Skip Navigation


Human Molecular Genetics Advance Access originally published online on March 25, 2004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/10/1081    most recent
ddh122v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (54)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Gueven, N.
Right arrow Articles by Lavin, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gueven, N.
Right arrow Articles by Lavin, M. F.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2004, Vol. 13, No. 10 1081-1093
DOI: 10.1093/hmg/ddh122
Human Molecular Genetics, Vol. 13, No. 10 © Oxford University Press 2004; all rights reserved

Aprataxin, a novel protein that protects against genotoxic stress

Nuri Gueven1,*, Olivier J. Becherel1, Amanda W. Kijas1, Philip Chen1, Orla Howe1, Jeanette H. Rudolph2, Richard Gatti3, Hidetoshi Date4, Osamu Onodera4, Gisela Taucher-Scholz2 and Martin F. Lavin1,5

1The Queensland Institute of Medical Research, Herston, Qld 4029, Australia, 2Gesellschaft für Schwerionenforschung mbH, Planckstr. 1, 64291 Darmstadt, Germany, 3UCLA School of Medicine, Department of Pathology, Los Angeles, CA 90095-1732, USA, 4Department of Neurology, Brain Research Institute, Niigata University, 1 Asahimachi, Niigata 951, Japan and 5Central Clinical School, University of Queensland, Brisbane, Qld 4006, Australia

Received February 17, 2004; Revised March 15, 2004; Accepted March 22, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Ataxia-oculomotor apraxia (AOA1) is a neurological disorder with symptoms that overlap those of ataxia-telangiectasia, a syndrome characterized by abnormal responses to double-strand DNA breaks and genome instability. The gene mutated in AOA1, APTX, is predicted to code for a protein called aprataxin that contains domains of homology with proteins involved in DNA damage signalling and repair. We demonstrate that aprataxin is a nuclear protein, present in both the nucleoplasm and the nucleolus. Mutations in the APTX gene destabilize the aprataxin protein, and fusion constructs of enhanced green fluorescent protein and aprataxin, representing deletions of putative functional domains, generate highly unstable products. Cells from AOA1 patients are characterized by enhanced sensitivity to agents that cause single-strand breaks in DNA but there is no evidence for a gross defect in single-strand break repair. Sensitivity to hydrogen peroxide and the resulting genome instability are corrected by transfection with full-length aprataxin cDNA. We also demonstrate that aprataxin interacts with the repair proteins XRCC1, PARP-1 and p53 and that it co-localizes with XRCC1 along charged particle tracks on chromatin. These results demonstrate that aprataxin influences the cellular response to genotoxic stress very likely by its capacity to interact with a number of proteins involved in DNA repair.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Ataxia-oculomotor apraxia (AOA) was first recognized in 1988 as a distinct human syndrome, characterized by early-onset cerebellar ataxia, oculomotor apraxia, early areflexia and late peripheral neuropathy (1). Ataxia, oculomotor apraxia and cerebral atrophy are major clinical features of AOA patients that are also observed in patients with ataxia-telangiectasia (A-T). A-T is caused by mutations in the ataxia-telangiectasia-mutated (ATM) gene (2,3). The most debilitating feature of this syndrome is the progressive neurodegeneration associated with loss of Purkinje cells and ectopic location of these cells in the molecular layer (4). The loss of Purkinje cells has also been associated with AOA (5). Although AOA shows strong similarities to A-T, extra-neurological features that are usually associated with A-T, such as telangiectasia, immunodeficiency and susceptibility to malignancies, have not been reported for AOA. One of the hallmarks of cells from A-T patients is their extreme sensitivity to ionizing radiation (IR) (6–8), the exact molecular basis of which remains unexplained but arises as a consequence of the failure of mutated ATM protein to recognize and signal breaks in DNA (9). ATM has multiple signalling functions to target DNA repair complexes to the sites of damage as well as to activate cell cycle checkpoints in response to genotoxic stress (9). Double-strand breaks in DNA activate ATM by autophosphorylation, which enables it to phosphorylate multiple substrates (10,11). ATM is intimately linked to the Mre11/Rad50/Nbs1 complex that is responsible for recognition and processing of double-strand breaks in DNA (12,13). Mutations in two members of this complex, Mre11 and Nbs1, also give rise to AT-like syndromes, A-TLD and Nijmegen breakage syndrome, respectively (14,15).

Two recent reports have linked one form of AOA (AOA1) to mutations in the APTX gene which is located on 9p13 (16,17). AOA1 is an autosomal recessive ataxia associated with hypoalbuminemia and hypercholesterolemia (18). This disease has been described primarily in Portuguese and Japanese families but is also found in other countries throughout the world (19). The APTX gene is proposed to code for a 342 amino acid protein named aprataxin. Sequence similarities link aprataxin to the family of histidine triad (HIT) domain proteins. Truncating and missense mutations have been described in APTX and are largely confined to the HIT domain (16,17). Members of the family of HIT proteins can be divided into two main groups: the Hint (histidine triad nucleotide binding)-related proteins, found in all forms of life, and the Fhit (fragile histidine triad)-related proteins, found in animals and fungi. Hint proteins bind nucleotides and display adenosine 5'-monophosphoramidase activity (20), whereas the Fhit proteins possess di-nucleoside-polyphosphate-hydrolase activity (21). In addition, the Fhit protein can act as a tumour suppressor (22). However, the exact mechanism involved remains unclear. Despite the similarity of aprataxin to the subgroup of Hint proteins, nothing is known about its cellular function or its role in preventing neurodegeneration. In addition to its central HIT domain, based on sequence similarities, the aprataxin protein also contains a putative forkhead associated (FHA) domain at its N-terminus (23) and a predicted C-terminal zinc-finger (ZF) domain (16,17). Given the phenotypic similarity between A-T and AOA1, it was of interest to determine whether AOA1 cells, with specific aprataxin mutations, would show increased sensitivity to genotoxic agents similar to that observed in A-T cells. In contrast to A-T cells, AOA1 cells were not hypersensitive to IR but, instead, displayed increased sensitivity to genotoxic agents that induce single-strand breaks in DNA. This sensitivity was corrected by reintroducing a wild-type aprataxin expression construct into cells from AOA1 patients. Consistent with this sensitivity profile we provide evidence that aprataxin is a nuclear protein that is also a component of the DNA single-strand break repair complex with PARP-1, XRCC1 and p53.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Sensitivity of AOA1 cells to genotoxic agents
Because of the overlap of the AOA1 and A-T phenotypes we determined whether AOA1 cells also show hypersensitivity to DNA-damaging agents. The results in Figure 1A demonstrate that three AOA1 cell lines showed no increased sensitivity to IR compared to control cells, whereas an A-T cell line was hypersensitive. In contrast, five AOA1 cell lines showed significantly greater sensitivity to hydrogen peroxide (H2O2) than either a control or an A-T cell line (Fig. 1A). AOA1 cells were also more sensitive to another agent that causes single-strand breaks in DNA, methyl methane sulfonate (MMS). However, there was no increased sensitivity for these cells after exposure to mitomycin C (MMC). As an additional criterion for agent sensitivity, we assayed for induced chromosome aberrations (ICA). Exposure of control cells to H2O2 resulted in an average of two ICAs per metaphase, while the values for three AOA1 cell lines were 4.32, 4.16 and 5.6 (Fig. 1B). These results suggest that aprataxin is involved in the response to or repair of DNA single-strand breaks.




View larger version (108K):
[in this window]
[in a new window]
  		Figure 1. Sensitivity to genotoxic agents and detection of aprataxin in AOA1 cells. (A) Cell survival of control (C3ABR), A-T (AT1ABR) and AOA1 (L136, L938, L939, L990, L991) cell lines to genotoxic agents (ionizing radiation, IR; methyl methane sulfonate, MMS; mitomycin C, MMC; hydrogen peroxide, H2O2). (B) Chromosome aberrations in response to H2O2. After treatment with 2 mM H2O2, 50 metaphases (10 for L136) were scored for each cell line for chromosome aberrations (sb, chromatid breaks; cb, chromosome breaks; Int, interchanges; ICA, induced chromosome aberrations per metaphase). (C) Schematic representation of the predicted functional domains of aprataxin (forkhead associated homology domain, FHA; nuclear localization signal, NL; histidine triad domain, HIT; zinc-finger motif, ZF) and the four aprataxin–GST constructs generated. Three of those (GST–FHA, GST–HIT, GST–ZF) were used to generate polyclonal antibodies (F1, F3 and F4). (D) Aprataxin, XRCC1 and PARP-1 expression in control (C3ABR), A-T (AT25ABR) and AOA1 (L136, L938, L939) cell lines. Recombinant aprataxin (A) (20 ng) expressed in S. cerevisiae was used as a positive control. Expression of PCNA was used as loading control.

 
Expression of aprataxin in AOA1 cells
Previous reports have revealed that there are two major mRNA species that are potentially capable of encoding short and long forms of aprataxin (16,17). To study the expression at the protein level and determine the status of aprataxin in AOA1 cells we generated antibodies against different regions of the predicted protein (Fig. 1C). Polyclonal antibodies were prepared in sheep using glutathione S-transferase (GST)–fusion proteins corresponding to the FHA, HIT and ZF homology domains. As is evident from Figure 1D (lane 2), a single band of ~45 kDa was detected in normal control cells (C3ABR) after immunoblotting with the anti-ZF antibody F4. This size is in general agreement with that derived from the cDNA nucleotide sequence which predicts a molecular weight of ~40 kDa. In addition, it migrates at approximately the same position as a human recombinant form of the protein (Fig. 1D, lane 1). Under these conditions no aprataxin was detected in three AOA1 cell lines (lanes 3–5): L136 (689insT/840delT), L938 (P206L/P206L) and L939 (P206L/V263G). Aprataxin was also not detected in two other AOA1 cell lines: L990 and L991 (data not shown). However, it was possible to detect low amounts of aprataxin in AOA1 cells carrying missense mutations in the gene when the protein was first immunoprecipitated with anti-aprataxin antibody (data not shown). Normal levels of aprataxin were detected in an A-T cell line (AT25ABR). It is clear from these data that missense mutations in the APTX gene are sufficient to destabilize the aprataxin protein (Fig. 1D). Since loss of aprataxin gave rise to increased sensitivity to agents that cause single-strand breaks in DNA, we determined whether this would alter the expression of other proteins involved in single-strand break repair. The results in Figure 1D show that two such proteins, XRCC1 and PARP-1, are present at normal levels in AOA1 cells.

Subcellular localization of aprataxin
In order to determine the cellular localization of aprataxin, we employed immunostaining, enhanced green fluorescent protein (EGFP)-tagging and subcellular fractionation. Immunofluorescent antibody staining (F4 antibody) of normal human foreskin fibroblasts (NFF) revealed that aprataxin is a nuclear protein, distributed both uniformly throughout the nucleus and localized to subnuclear structures of varying size (Fig. 2A, left panel). On the contrary, in fibroblasts derived from an AOA1 patient (FG4003) with two truncating mutations in APTX (689insT/840delT), only very low levels of background staining were detected (Fig. 2A, right panel). To confirm this distribution we transfected NFF cells with an EGFP–aprataxin construct and monitored the transient expression of the fusion protein (Fig. 2B, left panel). The data confirm that aprataxin is distributed throughout the nucleus and to subnuclear structures as observed with antibody staining in NFF cells. When the EGFP–aprataxin construct was transfected into HeLa cells a similar pattern of distribution was observed (Fig. 2B, right panel). Since the size and number of the larger subnuclear structures were reminiscent of nucleoli, we looked for co-localization of aprataxin with the nucleolus-specific marker protein, nucleolin, in stably transfected HeLa cells using anti-nucleolin antibody staining (Fig. 2C). When this fluorescence pattern was merged with the EGFP–aprataxin signal, co-localization of these proteins was observed. In addition to the uniform nuclear and nucleolar distribution, small intense regions of EGFP–aprataxin were also present in the nucleus (Fig. 2C, arrows). These were not localized to nucleoli since they did not co-localize with nucleolin. To confirm the observed nuclear and nucleolar localization, cells were subfractionated and immunoblotted with anti-aprataxin antibody. As expected, aprataxin was found both in the nucleoplasm and in nucleoli (Fig. 2D). Detection of nucleolin confirmed the integrity of the fractions.



View larger version (42K):
[in this window]
[in a new window]
  		Figure 2. Cellular localization of aprataxin. (A) Immunostaining of NFF. Examples of asynchronous NFF and fibroblasts from an AOA1 patient (FG4003) stained for aprataxin with antibody F4. (B) EGFP–aprataxin in NFF and HeLa cells. NFF and HeLa cells were transiently transfected with an EGFP–aprataxin expression construct and visualized after 48 h without fixation. (C) Co-localization with nucleolin. HeLa cells carrying a stable expression construct for EGFP–aprataxin were fixed and additionally stained for nucleolin. Arrow indicates small intense non-nucleolar region of EGFP–aprataxin fluorescence. (D) Cellular fractionation. Different fractions of control cells (C3ABR) were used to detect aprataxin by western blotting (A, 20 ng recombinant aprataxin, TE, total cell extract; C, cytoplasmic fraction; NP, nucleoplasmic fraction; NO, nucleolar fraction). Detection of nucleolin was used to determine the purity of the fractions.

 
Localization and expression of mutant forms of aprataxin
Since we demonstrated that missense mutations destabilize aprataxin, it was of interest to examine whether deletions of any of the putative domains in the molecule would also destabilize the protein and/or subcellular distribution. A series of EGFP–aprataxin constructs were generated (Fig. 3). As predicted from data with AOA1 cell lines (Fig. 1D), the V263G mutation in the HIT domain destabilized the protein (Fig. 3B). For this construct, it was only possible to detect residual perinuclear/cytoplasmic EGFP-fluorescence using a 10-fold longer exposure time and digital enhancement, compared to wild-type fluorescence, which was localized to the nucleus (Fig. 3A). A series of other constructs deleting the FHA, NL, HIT and ZF domains led to a similar reduction in the amount of protein as well as abnormal distribution (Fig. 3C–G).



View larger version (29K):
[in this window]
[in a new window]
  		Figure 3. Localization and expression of mutated EGFP–aprataxin. HeLa cells were transiently transfected with expression constructs coding for wild-type and mutant EGFP–aprataxin. EGFP–aprataxin was detected 48 h after transfection (A) wild-type; (B) V263G missense mutation; (C) aa 25–342; (D) aa 99–342; (E) aa 173–342; (F) aa 252–342; (G) aa 2–300. Exposure time for all mutant contructs was 10 times longer than for the wild-type aprataxin construct due to low fluorescence signals.

 
Localization of aprataxin in response to genotoxic agents
A typical feature of proteins involved in DNA repair is their re-localization to sites of damage (24). AOA1 cells are hypersensitive to H2O2 and since this agent gives rise to distinct XRCC1 foci (25) we looked for possible co-localization between a portion of the aprataxin and XRCC1. The results in Figure 4A, upper panel demonstrate the presence of multiple H2O2-induced XRCC1 foci in HeLa cells. There was no evidence of such foci for EGFP–aprataxin, but it is clear that both proteins are also uniformly distributed throughout the nucleus under these conditions. Exposure of cells to H2O2 failed to induce any PARP-1 foci (Fig. 4A, lower panel). We also showed that other genotoxic agents, IR, MMS and MMC, failed to alter the uniform nuclear distribution of aprataxin or the extent of nucleolar fluorescence (Fig. 4B).




View larger version (153K):
[in this window]
[in a new window]
  		Figure 4. Localization of aprataxin in response to genotoxic agents. (A) HeLa cells expressing EGFP–aprataxin (green) were treated with H2O2 for 30 min before fixation and immunostaining for XRCC1 (red, upper panels) and PARP-1 (red, lower panels). (B) HeLa cells expressing EGFP–aprataxin were untreated (–) or treated (+) with 10 Gy of IR, 1 µg/ml MMS and MMC and 10 mM H2O2. Cells were analyzed 30 min after treatment.

 
Involvement of aprataxin in XRCC1 foci formation
Since AOA1 cells show increased sensitivity to H2O2 and since treatment with this agent normally leads to formation of XRCC1 foci, we determined whether the loss of aprataxin in AOA1 cells would compromise the induction of XRCC1 foci. No foci were observed in untreated control fibroblasts, but at 10 min after exposure to H2O2, multiple, intensely labelled nuclear XRCC1 foci were observed in ~80% of those cells (Fig. 5). These foci decreased in number with time and by 60 min had formed larger aggregates. In contrast, multiple XRCC1 foci of more irregular shape and lower intensity were constitutively present in untreated AOA1 fibroblasts (Fig. 5), and this pattern of labelling was largely unchanged post-treatment. None of the small intensely stained foci was evident.



View larger version (71K):
[in this window]
[in a new window]
  		Figure 5. Induction of XRCC1 foci in control and AOA1 cells. NFF and AOA1 fibroblasts (FG4003) were immunostained for XRCC1 in response to H2O2 treatment (10 mM for 10 min). DNA damage-induced foci were visualized at time points after the treatment period as indicated.

 
Complementation of H2O2 sensitivity in AOA1 cell lines
To establish the importance of aprataxin in the cellular resistance to genotoxic agents, we transfected AOA1 cells (L938, L939) with full-length aprataxin. Transient transfections were carried out with the EGFP–aprataxin expression vector (A) and the empty vector as a control (V). Exposure of L938 AOA1 cells, transfected with EGFP–aprataxin, to H2O2 revealed a pattern of survival similar to that observed in control cells (Fig. 6A). Correction of H2O2 sensitivity was also observed in a second AOA1 cell line (L939) transfected with EGFP–aprataxin. AOA1 cells transfected with empty vector remained hypersensitive to H2O2. Complementation of H2O2 sensitivity in AOA1 cells by full-length aprataxin was also determined by correction of induced chromosome aberrations. The results in Figure 6B show that AOA1 cells transfected with full-length EGFP–aprataxin (L939/A) display similar numbers of chromosome aberrations when compared with H2O2-treated control cells (C1ABR). However, transfection of L939 AOA1 cells with an expression construct that encodes EGFP–aprataxin containing a point mutation within the HIT domain (V263G), previously described in an AOA1 patient (16), failed to normalize the number of H2O2-induced chromosome aberrations. Introduction of the EGFP–aprataxin construct into control cells did not alter the number of induced aberrations (data not shown).



View larger version (27K):
[in this window]
[in a new window]
  		Figure 6. Complementation of H2O2 sensitivity. (A) Control (C3ABR) and AOA1 cells (L938, L939) were transiently transfected with full-length aprataxin–EGFP expression vector (A; closed symbols) or empty vector (V; open symbols) and survival in response to H2O2 was determined. (B) For complementation of H2O2-induced chromosome aberrations, AOA1 cells (L939) were transiently transfected with either full-length aprataxin–EGFP expression vector (L939/A) or with the same construct containing a missense mutation (L939/V263G) within the HIT domain of aprataxin. Untransfected control (C1ABR) and AOA1 cells (L939) were used as controls.

 
Interaction of aprataxin with DNA repair proteins and mapping of interaction domains
While it was not possible to localize aprataxin to subnuclear structures with XRCC1 and PARP-1 it was still possible that these proteins interacted, given their uniform nuclear distribution and the sensitivity of AOA1 cell lines to agents that cause single-strand breaks in DNA. In addition, Caldecott (23) has reported an interaction between aprataxin and XRCC1 using the yeast two-hybrid system. Constitutive association was evident between aprataxin and XRCC1 when control extracts were immunoprecipitated with an anti-aprataxin antibody and subsequently immunoblotted with anti-XRCC1 (Fig. 7A, left panel). Exposure of cells to 10 mM H2O2 did not alter this association (Fig. 7B). This interaction was also seen in A-T (AT3ABR) cells. Co-immunoprecipitation also revealed that aprataxin was bound to both p53 and PARP-1 (Fig. 7A). Reverse co-immunoprecipitation using anti-XRCC1 and PARP-1 antibodies confirmed the association with aprataxin (Fig. 7A), and we also detected p53 in these co-immunoprecipitates. The H2O2-induced increase in p53 bound to aprataxin (Fig. 7B) is due to DNA-damage-induced p53 stabilization. As observed previously XRCC1 immunoprecipitates also contained PARP-1 (26). Exposure of cells to IR also failed to increase the extent of binding between aprataxin and the other DNA repair proteins (data not shown). None of the proteins was detected in immunoprecipitates with non-specific antisera. Because of the overlap in phenotype between AOA1 and A-T, we also screened for binding of aprataxin to ATM, the protein defective in A-T, but were unable to show any interaction using both co-immunoprecipitation and pull-down assays (data not shown). Since there are at least three putative functional domains in aprataxin, we determined, using GST affinity columns, which of these domains was responsible for binding to the repair proteins. GST–fusion proteins were prepared corresponding to the FHA domain, nuclear localization signal (NL), HIT and ZF homology domains (Fig. 1C). These were covalently bound to sepharose beads and protein binding was determined with total cell extract. The results in Figure 7C show that XRCC1 binds only to the FHA domain of aprataxin. Interestingly, PARP-1 bound to the FHA, NL and HIT domains, and p53 bound to both the FHA and HIT domains. Since aprataxin co-localizes with nucleolin in the nucleoli and since nucleolin is implicated in the genotoxic stress response (27), we determined whether nucleolin also interacted with aprataxin. The results in Figure 7C show that nucleolin bound most strongly to the FHA domain.



View larger version (83K):
[in this window]
[in a new window]
  		Figure 7. Interaction of aprataxin with single-strand break repair proteins. (A) Co-immunprecipitation of aprataxin with PARP-1, p53 and XRCC1 in control (C; C3ABR) and A-T (AT; AT3ABR) cells [10 ng of recombinant aprataxin (A) and 100 µg total cell extract (TE) were used as positive controls]. (B) Lysates from untreated (–) or H2O2 treated (+) control (C3ABR) and A-T cells (AT25ABR) were used to co-immunoprecipitate aprataxin with PARP-1, p53 and XRCC1. (C) Protein extracts from control cells (C3ABR) were incubated with either beads coupled to GST alone or GST–aprataxin fragments as indicated. Proteins eluted from the GST beads were separated by SDS–PAGE and nucleolin (Nucl). XRCC1, PARP-1 and p53 were detected by the appropriate antibodies.

 
Repair of breaks in DNA
We demonstrated that AOA1 cells are sensitive to agents that cause single-strand breaks in DNA and since we have also shown that aprataxin interacts with the repair protein XRCC1, it was possible that the increased sensitivity in these cells was due to a defect in single-strand break repair. We tested this possibility using exposure to H2O2, which gives rise primarily to single-strand breaks in DNA at concentrations >10 mM (28). Single-strand break repair was measured using analysis of intracellular NAD(P)H levels with a water soluble tetrazolium salt method, which represents a reliable method to monitor imbalance of break repair in base excision repair deficient cells (29). When cells were exposed to H2O2 a time-dependent depletion of NAD(P)H was observed and this occurred to approximately the same extent in control and AOA1 (L939) cells, indicating that there was no gross imbalance of single-strand break repair in the AOA1 cells (Fig. 8).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 8. Single-strand break repair of AOA1 cells. Depletion of intracellular NAD(P)H was measured in control (JHP) and AOA1 (L939) cells in response to H2O2 treatment at time points as indicated. Error bars represent standard deviation of six samples from one typical experiment.

 
Dynamics of aprataxin binding to chromatin
Biological imaging of charged particle tracks represents a new approach to investigating the association of repair proteins with chromatin (30,31). In this method a monolayer of cells is exposed to heavy ions where individual particle traversals can be visualized in single cells when a small angle (<5°) is utilized between the beam direction and the monolayer (30). Under these conditions the track structure leads to clustering of damage which includes single- and double-strand breaks and base damage (32). Thus this approach has the potential to visualize events associated with single-strand break repair. Based on data described above, our prediction was that aprataxin would co-localize with XRCC1 at the break sites. Exposure of a monolayer of HeLa cells expressing EGFP–aprataxin to uranium ions gave rise to EGFP–aprataxin aggregates distributed along the particle trajectory as discrete tracks (Fig. 9). It seemed likely that these aggregates localized to sites of DNA damage. In order to confirm this, we investigated whether XRCC1 was distributed in a similar fashion using immunostaining. Again, in this case, discrete tracks of XRCC1 aggregates were distributed across the nucleus. Merging the fluorescent images demonstrated that aprataxin and XRCC1 showed complete co-localization across the length of the tracks, indicating that they were both localizing to the same sites of DNA damage. We also demonstrated that aprataxin was tightly associated with chromatin in response to oxidative damage since these tracks were resistant to detergent extraction (data not shown).



View larger version (71K):
[in this window]
[in a new window]
  		Figure 9. Co-localization of aprataxin and XRCC1 charged particle tracks in chromatin. Co-localization of aprataxin and XRCC1 along heavy ion-induced tracks. HeLa cells expressing EGFP–aprataxin were unirradiated (–) or irradiated (+) with uranium ions at low angle (<5°) to generate particle track traversals through the nucleus. After fixation, XRCC1 was visualized by immunostaining. Co-localization of EGFP–aprataxin (green) and XRCC1 (red) is indicated by yellow colour in the merged image.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
The gene defective in the human genetic disorder AOA1 was predicted to code for a 342 amino acid protein, aprataxin (16,17). We have demonstrated here that aprataxin is predominantly expressed as a 45 kDa nuclear protein. Under these conditions it was not possible to detect a short form of the protein which is also predicted from the open reading frame (16). Immunoblotting failed to reveal the presence of the long form of aprataxin in extracts from five AOA1 cell lines but protein was detected in the cell lines carrying aprataxin missense mutations after immunoprecipitation. These data demonstrate that in the case of L938 (homozygous for P206L) and L939 (compound heterozygote for P206L/V263G) the presence of missense mutations destabilize the protein. This was further substantiated by the generation of an EGFP-tagged V263G aprataxin mutant that was expressed at almost undetectable levels in HeLa cells. The deletion of the various putative domains for the molecule also resulted in its destabilization. These results highlight the difficulty of expressing mutant forms of aprataxin for functional studies.

Use of immunofluorescence, EGFP-tagging, cellular fractionation and immunoblotting demonstrated that aprataxin was present both in the nucleoplasm and in the nucleolus. In the nucleoplasm, aprataxin was uniformly distributed with the exception of intense labelling in small discrete foci. These are not fragments of nucleoli since nucleolin did not localize in these foci (Fig. 2C). Nucleoli are multifunctional subnuclear organelles in which ribosome biogenesis and RNA processing take place (33). Other functions associated with these organelles include tumour suppression, cell cycle control, viral replication and DNA repair (33,34). Immunoblotting suggests that as much as 50% of the cellular aprataxin is localized to nucleoli, indicating a greater concentration per unit volume in this organelle than throughout the nucleoplasm as a whole. This is supported by antibody staining and EGFP fluorescence. Under conditions of asynchronous cell growth the relative amounts of nucleolar and nucleoplasmic aprataxin appear to be stable, and this distribution is not influenced by exposing cells to DNA damaging agents.

Early-onset ataxia with ocular motor apraxia is common to the autosomal recessive neurodegenerative disorders AOA1 and A-T (4,35). The gene product defective in A-T cells, ATM, recognizes double-strand breaks in DNA and signals this damage to cell cycle checkpoints (2,9). In the absence of ATM, a portion of induced double-strand breaks remain unrepaired which may contribute to the hypersensitivity to radiation that is characteristic of A-T cells (2,8). While AOA1 cells are not sensitive to radiation they show increased sensitivity to MMS and H2O2, agents that cause single-strand breaks in DNA (36,37). The different effects of H2O2 and radiation on AOA1 cell survival can be explained by the observation that H2O2 only gives rise to single-strand DNA breaks at concentrations up to 10 mM while radiation gives a mixture of lesions at any dose (28). Furthermore, the exact nature of the lesion may vary between the two treatments. In addition it is evident that H2O2 causes not only DNA damage but also membrane and mitochondrial damage which add further complexity to the outcome (38). It is also clear that the radiosensitive A-T cells show a normal response to H2O2 as has been reported previously (39).

The sensitivity to agents that cause single-strand breaks in DNA is a direct result of the loss of functional aprataxin since transfection of AOA1 cells with full-length aprataxin cDNA reversed this hypersensitivity to H2O2 and also decreased the genome instability observed in these cells after treatment with the same agent. Further evidence for a role for aprataxin in some aspect of DNA strand break repair is provided by its interaction with two proteins, XRCC1 and PARP-1, both of which have been shown to play central roles in repair of breaks in DNA (40–42). Interaction with XRCC1 was demonstrated by co-immunoprecipitation and GST pull-down with an N-terminal region containing the putative FHA domain of aprataxin. This domain is a phosphopeptide-binding module found in a range of proteins involved in signal transduction, RNA binding and DNA repair (43). We also showed that PARP-1 binds to a region of aprataxin extending from the FHA to the HIT domains. Since PARP-1, XRCC1, DNA polymerase ß and DNA ligase III{alpha} all form part of a complex involved in base excision repair (26,41) it suggests that aprataxin may also have a role in this repair process. However, the data described here provide no evidence for a gross defect in the rate of rejoining of DNA single-strand breaks in AOA1 cells. Nevertheless, it is clear that aprataxin accumulates along the path of heavy ion trajectories in the nuclei of irradiated cells, and the pattern of clustering coincides exactly with that of XRCC1. It is not possible to define the exact lesion recognized but track structure causes damage clusters that include single- and double-strand breaks (32). While XRCC1 is involved primarily in single-strand break repair there is evidence for a defect in rejoining of radiation-induced DNA double-strand breaks in the hamster cell lines EM9 and EM-C11, mutated in XRCC1 (44). It appears that mutation in XRCC1 leads to DNA ligase III deficiency. In the present case, loss of aprataxin does not destabilize XRCC1 or PARP-1. In addition another partner protein of XRCC1, PARP-1, binds to both single- and double-strand breaks and when disrupted in a defective Ku80 background leads to lethality (23). Thus it is possible that the aprataxin/XRCC1 association on chromatin is at single- and/or double-strand breaks in DNA.

The agent sensitivity phenotype of AOA1 and XRCC1-deficient cells is similar but not identical. XRCC1-deficient cells show strongest sensitivity to EMS and MMS, moderate sensitivity to H2O2 and MMC and weak sensitivity to IR and UV (23,45). Here we demonstrated that AOA1 cells were sensitive to MMS and H2O2 but not to radiation or MMC. These observations might well explain the difference in single-strand break rejoining in the two mutant types. Failure to observe increased sensitivity to IR and MMC suggests that aprataxin is not involved in recognizing double-strand breaks in DNA or breaks arising due to failure to resolve blocked replication forks. The weak sensitivity of XRCC1 mutants to MMC could be due to some partial involvement of this protein in double-strand break repair for which there is some evidence (44). A more specialized role of aprataxin in break recognition or processing or in relation to emphasis at specific stages of the cell cycle might not be manifested as a gross defect in rate of repair.

A characteristic of proteins recognizing damage in DNA is that they form discrete foci at either short and/or long times after DNA damage (46,47). It seems likely that these foci represent aggregates of proteins deposited to facilitate DNA repair or assist in the signalling of the DNA damage to cell cycle checkpoints (24). Exposure of CHO cells to H2O2 causes the appearance of XRCC1 foci which peak at ~30 min post-treatment (25). We have observed multiple small intensely labelled XRCC1 foci in 80% of control cells almost immediately after exposure to H2O2 and these appear to aggregate and form larger foci with time. In the case of AOA1 cells, less well-defined foci were constitutively present and showed little alteration after H2O2 treatment. Thus, in the absence of aprataxin there is an abnormal distribution of XRCC1 in the nucleus and the normal response to H2O2 is attenuated. Co-localization of aprataxin with XRCC1 at sites of DNA damage (tracks) after radiation exposure would appear to be in conflict with a failure to observe co-localization of aprataxin and XRCC1 in H2O2-induced foci. This might be explained by complex forms of DNA damage inflicted by high-LET (linear energy transfer) radiation. By necessity, high-LET radiation was used here to visualize track structure. In this context aprataxin and XRCC1 may recognize a specific form of damage only possible to detect with high-energy radiation. On the other hand, the oxidative damage caused by H2O2 would be expected to be recognized by XRCC1 and if only a subset of this damage requires the involvement of aprataxin, this protein would not be detected as co-localizing with XRCC1. This is supported by single-strand break repair data which appears to be grossly normal for AOA1 cells.

Based on the presence of several putative functional domains [FHA, polynucleotide kinase 3'-phosphatase (PNKP), HIT and ZF] in aprataxin, it may function in different aspects of single-strand break repair. As demonstrated, it binds to XRCC1 through its FHA domain which may tether it to the repair complex. This region of the molecule also shares homology with PNKP which again implicates aprataxin in single-strand break repair. The ZF domain might enable aprataxin to bind to a specific DNA structure (e.g. ends of the break) as part of the recognition of lesions in DNA. Aprataxin is also included in the Hint family of proteins (48). It possesses a minimum Hint domain which may confer adenosine 5'-monophosphoramidase (AMP) activity. Hint proteins hydrolyze the AMP–NH2 bond and AMP linked to a lysine side chain (49). It is conceivable that aprataxin, while not essential for the release of AMP from the ligase III–AMP complex during the process of sealing of breaks in DNA plays a role in enhancing the process. The appearance of p53 in co-immunoprecipitates with aprataxin, PARP-1 and XRCC1 is in keeping with a role for this molecule in DNA repair (50). A requirement for p53 in nucleotide excision repair has previously been demonstrated (51,52). Furthermore, there is also evidence for the direct involvement of p53 in base excision repair. Offer et al. (53) and Zhou et al. (54) have shown that p53 stimulated base excision repair in a reconstituted assay, possibly by binding to the APE-1 endonuclease. Finally, co-localization of nucleolin and aprataxin to the nucleoli, together with data revealing binding between nucleolin and the FHA domain of aprataxin, provide additional evidence for a role for aprataxin in the stress response, since Yang et al. (27) have demonstrated that activation of the RNA-binding properties of nucleolin is part of the cellular response to genotoxic stress. It is also intriguing to note that activation of nucleolin is achieved through phosphorylation by SAPK p38 which is consistent with binding of nucleolin to the FHA domain of aprataxin.

The similarity in clinical phenotype between AOA1 and A-T suggests that the two gene products have at least some degree of mechanistic overlap. It is still not clear what is responsible for the neurodegenerative phenotype in A-T but failure to completely repair double-strand breaks in DNA, aberrant apoptosis and oxidative stress resulting from loss of ATM may all contribute (2,55,56). It is clear that ATM is not primarily a repair protein since the vast majority of DNA breaks are repaired in A-T cells after radiation exposure (57,58). ATM signals these breaks to the cell cycle machinery to ensure stability of the genome (2). Our data suggest that aprataxin is also not directly involved in the rate of single-strand break rejoining but its association with proteins involved in DNA repair suggests that it may have a recognition or more subtle processing role in the cellular response to single-strand breaks. The defect in response in AOA1 cells to DNA single-strand breaks partially addresses that issue since there is evidence for association between repair genes and neurodegeneration (59–61). However, since the clinical defect in AOA1 is confined largely to the nervous system, it requires the aprataxin response to DNA strand breaks is more critical to this tissue. Any accompanying oxidative stress, confined largely to the brain as observed with A-T, would help to explain the specificity of the effect in AOA1. Failure to observe increased cancer incidence in AOA1 patients might be explained by differences in cell cycle regulation compared with that in other syndromes with cerebellar ataxia (e.g. A-T) or might be explained by the nature of the repair defect, i.e. single-strand breaks in AOA1 versus double-strand breaks in A-T. The recent description of senataxin, mutated in patients with AOA2, adds further support for an involvement of a DNA repair defect in the neurodegenerative phenotype (61). Senataxin is homologous to the yeast sen1p proteins that possess RNA helicase activity, which could also implicate this protein in DNA repair.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Cell culture and treatment
Lymphoblastoid cells lines (LCL) from control (C3ABR, C1ABR), A-T (AT1ABR, AT25ABR) and AOA1 (L136, L938, L939, L990, L991) patients were cultured in RPMI 1640 medium (GIBCO BRL) containing 10% fetal calf serum (FCS) (JRH Biosciences), 2 mM L-glutamine (Life Technologies), 100 U/ml penicillin (GIBCO BRL) and 100 U/ml streptomycin (GIBCO BRL) and maintained in a humidified incubator at 37°C/5% CO2. NFF, fibroblasts from an AOA1 patient (FG4003) and cervical adenocarcinoma cells (HeLa) were cultured under identical conditions. For all experiments LCL were used at a density of 1x106 cells/ml or at 75% confluency for adherent cell lines. Genotoxic agents MMS (Sigma) and MMC (Boehringer) were added to the cell culture medium using concentrations and time periods as indicated. Cells were treated with H2O2 (Sigma) for 30 min for survival and 10 min for cell staining before culture medium was replaced with fresh medium. All irradiations were performed at room temperature using a 137Cs source (Gammacell 40 Exactor, MDS Nordion, dose rate 1.1 Gy/min).

Cell survival and chromosome aberrations in response to genotoxic agents
Cell survival and chromosome aberrations were measured as described previously (62). For complementation of H2O2 sensitivity, a PCR product corresponding to full-length human aprataxin was cloned into the EGFP expression vector EGFP-C2 (Invitrogen). Control and AOA1 cell lines were transiently transfected with this construct and left for 24 h. Subsequently, cells were treated with H2O2 concentrations as indicated and the surviving fraction was measured after 48 h as described previously (63). For complementation of H2O2-induced chromosome aberrations, AOA1 cells were transiently transfected with full-length EGFP–aprataxin. A previously described AOA1 mutant, V263G, (16) was introduced into full-length EGFP–aprataxin expression vector as described previously (64). This construct was used as a negative control. EGFP–aprataxin expression was monitored by fluorescence microscopy.

Purification of recombinant human aprataxin
Full-length human APTX was cloned into a modified pCYB2 (New England Biolabs) plasmid, pMH1, containing a self cleaving 3'chitin-binding domain and a galactose-inducible promoter for overexpression in yeast (65). Specifically, the SalI–SmaI fragment of the pMH1 construct was replaced with SalI–human APTX coding region (long form). This construct, pMHA, was transformed into Saccharomyces cerevisiae, BJ2168, and overexpression induced using 2% galactose. Human aprataxin was affinity purified using a chitin resin (New England Biolabs) according to Hall and Kunkel (65).

Anti-aprataxin antibodies
Polyclonal antibodies against different domains of human aprataxin were generated in sheep using methods described previously (11,66). Briefly, GST–aprataxin fusion proteins corresponding to amino acids 1–110 (F1), 175–272 (F3) and 266–342 (F4) were expressed in Escherichia coli and used for immunization. The corresponding sheep antisera were pre-cleared on GST columns, before antibodies were affinity purified using the appropriate GST–aprataxin columns. Antibodies were tested using recombinant full-length human aprataxin protein.

Aprataxin–GST fusion proteins and binding studies
Four overlapping aprataxin–GST fusion proteins (GST–FHA, aa 1–110; GST–NL, aa 98–176; GST–HIT, aa 175–272; GST–ZnF, aa 266–342) and GST alone were generated by cloning the corresponding PCR products into pGEX5X-1 (Amersham). Recombinant proteins were induced and purified as recommended by the manufacturer (Amersham). For GST pull-down studies the fusion proteins were cross-linked to glutathione–sepharose beads as previously described (66). Total cell extracts were prepared by lysing 3x109 control cells (C3ABR) in 100 ml of lysis buffer L [50 mM Tris–HCl pH 7.4, 150 mM NaCl, 1 mM ethylene diamine tetra-acetic acid (EDTA), 1% NP-40, 5 mM dithiothreitol supplemented with protease and phosphatase inhibitors]. Cleared lysate was adjusted to 1 mg/ml and loaded onto the various columns simultaneously (GST, GST–FHA, GST–NL, GST–HIT and GST–ZF). Binding was performed overnight at 4°C. Columns were subsequently washed with 10 column volumes of phosphate-buffered saline (PBS) and subsequently 10 volumes of buffer PBS/0.15 M NaCl. Proteins were eluted with five volumes of buffer E (0.1 M glycine pH 2.5, 2 M NaCl, 1 mM phenylmethylsulfonyl fluoride). Eluted fractions were pooled, concentrated by centrifugual filtration and then separated by 10% SDS–PAGE. Proteins were transferred onto nitrocellulose membranes and detected using rabbit polyclonal anti-XRCC1 (AHP428, Serotec, 1/5000), mouse monoclonal anti-PARP-1 (MCA1522T, Serotec, 1/1000), mouse monoclonal anti-p53 (DO-1, BD PharMingen, 1/1000) and mouse monoclonal anti-nucleolin antibodies (M019-3, MBL, 1/1000) and the species-specific horseradish peroxidase conjugated secondary antibody from Chemicon (1/10 000).

Cellular fractionation, immunoprecipitation and western blotting
Cells were separated into cytoplasmic, nucleoplasmic and nucleolar fractions as previously described (67). For immunoprecipitations, cells were washed in PBS before resuspension in lysis buffer [50 mM Tris–HCl pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40, supplemented with protease and phosphatase inhibitors and Benzonase (Novagen)] for 1 h at 4°C. Insoluble components were removed by centrifugation (16 000 g for 10 min). For immunoprecipitation, 1 mg of total cell extract was incubated for 4 h at 4°C with 1 µg of antibody and 40 µl of protein G–Sepharose beads (Amersham). After washing the beads three times with lysis buffer and once with high- stringency buffer (50 mM Tris pH 7.5, 0.5 M NaCl) they were resuspended in 20 µl of sample loading buffer before separation of the proteins by SDS–PAGE. Proteins were transferred to nitrocellulose and immunoblots were performed using the appropriate antibody described above.

EGFP–aprataxin and immunostaining
APTX cDNA corresponding to full length (aa 2–342) or parts of aprataxin (aa 25–342, aa 99–342, aa 173–342, aa 252–342, aa 2–300; Fig. 3) were cloned into pEGFP-C2 (Clontech) via the XhoI and BamHI sites. HeLa cells were transfected with these constructs by electroporation (200 V, 10 ms) and stable cell lines selected using G418 and FACS-enrichment. For immunostaining, cells were grown on coverslips for 48 h before treatment with genotoxic agents for time intervals as indicated. Subsequently, coverslips were washed with PBS/5 mM MgCl2, fixed in 4% PFA/PBS/5 mM MgCl2 at room temperature for 10 min and permeabilized in 0.1% Triton X-100/PBS/5 mM MgCl2 for 10 min. Non-specific binding was blocked by incubating the coverslips in 5% FCS/PBS/5 mM MgCl2 for 1 h at 37°C. Coverslips were covered with various antibody solutions (all 1 : 200 in 5% FCS/PBS/5 mM MgCl2) overnight at 4°C before detection with species-specific AlexaFluor488-conjugate (Molecular Probes). Images were captured using a digital camera (Zeiss AxioCam MRc) attached to a fluorescence microscope (Zeiss Axiovert).

Single-strand break repair assay
Single-strand break repair was measured using a recently described method relying on analysis of intracellular NAD(P)H levels after treatment with H2O2 (29). The reduction of NAD(P)H to a brown coloured formazan dye was monitored using a water-soluble tetrazolium salt (CellTiter 96AQueous, Promega). Briefly, 96-well plates were seeded with control C3ABR and AOA1 (L939) cells (1x105 cells/well) and cultured in 100 µl of medium containing 10% FCS and antibiotics as described above. Cells were untreated or treated with 10 mM H2O2 as described above. Cells were then washed and resuspended in medium containing 20 µl of CellTiter 96 AQueous solution [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium salt (MTS) and 1-methoxy-5-methylphenazinium methylsulfate (PMS)]. The amount of formazan dye produced by the living cells in the medium was determined every 30 min by measuring absorbance at 490 nm in a 96-well plate spectrophotometer with 650 nm as a reference. A blank was prepared with only medium and CellTiter 96AQueous (MTS/PMS) solution. The depletion in NAD(P)H was assessed by comparing the absorbance of H2O2-treated cells with PBS-treated cells over time. Error bars represent the standard deviation of 12 parallel measurements of one typical experiment.

Generation of tracks of ion-induced DNA damage
HeLa cells were irradiated with low-energy ions entering the cell monolayer at a small angle (<5°) as described previously (30). The uranium ions used (3.5 MeV/u, 1x107 Pxcm–2 corresponding to an average of four traversals per nucleus) were generated at the UNILAC facility at the Gesellschaft für Schwerionenforschung (Darmstadt, Germany). Cells were grown for 72 h in 35 mm Petri dishes. Prior to irradiation the outer rim of each dish was removed and cells were irradiated for 2 min. After a subsequent incubation for 3 min in conditioned medium at 37°C the cells were fixed and immunostained as described above. Imaging was performed using confocal laser microscopy (LEICA TCS) as described previously (30).


   ACKNOWLEDGEMENTS
 
We thank Peter Parsons for the NFF cell strain, Aine Farrell for assistance with cell culture and Molly Cook and Thomas Kunkel for providing PMH1 and BJ2168. This work was supported by grants from the Queensland Cancer Fund (QCF), the Australian National Health and Medical Research Council (NHMRC) and the USPHS NS35322.


   FOOTNOTES
 
* To whom correspondence should be addressed at: Radiation Biology and Oncology, The Queensland Institute of Medical Research, 300 Herston Rd, Herston, Qld 4029, Australia. Tel: +61 733620341; Fax: +61 733620106; Email: nurig{at}qimr.edu.au


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Aicardi, J., Barbosa, C., Andermann, E., Morcas, R., Ghanem, Q., Fukuyama, Y., Awaya, Y. and Moe, P. (1988) Ataxia-ocular motor apraxia: a syndrome mimicking ataxia-telangiectasia. Ann. Neurol., 24, 497–502.[CrossRef][ISI][Medline]

  2. Lavin, M.F. and Shiloh, Y. (1997) The genetic defect in ataxia-telangiectasia. Annu. Rev. Immunol., 15, 177–202.[CrossRef][ISI][Medline]

  3. Perlman, S., Becker-Catania, S. and Gatti, R.A. (2003) Ataxia-telangiectasia: diagnosis and treatment. Sem. Ped. Neurol., 10, 173–182.[CrossRef]

  4. Boder, E. (1985) Ataxia-telangiectasia: an overview. Kroc. Found. Ser., 19, 1–63.[Medline]

  5. Sekijima, Y., Ohara, S., Nakagawa, S., Tabata, K., Yoshida, K., Ishigame, H., Shimizu, Y. and Yanagisawa, N. (1998) Hereditary motor and sensory neuropathy associated with cerebellar atrophy (HMSNCA): clinical and neuropathological features of a Japanese family. J. Neurol. Sci., 158, 30–37.[CrossRef][ISI][Medline]

  6. Taylor, A.M., Harnden, D.G., Arlett, C.F., Harcourt, S.A., Lehmann, A.R., Stevens, S. and Bridges, B.A. (1975) Ataxia telangiectasia: a human mutation with abnormal radiation sensitivity. Nature, 258, 427–429.[CrossRef][Medline]

  7. Chen, P.C., Lavin, M.F., Kidson, C. and Moss, D. (1987) Identification of ataxia telangiectasia heterozygotes, a cancer prone population. Nature, 274, 484–486.

  8. Sun, X., Becker-Catania, S.G., Chun, H.H., Hwang, M.J., Huo, Y., Wang, Z., Mitui, M., Sanal, O., Chessa, L., Crandall, B. and Gatti, R.A. (2002) Early diagnosis of ataxia-telangiectasia using radiosensitivity testing. J. Pediatr., 140, 724–731.[CrossRef][ISI][Medline]

  9. Shiloh, Y. and Kastan, M.B. (2001) ATM: genome stability, neuronal development, and cancer cross paths. Adv. Cancer. Res., 83, 209–254.[ISI][Medline]

  10. Bakkenist, C.J. and Kastan, M.B. (2003) DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature, 421, 499–506.[CrossRef][Medline]

  11. Kozlov, S., Gueven, N., Keating, K., Ramsay, J. and Lavin, M.F. (2003) ATP activates ataxia-telangiectasia mutated (ATM) in vitro importance of autophosphorylation. J. Biol. Chem., 278, 9309–9317.[Abstract/Free Full Text]

  12. Petrini, J.H., Maser, R.S. and Bressan, D.A. (2001) The Mre11–Rad50 complex. In Hoeckstra, M.F. (ed.), DNA Damage and Repair. Humana Press, Totowa, NJ, Vol. 3, pp. 147–172.

  13. D'Amours, D. and Jackson, S.P. (2002) The Mre11 complex: at the crossroads of DNA repair and checkpoint signalling. Nat. Rev. Mol. Cell. Biol., 3, 317–327.[CrossRef][ISI][Medline]

  14. Carney, J.P., Maser, R.S., Olivares, H., Davis, E.M., Le Beau, M., Yates, J.R., III, Hays, L., Morgan, W.F. and Petrini, J.H. (1998) The hMre11/hRad50 protein complex and Nijmegen breakage syndrome: linkage of double-strand break repair to the cellular DNA damage response. Cell, 93, 477–486.[CrossRef][ISI][Medline]

  15. Stewart, G.S., Maser, R.S., Stankovic, T., Bressan, D.A., Kaplan, M.I., Jaspers, N.G., Raams, A., Byrd, P.J., Petrini, J.H. and Taylor, A.M. (1999) The DNA double-strand break repair gene hMRE11 is mutated in individuals with an ataxia-telangiectasia-like disorder. Cell, 99, 577–587.[CrossRef][ISI][Medline]

  16. Date, H., Onodera, O., Tanaka, H., Iwabuchi, K., Uekawa, K., Igarashi, S., Koike, R., Hiroi, T., Yuasa, T., Awaya, Y. et al. (2001) Early-onset ataxia with ocular motor apraxia and hypoalbuminemia is caused by mutations in a new HIT superfamily gene. Nat. Genet., 29, 184–188.[CrossRef][ISI][Medline]

  17. Moreira, M.C., Barbot, C., Tachi, N., Kozuka, N., Uchida, E., Gibson, T., Mendonca, P., Costa, M., Barros, J., Yanagisawa, T. et al. (2001) The gene mutated in ataxia-ocular apraxia 1 encodes the new HIT/Zn-finger protein aprataxin. Nat. Genet., 29, 189–193.[CrossRef][ISI][Medline]

  18. Le Ber, I., Moreira, M.C., Rivaud-Pechoux, S., Chamayou, C., Ochsner, F., Kuntzer, T., Tardieu, M., Said, G., Habert, M.O., Demarquay, G. et al. (2003) Cerebellar ataxia with oculomotor apraxia type 1: clinical and genetic studies. Brain, 126, 2761–2772.[Abstract/Free Full Text]

  19. Tranchant, C., Fleury, M., Moreira, M.C., Koenig, M. and Warter, J.M. (2003) Phenotypic variability of aprataxin gene mutations. Neurology, 60, 868–870.[Abstract/Free Full Text]

  20. Brenner, C., Garrison, P., Gilmour, J., Peisach, D., Ringe, D., Petsko, G.A. and Lowenstein, J.M. (1997) Crystal structures of HINT demonstrate that histidine triad proteins are GalT-related nucleotide-binding proteins. Nat. Struct. Biol., 4, 231–238.[CrossRef][ISI][Medline]

  21. Lima, C.D., Klein, M.G. and Hendrickson, W.A. (1997) Structure-based analysis of catalysis and substrate definition in the HIT protein family. Science, 278, 286–290.[Abstract/Free Full Text]

  22. Siprashvili, Z., Sozzi, G., Barnes, L.D., McCue, P., Robinson, A.K., Eryomin, V., Sard, L., Tagliabue, E., Greco, A., Fusetti, L. et al. (1997) Replacement of Fhit in cancer cells suppresses tumorigenicity. Proc. Natl Acad. Sci. USA, 94, 13771–13776.[Abstract/Free Full Text]

  23. Caldecott, K.W. (2003) DNA single-strand break repair and spinocerebellar ataxia. Cell, 112, 7–10.[CrossRef][ISI][Medline]

  24. Petrini, J.H. and Stracker, T.H. (2003) The cellular response to DNA double-strand breaks: defining the sensors and mediators. Trends Cell Biol., 13, 458–462.[CrossRef][ISI][Medline]

  25. El-Khamisy, S.F., Masutani, M., Suzuki, H. and Caldecott, K.W. (2003) A requirement for PARP-1 for the assembly or stability of XRCC1 nuclear foci at sites of oxidative DNA damage. Nucl. Acids Res., 31, 5526–5533.[Abstract/Free Full Text]

  26. Masson, M., Niedergang, C., Schreiber, V., Muller, S., Menissier-de Murcia, J. and de Murcia, G. (1998) XRCC1 is specifically associated with poly (ADP-ribose) polymerase and negatively regulates its activity following DNA damage. Mol. Cell. Biol., 18, 3563–3571.[Abstract/Free Full Text]

  27. Yang, C., Maiguel, D.A. and Carrier, F. (2002) Identification of nucleolin and nucleophosmin as genotoxic stress-responsive RNA-binding proteins. Nucl. Acids Res., 30, 2251–2260.[Abstract/Free Full Text]

  28. Dahm-Daphi, J., Sass, C. and Alberti, W. (2000) Comparison of biological effects of DNA damage induced by ionizing radiation and hydrogen peroxide in CHO cells. Int. J. Radiat. Biol., 76, 67–75.[CrossRef][ISI][Medline]

  29. Nakamura, J., Asakura, S., Hester, S.D., de Murcia, G., Caldecott, K.W. and Swenberg, J.A. (2003) Quantitation of intracellular NAD(P)H can monitor an imbalance of DNA single-strand break repair in base excision repair deficient cells in real time. Nucl. Acids Res., 31, e104.[Abstract/Free Full Text]

  30. Jakob, B., Scholz, M. and Taucher-Scholz, G. (2003) Biological imaging of heavy charged-particle tracks. Radiat. Res., 159, 676–648.[CrossRef][ISI][Medline]

  31. Aten, J.A., Stap, J., Krawczyk, P.M., van Oven, C.H., Hoebe, R.A., Essers, J. and Kanaar, R. (2004) Dynamics of DNA double-strand breaks revealed by clustering of damaged chromosome domains. Science, 303, 92–95.[Abstract/Free Full Text]

  32. Ward, J.F. (1994) The complexity of DNA damage: relevance to biological consequences. Int. J. Radiat. Biol., 66, 441–445.[ISI][Medline]

  33. Leung, A.K. and Lamond, A.I. (2003) The dynamics of the nucleolus. Crit. Rev. Eukaryot. Gene Expr., 13, 39–54.[CrossRef][ISI][Medline]

  34. Daniely, Y., Dimitrova, D.D. and Borowiec, J.A. (2002) Stress-dependent nucleolin mobilization mediated by p53-nucleolin complex formation. Mol. Cell. Biol., 22, 6014–6022.[Abstract/Free Full Text]

  35. Shimazaki, H., Takiyama, Y., Sakoe, K., Ikeguchi, K., Niijima, K., Kaneko, J., Namekawa, M., Ogawa, T., Date, H., Tsuji, S., Nakano, I. and Nishizawa, M. (2002) Early-onset ataxia with ocular motor apraxia and hypoalbuminemia: the aprataxin gene mutations. Neurology, 59, 590–595.[Abstract/Free Full Text]

  36. Solveig Walles, S.A. and Erixon, K. (1984) Single-strand breaks in DNA of various organs of mice induced by methyl methanesulfonate and dimethylsulfoxide determined by the alkaline unwinding technique. Carcinogenesis, 5, 319–323.[Abstract/Free Full Text]

  37. Ward, J.F., Evans, J.W., Limoli, C.L. and Calabro-Jones, P.M. (1987) Radiation and hydrogen peroxide induced free radical damage to DNA. Brit. J. Cancer., 55 (Suppl. VIII), 105–112.[ISI][Medline]

  38. Gamaley, I.A. and Klyubin, I.V. (1999) Roles of reactive oxygen species: signaling and regulation of cellular functions. Int. Rev. Cytol., 188, 203–255.[ISI][Medline]

  39. Cantoni, O., Sestili, P., Fiorilli, M., Santoro, M.P., Tannoia, M.C., Novelli, G., Cattabeni, F. and Dallapiccola, B. (1989) Identification of 4 ataxia telangiectasia cell lines hypersensitive to gamma-irradiation but not to hydrogen peroxide. Mutat. Res., 218, 143–148.[CrossRef][ISI][Medline]

  40. Thompson, L.H., Brookman, K.W., Jones, N.J., Allen, S.A. and Carrano, A.V. (1990) Molecular cloning of the human XRCC1 gene, which corrects defective DNA strand break repair and sister chromatid exchange. Mol. Cell. Biol., 10, 6160–6171.[Abstract/Free Full Text]

  41. Caldecott, K.W., Aoufouchi, S., Johnson, P. and Shall, S. (1996) XRCC1 polypeptide interacts with DNA polymerase beta and possibly poly(ADP-ribose) polymerase, and DNA ligase III is a novel molecular ‘nick-sensor’ in vitro. Nucl. Acids Res., 24, 4387–4394.[Abstract/Free Full Text]

  42. Burkle, A. (2001) Physiology and pathophysiology of poly(ADP-ribosyl)ation. Bioassays, 23, 795–806.[CrossRef][ISI][Medline]

  43. Durocher, D. and Jackson, S.P. (2002) The FHA domain. FEBS Lett., 513, 58–66.[CrossRef][ISI][Medline]

  44. Nocentini, S. (1999) Rejoining kinetics of DNA single- and double-strand breaks in normal and DNA ligase-deficient cells after exposure to ultraviolet C and gamma radiation: an evaluation of ligating activities involved in different DNA repair processes. Radiat. Res., 151, 423–432.[ISI][Medline]

  45. Shen, M.R., Zdzienicka, M.Z., Mohrenweiser, H., Thompson, L.H. and Thelen, M.P. (1998) Mutations in hamster single-strand break repair gene XRCC1 causing defective DNA repair. Nucl. Acids Res., 26, 1032–1037.[Abstract/Free Full Text]

  46. Nelms, B.E., Maser, R.S., MacKay, J.F., Lagally, M.G. and Petrini, J.H. (1998) In situ visualization of DNA double-strand break repair in human fibroblasts. Science, 280, 590–592.[Abstract/Free Full Text]

  47. Goldberg, M., Stucki, M., Falck, J., D'Amours, D., Rahman, D., Pappin, D., Bartec, J., and Jackson, S.P. (2003) MDC1 is required for the intra-S-phase DNA damage checkpoint. Nature, 421, 952–956.[CrossRef]