Ataxia-telangiectasis occurs in individuals who completely lack functional ataxia-telangiectasia mutated (ATM) protein. Originally known as Louis-Bar syndrome (2), this autosomal recessive disorder occurs worldwide, with an prevalence of 1:40,000-1:100,000 in outbred populations (reviewed in Ref. 3). The pleiotropic clinical phenotype of individuals with AT is summarized below.
AT patients rarely have functional abnormalities as infants. However, the vast majority develop gait abnormalities in their second year of life. All patients lose cere-bellar function over time, resulting in progressive ataxia, ocular apraxia, dysarthric speech, drooling, and choreoathetoid movements (4,5). Most index cases are initially thought to have a form of cerebral palsy (6). The classic AT patient is wheelchair-bound by the end of the first decade of life, and older AT patients may develop intellectual arrest. Dysphagia and silent aspiration become major problems when the neurodegeneration reaches an advanced stage (7). Functional neurological abnormalities in AT patients are accompanied by a continual loss of neurons from both the central and peripheral nervous system (reviewed in Ref. 4). The cerebellum is particularly affected owing to the cumulative effects of ongoing Purkinje cell death (Fig. 1).
Most, but not all, AT homozygotes express clinically significant, but nonprogressive, humeral and cellular immune defects. These can include one or more of the following: thymic hypoplasia, low numbers of circulating T cells, functional impairment of T-cell-mediated immunity, abnormally high levels of IgM, oligo-clonal expansions, and/or selective deficiencies of IgA, IgE, IgG2, and IgG4 (8-10). Opportunistic infections are rare; however, otitis media and sinus infections are frequent. The risk of lower respiratory infections (pneumonia and bronchitis) increases with age, and the combination of immunodeficiency and progressive loss of cerebellar function makes aspiration pneumonia the leading cause of death in AT patients, whose median life expectancy was estimated in a recent survey to be ~30 years (11).
Cancer is a frequent complication of AT, with the lifetime risk estimated to be 30-40% (12). The most common malignancies are tumors of the immune system that occur in the first 15 years of life (Fig. 2). More than 40% of all tumors in AT patients are non-Hodgkin's lymphomas, another ~20% are acute lymphocytic leukemias, and ~5% are Hodgkin's lymphomas (4,13-15). Older AT patients are at risk to develop solid tumors and chronic T-cell leukemias (4,14,16). A wide range of solid tumors have been reported (4,13,17), the most frequent being gastric carcinoma, breast carcinoma, medulloblastoma, basal cell carcinoma, ovarian dysger-minoma, hepatoma, and uterine leiomyoma (in approximate order of frequency).
Many female AT patients have congenital hypoplasia of the ovaries and menarche may be delayed or absent (4,18). Male patients have been shown to have histological abnormalities of their testes, and incomplete spermatogenesis has been reported (19,20), although hypogonadism is less frequent and milder than in affected females.
In addition to an increased susceptibility to malignancy and occasional insulin-resistant diabetes (21), AT homozygotes also show multiple dermatological changes associated with premature aging. Telangiectasias usually appear on the sclera, face, and antecubital/popliteal fossae by 6-7 years of age (6). In the second and third decades of life, AT homozygotes develop additional cutaneous changes, including graying of the hair, senile keratoses, skin atrophy, and areas of hyper-pigmentation and hypopigmentation (4).
Atherosclerotic heart disease and stroke are not a major problem for AT homozygotes, perhaps because of their early death. However, a recent epidemiological study found that AT carriers died an average of 7.8 years earlier than noncar-riers in the same families, a difference that was found to be highly significant (P<.001). Although cancer caused most of the excessive deaths in this study, a 2.6 relative risk of ischemic heart disease caused the remainder (22).
Serum alpha-fetoprotein (AFP) levels are elevated in >90% of older children with AT (18). AT patients usually exhibit mild postnatal growth retardation, and nutrition is a problem in those older individuals with dysphagia and advanced cerebel-lar degeneration. AT homozygotes tend to have sad, hypotonic facies, but congenital malformations are not a feature of the syndrome.
Genetic instability is a hallmark of the AT phenotype. Spontaneous in vivo chromosomal aberrations occur frequently in both lymphoid and nonlymphoid cells from AT patients (reviewed in Refs. 3, 23, and 24). The range of spontaneous karyotypic aberrations includes chromosomal breaks, acentric fragments, dicen-tric chromosomes, structural rearrangements, and aneuploidy. Although the breakpoints in nonlymphoid cells appear to be distributed randomly, AT lymphocytes also have 30- to 50-fold increases in the frequency of rearrangements involving four specific sites: 7p14, 7q35, 14q11.2 and 14q32 (25). These rearrangements interrupt T-cell receptor (TCR) and Ig heavy chain genes, genes that sustain obligatory double-strand breaks (DSBs) and subsequent recombination during the maturation of the immune system (24). Genetic instability has been documented at individual loci (26-30), and spontaneous intrachromosomal recombination is elevated 30- to 200-fold in AT cells grown in culture (31,32).
The first clue that defects in the ATM gene disrupt normal cellular responses to DNA damage were provided by case reports of AT patients who died from severe reactions to radiation therapy directed at their lymphoid tumors (33-37). Although not unusually susceptible to the induction of DNA breaks by ionizing radiation (38,39), cultured cells from AT patients are typically three to five times more sensitive than control cells to the cytotoxic effects of ionizing radiation as well as other agents that induce DSBs (40-44). AT radiosensitivity is also manifest by an increase in number of chromosomal aberrations induced by ionizing radiation or radiomimetic agents (reviewed in Ref. 23).
Cells from AT homozygotes can be exquisitely sensitive to the cytotoxic effects of ionizing radiation, but they differ from most other radiosensitive mammalian cells in that their ability to repair DNA damage appears to be largely intact. Base excision repair is apparently normal (45), and multiple biochemical studies have failed to detect gross abnormalities in the kinetics of single-strand and doublestrand break repair in AT cells (e.g., see Refs. 46, and 47). Other reports have found no evidence that AT cells are functionally defective in DNA repair (48,49). On the other hand, AT homozygotes appear to have subtle defects in their ability to process DNA breaks, as the accuracy of strand rejoining is reduced in AT cells (50-52). In addition, several studies have demonstrated increases in the fraction of DNA breaks left unrepaired in irradiated AT cells (38,53). This may indicate an inability to repair a small but critical fraction of DSBs (38,54-56) such as those that give rise to chromosomal breaks or that arise at replication forks (57). Alternatively, these persistent breaks may be early signs of incipient apoptosis (58).
In normal mammalian cells, multiple cell cycle checkpoints are triggered following exposure to ionizing radiation (reviewed in Refs. 59, and 60). These checkpoints also may restrain the cell cycle temporarily in response to the generation of strand breaks, shortened telomeres, stalled replication forks, and other sites of DNA damage that occurs spontaneously during the course of normal DNA metabolism (e.g., site-specific gene rearrangements, genetic recombination, and repair of replication errors). AT homozygotes lack the p53-mediated G1/S damCopyright © 2003 by Marcel Dekker, Inc. All Rights Reserved.
age-sensitive checkpoint (61), and AT cells do not have the S phase checkpoint, resulting in the phenomenon of radioresistant DNA synthesis (62). The G2/M and mitotic spindle checkpoints also appear to be defective in AT cells (63-66). However, the late-onset G2/M checkpoint that results in prolonged G2 arrest after x-irradiation is intact in AT cells (67).
Following exposure to ionizing radiation, intracellular concentrations of p53, p21, GADD45, c-jun, and ku70 normally increase and nuclear factor-kB (NF-kB), inhibitor kB kinase (IkB), and the p56lyn kinase are activated. In addition, a growing number of proteins have been shown to be phosphorylated following x-irradiation or other treatments that induce DSBs (S Table 1)]. These responses are all impaired in AT cells grown in culture, suggesting that they are ATM-dependent events (68-82).
L. Apoptosis, AT, and ATM Function
Several years ago, we noted that the sensitivity of AT cells to the killing effects of ionizing radiation was not due to a lack of cell cycle checkpoints or to a defect in repair of the bulk of DSBs. As an alternative, we proposed that ATM normally acts to protect cells from DNA damage-induced apoptosis, and that DSB-induced apoptosis was the primary cause of radiation sensitivity and gonadal abnormalities in AT (83,84). This was based, in part, on our finding that cultured AT fibroblasts and lymphoblasts are unusually susceptible to the induction of late-onset (>48 hr postirradiation) apoptosis by low doses of x-ray and streptonigrin (83,85). Our observations have since been confirmed by others (58,66,86-91). Indirect support for underlying hypothesis also has been provided by observations that x-irradiation-induced caspase 3 and 9 activities are particularly high in AT fibrob-lasts, ectopic expression of ATM exerts an inhibitory effect on c-myc-mediated apoptosis, and ATM is a target for proteolytic cleavage during apoptosis (92-95).
Subsequent work with Epstein-Barr virus (EBV)-transformed human lymphoblasts and Atm^/^ mice has made it clear, however, that the original hypothesis cannot fully explain the complex relationship between ATM, apoptosis, and DNA damage. As predicted, Atm~y~ mice are easily killed by ionizing radiation, and Atm~y~ spermatocytes and oocytes undergo spontaneous apoptosis during meiosis I, resulting in sterility (96-98). However, although epithelial cells from the gut and the parotid glands from Atm~y~ mice are exquisitely sensitive to radiation-induced apoptosis, lack of functional ATM protein does not appear to ra-diosensitize most tissues, including the brain (99). Surprisingly, inactivation of the ATM gene can actually protect mouse thymocytes and embryonic neurons from radiation-induced apoptosis and save HeLa cells from apoptosis induced by defective telomeres (100-103). In addition, several laboratories have shown that
Was this article helpful?