 15 families with Li-Fraumeni syndrome.  Am J Hum Genet1995; 56:608-615.
  149. Zhang J, Powell SN: The role of the BRCA1 tumor suppressor in DNA double-strand break repair.  Mol Cancer Res2005; 3:531-539.
  150. Hartman AR, Ford JM: BRCA1 and p53: compensatory roles in DNA repair.  J Mol Med2003; 81:700-707.
  151. Bryant HE, Schultz N, Thomas HD, et al: Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase.  Nature2005; 434:913-917.
  152. Farmer H, McCabe N, Lord CJ, et al: Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy.  Nature2005; 434:917-921.
  153. Horton JK, Wilson SH: Hypersensitivity phenotypes associated with genetic and synthetic inhibitor-induced base excision repair deficiency.  DNA Repair (Amst)2007; 6:530-543.
  154. Lord CJ, Garrett MD, Ashworth A: Targeting the double-strand DNA break repair pathway as a therapeutic strategy.  Clin Cancer Res2006; 12:4463-4468.
  155. Ratnam K, Low JA: Current development of clinical inhibitors of poly(ADP-ribose) polymerase in oncology.  Clin Cancer Res2007; 13:1383-1388.
  156. Ribic CM, Sargent DJ, Moore MJ, et al: Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer.  N Engl J Med2003; 349:247-257.
  157. Jasin M: Homologous repair of DNA damage and tumorigenesis: the BRCA connection.  Oncogene2002; 21:8981-8993.
  158. Venkitaraman AR: Cancer susceptibility and the functions of BRCA1 and BRCA2.  Cell2002; 108:171-182.
  159. Venkitaraman AR: A growing network of cancer-susceptibility genes.  N Engl J Med2003; 348:1917-1919.
  160. Xu B, Kim S, Kastan MB: Involvement of Brca1 in S-phase and G(2)-phase checkpoints after ionizing irradiation.  Mol Cell Biol2001; 21:3445-3450.
  161. Wang Y, Cortez D, Yazdi P, et al: BASC, a super complex of BRCA1-associated proteins involved in the recognition and repair of aberrant DNA structures.  Genes Dev2000; 14:927-939.
  162. Moynahan ME, Chiu JW, Koller BH, Jasin M: Brca1 controls homology-directed DNA repair.  Mol Cell1999; 4:511-518.
  163. Harkin DP, Bean JM, Miklos D, et al: Induction of GADD45 and JNK/SAPK-dependent apoptosis following inducible expression of BRCA1.  Cell1999; 97:575-586.
  164. Pellegrini L, Yu DS, Lo T, et al: Insights into DNA recombination from the structure of a RAD51-BRCA2 complex.  Nature2002; 420:287-293.
  165. Yang H, Jeffrey PD, Miller J, et al: BRCA2 function in DNA binding and recombination from a BRCA2-DSS1-ssDNA structure.  Science2002; 297:1837-1848.
  166. Alter BP: Cancer in Fanconi anemia, 1927–2001.  Cancer2003; 97:425-440.
  167. D'Andrea AD, Grompe M: The Fanconi anaemia/BRCA pathway.  Nat Rev Cancer2003; 3:23-34.
  168. Kennedy RD, D'Andrea AD: DNA repair pathways in clinical practice: lessons from pediatric cancer susceptibility syndromes.  J Clin Oncol2006; 24:3799-3808.
  169. Gurtan AM, D'Andrea AD: Dedicated to the core: understanding the Fanconi anemia complex.  DNA Repair (Amst)2006; 5:1119-1125.
  170. Howlett NG, Taniguchi T, Olson S, et al: Biallelic inactivation of BRCA2 in Fanconi anemia.  Science2002; 297:606-609.
  171. Garcia-Higuera I, Taniguchi T, Ganesan S, et al: Interaction of the Fanconi anemia proteins and BRCA1 in a common pathway.  Mol Cell2001; 7:249-262.
  172. Taniguchi T, Garcia-Higuera I, Xu B, et al: Convergence of the fanconi anemia and ataxia telangiectasia signaling pathways.  Cell2002; 109:459-472.