An Ultraviolet-activated K+ Channel Mediates Apoptosis Of Myeloblastic Leukemia Cells*

Exposure of mammalian cells to UV light causes initial changes in the cell membrane, induces phosphorylation and clustering of growth factor/cytokine receptors, and activates the Jun N-terminal kinase/stress-activated protein kinase (JNK/SAPK) signaling pathway leading to programmed cell death (apoptosis). In this study, we found that an early event in the cell membrane of myeloblastic leukemia (ML-1) cells was the vigorous activation of the voltage-gated K+ channel by UV irradiation. The strong enhancement by UV irradiation of K+ channel activity in the cell membrane subsequently activated the JNK/SAPK signaling pathway and resulted in myeloblastic leukemia cell apoptosis. Suppression of UV-induced K+ channel activation with specific channel blockers prevented UV-induced apoptosis through inhibition of UV-induced activation of the proteins SEK (SPAK kinase) and JNK. However, suppression of K+ channel activity could not protect cells from etoposide-induced apoptosis, which bypasses the membrane event. Elimination of extracellular Ca2+ had no effect on the UV-induced and K+ channel-mediated JNK/SAPK activation. Thus, we have identified a novel mechanism in which activation of K+ channels by UV-irradiation upstream of SEK and SAPK/JNK mediates UV-induced myeloblastic cell apoptosis.

Exposure of mammalian cells to UV irradiation causes programmed cell death and cancers. Early responses to UV irradiation include the activation of transcription factors, Ap-1 and NF-B (1, 2), and of immediate early genes, c-fos and c-jun (3,4). This activation, which is known as the UV response, is mediated by activation of intracellular signaling pathways that are shared with growth factors. Erk 1/2 , JNK/SAPK 1 , and p38 are three mitogen-activated protein kinase pathways that can be activated by UV irradiation (5,6). The degree of activation, however, varies. UV irradiation strongly increases JNK/SAPK activity but only modestly increases Erk 1/2 activity in opposed to growth factors such as epidermal growth factor (7)(8)(9)(10).
What upstream events of this signaling transduction are induced by UV irradiation? Ras, a small G protein and transmitting signal from membrane to cytoplasm, is activated by UV irradiation and mediates UV-induced activation of JNK, Erk, and transcription factors (6,(11)(12)(13)(14). Src, a nonreceptor tyrosine kinase, is stimulated by UV irradiation (13). Exposure to UV light induces clustering and internalization of cell surface receptors for epidermal growth factor, tumor necrosis factor, and interleukin 1. Inhibition of clustering or receptor down-regulation attenuates the response to UV (7). UV irradiation induces ligand-independent activation of numerous receptor tyrosine kinases such as epidermal growth factor, platelet-derived growth factor and insulin receptors, and protein-tyrosine kinases at the inner side of the plasma membrane (13,(15)(16)(17)(18). Other membrane-associated proteins, the protein-tyrosine phosphatases, can be inhibited with UV irradiation by targeting an essential -SH group in the tyrosine phosphatase. This results in inhibition of dephosphorylation and enhancement of autophosphorylation of epidermal growth factor receptor and platelet-derived growth factor receptor (18). It has been demonstrated in enucleated cells that UV activation of NF-B and JNK does not require a nuclear signal (19). These two observations, the full response of enucleated cells to UV irradiation and the involvement of membrane-associated proteins in the UV response, strongly suggest that important UV-induced cell events probably occur through initiation of conformational changes in the plasma membrane.
Several recent studies have implied that potassium (K ϩ ) plays an important role in the regulation of programmed cell death. A bacterial pore-forming cytolysin, staphylococcal ␣-toxin, which selectively permeabilizes plasma membranes for monovalent ions, appeared to induce apoptosis (20). In contrast, diminishing the normal K ϩ electrochemical gradient completely nullified the ability of the anti-Fas antibody to induce apoptosis in Jurkat cells (21). Apoptotic cells and shrunken cells have a much lower intracellular K ϩ concentration compared with normal cells (22)(23). Furthermore, both DNA autodigestion and nuclease activity of thymocytes are suppressed by an increase in extracellular K ϩ concentration in a dose-dependent manner. The complete inhibition can be reached at a concentration of 150 mM extracellular K ϩ , close to the intracellular K ϩ concentration found in normal cells (23).
The question that naturally arises, then, is in regard to the existence of a relationship between K ϩ channel activity and UV irradiation-induced apoptosis in ML-1 cells. Because voltagegated K ϩ channel activity involved in cell proliferation (24,25) is regulated by growth factors (26) and is associated with Srctyrosine kinase (27), we propose that activation of the K ϩ channel can mediate UV irradiation-induced apoptosis. To address this hypothesis, we first observed the effects of UV-C light on the K ϩ channel in myeloblastic leukemia cells (ML-1) by using whole-cell and cell-attached patch recording techniques. Then, we investigated the role of K ϩ channels in UVinduced cell death. Finally, we examined the effect of K ϩ channel activity on UV-stimulated JNK pathway. Our results show that UV irradiation activated K ϩ channels at both whole-cell and single-channel levels. Blockade of K ϩ channels with 4-aminopyridine (4-AP) almost completely prevented UV-induced apoptosis and suppressed UV-stimulated JNK pathway, indicating that UV-activated K ϩ channels do mediate apoptosis in myeloblastic leukemia cells.

MATERIALS AND METHODS
Cell Culture-ML-1 cells originally isolated from an acute myeloblastic leukemia patient were received as a generous gift from Dr. R. W. Craig (Dartmouth Medical School, NH). All myeloblastic leukemia cells were grown in Roswell Park Memorial Institute (RPMI) 1640 culture medium containing 7.5% heat-inactivated fetal bovine serum and 25 mM HEPES buffer. Cells were grown in suspension culture in a humidified incubator at 5% CO 2 , 37°C. FDC-P1 murine myeloid progenitor cells were maintained by incubating at 37°C (5% CO 2 ) in RPMI 1640 medium supplemented with 25 mM HEPES buffer, 0.0004% (v/v) ␤-mercaptoethanol, 10% fetal bovine serum, and 10% conditioned medium from WEHI-3b cells as a source of interleukin 3.
Patch Clamp Experiments-Patch pipettes with a resistance of 3-4 M⍀ when filled with 150 mM KCl solution were manufactured with a two-stage puller (PP-83, Narishige). For whole cell K ϩ current recording, the nystatin perforated patch technique was used. The pipette tip was filled with a solution containing 140 mM KCl, 0.05 mM CaCl 2 , 2 mM Apoptosis Induction-Myeloblastic leukemia cells at a concentration of 3 ϫ 10 5 cells/ml were incubated with complete culture medium. K ϩ channel blockers were added into the culture medium to a final concentration of 2.0 mM. For UV irradiation experiments, cells were placed in a tissue culture hood at a distance of 60 inches from the UV-C light source and exposed at an intensity of 40 mW/cm 2 for 3 to 8 min (60 to 72 J/m 2 ). For exposure to etoposide (an apoptosis inducer), a stock solution of 10 mg/ml etoposide was added to the culture medium at a final concentration of 20 g/ml. After etoposide and UV treatments, cells were incubated at 37°C in 5% CO 2 for 15 to 24 h. Cell viability was measured using the trypan blue dye exclusion method.
Apoptosis Detection Assays-Cell apoptosis was detected by DNA fragmentation and nuclear staining with ethidium bromide/acridine orange. To determine internuclosomal DNA cleavage, myeloblastic cells were washed twice with phosphate-buffered saline. Lysis buffer (200 mM Tris-HCl, pH 8.0, 100 mM EDTA, 1% SDS, and 100 g/ml proteinase K) was added, and cells were then incubated for 4 h at 55°C. The nuclear lysates were extracted twice with an equal volume of phenol and then extracted with an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1). DNA was precipitated with 0.05 volumes of 5 M NaCl and 2.5 volumes of absolute ethanol, incubated overnight at Ϫ20°C, and centrifuged at 13,000 ϫ g for 10 min at 4°C. The DNA pellet was dried and dissolved in TE buffer (10 mM Tris-HCl, pH 7.5, 1 mM EDTA) containing 20 g/ml RNase A and incubated for 1 h at 37°C. The DNA was extracted with an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1). DNA samples were analyzed by electrophoresis on 1.5% agarose gels, and the results were visualized by staining with 1 g/ml ethidium bromide. PstI-digested DNA was used as a molecular weight marker. Nuclear staining with ethidium bromide and acridine orange was done by adding 2 l of dye mixture containing 100 g/ml each acridine orange and ethidium bromide to 25 l of cell suspension. Cell populations were scored according to color using a UV-fluorescence microscope (Nikon). Nuclei staining green have not lost membrane integrity. In contrast, myeloblastic cells in which the nuclei stained orange have lost membrane integrity. Apoptotic cells can be distinguished from nonapoptotic cells on the basis of the absence or presence of nuclear condensation/fragmentation.
Immunoblotting and Kinase Assays-SEK-1 (SPAK kinase 1) activity was determined by measuring the level of SEK-1 phosphorylation with Western blotting using monoclonal antibody (1:1000) against phosphorylated SEK-1 (New England Biolabs, Beverly, MA). After proper treatments, 1 ϫ 10 6 cells (5 ϫ 10 5 cells/ml) were lysed in 20 l of Laemmli buffer. Western blotting was performed using the same protocol as described below. Phospho-SEK-1 levels were quantified by measuring film densities with a densitometer.
JNK-1 or p38 protein levels were determined by Western blotting. Briefly, an equal volume of 2ϫ Laemmli buffer was added to 20 l of immunocomplex and boiled for 5 min. After fractionation on a 12% SDS-polyacrylamide gel electrophoresis gel, proteins were transferred to a polyvinylidene difluoride membrane (Millipore) and incubated with the same antibodies (1:5000) used for JNK-1 or p38 immunoprecipitations. The membranes were then incubated with goat anti-rabbit immunoglobulin IgG conjugated with alkaline phosphatase (1:10,000) (Santa Cruz Biotechnology). Secondary antibodies were detected with a Phototope-Star Western blot detection kit (New England Biolabs).

UV Irradiation
Stimulates K ϩ Channel Activity-Changes in cell membrane K ϩ channel activity can mediate functional adaptation to a variety of chemical and physical stresses through membrane voltage stabilization and maintenance of salt and water balance. We found that cytokine-mediated stimulation of proliferation in myeloblastic ML-1 cells is associated with increases in K ϩ channel activity. This channel activity is sensitive to inhibition by 4-AP but less sensitive to inhibition by Ba 2ϩ and tetraethyleneammonium (TEA) (24). Using the nystatin-perforated whole cell technique, the whole-cell current was activated by depolarization of the membrane potential from a holding potential of Ϫ60 to ϩ80 mV in 20-mV increments. Upon exposure of ML-1 cells to UV-C light for 1 min, the amplitude of the K ϩ current increased markedly. UV-evoked K ϩ current was sensitive to 4-AP and was completely blocked by 2 mM 4-AP (Fig. 1, A and B). The time course showed that the amplitude of the K ϩ current doubled within 1 min after exposure to UV light and reached the maximal amplitude within 5 min. In the presence of 4-AP, UV stimulation of K ϩ currents was blocked, following the time course shown (Fig. 1C).
To further confirm the effect of UV irradiation on single K ϩ channel activity, the cell-attached patch clamp was used. The single-channel current was recorded at a membrane potential of Ϫ60 mV in vivo (Fig. 1D). Exposure to UV-C irradiation (ϳ45 J/m 2 ) strongly stimulated K ϩ channel activity (NP o ). Activity increased from 9.6 Ϯ 1.6% to 68.0 Ϯ 5.6% within 30 s (Fig. 1E). In seven independent patches with 100 M 4-AP in the patch pipette, UV irradiation failed to activate K ϩ channel activity, and NP o remained unchanged at 12.2 Ϯ 3.5% (Figs. 1, D and E). The addition of 20 mM TEA in the patch pipette reduced the UV-activated channel activity to 28.8 ϩ 4.9% (n ϭ 4, Fig. 1E). These results suggest that an early effect of UV irradiation is the direct stimulation of cell membrane K ϩ channel activity in ML-1 cells.
Effect of Suppressing K ϩ Channel Activity on UV-induced Apoptosis-To determine whether UV-induced K ϩ channel hyperactivity is a component of the cell signaling pathway mediating UV-induced apoptosis, the effect of blocking K ϩ channel activity with the K ϩ channel inhibitor 4-AP was determined by measuring cell viability after UV irradiation in the presence or absence of 4-AP. In the presence of 2 mM 4-AP, cell viability was protected from UV irradiation (99.5 Ϯ 0.5% for control, 96.1 Ϯ 0.7% for 4-AP-treated, 32.0 Ϯ 6.7% for UV-induced, and 92.6 Ϯ 1.6% for UV plus 4-AP) ( Fig. 2A). In contrast, 4-AP had no protective effect on cells treated with another apoptosis-inducer, etoposide (an inhibitor of topoisomerase II). With etoposide alone, viability decreased to 59.4 Ϯ 3.2%. This decline was indistinguishable from the effect of etoposide measured in the presence of 4-AP (61.5 Ϯ 0.4%) (Fig. 2A).
The protective effect of 4-AP on UV-induced cells was a time-dependent process. The addition of 4-AP before or at the onset of UV irradiation completely prevented cell death from UV irradiation. However, when 4-AP was added 5 s after the onset of UV irradiation, 30% of the cells died (Fig. 2B). We also found that four other types of myeloblastic cells (FDC-P1, U937, HL-60, and Himeg-1) exhibited 4-AP-sensitive K ϩ channel activity (data not shown). After exposure of these cells to UV irradiation, their viabilities decreased to 57.7 Ϯ 3.6%, 8.3 Ϯ 0.6%, 27.0 Ϯ 0.3%, and 31.0 Ϯ 7.0%, respectively. The suppres- Whole-cell currents were activated by exposure of ML-1 cells to UV light in the absence and presence of 4-AP. The membrane potential was depolarized from a holding potential of -60 mV to ϩ80 mV at 20-mV increments. B, current-voltage relationship of the 4-AP-sensitive K ϩ current activated by UV light. C, time course of UV-activated K ϩ current in the absence and presence of 4-AP. Currents were normalized as I UV /I C where I UV and I C represent amplitudes of the K ϩ current measured before and after UV irradiation, respectively. D, single channel recording of K ϩ channel in ML-1 cells. Outward current recorded as an upward deflection was obtained from cell-attached patches at a membrane potential of Ϫ60 mV. UV-C irradiation was directly applied to the patch chamber to activate K ϩ channels in the same patch. The bottom trace demonstrates that UV irradiation induced an increase in K ϩ channel activity that was blocked by application of 100 M 4-AP in the patch pipette. Channel activity (NP o ) was plotted as a function of time in the lower panel. E, statistics of K ϩ channel activity stimulated by UV irradiation and blocked by 100 M 4-AP or by 20 mM TEA. Vertical bars represent mean K ϩ channel activity (horizontal bars represent S.E.). An asterisk represents a significant difference (statistical tests: ANOVA and Tukey, p Ͻ 0.001). Data were collected from seven independent experiments. sion of K ϩ channel activity with 4-AP had the same protective effect on UV-induced apoptosis in these cells as it did in ML-1 cells (Fig. 2C).
The effects of blocking channel activity on UV-and etoposide-induced cell death were evaluated using DNA fragmentation and nuclear staining methods. Suppression of K ϩ channel activity with 4-AP completely prevented UV-induced DNA fragmentation in ML-1 cells but did not prevent etoposideinduced DNA fragmentation (Fig. 2D). The suppression of UVinduced DNA fragmentation was also observed in four other myeloid leukemia cell lines (Fig. 2E). The protective effect of 4-AP against UV-and etoposide-induced apoptosis was evaluated based on the extent of nuclear staining with ethidium bromide/acridine orange. Exposure of cells to UV irradiation and etoposide resulted in orange-stained nuclei indicating nuclear death (Fig. 2F). Suppression of K ϩ channel activity with 4-AP protected the cells against UV irradiation-induced nu-clear death, whereas it was ineffective in preventing etoposideinduced nuclear death. These results reveal that UV irradiation elicits K ϩ channel hyperactivity, which, in turn, mediates apoptosis. The ability of etoposide to induce apoptosis despite the presence of 4-AP is consistent with its known inhibition of topoisomerase II activity at the level of the nucleus.
Effect of Suppressing K ϩ Activity on UV-activated JNK Signaling Pathway-Many cells respond to UV irradiation by activating their JNK signaling pathway. To determine whether UV irradiation-induced K ϩ channel hyperactivity is an essential component in the UV-activated JNK signaling pathway, JNK-1 activity was measured after suppression of K ϩ channel activity. JNK-1 was strongly activated after 5 min of UV irradiation (Fig. 3A). In contrast, activation of JNK-1 by UV irradiation was almost completely prevented when K ϩ channel activity was suppressed with 2 mM 4-AP. In addition, UVinduced JNK activation was partially inhibited, falling to 62 FIG. 1-continued and 24% of its original activity when K ϩ channel activity was suppressed with either 10 mM TEA or 5 mM Ba 2ϩ , respectively (Fig. 3A). The fact that 4-AP was the most potent protective agent is consistent with its rank as the most potent inhibitor of this particular type of K ϩ channel activity in these cells (24,26). The suppression by 4-AP of K ϩ channel activity and of JNK-1 activation was dose-dependent and reached its maxi-mum inhibitory effect at 1 mM (Fig. 3B). The suppressive effect of 4-AP on UV-induced JNK activity closely corresponds to its dose-dependent inhibition of the K ϩ current in ML-1 cells (26).
SEK is a specific protein kinase in the JNK signaling cascade that phosphorylates and activates JNK (30,31). To confirm the involvement of UV-stimulated K ϩ channel hyperactivity in mediating events upstream from SEK, the relationship between the phosphorylation state of SEK and K ϩ channel activity was characterized. The SEK-1 protein was strongly phosphorylated after a 5-min exposure to UV irradiation (Fig. 3C). Suppression of UV-induced K ϩ channel activity with either 4-AP or Ba 2ϩ inhibited SEK-1 phosphorylation by 70 and 16%, respectively. These results indicate that suppression of UV-induced K ϩ channel activity specifically inhibits the early events in the cell membrane upstream from JNK.
Another stress-activated MAP kinase, p38, is also activated in response to UV irradiation (32,33). To test for K ϩ channelmediated p38 activation in response to UV irradiation in ML-1 cells, the effect of UV irradiation was measured on p38 activity in the presence and absence of K ϩ channel blockers. Activation of p38 occurred irrespective of the presence or absence of 4-AP (Fig. 3D). Our results strongly suggest that UV-stimulated K ϩ channel hyperactivity is an essential upstream component of the JNK signaling pathway; however, p38 activation is not linked to stimulation of K ϩ channel activity. The mechanism underlying UV-induced p38 activation and apoptosis remains to be elucidated.
Effects of Osmotic Stress and Ca 2ϩ Influx on JNK Activity in K ϩ Channel-suppressed Cells-To further support the notion that 4-AP is not a nonspecific inhibitor of JNK pathways, we examined the effect of 4-AP on JNK-1 activation induced by hyperosmolarity in ML-1 cells. Hyperosmotic shock (600 mM sorbitol) strongly activated JNK-1. 4-AP had no effect on JNK activity (Fig. 4A). This result indicates that 4-AP has a specific inhibitory effect on UV-induced JNK activation. In addition, it FIG. 3. Effects of suppressing K ؉ channel activity on JNK and p38 kinase activities. A, protein immunoblot analysis and kinase assay of JNK-1 activation induced by UV irradiation. After incubation with 2 mM 4-AP, 10 mM TEA, or 5 mM Ba 2ϩ in normal culture medium at 37°C, ML-1 cells were exposed to UV irradiation for 5 min at room temperature. After an additional 10-min incubation at room temperature, cells were collected, and JNK-1 activity was measured by immunocomplex kinase assays. JNK-1 protein levels were analyzed by Western blot as shown in the lower panel. B, dose-dependent inhibition of UV-induced JNK-1 activity by suppression of K ϩ channel activity with 4-AP. JNK-1 kinase activity in ML-1 cells was measured after UV irradiation, and JNK-1 protein concentrations were measured by Western blot (lower panel). Nonirradiated cells served as controls. C, prevention of UV irradiation-induced SEK-1 phosphorylation by suppression of K ϩ channel activity. Western blotting was performed using the anti-phospho-SEK antibody. SEK-1 activity was determined by measuring the level of SEK-1 phosphorylation with Western blotting. Phospho-SEK-1 levels were quantified by densitometry. All of the kinase activity data were repeated in three to four independent experiments and were quantified on the basis of band density. D, effect of suppressing K ϩ channel activity with 4-AP on p38 kinase activity in response to UV irradiation. Kinase activity of p38 was determined by immunocomplex kinase assays; protein concentration of p38 is shown in the lower panel. JNK-1 and P38 activities were measured by immunocomplex kinase assay with GST-ATF-2 as the substrate. ATF-2 phosphorylation was quantified by densitometry. JNK-1 or p38 protein levels were determined by Western blotting. Data shown are representatives from three to four independent experiments.
suggests that the effect of 4-AP on UV-induced JNK-1 activation is likely through blockage of K ϩ channel activity and that activation of JNK-1 by hyperosmotic shock in these cells may not be a K ϩ channel activity-mediated process.
Recent studies have shown that an increase of Ca 2ϩ influx may be a component of the signaling mechanism for mediating UV-induced apoptosis. Accordingly, we examined whether UV irradiation could induce JNK stimulation when extracellular Ca 2ϩ concentration was reduced by the addition of EGTA. Our results showed that at a very low Ca 2ϩ concentration (0.5 mM EGTA) or in a nominally Ca 2ϩ -free medium (5 mM EGTA) JNK activation still occurred in response to UV irradiation (Fig. 4B). In addition, suppression of K ϩ channel activity with 4-AP inhibited UV-induced JNK activation in Ca 2ϩ -free medium. Therefore, Ca 2ϩ influx did not play a significant role in the JNK signaling pathway mediating UV-induced apoptosis in ML-1 cells. DISCUSSION We provide evidence for a novel mechanism of UV irradiation-induced apoptosis in myeloblastic leukemia cells. An important early component of the signaling process mediating UV-induced apoptosis is strong activation of cell membrane K ϩ channels. There is growing evidence showing that K ϩ channel activities are probably involved in programmed cell death. Various investigations have shown that K ϩ channel activity can be affected by apoptosis inducers, including reactive oxygen species (34 -37), Fas ligand and tumor necrosis factor (38,39), and anticancer drugs (40,41). The K ϩ channel blocker 4-AP inhibits the shrinkage of human eosinophils undergoing apoptosis induced by cytokine withdrawal (42), and a combination of two K ϩ channel blockers, TEA and 4-AP, inhibited interleukin 1b release from lipopolysaccharide-stimulated monocytes (43). Neurons undergoing apoptosis exhibited an up-regulation of outward K ϩ currents. This enhancement of outward K ϩ current, induced by serum deprivation and staurosporine, can be prevented by the K ϩ channel blocker TEA and by increasing the extracellular K ϩ concentration. It has also been observed that the K ϩ channel opener cromakalim induces neuronal apoptosis (44). Thus, it appears that the activation of K ϩ channels is responsible for K ϩ efflux and the consequent membrane hyperpolarization and decrease in cell volume, thereby activating a particular signaling system leading to 1␤-converting enzyme activation and apoptosis.
The stimulation of K ϩ channel activity could result in the quick loss of intracellular K ϩ . The loss of intracellular K ϩ activates interleukin 1␤-converting enzyme (21,43,45). Some evidence suggests that interleukin 1␤-converting enzyme can affect upstream events in the JNK pathway at the JNK level (46). UV-induced activation of interleukin 1␤-converting enzyme and JNK-1 could occur subsequent to the stimulation of K ϩ channel activity and the loss of intracellular K ϩ . This mechanism has been implicated in apoptosis in neuronal cells (47,48). Alternatively, cell shrinkage that occurs as a result of a quick fall in intracellular K ϩ concentration, may trigger apoptosis. Accordingly, suppression of K ϩ channel activity may prevent a quick loss of intracellular K ϩ ions resulting from UV-induced K ϩ channel hyperactivity. This possibility is supported by recent findings that UV irradiation-induced JNK activation can be mimicked by hypertonic stress in HeLa cells (7). In addition, cytokine receptors can be activated by either UV irradiation or hypertonic stress. It is speculated that cytokine receptor activation induced by hypertonic stress occurs as a consequence of cell shrinkage.
UV-induced apoptosis in ML-1 cells is dependent on stimulation of SEK/JNK and p38 pathways in ML-1 cells. We found that activation of 4-AP-sensitive K ϩ channel activity occurs upstream from the stimulation of the SEK/JNK pathway and that p38 stimulation is a component of the cell signaling systems responsible for UV-induced apoptosis. Activity of p38, however, resides in a signaling pathway parallel to that of SEK/JNK, as shown by the observation that inhibition of K ϩ channel activity with 4-AP has no effect on p38 activity. This result suggests that p38 stimulation may not be linked to activation of this type of K ϩ channel activity. Our finding that UV-induced K ϩ channel hyperactivity precedes SEK and JNK stimulation documents for the first time a role for membrane ion channels in mediating radiation-and cytokine-induced signal transduction and apoptosis. This process, as well, has been shown to account for serum deprivation-induced neuronal cell apoptosis and occurs in the absence of Ca 2ϩ influx across the cell membrane (44). Some studies have shown that JNK acti- FIG. 4. Effects of hyperosmotic shock and extracellular Ca 2؉ on JNK-1 in ML-1 cells. A, effect of 4-AP on JNK activity induced by hyperosmotic shock. Hyperosmolarity shock (Hi osm) was performed by application of 600 mM sorbitol in culture medium for 10 min in the presence and absence of 2 mM 4-AP. Cells were then harvested and measured for JNK-1. Height of bars represents mean values of JNK-1 activity and is quantified on the basis of band density. B, effect of extracellular Ca 2ϩ on JNK-1 activation in response to UV irradiation. ML-1 cells were exposed to UV irradiation in the presence (4th and 6th lanes from the left) or absence of 4-AP (1st-3rd and fifth lanes from the left). Extracellular Ca 2ϩ was removed by preincubation of cells with either 0.5 mM or 5 mM EGTA (3rd to 6th lanes) 30 min before UV irradiation. JNK-1 activity was measured under these different conditions by immunocomplex kinase assays. vation is dependent on extracellular Ca 2ϩ influx (30,31). Such an increase in Ca 2ϩ influx could occur as a consequence of UV-induced stimulation of K ϩ channel activity, which increases the electrical driving force for Ca 2ϩ influx through membrane hyperpolarization (26,50). This appears to be true in lymphocytes and neuronal cells where JNK activation and apoptosis are Ca 2ϩ -dependent (50 -53). There are other recent studies, however, suggesting that the activation of JNK is Ca 2ϩ -independent (49). Our result reveal that Ca 2ϩ influx does not appear to be a component of the JNK signaling pathway mediating UV-induced apoptosis in ML-1 cells.