GENOMIC INSTABILITY SYNDROMES

Contents

• Ataxia Telangiectasia and Related Disorders
• RecQ DNA Helicase-associated Syndromes
• Fanconi anemia
• Other Disorders
• Summary
• References

The human genome is exposed to potentially deleterious genotoxic events during every cell division cycle.

The endogenous sources of DNA damage result from cellular metabolism or routine errors in DNA replication and recombination. In addition, cellular and organismal exposure to exogenous genotoxic agents such as ultraviolet (UV) light, oxidative stress, and chemical mutagens, can lead to a variety of nucleotide modifications and DNA strand breaks. In order to combat these attacks on the genome, the cell has evolved a response system that induces cell cycle arrest, allowing sufficient time for specialized groups of proteins to repair the incurred damage. The DNA damage response system activates the appropriate DNA repair pathway, or, in the case of irreparable damage, induces apoptosis.

In the last decade, the characterization of many proteins involved in sensing and responding to DNA damage has enhanced our understanding of genotoxic stress responses. Mutations in the genes that encode DNA damage response proteins can result in a number of genomic instability syndromes, disorders that often result in a heightened predisposition to cancer.¹ Another hallmark of these disorders is immunodeficiency, a phenotype caused by an inability to properly repair DNA strand breaks that occur during development of the immune system. The phenotypes exhibited by genomic instability syndromes highlight the significance of proteins that sense, relay, or transduce signals associated with the genotoxic stress response (Table 1).

Ataxia Telangiectasia and Related Disorders

Ataxia telangiectasia (A-T) is a devastating genetic disease that presents in the first decade of life. A-T is a multi-symptom disorder characterized by immunodeficiency, neurodegeneration, premature aging, sensitivity to ionizing radiation, and susceptibility to certain types of cancer. A-T is ultimately fatal in the second or third decade of life.²

Figure 1. Phosphorylated histone H2AX (?-H2AX) nucleates the formation of genotoxic stress response complexes. The phosphorylation of ?-H2AX, a very early event in genotoxic stress responses, requires ATM or ATM-related protein kinase activity. ?-H2AX marks the damaged DNA and recruits BRCT-containing proteins, such as 53BP1, MDC1, BRCA1, and Nbs1. ATM or ATM-related kinases can phosphorylate many of these proteins and other downstream targets to relay signals for cell cycle checkpoint activation, DNA repair, or apoptosis. While details regarding protein-protein interactions are only beginning to be elucidated, it is known that 53BP1 recruitment is important for coupling ATM to p53 and SMC1, but not Chk2 activation.

The protein product of the gene mutated in A-T, A-T-mutated (ATM), is a large protein kinase involved in cell cycle checkpoint and genotoxic stress responses.^3-5 ATM is a member of the phosphatidylinositol 3-kinase-related kinase (PIKK) family. All PIKKs have a highly conserved catalytic domain that bears similarity to the catalytic domain of the lipid kinase phosphatidylinositol 3-kinase (PI 3-K). Despite this similarity, all of the PIKK family members exhibit protein rather than lipid kinase activity. ATM and related protein kinases participate in one of the earliest events that occurs in response to DNA insult by phosphorylating the C-terminal tail of the core histone H2AX protein [?-H2AX (phosphorylated form); Figure 1].^6-8 ?-H2AX marks the site of damage and nucleates the formation of damage response and repair complexes.^6,8 BRCA1 C-terminal (BRCT) domain-containing proteins are recruited to ?-H2AX creating a protein scaffold for further assembly of signaling complexes that include molecules such as p53-binding protein 1 (53BP1),^9-12 mediator of DNA damage checkpoint protein 1 (MDC1),^13-15 and the BRCA1 tumor suppressor protein.⁹ Additionally, the BCRT domain-containing protein Nbs1, the Rad50 and Rad51 DNA repair factors, and others are recruited to the damage site. For many of these proteins, specific details regarding direct protein-protein interactions and order of complex assembly are only beginning to be elucidated. For example, a recent study suggests that Nbs-1 can directly bind to γ-H2AX.¹⁶ Furthermore, many of the these proteins are substrates for ATM and related protein kinases, underscoring the critical role of this protein kinase in genotoxic stress responses.^3-5

While the mutation of both ATM alleles, which results in A-T, occurs at low frequency (approximately 1 in 40,000 births), it is estimated that heterozygous carriers constitute 1-2% of the population.³ Carriers may exhibit intermediate sensitivity to radiation and predisposition to cancer, and thus knowledge of ATM protein function may have broad implications for public health.³ For instance, if a radiation-sensitive carrier is diagnosed with cancer, awareness of the ATM mutation may cause the physician and patient to consider treatment options other than radiation therapy.

Patients with Nijmegen breakage syndrome (NBS) are immunodeficient and exhibit increased cancer incidence, but unlike A-T, are not ataxic.¹⁷ NBS does affect the nervous system however, as these patients present with microcephaly and mental retardation.¹⁷ The gene mutated in NBS, Nbs1, encodes a protein that is a downstream substrate of the ATM protein kinase. Nbs1 is part of a DNA double strand break-binding complex that also includes the Mre11 and Rad50 gene products, γ-H2AX¹⁶ nucleating the formation of the Nbs1/Mre11/Rad50 trimer. The Mre11 gene is mutated in another related syndrome, A-T-like disorder (ATLD).¹⁹ Patients with ATLD exhibit chromosomal instability, ionizing radiation sensitivity, and cerebellar degeneration, although they are not immunodeficient and show no increased incidence of cancer.

Interestingly, while the clinical phenotypes of these diseases are variable, cultured cells derived from patients exhibit very similar characteristics. A-T, NBS, and ATLD cells are all sensitive to ionizing radiation, exhibit chromosomal instability, and possess multiple cell cycle checkpoint defects.³ One common feature of all three cell types is radioresistant DNA synthesis (RDS), or the inability of the cell to suppress DNA synthesis following irradiation.³ When a normal cell is exposed to genotoxic stress, DNA synthesis stops until the damage is repaired and synthesis can resume. The presence of RDS implies a defect in S phase, the DNA synthesis checkpoint of the cell cycle, whereby the cell continues to synthesize damaged DNA and has the potential to pass on unrepaired errors to daughter cells.

The direct relationships between the mutated genes encoding the proteins associated with A-T, NBS, and ATLD underscore their participation in common signaling pathways in response to assaults on the genome. However, differences in the clinical presentation of these disorders also suggest that their functions are distinct. There are differences in the phenotypes of the related mouse models as well. ATM^-/- mice are viable and exhibit many features of A-T, with the noted exception of neurodegenerative effects. In contrast, Nbs1^-/-, Mre11^-/-, and Rad50^-/- mice are not viable and expire at various times during embryonic development.^17,20 However, mice bearing a truncation in the Nbs1 gene that mimics a mutation found in humans are viable and exhibit some similarities to NBS patients, albeit with fewer malignancies.¹⁷ Rad50 mutations are not documented in humans, which may indicate a crucial function for this protein.²⁰

RecQ DNA Helicase-associated Syndromes

The human RecQ helicase family has 5 members. These proteins bear a high degree of similarity to the helicase domain of the E.coli RecQ helicase, an enzyme involved in DNA recombination and repair.²¹ Three of the human RecQ proteins are defined as tumor suppressor genes. This is evidenced by their mutation in three distinct genomic instability syndromes that share a predisposition to cancer. BLM, WRN, and RTS, the genes mutated in Bloom's syndrome, Werner's syndrome, and Rothmund-Thompson syndrome, respectively, encode proteins that contain helicase activities stimulated by single-stranded DNA binding proteins.^22-24 These helicases play an important role in regulating recombination, a fact that translates into elevated levels of recombination, or a "hyperrecombination" phenotype, in eukaryotic cells deficient in one of these proteins.²¹ Hyperrecombination can lead to large deletions that may have an impact on tumor formation.²⁵ All three enzymes have unique substrate specificities for particular types of DNA, and exhibit very little sequence similarity outside of the helicase domain. This variation in substrate specificity is consistent with syndrome-specific clinical features. For instance, Bloom's syndrome patients are susceptible to a wide range of epithelial and leukemic cancers,²⁶ while Werner's syndrome leads to a predominance of thyroid cancer, melanoma, and various sarcomas,^21,27 and Rothmund-Thompson syndrome patients exhibit an increased prevalence of skin cancer and osteosarcoma.^21,28

Of the three RecQ helicases described above, the functions and interactions of the WRN and BLM proteins are the best defined. WRN and BLM are both substrates of the ATM protein kinase and defects in either enzyme result in the loss of p53-mediated apoptosis.^21,29,30 One unique feature of cells from Werner's patients is accelerated senescence, implying impaired telomere function. Indeed, the WRN protein colocalizes with telomere-binding proteins, and demonstrates substrate specificity for telomeric DNA.²¹ Consistent with the WRN cellular phenotype, patients with Werner's syndrome exhibit signs of premature aging.²¹

Fanconi anemia

Fanconi anemia (FA) is characterized by congenital abnormalities in multiple organ systems, pancytopenia from bone-marrow failure, chromosomal instability, and increased cancer susceptibility.^31,32 One defining feature of FA is chromosomal sensitivity to DNA crosslinking agents such as mitomycin C and diepoxybutane.^31,32 Chromosomal breakage induced by these compounds has been used as a diagnostic test for FA.³²

In contrast to the diseases resulting from A-T-related and RecQ helicase mutations, relatively little is known regarding the function of the proteins mutated in FA. Patients with FA are categorized into 8 different complementation groups, and 7 different FA genes including, FANC-A, FANC-C, FANC-D1, FANC-D2, FANC-E, FANC-F, and FANC-G have been cloned to date.³² These genes are not clustered on a single chromosome, but are distributed widely throughout the genome. Of the FA proteins, FANC-A, FANC-C, FANC-E, FANC-F, and FANC-G contain no obvious catalytic domains and exist in a nuclear complex (A/C/E/F/G) that possesses no enzymatic activity.³² However, it has been shown that the A/C/E/F/G complex mediates the monoubiquitination of FANC-D2 in response to DNA damage. This post-translational modification results in translocation of FANC-D2 to nuclear foci where DNA-repair complexes including the proteins BRCA1, BRCA2, and Rad51 are localized.³² It is interesting to note that the FANC-D1 protein was recently identified as being identical to the BRCA2 protein, and that patients with BRCA2/FANC-D1 mutations share clinical features with FA.³³

FANC-A, FANC-C, FANC-D2, and FANC-G genes have been targeted in mice, and the resulting phenotype is not as severe as that found in human FA.^32,34-36 These mice are sensitive to DNA crosslinking agents and ionizing radiation, but do not exhibit anemia. Furthermore, only FANC-D2-deficient mice show an increased cancer risk, exhibiting a high incidence of epithelial tumors.

Other Disorders

The nucleotide-excision repair (NER) process employs an array of more than 30 proteins that detect DNA damage, excise the damaged strand, and synthesize new DNA using the complementary strand as a template.^37,38 Types of DNA lesions repaired by NER include cyclobutane pyrimidine dimers and 6-4 photo products resulting from UV light exposure.³⁸ Two autosomal recessive disorders, Xeroderma pigmentosum (XP) and Cockayne's syndrome (CS), occur as a result of mutations in NER genes. XP results from defects in one of seven genes (XPA through XPG), while CS results from mutations in the genes CSA or CSB.^3,38 Patients with either disorder are extremely sensitive to UV light while those with XP, specifically, exhibit a predisposition to cancer.^3,38 Mutations in the XPB and XPD are also associated with Tricothiodystrophy (TTD), a disorder in which patients present with mental retardation and UV light sensitivity, but no increased incidence of cancer.^3,38

Summary

These groups of genomic instability disorders have united physicians treating these patients with researchers interested in the genotoxic stress response, cell cycle, and DNA repair. As current treatment options for many of these diseases are limited, a more thorough understanding of defects at the molecular level will hopefully lead to new drug targets and ultimately therapies and treatments for these patients. New insights into genotoxic stress pathways will likely yield additional treatment options for cancer patients as well.

References

1. Duker, N.J. (2002) Am. J. Med. Gen. 115:125.
2. Rotman, G. & Y. Shiloh (1998) Hum. Mol. Gen. 7:1555.
3. Shiloh, Y. (2003) Nat. Rev. Cancer 3:155.
4. Abraham, R.T. (2001) Genes Dev. 15:2177.
5. Kastan, M.B. & D.-S. Lim (2000) Nat. Rev. Mol. Biol. 1:179.
6. Paull, T.T. et al. (2000) Curr. Biol. 10:886.
7. Downs, J.A. et al. (2000) Nature 408:1001.
8. Celeste, A. et al. (2002) Science 296:922.
9. Abraham, R.T. (2002) Nat. Cell Biol. 4:E277.
10. Wang, B. et al. (2002) Science 298:1435.
11. DiTullio, Jr., R.A. et al. (2002) Nat. Cell Biol. 4:998.
12. Fernandez-Capetillo, O. et al. (2002) Nat. Cell Biol. 4:993.
13. Lou, Z. et al. (2003) Nature 421:957.
14. Goldberg, M. et al. (2003) Nature 421:953.
15. Stewart, G.S. et al. (2003) Nature 421:961.
16. Kobayashi, J. et al. (2002) Curr. Biol. 12:1846.
17. Tauchi, H. et al. (2002) Oncogene 21:8967.
18. Carney, J.P. et al. (1998) Cell 93:477.
19. Stewart, G.S. et al. (1999) Cell 99:577.
20. de Jager, M. & R. Kanaar (2002) Genes Dev. 16:2173.
21. Nakayama, H. (2002) Oncogene 21:9008.
22. Ellis, N.A. et al. (1995) Cell 83:655.
23. Yu, C.-E. et al. (1996) Science 272:258.
24. Kitao, S. et al. (1998) Genomics 54:443.
25. Tachibana, A. et al. (1996) Mol. Carcinog. 17:41.
26. German, J. et al. (1997) Cancer Genet. Cytogenet. 93:100.
27. Goto, M. et al. (1996) Cancer Epidemiol. Biomarkers Prev. 5:239.
28. Vennos, E.M. et al. (1992) J. Am. Acad. Dermatol. 27:750.
29. Pichierri, P. et al. (2003) Oncogene 22:1491.
30. Beamish, H. et al. (2002) J. Biol. Chem. 277:30515.
31. Ahmad, S.I. et al. (2002) BioEssays 24:439.
32. D'Andrea, A.D. & M. Grompe (2003) Nat. Rev. Cancer 3:23.
33. Howlett, N.G. et al. (2002) Science 297:606.
34. Carreau, M. et al. (1998) Blood 91:2737.
35. Cheng, N.C. et al. (2000) Hum. Mol. Genet. 9:1805.
36. Yang, Y. et al. (2001) Blood 98:3435.
37. Wood, R. et al. (2001) Science 291:1284.
38. Friedberg, E.C. (2001) Nat. Rev. Cancer 1:22.