Protective role for c-Jun in the cellular response to DNA damage

c-Jun, a member of the AP-1 family of transcription factors, has been implicated in the regulation of many important biological processes including cell cycle progression, transformation, differentiation, and apoptosis. Accordingly, its expression and function are upregulated in response to diverse stimuli including mitogens and a wide range of stresses. Transcriptional activation of the c-Jun protein is dependent on its phosphorylation at Ser63 and Ser73, a process mediated by c-Jun N-terminal kinase (JNK). Active c-Jun is required for AP-1 transactivation and c-Jun-mediated transformation, but its role during stress remains unclear as both pro-apoptotic and pro-survival effects of c-Jun have been observed. Here we investigated the importance of c-Jun N-terminal phosphorylation in influencing the sensitivity of human T98Gglioblastoma cells to a variety of cytotoxic agents. Stable expression of a nonphosphorylatable dominant negative protein c-Jun(S63A,S73A) markedly inhibited activation of AP-1-driven transcription and greatly increased the cytotoxic effects of DNA-damaging agents associated with enhanced apoptosis. However, the same cells expressing the mutant Jun protein did not differ from parental cells in their sensitivity to several non-DNA-damaging cytotoxic agents. Our results suggest that activated c-Jun serves a selective role in protecting human tumor cells from apoptosis induced by DNA damage. JNK 32


INTRODUCTION
c-Jun, a prominent member of the AP-1 transcription family, has been implicated in the regulation of a wide range of biological processes including development, differentiation, transformation, and apoptosis (1)(2)(3)(4). Its transcriptional activities are regulated by changes in the level of c-jun expression as well as posttranslational modifications of the c-Jun protein. In particular, the transactivation potential of c-Jun critically depends on its phorphorylation of amino acid residues Ser63 and Ser73 (5, 6) mediated by c-Jun N-terminal kinase (JNK 1 ) (reviewed in (1)). Not surprisingly, both c-jun expression and phosphorylation of c-Jun protein are highly induced in response to environmental cues including mitogenic stimuli as well as various stresses (7,8). While the growth regulatory functions of c-Jun have been firmly established, its role in regulating responses to stress is more controversial. where inhibition of c-Jun activity was found to protect cells from nerve growth factor (NGF) withdrawal (9). Consistent with this finding, overexpression of c-Jun in 3T3 fibroblasts resulted in the induction of apoptosis (10). Additional studies utilizing strategies to either over-or under-express c-Jun protein, or block its phosphorylation, have provided evidence that c-Jun is an important mediator of apoptosis (reviewed in (11) induced by various cytotoxic agents including some chemotherapeutic drugs (12) and alkylating agents (13). However, a growing number of studies employing similar strategies to modulate c-Jun expression/activity have suggested that elevated expression and/or activation of c-Jun enhances cell survival during certain conditions of stress. In particular, a protective influence of c-Jun against UVCinduced apoptosis was supported by a recent study employing c-jun -/fibroblasts (14,15). of human T98G glioblastoma cells exposed to a variety of DNA-damaging as well as non-DNA-damaging cytostatic/cytotoxic compounds. We report that T98G cells expressing dnJun exhibit enhanced sensitivity to DNA-damaging treatments, but not to other stressors that act independent of DNA damage to inhibit growth and/or induce cell death. These findings suggest that N-terminus phosphorylation-dependent activation of c-Jun has an important role in protecting human tumor cells against DNA damage-induced apoptosis.

EXPERIMENTAL PROCEDURES
Cell Culture, Treatments, and Viability Assay -Human glioblastoma T98G cells and their dnJun derivatives were previously described (20). Cells were cultured in DMEM (Gibco BRL), supplemented with 10% fetal bovine serum (HyClone). To assess their sensitivity to various stresses, cells were seeded at a density of 1,000 cells/well in 96-well plates and on the following day were treated in the same medium (except in the case of UVC irradiation) with cytotoxic agents for 1 h. Following all treatments cells were washed with PBS and supplemented with fresh complete medium. In the case of UVC treatment, cells were washed with PBS prior to irradiation and fresh medium was added immediately after treatment.
CAT assay -Transient transfections were carried out using the standard calcium phosphate precipitation method as described (22). Briefly, T98G cells were seeded into 24well tissue culture plates at a density of 5x10 3 cells/well and transfected the following day with In vitro kinase activity assay -Cells were treated with various agents and lysates were prepared 30 min later in whole-cell extract buffer (25 mM HEPES pH 7.7, 0.3 M NaCl, 1.5 mM MgCl 2 , 0.1% Triton X-100, 100 .g/ml PMSF, 0.1 mM Na 3 VO 4 , 20 mM aglycerophosphate, 2 .g/ml leupeptin and 0.54 mM freshly added dithiothreitol). The JNK kinase assay was performed as described before (23)  These disks were subsequently washed and the amount of radioactive DNA was quantitated by scintillation counting using Biosafe-II scintillation liquid.

Expression of phosphorylation-deficient c-Jun(S63A,S73A) in T98G cells results in growth inhibition and reduced AP-1 transactivation -To investigate the importance of c-Jun
N-terminal phosphorylation in determining the sensitivity of human tumor cells to various stresses, T98G cells that overexpress a nonphosphorylatable dominant negative mutant c-Jun(S63A,S73A) protein, dnJun, were generated (20). The levels of c-Jun protein in two such dnJun cell lines (clones 10-6 and 10-10) as well as parental T98G cells transfected with empty vector (T98GX) are shown in Fig. 1A. The T98GdnJun clones showed approximately 5-6 fold higher amounts of total c-Jun protein (endogenous c-Jun and dnJun proteins combined) relative to T98GX cells, indicating a high level of dnJun expression. The expression of the dnJun in T98G cells was associated with growth inhibition in clonal cell lines, characterized by slower rates of proliferation and reduced saturation densities (Fig. 1B).
The effect of dnJun on cell growth could result from at least two different mechanisms.
Nonphosphorylatable dnJun could act as a dominant negative inhibitor of AP-1 transcriptional activity, as has been previously suggested (5,6). Alternatively, given that JNK is proposed to bind tightly to c-Jun and release it only after phosphorylation (24), dnJun could block JNK activity directly by acting as a pseudo-substrate and tying up JNK protein. To explore the first possibility, the AP-1 transcriptional activity of T98G cells expressing dnJun was examined using transiently-transfected CAT-based reporter constructs linked to various AP-1 binding elements to which c-Jun has been shown to interact (25,26). The JNK pathway was then specifically stimulated by expression of a constitutively active mutant form of the upstream kinase, MEKK1 (27). This resulted in strong activation of AP-1 (6xAP-1 and 3xTRE) and ATF/CREB (5xjun2) binding site reporter constructs (Fig. 1C, open bars) without affecting CAT expression driven by a control promoter 56Fos∆E, which lacks a functional AP-1 site Expression of phosphorylation-deficient dnJun sensitizes T98G cells to DNAdamaging agents, but not to other cytotoxic compounds -Assessment of long-term (5-7 days) growth effects in cells subjected to 1-h exposure to different cytotoxic compounds was performed using an MTS tetrazolium dye conversion assay measuring viable cell mass (Promega). A reduction in viable cell mass can occur as result of growth inhibition or cell death. A variety of agents known to act via different mechanisms to damage DNA were tested ( Table 1). As shown in Fig. 2, cells harboring dnJun were more sensitive than their normal counterparts to the growth inhibitory effects of all of the DNA-damaging agents tested, regardless of the compound's mode of action. Parental and dnJun-expressing T98G cells were also examined for their sensitivity to several other agents known to exert growth inhibitory or toxic effects on cells unrelated to DNA damage. These included the microtubule-disrupting agent taxol, the anti-estrogen agent tamoxifen, and two agents known to cause stress to the endoplasmic reticulum, thapsigargin and tunicamycin ( Table 1). As seen in Fig. 3, no differences in sensitivity to any of these agents were seen when comparing the results for T98GX and T98GdnJun cells.

DNA synthesis, cell cycle progression and apoptosis in T98GX and T98GdnJun cells
-A number of different mechanisms such as (i) impaired DNA synthesis, (ii) perturbations in cell cycle progression, and (iii) induction of apoptosis could contribute to the reduction in viable cell mass seen following treatment with cytotoxic agents. c-Jun activity could influence such effects thereby accounting for the differential sensitivity of T98GX and T9GdnJun cells  Fig. 6A), although a small but significant (p < 0.02, Student's t-test) difference in the proportion of cells in the G1 and S phases was noted (Fig. 6B). The greater proportion of T98GdnJun cells in G1 with a reduced number in the S phase compartment could reflect the slower growth rate of the T98GdnJun cells. Cisplatin treatment resulted in an increased accumulation of cells in S phase and loss of cells in the G2 compartment in both T98GdnJun and T98GX cultures. However, consistent with the morphological analysis above, cisplatin treatment of dnJun-expressing cells resulted in the appearance of a significant sub-G1 compartment that was much less apparent in T98GX cells ( Fig. 6C, upper panel). A similar effect was noted following etoposide treatment (data not shown). Treatment of T98GX and T98GdnJun cells with thapsigargin resulted in a very different profile from that seen with cisplatin. Both parental T98GX and dnJun-expressing T98GdnJun cells became arrested in the G1 phase of the cell cycle with and no accumulation of cells in sub-G1 region (Fig. 6C, lower panel).
Additional confirmation of the differential extent of apoptosis in T98GX and T98GdnJun cells following treatment with DNA-damaging agents was obtained using an ELISA-based assay that detects the presence of histone-associated DNA fragments in the cytoplasm. As shown in Fig. 6D for the T98GdnJun10-10 clone, all three of the DNAdamaging agents tested caused greater amounts of apoptosis in T98GdnJun cells relative to T98GX, while no differences in the levels of apoptotic signals were detected in T98GX and T98GdnJun cultures treated with taxol, thapsigargin or tunicamycin (data not shown).
In summary, both long-term viability and short-term apoptosis assays indicated that overexpression of phosphorylation-deficient dnJun protein renders human T98G glioblastoma cells much more sensitive to from DNA-damaging agents relative to their normal counterparts.
These results support an important role for c-Jun in cellular protection against such insults.  (Fig. 1C). It is interesting to note that overexpression of phosphorylation-deficient dnJun had no inhibitory effect on JNK activation under conditions of genotoxic stress (Fig. 1D) suggesting that the effects described Examining the responsiveness of normal and dnJun-expressing T98G cells (T98GX and T98GdnJun, respectively) to different cytotoxic agents we observed no differences in their sensitivities to any of four compounds known to act independent of DNA damage to exert the cytotoxic effects (Fig. 3). This result is consistent with the previous observation that activation of the MEKK1 pathway has no role in the induction of apoptosis by microtubule altering drugs (16). However, T98GdnJun cells displayed much greater sensitivity to all of twelve different DNA-damaging agents tested (Fig.2, data not shown). Interestingly, this effect was seen regardless of the agent's particular mode of action; i.e., agents capable of causing single and double strand DNA breaks such as UVC (Fig. 2B) and γ-irradiation (data not shown), agents creating DNA adducts (cisplatin), alkylating agents (MMS, MNNG), topoisomerase inhibitors (etoposide), and agents that cause extensive oxidative damage (streptonigrin) (summarized in Table 1). This suggests that c-Jun or some other AP-1 component whose activity is subject to interference by the presence of the mutant c-Jun protein serves a general protection function during the response to DNA damage. These findings appear to contradict the more widely held view of c-Jun as a mediator of apoptosis (10)(11)(12). However, like ours, several previous studies have suggested a protective function for c-Jun following DNA damage. These include reports demonstrating that abrogation of c-Jun induction renders cells more sensitive to DNA damage (21,29), and that elevated c-Jun expression render cells more resistant to doxorubicin-induced cell death (30).
One possible way in which c-Jun could influence cell survival during the response to DNA damage is through affecting DNA repair capacity. Indeed, in a previous study (20), we provided evidence suggesting that inhibition of the JNK pathway affects the ability of T98G by guest on March 24, 2020 http://www.jbc.org/ Downloaded from cells to repair cisplatin-generated DNA adducts (20). In association with the present study we examined the efficiency of DNA repair following treatment with 50-100 µg/ml etoposide in T98GX and T98GdnJun cells using a Comet assay (31). This single cell DNA repair assay allows assessment of DNA repair of double strand DNA (dsDNA) breaks. While one of the two T98GdnJun cell clones (clone 10-6) employed in the current studies exhibited reduced repair of etoposide-induced dsDNA breaks compared to control cultures 2 h following the treatment, the other T98GdnJun cell line did not (data not shown). Thus, it appears unlikely that differential repair capacity is the explanation for the enhanced sensitivity of T98GdnJun cells to DNA-damaging treatments. However, it is worth noting that although dsDNA breaks is the prevailing form, it is not the only type of DNA damage caused by etoposide, and it is conceivable that other repair activities could also influence the response. Human cells have multiple systems responsible for repair of various types of DNA damage and obviously more detailed investigations will be necessary to determine what, if any, relationship exists between c-Jun and DNA repair in influencing sensitivity of dnJun-expressing T98G cells to DNA damaging agents.
As reported here, T98GdnJun cells exhibit greater apoptotic changes following treatment with DNA-damaging treatments than do their wild type counterparts. Evidence of enhanced apoptosis in T98GdnJun cells following treatment with the DNA damaging agents included (i) general morphological appearance and detachment of cells from the tissue culture plates (Fig. 5, data not shown), (ii) higher levels of fragmented DNA as assessed in an ELISAbased assay (Fig. 6D); (iii) higher numbers of fragmented nuclei as visualized by DAPI staining (Fig. 5), and (iv) a greater proportion of cells in the sub-G1 compartment as determined by FACS analysis (Fig. 6C). No such differences were noted in cultures treated with non DNA-damaging agents. For example, treatment with thapsigargin resulted in a   with the designated agents. Viable cell mass was assessed 5-7 days later using an MTS dye assay, and is expressed as the percentage of the value obtained in a parallel untreated culture.
All assays were carried out in triplicate and repeated at least four times (shown is a representative experiment for each treatment).    Cells were seeded at a density of 5x10 4 cells/cm 2 , and treated the following day for 1 h with either 100 µM cisplatin, 1 µg/ml doxorubicin, or 50 µg/ml etoposide. Apoptosis was assessed 24 h later using an ELISA-based assay that measures levels of degraded nuclear DNA in the cytoplasm. The assay was performed as specified by the manufacturer (Boehringer Mannheim) and each treatment point was assayed out in triplicate. The results are expressed as percentage of values seen in untreated parallel cultures of the same origin.