(CANCERRESEARCH55. 5991-6001. December 15, 19951 Review Ataxia..Telangiectasia and Cellular Responses to DNA Damage' M. Stephen Meyn2 Departments of Genetics and Pediatrics, Yale University School of Medicine, New Haven, connecticut 06510 Abstract elevated frequencies of spontaneous and induced chromosome aber rations, high spontaneous rates of intrachromosomal recombination, Ataxia-telangiectasia (A-T) is a human disease characterized by high aberrant immune gene rearrangements, and inability to arrest the cell cancer risk, immune defects, radiation sensitivity, and genetic instability. Although A-T homozygotes are rare, the A-T gene may play a role in cycle in response to DNA damage (3—6)]. sporadic breast cancer and other common cancers. Abnormalities of DNA The nature of the A-T defect has been the subject of much repair, genetic recombination, chromatin structure, and cell cycle check speculation; most hypotheses focus on the radiation sensitivity of point control have been proposed as the underlying defect in A-T; how A-T cells. Early reports that A-T fibroblasts were unable to excise ever, previous models cannot satisfactorily explain the plelotropic A-T radiation-induced DNA adducts prompted suggestions that the phenotype. radiation sensitivity of A-T cells was due to an intrinsic defect in Two recent observations help clarify the molecular pathology of A-T: DNA repair (7). However, subsequent work indicated that not all (a) inappropriate p53-mediated apoptosis is the major cause of death in A-T fibroblasts have a defect in DNA adduct excision (8), and that A-T cells irradiated in culture; and (b) ATM, the putative gene for A-T, has extensive homology to several celi cycle checkpoint genes from other the kinetics of repair of DNA breaks and chromosome aberrations organisms. Building on these new observations, a comprehensive model is is grossly normal (reviewed in Ref. 3). Functional abnormalities of presented in which the ATM gene plays a crucial role in a signal trans specific repair enzymes have been proposed [e.g. , topoisomerase II duction network that activates multiple cellular functions In response to (9) and poly(ADP-ribosylase) (10)], but conclusive evidence of DNA damage. In this Damage Survefflance Network model, there is no repair enzyme defects in A-T is lacking. Structural abnormalities intrinsic defect in the machinery of DNA repair in A-T homozygotes, but of chromatin, subtle defects in DNA repair that affect repair their lack of a functional ATM gene results In an Inabifity to: (a) halt at multiple cell cycle checkpoints in response to DNA damage; (b) activate quality, and abnormalities of differentiation also have been offered damage-inducible DNA repair; and (c) prevent the triggering of pro as explanations for the A-T phenotype (11—15). grammed cell death by spontaneous and induced DNA damage. Absence Several investigators have suggested that a defect in genetic recom of damage-sensitive cell cycle checkpoints and damage-induced repair bination, resulting in an inability to productively rearrange and repair disrupts immune gene rearrangements and leads to genetic instability and genes, would provide a unifying explanation for radiation sensitivity, cancer. Triggering of apoptosis by otherwise nonlethal DNA damage is immune defects, and karyotypic abnormalities in A-T (5, 6, 16—18). primarily responsible for the radiation sensitivity ofA-T homozygotes and However, a defect in genetic recombination is difficult to resolve with results in an ongoing loss of cells, leading to cerebellar ataxia and neuro reports of near-normal frequencies of extrachromosomal recombina logical deterioration, as well as thymic atrophy, lymphocytopenla, and a paucity of germ cells. lion (4, 19, 20) and the observation that spontaneous rates of chro Experimental evidence supporting the Damage Surveffiance Network mosomal recombination in A-T fibroblast lines are 30- to 200-fold model is summarized, followed by a discussion of how defects in the higher than normal (4). Other investigators, struck by the inability of ATM-dependent signal transduction network might account for the A-T irradiated A-T cells to temporarily halt DNA replication and cell cycle phenotype and what insights this new understanding of A-T can offer progression, have proposed that A-T cells cannot recognize or respond regarding DNA damage response networks, genomic instability, and can to DNA damage (1 1, 12, 21—23).In these models, the radiation cer. sensitivity of A-I cells generally is assumed to result from an inability Previous Models for A-T3 to delay the cell cycle to allow sufficient time to repair DNA damage. These models could explain why A-T cells are radiation sensitive A-T is an autosomal recessive disease with a pleiotropic phenotype despite grossly normal DNA repair, but they cannot readily account that includes progressive cerebellar ataxia, cellular and humoral im for several studies in which experimental conditions that prolonged or mune defects, progenc changes of the skin, endocrine disorders, temporarily halted the cell cycle did not improve the survival of gonadal abnormalities, and a high incidence of cancer; the relative risk irradiated A-I cells (24—26). Neither do these models explain another of developing some tumors (e.g. , lymphoma) is several hundredfold puzzling feature of A-I: unlike most normal mammalian cells, fatally higher than normal ( 1, 2). Heterozygote carriers are also at increased irradiated A-I cells typically die before they can complete the first risk for cancer, particularly breast carcinoma (2). A-T cells are sen postirradiation mitosis (113—115). sitive to the killing effects of ionizing radiation, and they exhibit in The finding that the radiosensitivity of A-I cells in culture is vivo and in vitro abnormalities consistent with a defect involving primarily the result of inappropriate apoptosis, together with recent DNA metabolism and/or maintenance of genomic integrity [e.g., observations regarding the role that the p53 tumor suppressor protein plays in mediating cellular responses to DNA damage, has led to the Received 8/21/95; accepted 10/16/95. development of a new model regarding the nature of the A-I defect. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with Initial analysis of the A-T disease gene supports this model, which is 18 U.S.C. Section 1734 solely to indicate this fact. related to previous proposals in which the A-T gene product normally I This work has been supported by grants from the NIH and the A-I Children's Project. mediates cellular responses to DNA damage (e.g., see Ref. 22), but 2 lo whom requests for reprints should be addressed, at Yale University School of overcomes the objections to those models, cited above, while provid Medicine, 333 Cedar Street, P.O. Box 208005, New Haven. CI 06520-8005. 3 The abbreviations used are: A-I, ataxia-telangiectasia; LOH, loss of heterozygosity; ing a unifying explanation for the pleiotropic manifestations of the ICR, I-cell receptor;P13-kinase.phosphatidylinositol3-kinase. disease. 5991 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1995 American Association for Cancer Research. A COMPREHENSIVE MODEL FOR ATAXIA-TELANGIECTASIA Ihe Damage Surveillance Network Model a critical role. The detection of certain types of spontaneous or induced DNA damage triggers this signal transduction network, resulting in the Biological organisms are not passive targets ofDNA-damaging agents; activation of a group of cellular functions that promote genetic stability they actively respond to DNA damage [e.g., the SOS system in Esche by temporarily arresting the cell cycle and enhancing DNA repair. At the richia coli (30)]. A growing body of evidence indicates that the response of mammalian cellstoDNA damage iscomplex,perhapsinvolving same time, the A-I-dependent network promotes cellular survival by several interrelated signal-transductionnetworks that detect DNA dam inhibiting execution of damage-induced programmed cell death. In ad age, activate DNA repair, and alter the cell cycle (22, 27—29).Fig. IA dition to the five responses illustrated in Fig. 1A, there also may be other, depicts a damage response network in which the A-T gene product plays asyetundefined,A-I-dependentfunctions. A. NormalIndividuals Activated DNARepair @ @GADD45 ,t p21 ‘t'@@ GuS p53 Checkpoint Spontaneous and f Induced DNADamage @. ssandds ___________ S phase @ ImmuneGene DNAbreaks ATM Checkpoint @ Rearrangments \I ShortTelomeres p53 GRIM Checkpoint B. A-T Homozygotes No enhanced reactIvation ‘V d2 of Irradiated virus D air No enhanced mutagenesis @ @GADD45 of irradiated virus @ /@ p21 increased genomic instability chromosomal translocations p53 epithelial cells with micronudel aneuploid cells in muftipletissues allelelossin erythrocytes Spontaneousand I ICR transrearrangements Induced recombination between repeated genes induced chromosome aberrations DNADamage solid tumors @ 55 and ds @. *- Disruption of Immune gene rearrangement ImmuneGene DNAbreaks 9@PQk@t Lowlevelsof lgA, IgE, g02 and lgG4 @ Rearrangments %4@ Lowproportionof I cells expressingW@3TCRs Frequent translocatlons near immune genes ShortTelomeres Increased risk of lymphoma and leukemia p53 Abnormal cell cycle kinetics following DNAdamage C ck nt RadioresistantDNAsynthesis Programmed Cell Death Increased spontaneous cell death progressive loss of neurons thymic hypoplasia