Karen Hobson

ATM associates with and phosphorylates p53: mapping the region of interaction

The human genetic disorder ataxia-telangiectasia (AT) is characterized by immunodeficiency, progressive cerebellar ataxia, radiosensitivity, cell cycle checkpoint defects and cancer predisposition. The gene mutated in this syndrome, ATM (for AT mutated), encodes a protein containing a phosphatidyl-inositol 3-kinase (PI-3 kinase)-like domain. ATM also contains a proline-rich region and a leucine zipper, both of which implicate this protein in signal transduction. The proline-rich region has been shown to bind to the SH3 domain of c-Abl, which facilitates its phosphorylation and activation by ATM (Refs 4,6). Previous results have demonstrated that AT cells are defective in the G1/S checkpoint activated after radiation damage and that this defect is attributable to a defective p53 signal transduction pathway. We report here direct interaction between ATM and p53 involving two regions in ATM, one at the amino terminus and the other at the carboxy terminus, corresponding to the PI-3 kinase domain. Recombinant ATM protein phosphorylates p53 on serine 15 near the N terminus. Furthermore, ectopic expression of ATM in AT cells restores normal ionizing radiation (IR)-induced phosphorylation of p53, whereas expression of ATM antisense RNA in control cells abrogates the rapid IR-induced phosphorylation of p53 on serine 15. These results demonstrate that ATM can bind p53 directly and is responsible for its serine 15 phosphorylation, thereby contributing to the activation and stabilization of p53 during the IR-induced DNA damage response.

Cellular localisation of the ataxia-telangiectasia (ATM) gene product and discrimination between mutated and normal forms.

The recently cloned gene (ATM) mutated in the human genetic disorder ataxia-telangiectasia (A-T) is involved in DNA damage response at different cell cycle checkpoints and also appears to have a wider role in signal transduction. Antibodies prepared against peptides from the predicted protein sequence detected a ∼ 350 kDa protein corresponding to the open reading frame, which was absent in 13/23 A-T homozygotes. Subcellular fractionation, immunoelectronmicroscopy and immunofluorescence showed that the ATM protein is present in the nucleus and cytoplasmic vesicles. This distribution did not change after irradiation. We also provide evidence that ATM protein binds to p53 and this association is defective in A-T cells compatible with the defective p53 response in these cells. These results provide further support for a role for the ATM protein as a sensor of DNA damage and in a more general role in cell signalling, compatible with the broader phenotype of the syndrome.

Transactivation-deficient p73α (p73Δexon2) inhibits apoptosis and competes with p53

p73 has recently been identified as a structural and functional homolog of the tumor suppressor protein p53. Overexpression of p53 activates transcription of p53 effector genes, causes growth inhibition and induced apoptosis. We describe here the effects of a tumor-derived truncated transcript of p73α (p73Δexon2) on p53 function and on cell death. This transcript, which lacks the acidic N-terminus corresponding to the transactivation domain of p53, was initially detected in a neuroblastoma cell line. Overexpression of p73Δexon2 partially protects lymphoblastoid cells against apoptosis induced by anti-Fas antibody or cisplatin. By cotransfecting p73Δexon2 with wild-type p53 in the p53 null line Saos 2, we found that this truncated transcript reduces the ability of wild-type p53 to promote apoptosis. This anti-apoptotic effect was also observed when p73Δexon2 was co-transfected with full-length p73 (p73α). This was further substantiated by suppression of p53 transactivation of the effector gene p21/Waf1 in p73Δexon2 transfected cells and by inhibition of expression of a reporter gene under the control of the p53 promoter. Thus, this truncated form of p73 can act as a dominant-negative agent towards transactivation by p53 and p73α, highlighting the potential implications of these findings for p53 signaling pathway. Furthermore, we demonstrate the existence of a p73Δexon2 transcript in a very significant proportion (46%) of breast cancer cell lines. However, a large spectrum of normal and malignant tissues need to be surveyed to determine whether this transdominant p73 variant occurs in a tumor-specific manner.

Chk1 complements the G2/M checkpoint defect and radiosensitivity of ataxia-telangiectasia cells.

Cells from patients with the human genetic disorder ataxia-telangiectasia (A-T) are defective in the activation of cell cycle checkpoints in response to ionizing radiation damage. In order to understand the role of ATM in checkpoint control we investigated whether Schizosaccaromyces pombe chk1, a protein kinase implicated in controlling the G2 DNA damage checkpoint, might alter the radiosensitive phenotype in A-T cells. The fission yeast chk1 gene was cloned into an EBV-based vector under the control of a metallothionein promoter and transfected into A-T lymphoblastoid cells. Induction of chk1 enhanced the survival of an A-T cell line in response to radiation exposure as determined by cell viability and reduction of radiation-induced chromosome aberrations. This can be accounted for at least in part by the restoration of the G2 checkpoint to chk1 expressing cells. There was no evidence that chk1 expression corrected either the G1/S checkpoint or radioresistant DNA synthesis in S phase in these cells. These results suggest that chk1 when overexpressed acts downstream from ATM to restore the G2 checkpoint in these cells and correct the radiosensitive phenotype. These data allow us to dissociate individual checkpoint events and relate them to the radiosensitive phenotype in A-T cells.

Defect in Radiation Signal Transduction in Ataxia-telangiectasia

Exposure of mammalian cells to ionizing radiation causes a delay in progression through the cycle at several checkpoints. Cells from patients with ataxia-telangiectasia (A-T) ignore these checkpoint controls postirradiation. The tumour suppressor gene product p53 plays a key role at the G1/S checkpoint preventing the progression of cells into S phase. The induction of p53 by radiation is reduced and/or delayed in A-T cells, which appears to account for the failure of delay at the G1/S checkpoint. We have investigated further this defect in radiation signal transduction in A-T. While the p53 response was defective after radiation, agents that interfered with cell cycle progression such as mimosine, aphidicolin and deprivation of serum led to a normal p53 response in A-T cells. None of these agents caused breaks in DNA, as determined by pulse-field gel electrophoresis, in order to elicit the response. Since this pathway is mediated by protein kinases, we investigated the activity of several of these enzymes in control and A-T cells. Ca+2-dependent and -independent protein kinase C activities were increased by radiation to the same extent in the two cell types, a variety of serine/threonine protein kinase activities were approximately the same and anti-tyrosine antibodies failed to reveal any differences in protein phosphorylation between A-T and control cells. It is not evident what is the nature of the defect in signal transduction in A-T cells. However, it is clear that the p53 response is normal in these cells after exposure to some agents and it is mediated through protein kinase C or another serine/threonine kinase.

Use of DNA-protein interaction to isolate specific genomic DNA sequences

We describe a simple and rapid method for the isolation of specific genomic DNA sequences recognized by DNA-binding proteins. This procedure consists of four steps: (1) restriction enzyme digestion and size fractionation of genomic DNA; (2) DNA--protein binding using the gel mobility-shift assay; (3) ligation of isolated DNA fragments followed by transformation of Escherichia coli; and (4) screening of recombinant clones for inserts containing specific DNA--protein binding sequences. We have used this protocol to isolate human DNA sequences, 100-200 bp in size, that are recognized by both partially purified and affinity purified proteins. Unlike other procedures designed to identify genomic target sequences, the method described does not require polymerase chain reaction or successive immunoprecipitations.

Comparative study of radiation-induced G2 phase delay and chromatid damage in families with ataxia-telangiectasia

Two assay systems, radiation-induced chromosome aberrations and flow cytometry, were compared for the detection of ataxia-telangiectasia (A-T) heterozygotes. In three A-T families, the frequencies of chromatid aberrations in phytohemagglutinin-stimulated blood lymphocytes after 1 Gy of gamma-irradiation were twofold higher in A-T homozygotes than in obligate A-T heterozygotes, which were in turn approximately twofold higher than in normal control cells. Other consanguineous relatives of A-T patients had intermediate levels of induced chromatid aberrations, suggesting that they were carriers of the gene. Matched Epstein-Barr virus-transformed lymphoblastoid cell lines from A-T homozygotes showed a greater radiation-induced accumulation in the G2 phase of the cell cycle than did control cells. In family B, both obligate heterozygotes had increased G2 delay, as did the one heterozygote available for family C, and two of the grandparents in that family were in the high range for G2 delay. Neither parent in family A had high G2 phase delay after irradiation although the induced chromatid aberrations were in the heterozygote valve range. These results show a good concordance between the two assay systems for A-T heterozygotes, with the chromatid aberrations somewhat more consistent.

Cloning and characterization of cDNA encoding a human arginyl-tRNA synthetase.

Arginyl-tRNA synthetase (ArgRS) plays a key role in protein synthesis as part of a multienzyme complex with a number of other aminoacyl-tRNA synthetase (aaRS) enzymes. We have isolated a full-length cDNA encoding ArgRS as part of a project on complementation of radiosensitivity in human cells with an Epstein-Barr Virus (EBV) vector-based human cDNA library. DNA sequence analysis identified an open reading frame of 1983 nucleotides with 87% homology to other mammalian ArgRS genes. The deduced amino acid (aa) sequence (661 aa) showed 87.7% identity to the Chinese hamster ovary (CHO) enzyme and 37.7% identity to the homologous Escherichia coli enzyme. Northern blot analysis revealed the presence of a single mRNA species of approx. 2.2 kb. The results described here demonstrate that ArgRS is highly conserved in mammalian cells and confirm the presence of a hydrophobic N-terminal region in the higher-molecular-weight complexed form of ArgRS.

Characterization of the breast cancer associated ATM 7271T > G (V2424G) mutation by gene expression profiling

Mutations in ATM are responsible for the autosomal recessive disorder ataxia telangiectasia. Heterozygous mutations in ATM have been associated with an elevated risk of breast cancer. We previously reported one breast cancer family in which ATM 7271T>G (V2424G) segregated with disease, and apparently acted in a dominant negative manner. We now report the screening of 782 multiple‐case breast cancer families that identified two additional index cases with ATM 7271T>G. Phylogenetic sequence analysis showed that V2424 is a highly conserved residue, and that the 2424G variant is likely to interfere with function. To elucidate the consequences of this mutation, we expression profiled wild‐type, heterozygous, and homozygous lymphoblastoid cell lines (LCLs) from Scottish and Australian families using an oligonucleotide microarray. Cluster analysis revealed 77 genes that were differentially expressed in homozygous and heterozygous V2424G cells (compared to wild‐type) and 11 genes differentially expressed in the homozygous cells. We also evaluated the profiles of LCLs after exposure to ionizing radiation (IR) and identified 77 genes that were differentially expressed in wild‐type cells, but not in homozygous or heterozygous V2424G cells. We validated the expression differences by RT‐PCR in additional heterozygous V2424G LCLs from another breast cancer family. We found no consistent cytotoxicity or abrogation of ATM kinase activity after IR in seven heterozygous V2424G LCLs, compared to wild‐type LCLs, but did find an increase in the number of chromosomal aberrations. These data suggest that the V2424G missense mutation acts largely as a dominant negative in terms of the associated expression profiles.

Prediction of BRCA1 and BRCA2 mutation status using post-irradiation assays of lymphoblastoid cell lines is compromised by inter-cell-line phenotypic variability

Background and purposeAssays to determine the pathogenicity of unclassified sequence variants in disease-associated genes include the analysis of lymphoblastoid cell lines (LCLs). We assessed the ability of several assays of LCLs to distinguish carriers of germline BRCA1 and BRCA2 gene mutations from mutation-negative controls to determine their utility for use in a diagnostic setting.Materials and methodsPost-ionising radiation cell viability and micronucleus formation, and telomere length were assayed in LCLs carrying BRCA1 or BRCA2 mutations, and in unaffected mutation-negative controls.ResultsPost-irradiation cell viability and micronucleus induction assays of LCLs from individuals carrying pathogenic BRCA1 mutations, unclassified BRCA1 sequence variants or wildtype BRCA1 sequence showed significant phenotypic heterogeneity within each group. Responses were not consistent with predicted functional consequences of known pathogenic or normal sequences. Telomere length was also highly heterogeneous within groups of LCLs carrying pathogenic BRCA1 or BRCA2 mutations, and normal BRCA1 sequences, and was not predictive of mutation status.ConclusionGiven the significant degree of phenotypic heterogeneity of LCLs after γ-irradiation, and the lack of association with BRCA1 or BRCA2 mutation status, we conclude that the assays evaluated in this study should not be used as a means of differentiating pathogenic and non-pathogenic sequence variants for clinical application. We suggest that a range of normal controls must be included in any functional assays of LCLs to ensure that any observed differences between samples reflect the genotype under investigation rather than generic inter-individual variation.