The cell cycle-regulated protein human GTSE-1 controls DNA damage-induced apoptosis by affecting p53 function.

GTSE-1 (G2 and S phase-expressed-1) protein is specifically expressed during S and G2 phases of the cell cycle. It is mainly localized to the microtubules and when overexpressed delays the G2 to M transition. Here we report that human GTSE-1 (hGTSE-1) protein can negatively regulate p53 transactivation function, protein levels, and p53-dependent apoptosis. We identified a physical interaction between the C-terminal regulatory domain of p53 and the C-terminal region of hGTSE-1 that is necessary and sufficient to down-regulate p53 activity. Furthermore, we provide evidence that hGTSE-1 is able to control p53 function in a cell cycle-dependent fashion. hGTSE-1 knock-down by small interfering RNA resulted in a S/G2-specific increase of p53 levels as well as cell sensitization to DNA damage-induced apoptosis during these phases of the cell cycle. Altogether, this work suggests a physiological role of hGTSE-1 in apoptosis control after DNA damage during S and G2 phases through regulation of p53 function.

Cellular response to DNA damage efficiently induces cell growth arrest or apoptosis through a complex network of regulatory pathways. A key element in the integration of these pathways is the tumor suppressor protein p53. The function of p53 as tumor suppressor is mainly ascribable to its activity as a transcription factor that specifically activates genes in response to various types of stress, including DNA damage, oncogene activation, hypoxia, and cell adhesion (1). When the p53 response is activated, p53-inducible genes can mainly induce growth arrest, the repair of damaged DNA, or apoptosis (2). The major mechanisms that govern p53 activity appear to be exerted at the protein level. This includes regulation of p53 protein stability, control of its subcellular localization, and post-translational modifications as well as conformational changes that allow activation of the DNA-binding ability of p53 (3). The key negative regulator of p53 is the proto-oncogene Mdm2. The proposed mechanism by which Mdm2 negatively regulates p53 function involves Mdm2 binding to the N-terminal region of p53 and its ubiquitination (4,5), which promotes p53 nuclear export and subsequent degradation through the ubiquitin-proteasome pathway.
The murine GTSE-1 1 (G 2 and S phase-expressed) gene, previously named B99, was cloned in our laboratory during a screening of p53-inducible genes from a murine cell line that stably expresses a temperature-sensitive p53 allele (6). Wildtype p53 (wt-p53) induces GTSE-1 transcription by an active p53-binding site located in the promoter region; however, similar to other p53 target genes, GTSE-1 is also induced by DNA damage independent of p53 status (6). Further characterization of murine GTSE-1 protein showed that is mainly localized to the microtubules (6), although clear evidence of nuclearcytoplasmic shuttling of the protein has been observed in our laboratory. 2 GTSE-1 protein is cell cycle-regulated with increased expression during S and G 2 phases. It becomes phosphorylated in mitosis and markedly reduced in G 1 (7). Overexpression of GTSE-1 resulted in a delay of the G 2 to M phase transition independent of p53 (6). The same cell cycle regulation, cellular localization, and the ability to delay the G 2 phase have been also observed for the GTSE-1 human homologue protein, hGTSE-1 (8).
Here we report that hGTSE-1 can control DNA damageinduced apoptosis by regulating p53 stability and function. We provide evidence that hGTSE-1 protein can negatively regulate p53 levels and activity, thus affecting DNA damage-induced apoptosis in wt-p53 cells. The proposed mechanism by which hGTSE-1 regulates p53 activity involves a physical interaction between the C-terminal region of hGTSE-1 and the C-terminal regulatory domain of p53. Moreover, hGTSE-1 knock-down (KD) by small interference RNA (siRNA) resulted in S/G 2specific increase of p53 levels and cell sensitization to DNA damage-induced apoptosis during these cell cycle phases, thus implying that endogenous hGTSE-1 plays a role in p53-dependent apoptosis control during the cell cycle window where it is physiologically expressed.

Cell Lines and Treatments
U2OS (wt-p53) and MG-63 (p53 null) human osteosarcoma cell lines were cultured at 37°C in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, penicillin (100 units/ml), and streptomycin (100 g/ml). Cells treated with methyl methanesulfonate (MMS) (Fluka) were incubated for 4 h with the drug, followed by extensive washing and medium replacement until the end of the treatment. Etoposide (ET) (Sigma) treatment consisted of the addition of ET for the indicated time.

Transfection and Reporter Gene Assay
DNA-Expression vectors were transfected in mid-log growth phase U2OS cells using the calcium phosphate method. Unless stated otherwise, cells were analyzed 24 h after transfection.
siRNA-Transfection of siRNA (Dharmacon Research) was performed as described by Elbashir et al. (9) using Oligofectamine (Invitrogen). Reporter plasmid and siRNA cotransfection was carried out using LipofectAMINE 2000 (Invitrogen) as recommended by the manufacturer for siRNA in adherent cells. In all cases, medium without antibiotics was used. The hGTSE-1 mRNA targeted sequence was AAAUUU-GACUUCGAUCUUUCA (sihGTSE-1). The control siRNA used was AACCUUUUUUUUUGGGGAAAA (siCONT). siRNA for human p53 was designed as described by Brummelkamp et al. (10).
For luciferase assay, cells were transfected with the reporter plasmid (pG13-Luc) and a plasmid encoding Renilla luciferase, pRL-CMV. The assay was performed with a dual luciferase kit (Promega). Renilla luciferase activity was used to standardize p53 activation values for transfection efficiency.

Immunoprecipitation and Western Blot Analysis
Cells were harvested in ice-cold lysis buffer containing 50 mM Tris-HCl, pH 8, 150 mM NaCl, 1% Nonidet P-40, 0.1 mM sodium orthovanadate, 2 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, 5 mM EDTA, and 10 g/ml each of chymostatin, leupeptin, antipain, and pepstatin. Lysis was performed at 4°C for 10 min. The lysates were then clarified by centrifugation and precleared with 20 l of Protein A-Sepharose CL-4B (Amersham Biosciences). Then antibody prebound to 20 l of Protein A-Sepharose CL-4B was added and incubated at 4°C for 4 h. The resin was washed, and bound proteins were eluated in SDS-PAGE sample buffer.
Western blot analysis was performed according to standard procedures using the following primary antibodies: affinity-purified LF1 anti-hGTSE-1 polyclonal antibody, DO-1 anti-p53 monoclonal antibody (Santa Cruz Biotechnology), anti-actin polyclonal antibody (Sigma), and N20 anti-Bax polyclonal antibody (Santa Cruz Biotechnology). Bound primary antibodies were visualized by enhanced chemiluminescence (ECL; Amersham Biosciences) after addition of horseradish peroxidaseconjugated secondary antibodies.

Immunofluorescence, Apoptosis, Flow Cytometry, and in Vitro
Binding Assay Immunofluorescence-Cells were fixed in 3% paraformaldehyde, treated with Triton X-100, and stained using specific antibodies followed by a fluorescein isothiocyanate or RITC-conjugated secondary antibody (Sigma). Images were analyzed with a laser scan confocal microscope (Leica).
Apoptosis-Cell death was assessed by scoring apoptotic morphology of the nucleus stained with propidium iodide (P.I.) by counting at least 200 cells on each experiment.
Flow Cytometry-After siRNA transfection and ET treatment, cells were harvested by trypsin treatment and fixed with cold 70% ethanol.
After washing with phosphate-buffered saline, cells were treated with RNase A, and then DNA was stained with propidium iodide (25 g/ml). Cytometric analysis was performed on a Bryte HS (Bio-Rad) cytofluorimeter. Cell sorting was performed by a FACSventage instrument (Becton-Dickinson) as reported by Delia et al. (11).
In Vitro Binding Assay-35 S-labeled proteins were in vitro translated using reticulocyte lysates (TNT system; Amersham Biosciences) and incubated with purified GST or GST⅐p53 or GST⅐p53 deletion mutants immobilized on glutathione-Sepharose 4B beads (Amersham Biosciences). Bound proteins were eluted and resolved in SDS-PAGE.

hGTSE-1 Alleviates DNA Damage-induced Apoptosis in Cells
Harboring Functional p53-We have previously reported that mouse or human GTSE-1 was able to delay the G 2 progression when overexpressed (6,8). Here we investigated whether this effect on the cell cycle was also accompanied by interference in DNA damage-induced cell death. To approach this topic we overexpressed hGTSE-1 in two human osteosarcoma cell lines containing wt-p53 (U2OS) and p53 null (MG-63). We first compared the effect of hGTSE-1 ectopic expression in both cell types in terms of cell cycle profile. As shown in Fig. 1A, U2OS and MG-63 cells responded similarly to hGTSE-1 overexpression, delaying the G 2 progression. Then MG-63 and U2OS cells were transfected with GFP or GFP-tagged hGTSE-1 expression vectors. 24 h later cells were treated with DNA damaging agents, such as ET or methyl methanesulfonate. ET (final concentration 100 M) was added to the culture medium for 18 h. Methyl methanesulfonate (final concentration 100 M) was added for 4 h, after which it was removed and replaced with normal medium for an additional 14 h. Apoptosis was then assessed by scoring nuclear alteration in transfected cells as we described previously (12). As shown in Fig. 1B, GFP⅐hGTSE-1 expression decreased the apoptotic response induced by both ET and methyl methanesulfonate in U2OS cells, whereas no evidence of apoptotic protection was observed in MG-63 cells. Fig. 1C shows an indicative field of the difference in nuclear morphology in cells treated with ET overexpressing, or not, GFP⅐hGTSE-1. These results suggest a potential role of hGTSE-1 in regulating the p53-dependent apoptotic process. As can be noted in Fig. 1C, hGTSE-1, which mainly localizes to the microtubules in unstressed cells (6,8), accumulates into the nucleus after damage. This is in agreement with recently obtained data indicating that hGTSE-1 is a nucleocytoplasmic shuttling protein that can be relocalized to the nucleus in response to DNA damage signals, with potential consequences in controlling p53-dependent functions. 2 To directly assess the role of hGTSE-1 in p53-induced apoptosis, we used a well characterized model based on wt-p53 overexpression in Saos-2 cells. wt-p53-expressing vector was transfected together with GFP⅐hGTSE-1 or GFP alone. 48 h later, cells were fixed and stained using an anti-p53 polyclonal antibody, and the apoptotic nuclei were scored in cotransfected cells. As shown in Fig. 1D, GFP⅐hGTSE-1 significantly reduced p53-induced apoptosis when compared with cells transfected with GFP alone, thus confirming that hGTSE-1 overexpression could efficiently interfere with p53 proapoptotic activity.
We next investigated the role of endogenous hGTSE-1 in apoptosis induced by DNA damage. For this purpose we designed siRNA to KD hGTSE-1 protein expression in human cells. The ability of the specific siRNA for hGTSE-1 (named sihGTSE-1) to down-regulate hGTSE-1 protein levels was tested in U2OS cells. As shown in Fig. 2A, 40 h after sihGTSE-1 transfection, expression of endogenous hGTSE-1 protein was almost completely abolished. A scrambled siRNA (named siCONT) was used as control.
To address the physiological role of hGTSE-1 in DNA-damaged cells, U2OS cells were transfected with siCONT or sih-GTSE-1. 40 h after siRNA transfection, cells were treated with 50 M ET. Apoptosis was then determined as subG 1 population by using flow cytometry analysis. As observed in Fig. 2B, hGTSE-1/knock-down (hGTSE-1/KD) cells displayed an increased subG 1 population after ET when compared with control cells. Fig. 2C shows a representative field evidencing the enhancement of apoptotic nuclei induced by ET treatment in hGTSE-1/KD cells when compared with cells transfected with control siRNA. The obtained data suggest that hGTSE-1 knockdown significantly sensitizes cells to undergo apoptotic cell death upon treatment with DNA-damaging agents.
Finally, to analyze the involvement of p53 in apoptosis induced by DNA damage in hGTSE-1/KD cells, the effect of ET treatment was evaluated in hGTSE-1/p53 double knock-down cells. U2OS cells were transfected with sihGTSE-1, sip53 (siRNA specific for p53 silencing, Brummelkamp et al.,Ref. 10), and their combination as indicated in Fig. 2D. sip53 efficiency to down-regulate p53 levels was about 80 -90% as determined by Western blotting (data not shown). Although in the absence of damage no significant apoptosis was detected, hGTSE-1/KD cells displayed enhanced cell death compared with cells transfected with control siRNA when treated with ET. Interestingly, p53 silencing significantly rescued susceptibility of hGTSE-1/K D cells to ET-induced apoptosis. These data and those presented in Fig. 1B indicate that p53 is a key element for hGTSE-1-dependent control of apoptosis upon DNA damage.
hGTSE-1 Expression Regulates p53 Activity and Protein Levels-Because we obtained evidence pointing to a role of hGTSE-1 in p53-dependent apoptosis after DNA damage, we investigated the functional activity of p53 in cells expressing different levels of hGTSE-1. We first established a U2OS cell line with hGTSE-1 expression under the control of tetracycline. p53 protein levels were determined in cell lysates of all seven hGTSE-1-inducible clones obtained after 24 h of tetracycline removal. As shown in two representative clones (Fig. 3A), p53 levels strongly decreased after hGTSE-1 induction, suggesting that overexpressed hGTSE-1 is able to reduce endogenous levels of p53. We then determined p53 transactivation activity in U2OS cells transiently cotransfected with a synthetic promoter containing multiple p53 binding sites (pG13-Luc) together with hGTSE-1 or an empty vector. As shown in Fig. 3B, overexpression of hGTSE-1 efficiently down-regulated endogenous p53driven transactivation activity, altogether suggesting that hGTSE-1 could exert resistance to DNA damage-induced apoptosis at least in part by regulating p53.
p53 levels and activity were then determined in cells expressing lower levels of hGTSE-1 protein by siRNA-hGTSE-1 silencing. In agreement with hGTSE-1 overexpression data, down-regulation of endogenous hGTSE-1 resulted in a moderate increase of p53 protein level (Fig. 3C). Consistently, enhancement of p53 transactivation was detected after quantification of the pG13-Luc reporter activity when cotransfected with sihGTSE-1 (Fig. 3D).
Although silencing of hGTSE-1 increased p53 levels and activity in unstressed cells, it failed to stimulate apoptotic cell death. This fact could indicate that the amount of accumulated p53 in hGTSE-1/K D cells is not sufficient to induce apoptosis or that accumulated p53 is not fully active or efficiently modified to trigger the apoptotic process. To investigate this hypothesis, we analyzed p53 levels and activity in hGTSE-1/K D cells exposed or not to ET treatment (25 M for 18 h). Phosphorylation of p53 in serine 15 is a key event during its full activation (13). Using a p53-Ser-15 phospho-specific antibody, we observed that hGTSE-1 knock-down did not induce detectable p53 phosphorylation in the absence of damage, whereas it followed p53 increments in ET-treated cells. Consistently, levels of the p53induced proapoptotic protein Bax remained unchanged in unstressed hGTSE-1/K D cells, whereas its up-regulation was clearly detected after ET treatment (Fig. 3E). Finally, when p53 transactivation function was assessed using the pG13-Luc reporter, we observed that hGTSE-1 silencing could significantly enhance p53 activity even in cells treated with ET (Fig.  3F), thus concluding that increased p53 levels due to hGTSE-1 silencing should synergize with DNA damage-dependent p53 stabilization and activation, therefore providing a more robust p53 response when compared with cells expressing normal levels of hGTSE-1 protein.
hGTSE-1 Physically Interacts with p53-Looking for possible mechanisms linking hGTSE-1 expression with p53 regulation, we investigated whether hGTSE-1 could physically interact with p53. hGTSE-1/p53 interaction was analyzed in vivo by coimmunoprecipitation assay. U2OS cells were cotransfected with vectors expressing hGTSE-1 and the GFP⅐p53 fusion protein. Cell lysates were immunoprecipitated with hGTSE-1 antiserum and immunoblotted using the DO-1 anti-p53 monoclonal antibody. The results demonstrate that hGTSE-1 and p53 can bind to each other because the 90-kDa GFP⅐p53 protein specifically immunoprecipitated with the anti-hGTSE-1 antiserum (Fig. 4A). To obtain evidence that this association could occur between the respective endogenous products, U2OS cells lysates were immunoprecipitated with DO-1 anti-p53 antibody or 9E10 anti-Myc antibody as control. Fig. 4B shows that endogenous p53 and hGTSE-1 can physically interact, because hGTSE-1 coprecipitated when DO-1 antibody was used. Similar results were also obtained in HCT116 cells (data not shown).
We then performed in vitro binding experiments to define the interaction region between hGTSE-1 and p53. In vitro translated (IVT) 35 S-labeled hGTSE-1 protein was incubated with different GST⅐p53 deletions described in Fig. 4C. As shown in Fig. 4D, hGTSE-1 was bound by the C-terminal region of p53 (amino acids 355-393), which has been established as a p53 regulatory domain crucial for the control of its transactivation function.
The region of hGTSE-1 that is able to bind p53 was then determined by using IVTs of two hGTSE-1 constructs coding from amino acid 1 to 476 (NT-hGTSE-1) and from amino acid 476 to 720 (CT-hGTSE-1). These deletions were tested for in vitro binding to the full-length p53 (GST⅐p53) as described above. As shown in Fig. 4E, CT-hGTSE-1 was preferentially retained on the beads containing GST⅐p53, supporting the existence of a direct interaction between the C-terminal regions of hGTSE-1 and p53.
To confirm the data obtained in in vitro assays, the effect of NT-hGTSE-1 and CT-hGTSE-1 deletion mutants on endogenous p53 transactivation function was assessed in U2OS cells.  3. Regulation of p53 by hGTSE-1 expression. A, p53 protein levels after induction of hGTSE-1 expression in two representative hGTSE-1 tet-inducible clones. B, effect of hGTSE-1 ectopic expression on p53 transactivation activity using the pG13-Luc reporter in U2OS cells. pRL-CMV vector (Renilla luciferase) was cotransfected to normalize transfection efficiency. Arbitrary units of luciferase corresponding to each transfection are indicated. C, determination of p53 protein levels after hGTSE-1 silencing by transfecting sihGTSE-1 in U2OS cells. D, analysis of p53 transactivation activity in knock-down hGTSE-1 cells using pG13-Luc as reporter as described in panel B. E, p53, phosphorylated-p53, and Bax protein levels in cells with normal or down-regulated levels of hGTSE-1 protein treated or not with ET. F, p53 transactivation activity in cells containing normal or down-regulated levels of hGTSE-1 protein treated or not with ET using the pG13-Luc reporter.
p53 activity was analyzed in transiently transfected cells with the indicated hGTSE-1 constructs together with the pG13-Luc reporter. As shown in Fig. 4F, the construct lacking the region involved in p53 binding (NT-hGTSE-1) completely lost the ability to down-regulate p53 activity when compared with the full-length protein. Interestingly, the CT-hGTSE-1 deletion FIG. 4. Interaction between hGTSE-1 and p53. A, hGTSE-1 and GFP⅐p53 proteins were overexpressed in U2OS cells and immunoprecipitated using hGTSE-1 antiserum or normal rabbit serum (NRS) as control. p53 was detected using the DO-1 anti-p53 monoclonal antibody (upper panel). Immunoprecipitated hGTSE-1 was determined in the same membrane (lower panel). B, endogenous hGTSE-1 and p53 complex was assessed in U2OS cells lysates immunoprecipitated with anti-p53 DO-1 monoclonal antibody. 9E10 anti-Myc monoclonal antibody was used as control. hGTSE-1 protein was detected using the LF1b anti-hGTSE-1 antibody (upper panel). Immunoprecipitated p53 was determined in the same membrane (lower panel). C, the GST⅐p53 and hGTSE-1 constructs used for in vitro binding experiments. D, In vitro binding assay using 35 S-IVT full-length hGTSE-1 (Input) incubated with purified GST⅐p53 constructs coupled to glutathione-Sepharose matrix. Bound material was separated on SDS-PAGE and visualized by autoradiography. E, in vitro binding assay using 35 S-IVT NT-hGTSE-1 or CT-hGTSE-1 (Inputs) incubated with GST or GST⅐p53 coupled to glutathione-Sepharose matrix. F, U2OS cells transfected with the indicated forms of hGTSE-1 together with the pG13-Luc reporter. The control was performed by transfecting the same amount of an empty pcDNA3 vector together with pG13-Luc. 5. hGTSE-1 control of p53 and p53-dependent apoptosis is cell cycle-regulated. A, analysis of p53 levels in G 1 and S/G 2 -sorted cells transfected or not with sihGTSE-1. Cyclin A levels and actin were determined as S/G 2 cell cycle marker and loading control, respectively. B, images of U2OS cells transfected with sihGTSE-1 or siCONT and treated with ET. Panels show corresponding fields of cells stained for cyclin A in green (1, 3, and 5, 7) and their respective nuclei with P.I. (2, 4, and 6, 8). Arrows indicate apoptotic cells positive or not for cyclin A staining. C, scoring of apoptosis induced by ET in cells transfected with sihGTSE-1 or siCONT, depending on the staining for cyclin A as indicated under "Results." Data shown here correspond to a single representative experiment of four experiments performed. D, Western blot showing cyclin A protein levels upon hGTSE-1 silencing and ET treatment. mutant maintained full ability to repress p53 transactivation activity, thus suggesting that hGTSE-1 regulation of p53 activity correlates with its ability to bind p53. Moreover, the hGTSE-1 region involved in p53 binding is necessary and sufficient to repress p53 transactivation function.

FIG.
hGTSE-1 Plays a Role in DNA Damage-induced Cell Death during the S and G 2 Phases-hGTSE-1 protein is specifically expressed during S and G 2 phases of the cell cycle (8), with an expression profile similar to that of cyclin A. This behavior could indicate that endogenous hGTSE-1 may regulate p53 in a cell cycle-specific manner. To address this topic, U2OS cells were transfected with sihGTSE-1 or siCONT, and 40 h later the DNA dye Hoechst 33342 was added to the culture medium. Using a fluorescence-activated cell sorter (FACS, Becton-Dickinson), cells in G 1 phase (DNA content, 2N) and those in S/G 2 /M phases (DNA content, more than 2N to 4N) were separated, and p53 levels were determined in all fractions. As shown in Fig. 5A, although siRNA-dependent hGTSE-1 silencing did not affect p53 levels of G 1 phase cells, it specifically increased p53 protein levels in cells passing S/G 2 /M phases, indicating that hGTSE-1 is a cell cycle-specific regulator of p53.
This result prompted us to investigate whether cells in the course of the S and G 2 phases could be more sensitive to apoptosis when hGTSE-1 was silenced. To address this issue, U2OS cells were transfected with sihGTSE-1 or siCONT and treated with ET as described for Fig. 2B. 18 h after ET treatment, cells were fixed and stained with an anti-cyclin A antibody as S/G 2 marker and P.I. A similar number of apoptotic cells were counted in hGTSE-1/K D and control cells and then scored for the reactivity to cyclin A, as indicated in the fields shown in Fig. 5B. Interestingly, among the hGTSE-1/K D cells, the cyclin A-positive subpopulation with apoptotic morphology displayed a considerable increase with respect to the control (Fig. 5C). Cyclin A protein levels were not changed by hGTSE-1 silencing or ET treatment (Fig. 5D), implying that the increased ratio of apoptotic hGTSE-1/K D cells positive to cyclin A was not because of increased cyclin A-positive population. All together these results suggest that by regulating p53, endogenous hGTSE-1 plays a role in apoptosis control during S and G 2 , the cell cycle window where it is physiologically expressed.

DISCUSSION
In this work we have presented evidence that hGTSE-1 protein participates in the cellular response to DNA-damaging agents by regulating p53 function and stability during the S and G 2 phases of the cell cycle. We demonstrated that hGTSE-1 protein can regulate DNA damage-induced apoptosis by controlling p53 transactivation activity and protein levels. A physiological role for hGTSE-1 protein in DNA damage-induced apoptosis was established by siRNA-dependent hGTSE-1 silencing, which sensitizes cells to apoptotic cell death after such types of stress. Moreover, we observed that the ability of hGTSE-1 to control apoptosis appeared to be restricted to the S and G 2 phases of the cell cycle, in accordance with its cell cycle expression profile and its ability to regulate p53 stability.
The potential mechanism by which hGTSE-1 regulates p53 activity involves a physical interaction between the C-terminal region of hGTSE-1 and the C-terminal regulatory domain of p53. Moreover, the C-terminal region of hGTSE-1 is sufficient to repress p53 transactivation function. The C-terminal regulatory domain of p53 is the target for several post-translational modifications, such as phosphorylation (14,15) and acetylation (16,17), regulating p53 activity and stability, thus suggesting that the interaction of hGTSE-1 within this domain could interfere with one or more modifications required for p53 function.
Recent observations obtained in this laboratory on the sub-cellular localization of GTSE-1 suggest that it can shuttle from cytoplasm (where it localizes to the microtubules) to the nucleus relying on conserved nuclear localization and nuclear export signals. 2 In addition, DNA damage induces a p53-independent stabilization of hGTSE-1 protein as well as its accumulation in the nucleus (see Fig. 1C), thus targeting p53 nuclear activities. 2 DNA damage causes G 1 and G 2 /M cell cycle arrest. p53-dependent arrest of cells in G 1 through p21 waf1 expression is an important component of cellular response to stress, although p53-independent mechanisms involving cyclin D1 degradation have been recently reported (18). DNA damage causes an immediate p53-independent G 1 arrest due to a rapid proteolysis of cyclin D1, thus suggesting that DNA damage is at least a two-step process: a fast p53-independent phase mediated by cyclin D1 degradation and a slower maintenance of arrest depending on increased p53 stability. Similarly, it has been shown that genotoxic stress activates p53-dependent and p53independent pathways that contribute to G 2 /M cell cycle arrest (reviewed in Ref. 19). Using different approaches it has been demonstrated that p53 is not required for the initial arrest of human cells in G 2 but plays a crucial role for the long-term maintenance of the arrest (20,21). A role of p53 in maintaining G 2 arrest may be attributable to its ability to transcriptionally repress both cdc2 and cyclin B1 genes (22,23) and to inhibit cdc2 via the p53 target genes: Gadd45, 14 -3-3, and p21 waf-1 (19). The mentioned data explain some mechanisms involved in p53-dependent G 2 arrest. However, emerging evidence could indicate a cell cycle-dependent regulation of p53 function. It has been shown that Mdm2⅐p53 complex was preferentially found in S/G 2 phases of the cell cycle (24). Moreover, as we described for hGTSE-1, 14 -3-3 overexpression induces G 2 cell cycle arrest (25). 14 -3-3 regulates the cdc2⅐cyclin B1 complex and, therefore, the entry into mitosis (26). It has been reported that 14 -3-3 is also able to delay the apoptotic program through sequestration of Bax (27). More recently, another protein with S/G 2 -specific expression, PTTG1/securin, has been shown to be involved in the regulation of p53 (28). Similar to hGTSE-1, PTTG1/securin binds the C-terminal regulatory domain of p53 and represses p53 transactivation and apoptosis. The behavior of proteins such as hGTSE-1, 14 -3-3, and PTTG1/securin involved in both G 2 cell cycle regulation and apoptosis delay imply that DNA damage-induced G 2 arrest may be accompanied by specific mechanisms that control the triggering of the apoptotic program within this phase of the cell cycle.
In conclusion, we observed that hGTSE-1 plays a specific role after DNA damage by controlling p53-dependent functions during S/G 2 . This is supported by the fact that hGTSE-1 silencing resulted in enhanced p53 levels and augmented susceptibility to DNA damage-induced apoptosis in S/G 2 cell population. In this context, we propose that after DNA damage, hGTSE-1 could play a dual role during the G 2 checkpoint, promoting the delay of G 2 to M transition and protecting, at the same time, these cells from p53-dependent apoptosis.