Persistent Activation of c-Jun N-terminal Kinase 1 (JNK1) in γ Radiation-induced Apoptosis

The c-Jun N-terminal kinases (JNK) are activated by various stimuli, including UV light, interleukin-1, tumor necrosis factor-α (TNF-α), and CD28 costimulation. Induction of JNK by TNF-α, a strong apoptosis inducer, implies a possible role of JNK in the regulation of programmed cell death. Present studies show that lethal doses of γ radiation (GR) induced JNK activities at the early phase of apoptosis in Jurkat T-cells. We demonstrate that JNK1 was activated by either the T-cell activation signals, anti-CD28 monoclonal antibody plus phorbol 12-myristate 13-acetate (PMA), or the apoptosis-inducing treatment, GR; however, the induction patterns were different. In contrast to the rapid and transient JNK1 activation caused by CD28 signaling plus PMA, GR induced a delayed and persistent JNK1 activation. This implies a distinct regulatory mechanism and specific function of JNK1 in irradiated cells. The nuclear and cytosolic JNK1 activities were simultaneously increased in the irradiated cells without an evident change in the protein levels. The abilities of GR to induce JNK1 activation and DNA fragmentation were correlated. Peripheral blood lymphocytes were more sensitive to GR than Jurkat cells in JNK1 induction. The responsiveness of JNK1 to GR suggests the involvement of JNK1 in the initiation of the apoptosis process.

Apoptosis is the unique morphological pattern of cell death characterized by chromatin condensation, membrane blebbing, and cell fragmentation. The most prominent event in the early stages of apoptosis is internucleosomal DNA cleavage by undefined endonuclease activities. This programmed cell death is widely observed in different cells of various organisms, from nematodes to mammals. It is generally accepted that apoptosis plays important roles in developmental processes, maintenance of homeostasis, and elimination of cells that have suffered serious damage (reviewed in Refs. 1 and 2).
␥ radiation (GR) 1 is one of many stimuli that induce cellular damage and apoptosis (2). GR can cause single-stranded and double-stranded breaks in the genomic DNA (3). Hydroxyl radicals generated by radiolytic attack on H 2 O in the cellular aqueous environment can cause oxidative damages to macromolecules (4). Although the direct damaging effects of GR have been well studied, the biochemical and genetic mechanisms that initiate the active programmed cell death in radiationdamaged cells remain largely unknown. The p46/p54 serine/threonine kinases, c-Jun N-terminal kinases (JNK1 and JNK2), are emerging members of the MAP kinase-related family (5). Similar to MAP kinase, JNK activation requires phosphorylation at 2 residues, Thr-183 and Tyr-185, by MAP kinase kinase 4/JNK kinase (5), a dual specificity kinase, which is structurally related to MAPK/ERK kinases (MEKs). MAP kinase kinase 4/JNK kinase itself is phosphorylated and activated by the upstream kinase MAPK/ERK kinase kinase 1 (MEKK1) (5). The JNK kinase cascade can be induced by various mitogenic factors including growth factors, oncogenic Ras, phorbol esters, and T-cell activation signaling (5,6). JNK activity is also induced by stimuli such as UV light, protein synthesis inhibitors, osmotic shock, and proinflammatory cytokines (5,7). This kinase cascade was shown to be the common pathway shared by cell proliferation and stress-response signaling. The exact mechanism of how JNK kinase cascade integrates with other signaling pathways to achieve specific response to different stimuli remains to be elucidated. There are three cellular proteins currently known to be phosphorylated by JNK, which are the transcription factors c-Jun (8), ATF-2 (9), and Elk-1 (10). Transcription activities driven by the responsive elements binding these transcription factors are strongly enhanced after JNK activation (5,9,10).
Prominent JNK activation was observed in cells treated with TNF-␣ (7), a potent inducer of apoptosis (2). Moreover, the tumor suppressor p53, which causes apoptosis on some occasions (1), was suggested as an in vivo substrate of JNK1 (11). Based on these findings and the general involvement of JNK in responses to various stresses, we proposed that JNK may be activated by GR. In these studies, we demonstrated that JNK1 was activated in cells exposed to lethal doses of GR. The radiation-induced JNK1 activation showed a unique kinetics in comparison with that induced by other stimuli.
␥ and UV Irradiation and Cell Extract Preparation-Cultured cells were ␥-irradiated by using a 137 Cs source Gammacell 1000 (10 Gy/min) at various dosages. UV irradiation was performed by using a UV Stratalinker 1800 (Stratagene). Whole cell extracts were prepared according to the procedure of Kyriakis et al. (13). Nuclear and cytosolic fractions were collected as described previously (12).
DNA Fragmentation and Western Blot Analysis-10 6 irradiated cells were lysed in 50 l of NTE buffer (100 mM NaCl, 40 mM Tris-Cl, pH 7.4, 20 mM EDTA) containing 0.5% SDS. The lysate was heated at 65°C for 10 min to inactivate nucleases and digested by a 2-h incubation with 0.5 mg/ml proteinase K followed by a 2-h incubation with 0.2 mg/ml RNase A at 50°C. The DNA fragmentation was analyzed on a 1.8% agarose gel in the presence of 0.5 g/ml ethidium bromide. Western blot using anti-JNK1 Ab (1:1,000 dilution) was performed as described previously (12).
Solid-phase, Immunocomplex, and In-gel Kinase Assays-The JNK activities were precipitated by GST-Jun-(1-79), and solid-phase kinase assay was performed as described by Hibi et al. (8). Immunocomplex kinase assay was carried out as described by Kyriakis et al. (13) with some modifications. Cellular JNK1 activities were immunoprecipitated by rabbit anti-JNK1 Ab, Ab101. The precipitates were washed twice with lysis buffer, twice with LiCl buffer (500 mM LiCl, 100 mM Tris-Cl, pH 7.6, and 0.1% Triton X-100), and twice with kinase buffer (20 mM Mops, pH 7.2, 2 mM EGTA, 10 mM MgCl 2 , 1 mM dithiothreitol, 0.1% Triton X-100, and 0.1 mM Na 3 VO 4 ). Then the pellets were mixed with 5 g of GST-Jun-(1-331), 15 M ATP, and 20 Ci of [␥-32 P]ATP in 30 l of kinase buffer. The kinase reaction was performed at 30°C for 30 min and terminated with an equal volume of Laemmli sample buffer. The reaction products were analyzed by 10% SDS-PAGE and autoradiography. The in-gel kinase assay using GST-Jun-(1-331) as the substrate was performed as described by Hibi et al. (8).

GR Induces the JNK Activity in the Early Phase of Apoptosis-Jurkat cells irradiated by 100
Gy were collected at different time points for viability count, DNA fragmentation analysis, and cell lysate preparation. Cellular DNA was analyzed by 1.8% agarose electrophoresis and ethidium bromide staining. The typical nucleosomal DNA ladders appeared 4 h after irradiation, and the intensity of fragmented DNA was continuously increased, indicating the progression of massive fragmentation of chromosomal DNA (Fig. 1A). The viability of irradiated cells dropped below 50% 48 h after irradiation as determined by trypan blue exclusion, with no viable cells observed 4 days after irradiation (data not shown).
To study the possible JNK activation induced by GR, we used GST-Jun-(1-79) bound to GSH-Sepharose beads to precipitate JNK or JNK-like activities from Jurkat cell lysate. The precipitated complexes were washed and subjected to solid-phase kinase assay. As shown in Fig. 1B, GST-Jun phosphorylation activity dramatically increased between 1 and 2 h after irradiation and peaked at 3 h; then it slightly decreased but remained constant through the 6-h time point. The kinase activity was still higher than the basal level 12 h after irradiation. We did not, however, detect a significant increase of GST-Jun phosphorylation activity before the 1-h time point (data not shown). In comparison with the immediate and transient JNK activation induced by other stimuli such as TNF-␣ (7) and UV light (14), the induction of JNK activity by GR was late and sustained.

GR and T-cell Activation Signals Induce Different Kinetics of JNK1
Activation-To determine if the GST-Jun-(1-79) phosphorylation detected by solid-phase kinase assay is due to JNK or other JNK-related kinases, JNK1-specific antiserum (Ab101) was used for the detection of JNK1 activity. As shown in Fig. 2, this antiserum recognizes the 46-kDa JNK1 protein in Jurkat cell extract ( Fig. 2A, lane 1). The anti-JNK1 Ab also recognized the affinity-purified GST-JNK1 protein ( Fig. 2A,  lanes 2 and 3). To further determine the specificity of the anti-JNK1 Ab, we incubated Ab101 and protein A-conjugated agarose beads with cell lysate prepared from Jurkat cells exposed to various JNK stimuli. The precipitated kinase activities were resolved in SDS-PAGE copolymerized with GST-Jun-(1-331) protein, and the in-gel kinase reaction was performed after denaturing and renaturing of the protein gel. As shown in Fig. 2B, single kinase activity around 46-kDa, which phosphorylated GST-Jun, was precipitated by anti-JNK1 from UV light, anisomycin, and GR-treated cell lysates (lanes 2-4, respectively). These results clearly showed that Ab101 specifically recognized only one kinase activity, JNK1, which can phosphorylate the c-Jun protein.
Because the kinetics of c-Jun N-terminal phosphorylation activities induced by GR is very different in comparison with those induced by other stimuli, it is possible that GR may induce several kinases, in addition to JNK, that will phosphorylate the N terminus of c-Jun. The kinetics pattern detected by solid-phase kinase assay could be the additive effect of different kinase activities. To determine the JNK1 activation kinetics, we used the anti-JNK1 Ab for immunocomplex kinase assay. As shown in Fig. 3A, the precipitated JNK1 activity from irradiated cells has the same kinetics as that determined by solid-phase kinase assay. This shows that GR induced a delayed and sustained JNK1 activation. To exclude the possibility that this unique kinetics is the specific property of certain Jurkat cell clones, we used anti-CD28 mAb plus PMA, which are the potent stimuli for T-cell activation (12), to induce JNK activation. As shown in Fig. 3B, after addition of anti-CD28 plus PMA, JNK1 activity rapidly increased within 15 min and reached the peak at the 30-min time point. However, the JNK1 activity induced by anti-CD28 mAb plus PMA diminished sig- GR Induces Nuclear JNK1 Activation without Inducing Nuclear Translocation of JNK1-The mechanism of JNK1 activation after irradiation could be the de novo synthesis of the JNK1 protein or the activation of the pre-existing kinase. Using anti-JNK1 Ab in Western blots, we found that the JNK1 protein was constitutively expressed in non-stimulated Jurkat cells and remained at the same levels after irradiation (data not shown). This indicates that the elevation of kinase activity is not due to an increased production of JNK1 protein.
The known JNK kinase cascade was initiated in the cytosol (5), and it was shown that nuclear events are not required for UV-induced JNK activation (15). Since most of the possible JNK substrates are nuclear transcription factors, we thought that nuclear translocation may be a necessary step for JNK function. To study the translocation of JNK1, we examined the distribution of JNK1 proteins in nuclear and cytosolic fractions after irradiation. In the absence of kinase activity, we detected the constitutive expression of JNK1 protein in the nuclear fraction of unstimulated cells. There is no significant change in JNK1 protein levels in either nuclear or cytosolic fractions after irradiation (Fig. 4A). Although we did not detect an apparent nuclear translocation of JNK1, we did observe a simultaneous increase of JNK1 activity in the nuclear and cytosolic fractions after irradiation (Fig. 4B). This nuclear JNK1 activation is consistent with the expected function of JNK1.
The Ability of GR to Induce JNK1 Activation and Apoptosis Is Correlated-To study the relation between JNK1 activation and induction of apoptosis, Jurkat cells and normal peripheral lymphocytes were exposed to various dosages of GR. If JNK activation is a paralleled phenomenon of apoptosis, there should be a dose-dependent response of JNK activation. Since normal lymphocytes are more susceptible to GR than the Jurkat cells, JNK1 activity should be induced by a lower dose of GR in normal cells than in Jurkat cells. As shown in Fig. 5A, low doses of radiation (5-20 Gy), which did not cause significant DNA fragmentation in Jurkat cells (data not shown), also failed to induce JNK1 activation. JNK1 induction increased as the dosage of irradiation increased and reached the plateau at 60 Gy. However, in normal peripheral lymphocytes, 5 Gy of GR induced a significant JNK1 activation (Fig. 5B). This dosage of radiation, which did not cause detectable DNA fragmentation in Jurkat cells, induced a significant DNA fragmentation in normal lymphocytes (data not shown). These results show that the abilities of irradiation to induce apoptosis and JNK activation are integrated.

DISCUSSION
Because (i) overexpression of MAPK/ERK kinase kinase (MEKK), the JNK-activating kinase, has a lethal effect on fibroblasts (16) and (ii) TNF-␣ strongly induces JNK activity (7), we suspected that the JNK kinase cascade may be involved in the induction of cell death. In these studies we show that JNK1 activity is strongly activated during the early phase of apoptosis caused by GR. Because of the correlation between JNK1 activation and apoptosis induced by GR, we propose that JNK1 activation may be involved in the initiation of programmed cell death in response to radiation damages.
Immediate and transient kinetics of activation is universal in JNK responses to various stimuli (14). However, the JNK1 induction by GR in Jurkat cells was delayed and persistent. Since the T-cell activation signals, anti-CD28 mAb plus PMA, induced a rapid and transient JNK1 activation in the same cell line, the unique pattern of JNK1 induction should be a specific cellular response to GR rather than a unique property of Jurkat cells. The induction of JNK1 in both T-cell activation and apoptosis indicates that JNK1 is the common kinase shared by these two distinct phenomena. However, the opposite outcomes imply that the presence or absence of other co-activators at different times may be important. In addition, prolonged JNK1 induction in irradiated cells may cause the persistent activation of some cellular factors (e.g. c-Jun or p53) and results in detrimental effects to the cells. Therefore, the different timing and/or duration of JNK1 induction may lead to opposite outcomes, T-cell proliferation or apoptosis.
JNK (JNK1/JNK2) and p38-Mpk protein kinases are coordinately regulated, although to a different extent, by proinflammatory cytokines, UV light, and other environmental stresses (7,14). Since the substrate specificities of these kinases are different (14), they may have distinct functions in response to the stimuli. It will be important to determine whether GR induces all of these stress-responsive kinases. It is possible that different combinations of the kinase members, with various kinetics, may mediate diverse cellular signaling and dictate The delayed kinetics of JNK1 induction in irradiated cells also implies the existence of a distinct activation or regulation mechanism. In the known JNK activation mechanism, a nuclear translocation of JNK after activation at the proximity of the plasma membrane is required for nuclear function of the kinase (5). However, we show that nuclear JNK1 is constitutively present and can be activated without evident, nuclear translocation in irradiated cells. The delay of JNK1 activation may be due to the time needed for accumulation of cellular damage to certain threshold levels. Furthermore, the existence of unrepairable damage could be the reason for persistent JNK1 activation, probably through the continued activation of upstream kinases. The other possible explanation for the prolonged JNK1 induction after ionizing radiation is the absence of dual specificity phosphatase activities, which were shown to dephosphorylate ERK and JNK causing down-regulation of the kinase activities (17,18). The imbalance between kinase and phosphatase activities may have detrimental effects and lead to cell death. Finally, identification of JNK1 as a potential signaling molecule in mediating apoptosis will open a new avenue for unraveling the signal transduction mechanisms of apoptosis.