Gene name - telomere fusion
Synonyms - ATM, atm, dATM
Cytological map position - 88E3--4
Function - enzyme
Symbol - tefu
FlyBase ID: FBgn0045035
Genetic map position - 3-
Classification - Phosphatidylinositol 3- and 4-kinase
Cellular location - cytoplasmic
ATM is a large, multifunctional protein kinase that regulates responses required for surviving DNA damage, including functions such as DNA repair, apoptosis, and cell cycle checkpoints. Drosophila ATM (accepted FlyBase name: Telomere fusion) function is essential for normal adult development. Extensive, inappropriate apoptosis occurs in proliferating atm mutant tissues, and in clonally derived atm mutant embryos, frequent mitotic defects are seen. At a cellular level, spontaneous telomere fusions and other chromosomal abnormalities are common in atm larval neuroblasts, suggesting a conserved and essential role for dATM in the maintenance of normal telomeres and chromosome stability. Evidence from other systems supports the idea that DNA double-strand break (DSB) repair functions of ATM kinases promote telomere maintenance by inhibition of illegitimate recombination or fusion events between the legitimate ends of chromosomes and spontaneous DSBs. Drosophila will be an excellent model system for investigating how these ATM-dependent chromosome structural maintenance functions are deployed during development. Because neurons appear to be particularly sensitive to loss of ATM in both flies and humans, this system should be particularly useful for identifying cell-specific factors that influence sensitivity to loss of dATM and are relevant for understanding the human disease, ataxia-telangiectasia (Silva, 2004).
The phosphoinositide 3-kinase-related kinases (PIKK) family in mammals includes six subfamilies based on sequence homology and function. Human ataxia-telangiectasia mutated (ATM) and ATM and Rad-3-related (ATR) proteins are the key kinases that transduce signals in response to various types of DNA damage. Drosophila mei-41 was originally reported as the ATM ortholog. However, sequence analysis reveals that CG6535, a gene predicted by the annotated Drosophila genome, is more closely related to ATM, and mei-41 actually belongs to the ATR subfamily. A related kinase ATX (CG32743 in Drosophila) plays a role in eliminating RNA species containing premature termination codons in C. elegans, whereas in humans it may have a role in the DNA damage response. Two other PIKK family members Drosophila CG2905 and CG5092 are closely related to TRRAP and mTOR, respectively, whereas the catalytic subunit of DNA-dependent protein kinase is not present in Drosophila. Among these kinases, flies with mutations in mei-41 and mTOR have been studied in detail, whereas the functions of CG2905, CG6535, and CG32743 in Drosophila remain unknown (Song, 2004 and references therein).
When a eukaryotic cell is exposed to a genotoxic treatment such as ionizing radiation that induces a DNA double-strand break, the response is led by ATM and ATR, which activate DNA repair and cell-cycle arrest in the damaged cell. ATM and ATR proteins were conserved in all eukaryotes examined. Components of the so-called MR protein complex, Rad50 and the less-well conserved Nbs1, cooperate with ATM/ATR homologs in DNA damage responses, with MR proteins activating ATM/ATR homologs and vice versa (Purdy, 2004 and references therein).
ATM/ATR homologs and MR proteins also respond to the naturally occurring DNA double-strand 'breaks'-- telomeres -- at the ends of linear chromosomes. Human, mouse and yeast cells that are deficient in ATM/ATR homologs have shorter telomeres which often fuse to other telomeres or induced double-strand breaks. Likewise, mutations in MR proteins led to shortened and fused telomeres in yeast. In these organisms, telomere length is maintained by telomerase, a reverse transcriptase which restores the GC-rich repeats at chromosome ends that would otherwise shorten after each round of DNA replication. The ends are further protected by the binding of proteins such as TRF2 homologs, and by DNA secondary structures. Genetic analyses suggest that ATM/ATR and MR proteins function to maintain telomere length or to prevent telomere fusion by telomerase-dependent and telomerase-independent mechanisms. The exact nature of these mechanisms, and their relative contribution to telomere maintenance, needed clarification, and hence, these studies (Purdy, 2004 and references therein).
Six papers have now reported that the Drosophila homologs of ATM (allelic to the previously described gene telomere fusion), Mre11 and Rad50 play a role in telomere maintenance. Mutants defective in each of these proteins show increased fusion of telomeres, with the proportion of mitotic cells showing at least one fusion event ranging from 50% to over 90%. Drosophila telomeres are unusual; they are made up of repeats of Het-A and TERT retrotransposons that are maintained by transposition rather than telomerase activity, and are protected by binding of two proteins, heterochromatin protein 1 (HP1) and HP1, ORC2 associated protein (HOAP). The new studies (Bi, 2004; Ciapponi, 2004; Oikemus, 2004; Silva, 2004; Song, 2004; Gorski, 2004) clearly define telomerase-independent contributions of ATM and MR proteins to telomere maintenance and, specifically, in the prevention of fusions. Furthermore, it now appears that the role of ATM/ATR and MR proteins in telomere protection is universal, and occurs regardless of other mechanisms that lengthen telomeres or protect them. The consequences of the loss of this protection in Drosophila are devastating; fused telomeres seem to induce a breakage-fusion cycle and lead to apoptosis and organismal death. Thus, paradoxically, DNA damage checkpoints that have been shown to promote apoptosis in response to DNA damage are fulfilling an anti-apoptotic role indirectly through telomere protection (Purdy, 2004 and references therein).
A possible mechanism for telomere protection by ATM and MR proteins in Drosophila is suggested by two findings. (1) Het-A sequences are still present at telomeres in mutants, even in those participating in fusion. Although it was difficult to quantify the extent of sequences present, it is clear that total recession of Het-A repeats is not necessary for fusion. (2) Two telomere-associated proteins, HOAP and HP1, are missing or present at reduced levels from telomeres in ATM, Rad50 and Mre11 mutants. The reduction of HOAP is apparent in mitotic cells in Mre11 and Rad50 mutants, while both HOAP and HP1 are reduced on telomeres of polytene chromosomes in ATM, Mre11 and Rad50 mutants. Because HOAP and HP1 are known to prevent telomere fusion in mitotically proliferating cells, reduction of these proteins may explain why telomeres fuse in ATM, Mre11 and Rad50 mutants. The fusions appear to be due, to a significant degree, to end-to-end joining of telomeres. Mutations in Drosophila ligase IV reduced, but did not abolish, telomere fusion in ATM mutants (Bi, 2004), similar to findings in budding yeast and mammals. Since ligase IV homologs are needed to repair a DNA double-stranded break by non-homologous end-joining, the latter may be the mechanism for at least some telomere fusion events (Purdy, 2004 and references therein).
Drosophila Mre11 and Rad50 mutants are defective in ionizing-radiation-induced repair of DNA (Ciapponi, 2004). Mre11 and Rad50 homologs also act in the repair of DNA double-strand breaks in yeast and mammals. Thus, a question posed by previous work is revisited: how can the same proteins act to repair DNA double-strand breaks induced, for example, by radiation, while acting to preserve them stably as telomeres if the breaks are chromosomal termini? This question becomes all the more pressing because internally deleted chromosome ends, such as those generated by mobilizing of a transposable element, can recruit HOAP and HP1 to function stably as telomeres for generations in Drosophila (Fanti, 1998; Cenci, 2003). Assuming that MR proteins are also needed for recruitment of telomeric proteins to internally deleted ends, what is different about those deletions that lead to protein recruitment and telomere formation, while other deletions are repaired through the activity of the same proteins? A related question is how HP1 and HOAP are recruited to telomeres in a MR-dependent manner but not to double-strand breaks caused by ionizing radiation, for instance? Conversely, if repair enzymes are recruited to a DNA double-strand break caused by damaging agents, why are they not recruited to chromosome ends (Purdy, 2004 and references therein)?
In addition to activating DNA repair, DNA double-strand breaks also activate a checkpoint that causes cell-cycle delays. ATM/ATR and MR proteins are needed for this checkpoint in yeast and mammals. The checkpoint role of Drosophila Mre11 and Rad50 remains to be examined, but Drosophila ATM may have a minor role that is apparent at shorter times after irradiation with mutants showing robust regulation of mitosis at later times. The question then is: if double-strand breaks produce cell-cycle delay, why do not telomeres? ATM and Mre11 were recently shown to become co-localized at telomeric foci in senescent human cells. Depletion of ATM caused these cells to re-enter the cell cycle. Thus, in senescent cells, ATM appears to recognize telomeres as DNA breaks and cause cell-cycle arrest. An intriguing possibility, then, is that onset of senescence may simply reflect a switch in which ATM goes from having a telomere-protective role that does not result in cell-cycle arrest, to a DNA-damage checkpoint role that does result in cell-cycle arrest and hence senesence. In any case, addressing how the same set of molecules, ATM and MR proteins, could initiate different sets of outcomes, depending upon where on the chromosome a 'break' forms, should prove to be challenging and rewarding (Purdy, 2004 and references therein).
For information on ATM structure see Harvester: Q13315 - Serine-protein kinase ATM (EC ) (Ataxia telangiectasia mutated) (A-T, mutated).
date revised: 30 September 2004
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.