Introduction

Top Summary Introduction Materials & Methods Results Discussion References

Synthetic ether phospholipids, characterized by the presence of an ether bond in position sn-1 of the glycerol backbone, are showing promise as a new class of clinical cancer chemotherapeutic drugs (Munder and Westphal, 1990; Houlihan et al., 1995). Some of these lipid molecules are being used as purging agents in autologous bone marrow transplantation (Koenigsmann et al., 1996) due to their antineoplastic activities and high selectivity for tumor cells. Several of these compounds are scheduled for, or currently undergoing, phase I/II clinical evaluation (Houlihan et al., 1995). The ether phospholipid ET-18-OCH₃ (1-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine; edelfosine) is a synthetic analogue of 2-lysophosphatidylcholine and shows a selective cytotoxic action against transformed cells (Munder and Westphal, 1990; Houlihan et al., 1995; Mollinedo et al., 1997). An important finding in the elucidation of the processes involved in the antineoplastic effect of ET-18-OCH₃ was its action as a potent inducer of apoptosis in tumor cells (Diomede et al., 1993, 1994; Mollinedo et al., 1993a, 1997), and this apoptotic action seems to account for the previously reported cytotoxic effects exerted by this ether phospholipid (Mollinedo et al., 1997). Although the direct apoptotic action of ET-18-OCH₃ on the cancer cell has been established (Diomede et al., 1993, 1994; Mollinedo et al., 1993b, 1997), the molecular mechanisms that initiate the active programmed cell death in ET-18-OCH₃-treated tumor cells remain largely unknown. One of the first putative targets for the action of ET-18-OCH₃ was considered to be the lipid metabolism, because this ether phospholipid affects lipid biosynthesis and reduces the levels of phosphatidylcholine (Modolell et al., 1979; Boggs et al., 1995a). In this regard, it has been recently reported that ET-18-OCH₃ behaves as a negative regulator of de novo phosphatidylcholine synthesis, acting at the CTP:phosphocholine cytidyltransferase step (Boggs et al., 1995b), and it also is a potent inhibitor of coenzyme A-independent transacylase (Winkler et al., 1996). Nevertheless, some studies show a lack of correlation between decreases in phospholipid metabolism and the sensitivity of different cell lines to ET-18-OCH₃, suggesting that the lipid perturbation by ET-18-OCH₃ is unlikely to be the underlying cause for its antineoplastic effect (Lu and Arthur, 1992a, 1992b).

The JNKs, also known as stress-activated protein kinases, are members of the MAPK-related family (Karin, 1995; Gupta et al., 1996) and are activated in response to a variety of cellular stresses (Verheij et al., 1996; Zanke et al., 1996). The JNK protein kinases phosphorylate the NH₂-terminal transactivation domain of c-Jun at Ser63 and Ser73, causing increased c-Jun transcriptional activity (Hibi et al., 1993). Thus, JNK can mediate the effect of extracellular stimuli on c-Jun and thereby acts as a physiologically relevant regulator of AP-1 transcriptional activity. JNK activation in turn requires its phosphorylation by mitogen-activated protein kinase kinase 4/c-Jun NH₂-terminal kinase kinase, also referred to as SEK1 (Lin et al., 1995). Mitogen-activated protein kinase kinase 4/c-Jun NH₂-terminal kinase kinase itself is phosphorylated and activated by the upstream kinase MAPK/ERK kinase kinase 1 (Yan et al., 1994). Thus, the stress-activated protein kinase/JNK pathway involves sequential activation of the proteins MAPK/ERK kinase kinase 1, SEK1, JNK, and c-Jun. A number of agents able to induce DNA degradation, including tumor necrosis factor, ceramide, cis-platinum, -radiation, or UV radiation, recently have been reported to activate JNK (Sluss et al., 1994; Chen et al., 1996 Verheij et al., 1996; Zanke et al., 1996). Furthermore, transient and stable expression of a dominant-negative kinase-inactive SEK1 construct impairs both activation of JNK and the apoptotic response triggered by various cell stressors (Verheij et al., 1996; Zanke et al., 1996). These observations are consistent with the involvement of JNK in the initiation of apoptosis.

We reported previously that ET-18-OCH₃ is able to induce the expression of fos and jun proto-oncogenes and to activate the transcription factor AP-1 in human leukemic cells (Mollinedo et al., 1994a). In this previous study, we observed that the steady state mRNA level of c-jun was dramatically increased on ET-18-OCH₃ treatment (Mollinedo et al., 1994a). On these grounds, and because JNK activation is involved in the induction of c-jun (Karin, 1995), we examined whether JNK stimulation and c-Jun could be implicated in the apoptotic response induced by ET-18-OCH₃.

In the current study, we found that ET-18-OCH₃ increases dramatically the steady state levels of c-jun mRNA and induces a persistent JNK activation. The results reported here establish an association between ET-18-OCH₃-induced apoptosis and JNK activation. Furthermore, we found that treatment of cells with antisense oligonucleotides directed against c-jun proto-oncogene protects cells from cell death induced by ET-18-OCH₃, indicating a role for this gene on the induction of apoptosis by ET-18-OCH₃.

	Materials and Methods

Top Summary Introduction Materials & Methods Results Discussion References

Chemicals and reagents. ET-18-OCH₃ was from R. Berthold (Biochemisches Labor, Bern, Switzerland) and from Laboratorios INKEYSA (Barcelona, Spain). ET-18-OCH₃ analogues were kindly provided by Dr. P. G. Munder (Max-Planck-Institut für Immunbiologie, Freiburg, Germany). ET-18-OCH₃ and its analogues were dissolved at 500 µg/ml as a stock solution in RPMI-1640 culture medium containing 10% (v/v) heat-inactivated fetal calf serum by heating at 50° for 30 min. The clear solutions were sterilized by filtration through a sterile filter (pore size, 0.22 µm) and stored at 4°. Human recombinant TNF- was kindly provided by Dr. G. Adolf (Boehringer Research Institute, Vienna, Austria). RPMI-1640 culture medium, fetal calf serum, and L-glutamine were purchased from GIBCO BRL (Gaithersburg, MD). Antibiotics were from Laboratorios Llorente (Madrid, Spain). Annexin-V-FLUOS to examine phosphatidylserine exposure was from Boehringer-Mannheim Biochemica (Mannheim, Germany). The Fluorescein Apoptosis Detection System kit for TUNEL assays was purchased from Promega (Madison, WI). [-³²P]dCTP (3000 Ci/mmol), [-³²P]ATP (3000 Ci/mmol), and D-threo-[dichloroacetyl-1-¹⁴C]chloramphenicol (54.6 mCi/mmol) were purchased from Amersham (Buckinghamshire, UK). PMA, GSH-agarose, and pyrithione were from Sigma Chemical (St. Louis, MO). Guanidine thiocyanate was from Fluka (Buchs, Switzerland). Formaldehyde was from J. T. Baker Chemicals B.V. (Deventer, Holland). Acrylamide, bisacrylamide, ammonium persulfate, and N,N,N',N'-tetramethylethylenediamine were from BioRad (Richmond, CA). All other chemicals were from Sigma or Merck (Darmstadt, Germany).

Cells and culture conditions. The human leukemic cell lines used in this study were promyelocytic HL-60, promonocytic U937, T lymphoid Jurkat, and chronic myelogenous K562. These cells were grown in RPMI-1640 culture medium supplemented with 10% (v/v) heat-inactivated fetal calf serum, 2 mM L-glutamine, 100 units/ml penicillin, and 24 µg/ml gentamicin. Cells were incubated at 37° in a humidified atmosphere of 5% CO₂/95% air. Ether phospholipids were added to the cell cultures at 3 or 5 µg/ml for the times indicated in the respective figures.

Analysis of DNA fragmentation in agarose gels. To assess apoptosis, we isolated fragmented DNA as described previously (Mollinedo et al., 1993a). In brief, 2 × 10⁶ cells were washed with phosphate-buffered saline and then lysed with 200 µl of hypotonic detergent buffer (10 mM Tris·HCl, pH 7.5, 1 mM EDTA, 0.2% Triton X-100) for 30 min at 4°. Nuclei and cell organelles were removed by centrifugation in a microfuge for 20 min, and the supernatant, containing the DNA released into the cytosol due to DNA fragmentation, was incubated with RNase A (75 µg/ml) for 45 min at 37° and then with proteinase K (200 µg/ml) in the presence of 0.5% SDS for additional 45 min at 37°. The DNA was extracted, precipitated, and analyzed by electrophoresis on 1% agarose gels as described previously (Mollinedo et al., 1993a).

TUNEL assay. Apoptosis also was analyzed in situ by the TUNEL technique using the Fluorescein Apoptosis Detection System Kit (Promega) according to the manufacturer's instructions, labeling the 3'-OH ends generated by DNA fragmentation through incorporation of fluorescein-12-dUTP (Gavrieli et al., 1992). Fluorescent cells were visualized and scored with a Zeiss LSM 310 laser scan confocal microscope.

Phosphatidylserine exposure. Phosphatidylserine exposure at the external surface of the cell was measured by the binding of FITC-labeled annexin V according to the protocol outlined by the manufacturers in the Annexin-V-FLUOS reagent (Boehringer-Mannheim). Then, cells were analyzed with Becton Dickinson (Le Pont de Claix, France) FACScan and FACStar-Plus flow cytometers.

[³H]Thymidine incorporation and TUNEL assay. Cell proliferation was monitored according to [³H]thymidine (Amersham) incorporation as described previously (Mollinedo et al., 1993a).

Northern blot. Total RNA was isolated by the acid guanidinium thiocyanate-phenol-chloroform extraction method. Samples of 20 µg of RNA were electrophoresed on 0.9% (w/v) agarose-formaldehyde gels and then transferred to Hybond-N nylon membranes (Amersham) as described previously (Mollinedo et al., 1994a). ³²P-labeled cDNA probes were prepared using the random hexanucleotide priming method (oligo-labeling kit; Pharmacia Biotech, Uppsala, Sweden) to a specific radioactivity 7 × 10⁸ cpm/mg of cDNA. cDNA probes for c-fos and c-jun (Mollinedo and Naranjo, 1991) were kindly provided by Dr. P. Sassone-Corsi (Laboratoire de Genetique Moleculaire des Eucaryotes, Center National de la Recherche Scientifique, Strasbourg, France) and Dr. R. Bravo (Squibb Institute, Princeton, NJ). The plasmid pAc 18.1, used as a probe for -actin, was used as a control probe as described previously (Mollinedo and Naranjo, 1991). Conditions for blot hybridization and washing have been described elsewhere (Mollinedo et al., 1994a). Quantitative analysis of the autoradiograms was performed by integration of peak areas after scanning with a PDI computing densitometer (Pharmacia).

Electroporation and CAT assay. Jurkat cells (20 × 10⁶) in the exponential phase of growth were electroporated as described previously (Mollinedo et al., 1994b) with 25 µg of the expression vector AP-1-TK-CAT plasmid (Angel et al., 1987), kindly provided by Dr. M. Karin (University of California, San Diego, CA). The AP-1-TK-CAT plasmid contains a single copy of the phorbol ester-responsive element (AP-1 site) inserted upstream to position 109 of the herpes simplex virus thymidine kinase promoter and fused to the structural gene coding for the CAT gene, used as reporter gene (Angel et al., 1987). Experimental conditions for CAT assay and thin layer chromatography have been described elsewhere (Mollinedo et al., 1994b). After autoradiogram development, the percentage of acetylation of each treatment was determined by scraping off both the acetylated and nonacetylated chloramphenicol and counting radioactivity.

Solid-phase JNK assay. Protein kinase assays were carried out using a fusion protein between GST and c-Jun (amino acids 1-223) as a substrate of JNK, as described previously (Hibi et al., 1993) with slight modifications. Cells (3-5 × 10⁶) were resuspended in 200 µl of extract buffer (25 mM HEPES, pH 7.7, 0.3 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.1% Triton X-100, 20 mM -glycerophosphate, 0.1 mM Na₃VO₄, 0.5 mM phenylmethylsulfonyl fluoride, 1 µg/ml leupeptin, 1 µg/ml aprotinin). Cells were incubated for 30 min in continuous rotation at 4° and then microfuged at 12,000 rpm for 10 min. The pellets were discarded, and the supernatants, representing cell extracts, were diluted with 600 µl of dilution buffer (20 mM HEPES, pH 7.7, 0.1 mM EDTA, 2.5 mM MgCl₂, 0.05% Triton X-100, 20 mM -glycerophosphate, 0.1 mM Na₃VO₄, 0.5 mM phenylmethylsulfonyl fluoride, 1 µg/ml leupeptin, 1 µg/ml aprotinin). Mixtures were incubated for 10 min on ice and then microfuged at 12,000 rpm for 10 min. The cell extracts were mixed with 20 µl of a suspension in dilution buffer of GSH-agarose beads, to which GST-c-Jun were freshly bound. Mixtures were rotated overnight at 4° in an Eppendorf tube and pelleted by centrifugation at 12,000 rpm for 1 min. After four 1-ml washes in dilution buffer containing 50 mM NaCl, to remove kinases that have weaker affinity to bind c-Jun(1-223) than JNK, the pelleted beads were resuspended in 30 µl of kinase buffer (20 mM HEPES, pH 7.7, 2 mM dithiothreitol, 20 mM -glycerophosphate, 20 mM MgCl₂, 0.1 mM Na₃VO₄, 20 mM ATP) and incubated with 4 µCi of [-³²P]ATP. After 30 min at 37°, the reaction was terminated by washing with dilution buffer containing 50 mM NaCl and microfugation at 12,000 rpm for 1 min. Then, the beads were boiled with 10 µl of 5× SDS-polyacrylamide gel sample buffer to elute the phosphorylated proteins, which were subsequently resolved in an SDS-10% polyacrylamide gel, followed by autoradiography. These conditions have been shown previously to enable specific binding of JNK to c-Jun NH₂-terminal domain (Hibi et al., 1993).

In-gel kinase assay for JNK activation. This assay was performed according to Hibi et al. (1993) with slight modifications. Cells were lysed with extract buffer as above and centrifuged at 12,000 rpm for 10 min, and the supernatants, representing the cell extracts, were saved. Protein content was determined in these extracts using a BioRad protein assay kit. Approximately 60 µg of protein was loaded onto an SDS-10% polyacrylamide gel, which was polymerized in the presence of GST-c-Jun (100 µg/ml). After electrophoresis, the gel was washed twice for 15 min each with 100 ml of 50 mM HEPES, pH 7.6, and 20% (v/v) 2-propanol to remove SDS and then twice for 15 min each with 50 mM HEPES, pH 7.6, 5 mM -mercaptoethanol, and 0.05% Tween 20, supplemented first with 3 M urea and then with 1.5 M urea and finally with 0.75 M urea. The proteins were renatured by washing the gel in 50 mM HEPES, pH 7.6, and 0.05% Tween 20 at 4°. The kinase reaction was performed by incubating the gel for 1 hr at 30° in kinase buffer supplemented with 20 µM ATP and 100 µCi [-³²P]ATP. Finally, the gel was washed with 100 ml of 5% trichloroacetic and 1% sodium pyrophosphate at room temperature, followed by drying and autoradiography.

Western blot. Mininuclear extracts were obtained from 3 × 10⁶ cells as described previously (Mollinedo et al., 1993b), and proteins (30 µg) were separated through an SDS-12% polyacrylamide gel under reducing conditions, transferred to nitrocellulose filters, and subjected to immunological detection. Low range prestained protein molecular mass standards (BioRad) also were run in the same gel. After electroblotting and blocking for 1 hr in 2% powdered nonfat dry milk in TBS buffer (50 mM Tris·HCl, pH 8.0, 150 mM NaCl), the nitrocellulose filters were incubated overnight with sheep anti-human/c-Jun antibody (Cambridge Research Biochemicals, Cheshire, UK) at a dilution of 1:250 in TBS containing 0.05% Tween 20. Signal was developed after incubation with a rabbit anti-sheep immunoglobulin coupled to peroxidase and using 1.7 mM 3,3'-diaminobenzidine with 0.03% (v/v) H₂O₂ in 50 mM Tris·HCl, pH 7.4, as a substrate solution. A rabbit antiserum to the Fos M peptide, kindly provided by Dr. J. Jain (Dana-Farber Cancer Institute, Boston, MA), was used for Fos detection.

Oligonucleotides. c-fos sense (5'-TTCTCGGGCTTCAACGCA-3'), c-fos antisense (5'-TGCGTTGAAGCCCGAGAA-3'), c-jun sense (5'-ACTGCAAAGATGGAAACG-3'), c-jun antisense (5'-CGTTTCCATCTTTGCAGT-3'), and a random oligonucleotide (5'-ACCGTTCGCTGTTATCTT-3') were synthesized by using phosphorothioate linkages. The antisense nucleotide sequences were complementary to the first 18 bases after the AUG sequences of human mRNAs for c-fos (accession no. K00650) and c-jun (accession no. J04111) obtained from the GenBank EMBL database. The corresponding sense oligonucleotides and the random oligonucleotide were used as a control. Oligonucleotides penetrated into the cells without any treatment, as described by Loke et al. (1989). HL-60 cells at an initial concentration of 1.2 × 10⁵ cells/ml were either untreated or treated with specific oligonucleotides in serum-free ITS medium composed of 5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml sodium selenite in RPMI-1640, containing 0.4% (w/v) bovine serum albumin, 2 mM glutamine, 100 units/ml penicillin, and 24 µg/ml gentamicin. Oligonucleotides were added at a final concentration of 50 µg/ml and incubated for 48 hr in serum-free ITS medium. Then, cells were incubated in the absence or presence of 5 µg/ml ET-18-OCH₃ for 4 hr in serum-free ITS medium and assayed for DNA fragmentation and Western blot analysis. In addition, cells pretreated for 48 hr with random, sense, and antisense c-jun oligonucleotides also were assayed for ET-18-OCH₃-induced JNK activation and [³H]thymidine incorporation as indicated above.

	Results

Top Summary Introduction Materials & Methods Results Discussion References

c-Jun expression induced by ET-18-OCH₃. We reported previously that ET-18-OCH₃ induces jun and fos proto-oncogenes and activates AP-1 transcription factor in HL-60, U937, and Jurkat human leukemic cells (Mollinedo et al., 1994a), which are highly sensitive to undergoing rapid apoptosis by this ether phospholipid (Mollinedo et al., 1993a, 1997).Because c-fos and c-jun have been implicated in the induction of apoptosis in various systems (Colotta et al., 1992; Ham et al., 1995), we analyzed and quantified the respective increases in the steady state mRNA levels of both c-jun and c-fos in three ET-18-OCH₃-sensitive human leukemic cell lines (U937, HL-60, and Jurkat) (Mollinedo et al., 1993a, 1994a, 1997) on treatment with ET-18-OCH₃. We corroborated our previous findings (Mollinedo et al., 1994a), showing that ET-18-OCH₃ induces a dramatic and sustained increase in the c-jun mRNA level, whereas the c-fos mRNA level was weakly or hardly increased (Mollinedo et al., 1994a). Because U937, HL-60, and Jurkat cells undergo apoptosis after only 6 hr of treatment with 3 µg/ml ET-18-OCH₃ (Mollinedo et al., 1993a, 1994a, 1997), we compared the relative increases in the steady state levels of c-jun and c-fos mRNA on treatment with ET-18-OCH₃ before the onset of apoptosis (Fig. 1A). After 4 hr of treatment with ET-18-OCH₃ (3 µg/ml), we observed a maximal induction in c-jun and c-fos expression occurring before the onset of apoptosis (Mollinedo et al., 1994a; data not shown). At this incubation time, we found relative inductions of 27-fold (U937), 38-fold (HL-60), and 17-fold (Jurkat) in the c-jun mRNA steady state levels on treatment with ET-18-OCH₃, whereas the relative increase in the c-fos mRNA steady state level was <5-fold in the three distinct human leukemic cell lines (Fig. 1A). Thus, we found that ET-18-OCH₃ is an extremely potent inducer of c-jun expression. We also found, corroborating our previous findings (Mollinedo et al., 1994a), that ET-18-OCH₃ activated AP-1 transcriptional activity when used at a concentration able to induce apoptosis, as assessed by CAT assays (Fig. 1B). This activation of AP-1 enhancer activity by ET-18-OCH₃ resulted in lower than that observed when cells were incubated with PMA (Fig. 1B), a strong activator of AP-1 transcription factor. PMA, unlike ET-18-OCH₃, induces a strong increase in the c-fos mRNA steady state level in these cell lines (Mollinedo et al., 1993b; data not shown). The relative weaker activation of AP-1 transcriptional activity in relation to the extremely high induction of c-jun proto-oncogene observed on ET-18-OCH₃ treatment (Fig. 1) could be explained by the very weak induction of c-fos in ET-18-OCH₃-treated cells (Fig. 1), because the AP-1 transcription factor containing c-Jun/c-Fos shows a higher DNA binding capacity and transcriptional activity than the c-Jun homodimers (Karin, 1995).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Expression of c-jun and c-fos proto-oncogenes and induction of AP-1 enhancer activity on ET-18-OCH3 treatment. A, Relative induction of c-fos and c-jun genes in ET-18-OCH3-treated human leukemic cells. c-fos () and c-jun () steady state mRNA contents were quantified by densitometric scanning of three independent autoradiograms of Northern blots performed after treatment of the indicated cell types with 3 µg/ml ET-18-OCH3 for 4 hr. Untreated control cells were run in parallel. Each value was corrected for the -actin mRNA content, and the fold induction was obtained by comparison with the corresponding value in untreated control cells. Data are shown as mean ± standard error of three independent experiments. B, Induction of AP-1 enhancer activity by ET-18-OCH3 in Jurkat cells as measured by CAT assay. Jurkat cells were electroporated with AP-1-TK-CAT plasmid and treated with PMA (10 ng/ml) or ET-18-OCH3 (3 µg/ml). Untreated control cells (C) were run in parallel. After 24 hr, cells were disrupted, and CAT activity was determined as described in Materials and Methods. Values are representative of three experiments and represent fold induction in AP-1 enhancer activity based on the ratio between the percentage of acetylation in each treatment and the corresponding percentage of acetylation in the control unstimulated cells (C).

ET-18-OCH₃-induced apoptosis is associated with JNK activation. The c-jun proto-oncogene is positively autoregulated by its own gene product, once it is properly phosphorylated (Karin, 1995). Because JNK activates c-Jun transcriptional activity (Hibi et al., 1993), candidate inducers of the JNK pathway include agents that cause sustained induction of c-Jun, such as the ether phospholipid ET-18-OCH₃ (Mollinedo et al., 1994a). On the other hand, several studies suggest that JNK activation may be related to apoptosis (Sluss et al., 1994; Chen et al., 1996; Verheij et al., 1996; Zanke et al., 1996). Fig. 2A shows that 5 µg/ml ET-18-OCH₃ induced a very rapid DNA internucleosomal fragmentation in HL-60 cells after only a 3-hr treatment. On these grounds, we examined JNK activation in HL-60 cells treated with ET-18-OCH₃. To this aim, we used a GST fusion protein containing amino acids 1-223 of c-Jun, GST-c-Jun-(1-223). This fusion protein was bound through its GST moiety to GSH-agarose beads to generate an affinity matrix to precipitate JNK activities from HL-60 cell lysates. The precipitated complexes were washed and subjected to solid-phase kinase assay. As shown in Fig. 2B, GST-c-Jun phosphorylation was observed after 30-60 min of ET-18-OCH₃ incubation, and the response was further increased with the incubation time and was found to be persistent (Fig. 2B). A strong JNK activation was obtained after 2 hr of treatment with ET-18-OCH₃ (Fig. 2B), before DNA fragmentation, occurring after 3 hr of treatment (Fig. 2A). Identical results were obtained using a GST-c-Jun-(1-79) fusion protein (data not shown). Both the apoptotic response and the JNK activation induced by ET-18-OCH₃ were found to be dose dependent and were well correlated (Fig. 3). Thus, HL-60 underwent apoptosis on treatment with 3 µg/ml ET-18-OCH₃, and this response was dramatically increased at 5 µg/ml ET-18-OCH₃ (Fig. 3A). ET-18-OCH₃-induced JNK activation was observed after cell incubation with a concentration of 1 µg/ml of the ether phospholipid, and it was particularly strong after treatment with 3 and 5 µg/ml ET-18-OCH₃ (Fig. 3B). In this regard, it is interesting to note that 1 µg/ml ET-18-OCH₃ has been reported as the threshold concentration in the inhibitory effect on DNA synthesis shown by this ether phospholipid (Mollinedo et al., 1993a). As shown in Figs. 2 and 3, ET-18-OCH₃-induced apoptosis as well as JNK activation were time and dose dependent, and JNK activation always preceded the induction of internucleosomal DNA fragmentation.

View larger version (51K): [in this window] [in a new window]   Fig. 2.   ET-18-OCH3 induces rapid apoptosis and persistent JNK activation in HL-60 cells. HL-60 cells were treated with 5 µg/ml ET-18-OCH3 for the indicated times and assayed for DNA fragmentation (A) or for JNK activation (B) as described in Materials and Methods. Control untreated cells were run in parallel in the same gels. Fragmented DNA from 6 × 105 cells was loaded in each lane of the agarose gel. A, 123-base pair DNA ladder was used as standard (Std). B, Position of phosphorylated GST-c-Jun-1-223 is indicated. Left, molecular masses (in kDa) of protein markers. Values are representative of three experiments.

View larger version (52K): [in this window] [in a new window]   Fig. 3.   Dose response of ET-18-OCH3-induced DNA fragmentation and JNK activation. HL-60 cells were incubated for 4 hr with increasing concentrations of ET-18-OCH3 and assayed for DNA fragmentation (A) or JNK activation (B) as described in Materials and Methods. Control untreated cells were run in parallel in the same gels. Fragmented DNA from 6 × 105 cells was loaded in each lane of the agarose gel. A, 123-base pair DNA ladder was used as standard. B, Position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated. Values are representative of three experiments.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Flow cytometric analysis of the binding of FITC-labeled annexin V to HL-60 cells treated with ET-18-OCH3. Untreated (a) and treated HL-60 cells with 5 µg/ml ET-18-OCH3 for 1 (b), 3 (c), 4 (d), 6 (e), or 9 (f) hr were stained with FITC-labeled annexin V and analyzed by fluorescence flow cytometry as described in Materials and Methods. Numbers at top right, percentage of cells that are positive for FITC-annexin V staining. Values are representative of three experiments.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Identification of the JNK polypeptides activated by ET-18-OCH3. HL-60 cells were incubated in the absence (C) or in the presence of 5 µg/ml ET-18-OCH3 for the indicated times. Then, cell extracts were prepared, and the JNK polypeptides were visualized after an in-gel kinase assay as described in Materials and Methods. Values are representative of three experiments.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Rapid and transient JNK activation induced by TNF- in HL-60 cells. HL-60 cells were treated with 100 units/ml TNF- or 5 µg/ml ET-18-OCH3 for the indicated times and assayed for JNK activation as described in Materials and Methods. Control untreated cells were run in parallel in the same gel. The position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated. Values are representative of three experiments.

View larger version (8K): [in this window] [in a new window]   Fig. 7.   Chemical structures of the ether phospholipids used in this study. The indicated sn-2 substitutions of the glycerol backbone give rise to the ether phospholipids ET-18-OCH3, ET-18-H, and ET-18-OH.

View larger version (45K): [in this window] [in a new window]   Fig. 8.   Molecular structure specificity in the ET-18-OCH3-induced apoptosis and JNK activation. HL-60 cells were incubated with 5 µg/ml concentration of ET-18-OCH3 or two closely related ET-18-OCH3 analogues, ET-18-H and ET-18-OH, for the indicated times and assayed for DNA fragmentation (A) or for JNK activation (B) as described in Materials and Methods. Control untreated cells (Control in A; C in B) were run in parallel in the same gels. Fragmented DNA from 6 × 105 cells was loaded in each lane of the agarose gels. A, 123-base pair DNA ladder was used as standard (Std). B, Position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated.*, ET-18-OCH3 and represents the JNK activation induced by treatment of HL-60 cells with 5 µg/ml ET-18-OCH3 for 2 hr. Values are representative of three experiments.

View larger version (15K): [in this window] [in a new window]   Fig. 9.   Lack of JNK activation by ET-18-OCH3 in K562 cells. HL-60 and K562 cells were treated with 5 µg/ml ET-18-OCH3 for the indicated times and assayed for JNK activation as described in Materials and Methods. Control untreated cells were run in parallel in the same gels. The position of phosphorylated GST-c-Jun-1-223 is indicated. Values are representative of three experiments.

View larger version (23K): [in this window] [in a new window]   Fig. 10.   Suppression of ET-18-OCH3-induced DNA degradation by PMA and Zn2+ and inhibition of ET-18-OCH3-induced JNK activation by PMA. A, HL-60 cells were incubated for 4 hr in the absence (Control) or presence of the indicated agents: 5 µg/ml ET-18-OCH3, 100 ng/ml PMA, 100 ng/ml PMA plus 5 µg/ml ET-18-OCH3, 20 µM Zn2+ plus 100 µM pyrithione, and 20 µM Zn2+ plus 100 µM pyrithione plus 5 µg/ml ET-18-OCH3. The Zn2+ ionophore pyrithione was added to facilitate cellular uptake of Zn2+. PMA, pyrithione, and Zn2+ were added 15 min before ET-18-OCH3. Fragmented DNA was obtained as described in Materials and Methods. Fragmented DNA from 6 × 105 cells was loaded in each lane. A 123-base pair DNA ladder was used as standard (Std). B, HL-60 cells were incubated for 2 hr in the absence (Control) or presence of the indicated agents (5 µg/ml ET-18-OCH3, 100 ng/ml PMA, or 100 ng/ml PMA plus 5 µg/ml ET-18-OCH3) and then assayed for JNK activation as described in Materials and Methods. PMA was added 15 min before ET-18-OCH3. The position of phosphorylated GST-c-Jun-1-223 is indicated. Values are representative of three experiments.

View larger version (36K): [in this window] [in a new window]   Fig. 11.   PMA induces a rapid and transient JNK activation but inhibits the sustained JNK activation induced by ET-18-OCH3 in HL-60 cells. Top, HL-60 cells were treated with 100 ng/ml PMA or 5 µg/ml ET-18-OCH3 for the indicated times and assayed for JNK activation as described in Materials and Methods. Control untreated cells were run in parallel in the same gel. The position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated. Experiment shown is representative of three performed. Bottom, HL-60 cells were treated with 100 ng/ml PMA and 5 µg/ml ET-18-OCH3 for the indicated times and assayed for JNK activation as described in Materials and Methods. HL-60 cells treated with 5 µg/ml ET-18-OCH3 alone for 2 hr were analyzed for JNK activation and run in parallel in the same gel to compare the inhibitory effect of PMA on ET-18-OCH3-induced JNK activation. Control untreated cells were run in parallel in the same gel. The position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated. Values are representative of three experiments.

Requirement of c-Jun in ET-18-OCH₃-induced apoptosis. To examine the role of c-jun in the induction of apoptosis by ET-18-OCH₃, we used antisense c-jun oligonucleotides. In addition, because c-fos has been implicated previously in the induction of apoptosis (Colotta et al., 1992), we used antisense c-fos oligonucleotides. Incubation of HL-60 cells in serum-free ITS medium for 48 hr with antisense oligonucleotides directed against c-jun or c-fos drastically reduced the respective c-Jun and c-Fos protein levels, whereas sense oligonucleotide-treated cells expressed the same protein levels found in untreated cells (Fig. 12D; data not shown). Thus, when HL-60 cells were treated with antisense c-jun oligonucleotides for 48 hr before the addition of ET-18-OCH₃, no c-Jun protein was detected (Fig. 12D). The treatment of HL-60 cells with ET-18-OCH₃ slightly increased the protein level of c-Jun, which was not affected by the presence of sense c-jun oligonucleotides (Fig. 12D). However, no c-Jun protein was detected when antisense c-jun oligonucleotide-treated HL-60 cells were incubated with ET-18-OCH₃ (Fig. 12D). Under these experimental conditions, we found that antisense c-jun oligonucleotides blocked the apoptotic response induced by ET-18-OCH₃ (Fig. 12A), whereas ET-18-OCH₃-induced apoptosis was not affected by the presence of antisense c-fos oligonucleotides (Fig. 12C). An additional scrambled and random oligonucleotide, with the same base composition as that used for antisense-c-jun oligonucleotide (see Materials and Methods), had no effect on ET-18-OCH₃-induced apoptosis or on c-Jun protein level (Fig. 12B; data not shown). These results indicate a role for c-Jun in ET-18-OCH₃-induced apoptosis.

View larger version (51K): [in this window] [in a new window]   Fig. 12.   ET-18-OCH3-induced apoptosis is prevented by antisense c-jun oligonucleotides. A-C, HL-60 cells were grown for 48 hr in serum-free ITS medium in the absence (Control) or presence of the indicated specific oligonucleotides as described in Materials and Methods. Then, cells were incubated for additional 4 hr in the absence or presence of 5 µg/ml ET-18-OCH3, and fragmented DNA was obtained and analyzed as described in Materials and Methods. Sense c-jun (S-c-jun), 50 µg/ml; antisense c-jun (As-c-jun), 50 µg/ml; sense c-fos (S-c-fos), 50 µg/ml; antisense c-fos (As-c-fos), 50 µg/ml; random oligonucleotide (Random), 50 µg/ml. Fragmented DNA from 6 × 105 cells was loaded in each lane. A 123-base pair DNA ladder was used as standard (Std). D, Immunoblot analysis of c-Jun protein. HL-60 cells were grown as in A for 48 hr in serum-free ITS medium in the absence (Control) or presence of specific oligonucleotides; then, cells were incubated for additional 4 hr in the absence or presence of 5 µg/ml ET-18-OCH3. Nuclear extracts were prepared as described in Materials and Methods. Equal amounts of nuclear proteins (30 µg) were run on an SDS-12% polyacrylamide gel and analyzed by immunoblotting using a specific anti-c-Jun antibody as described in Materials and Methods. The position of c-Jun protein is indicated. Left, molecular masses (in kDa) of protein markers. Sense c-jun (S-c-jun), 50 µg/ml; antisense c-jun (As-c-jun), 50 µg/ml.

View larger version (30K): [in this window] [in a new window]   Fig. 13.   JNK activation induced by ET-18-OCH3 is not inhibited by antisense c-jun oligonucleotides. HL-60 cells were grown for 48 hr in serum-free ITS medium in the absence (Control) or presence of the indicated specific oligonucleotides as described in Materials and Methods. Then, cells were incubated for additional 2 hr in the absence or presence of 5 µg/ml ET-18-OCH3 and assayed for JNK activation as described in Materials and Methods. Sense c-jun (S-c-jun), 50 µg/ml; antisense c-jun (As-c-jun), 50 µg/ml; random oligonucleotide (Random), 50 µg/ml. The position of phosphorylated GST-c-Jun-1-223 (GST-c-Jun) is indicated. Values are representative of three experiments.

	Discussion

Top Summary Introduction Materials & Methods Results Discussion References

Because (1) ET-18-OCH₃ strongly induces c-jun expression (Mollinedo et al., 1994a; current report); (2) JNK activates the transcriptional activity of c-Jun, stimulating c-jun transcription (Karin, 1995); and (3) JNK signaling has been reported to be involved in the induction of cell death by different agents in distinct systems (Sluss et al., 1994; Chen et al., 1996; Verheij et al., 1996; Zanke et al., 1996), we suspected that the JNK signaling route might be involved in the induction of apoptosis by ET-18-OCH₃. The results described here indicate for the first time the association of persistent JNK activation with ET-18-OCH₃-induced apoptosis, as well as the requirement of c-Jun for this apoptotic response. Thus, these data unveil a signaling route involved in the induction of apoptosis by ET-18-OCH₃. We found that ET-18-OCH₃ induces a persistent JNK activation in HL-60 cells, which undergo rapid apoptosis on treatment with this ether phospholipid (Fig. 2) (Mollinedo et al., 1993a). This JNK activation may mediate the increased AP-1 activity observed in ET-18-OCH₃-treated cells (Fig. 1) (Mollinedo et al., 1994a). However, ET-18-OCH₃-induced JNK activation is not detected in ET-18-OCH₃-resistant K562 cells. Subtle modifications in the molecular structure of ET-18-OCH₃ that abolish its apoptotic properties also abrogate its capacity to induce JNK activation. In this regard, ET-18-OCH₃ analogues, in which the methoxy group is substituted for an OH or H group in the sn-2 position of the glycerol backbone, lack the capacities to activate JNK and induce apoptosis in HL-60 cells. Thus, the results reported here indicate that ET-18-OCH₃-induced JNK activation is associated with the capacity of this ether phospholipid to induce apoptosis.

We also found that this JNK activation seems to precede the onset of apoptosis induced by ET-18-OCH₃. JNK activation was detected before DNA fragmentation and phosphatidylserine externalization, two early and well known processes involved in the apoptotic response. Furthermore, both dose-response and time course analyses indicated that JNK activation preceded DNA fragmentation in ET-18-OCH₃-treated HL-60 cells. On the other hand, our data suggest that in ET-18-OCH₃-treated HL-60 cells, DNA fragmentation, assessed through visualization of the typical internucleosomal DNA fragments in agarose gels and through TUNEL analysis, precedes phosphatidylserine exposure at the outer leaflet of the plasma membrane, as assessed through FITC-labeled annexin V binding.

PMA was found to block both ET-18-OCH₃-induced apoptosis and JNK activation. Although the molecular basis for these effects remains to be determined, the protective effect of PMA against apoptosis has been related to its ability to induce cellular alkalinization (Pérez-Sala et al., 1995). In this regard, an association has been reported between intracellular acidification and DNA degradation (Barry et al., 1993; Pérez-Sala et al., 1995). Thus, it will be of interest to study a putative relationship between intracellular pH and JNK activation.

The requirement of c-jun in ET-18-OCH₃-induced apoptosis was determined by the blockade of apoptosis by antisense c-jun oligonucleotides. These results are in agreement with previous reports showing a crucial role for c-Jun in the induction of apoptosis in different systems, as demonstrated by the use of c-Jun dominant-negative mutants, neutralizing antibodies, or antisense oligonucleotides (Colotta et al., 1992; Ham et al., 1995; Verheij et al., 1996). However, antisense c-fos oligonucleotides did not prevent ET-18-OCH₃-induced apoptosis, indicating that c-fos is not needed for the onset of ET-18-OCH₃-induced apoptosis. Although some previous reports have implicated c-fos in programmed cell death (Colotta et al., 1992), the results herein described are in agreement with recent evidence from c-Fos-deficient mice demonstrating that c-fos is not essential for the induction of apoptosis (Gajate et al., 1996).

We found that antisense c-jun oligonucleotides were unable to prevent ET-18-OCH₃-mediated JNK activation. In this regard, Verheij et al. (1996) reported that expression in U937 and BAE cells of a dominant-negative c-Jun mutant lacking the amino terminus, the portion that is phosphorylated by JNK, blocked stress-induced apoptosis but did not inhibit JNK activation. Taken together, these data indicate that inhibition of the apoptotic response to ET-18-OCH₃ or to stress, by antisense c-jun oligonucleotides or by a dominant-negative c-Jun mutant, occurs at a level downstream of JNK. These data also suggest a critical role for c-Jun in these apoptotic processes. The mechanism by which c-Jun could mediate apoptosis is unknown. It is tempting to speculate that activated c-Jun regulates apoptosis through modulation of certain proteins required in the apoptotic response or through sequestration of inhibitors of apoptosis. A putative hypothesis for the c-Jun action on apoptosis might stem from the capacity of c-Jun to complex with different proteins.

Immediate and transient kinetics of JNK activation in response to various stimuli that promote cell activation and proliferation have been reported (Karin et al., 1995; Chen et al., 1996). We also found that induction of monocyte/macrophage differentiation of HL-60 cells by TNF- or PMA was accompanied by a rapid and transient JNK activation, reaching a maximum JNK activation after 15 and 30 min of treatment, respectively. However, the JNK activation reported here on ET-18-OCH₃ treatment was delayed and persistent. JNK activation was detected after 30-60 min of ET-18-OCH₃ treatment; increased strongly with the incubation time, being very potent by 2 hr of treatment; and was sustained for a long period of time. Chen et al. (1996) recently reported that activation of Jurkat cells with different stimuli rendered a rapid and transient induction of JNK activity, whereas JNK remained persistently activated during -radiation- and UV-C-induced apoptosis. Thus, JNK activation occurs in different physiological processes, including cell activation, cell differentiation, and apoptosis. This indicates that JNK can function as a common kinase shared by different physiological processes. The different timing or duration of JNK induction, or both, may lead to opposite outcomes, namely cell activation/proliferation/differentiation (rapid and transient JNK activation) versus apoptosis (sustained JNK activation), as postulated previously (Chen et al., 1996). The participation of JNK activation in different physiological outcomes suggests that additional coactivators or JNK-interacting proteins could play a role in promoting a final cell response. The prolonged JNK activation in ET-18-OCH₃ may cause the persistent activation of some cellular factors, such as c-Jun, and results in detrimental effects on the cells.

The JNK protein kinases comprise a family of 10 isoforms, which correspond to alternatively spliced isoforms derived from the JNK1, JNK2, and JNK3 genes (Gupta et al., 1996). These JNK isoforms are detected as proteins of 46 kDa (JNK1), 55 kDa (JNK2), and 48 and 57 kDa (JNK3) (Gupta et al., 1996). With in-gel assays, we found that ET-18-OCH₃ induces activation of two JNK protein kinase forms with apparent molecular masses of 46 and 55 kDa in HL-60 cells. JNK protein kinases are members of the wide family of the MAPK-related family (Karin et al., 1995; Gupta et al., 1996), which includes two great subfamilies: MAPK/ERK, mainly involved in mitogenesis and differentiation; and JNK, supposedly involved in growth arrest and cell death. Mitogens are potent ERK activators but activate JNK poorly, and stress stimuli are powerful inducers of JNK but induce ERK relatively weakly (Karin et al., 1995; Gupta et al., 1996). Our current results demonstrate that ET-18-OCH₃ strongly activates the JNK route in a sustained way. Zhou et al. (1996) recently reported that this ether phospholipid inhibited the MAPK/ERK cascade. Thus, ET-18-OCH₃ arrests mitogenic signals (MAPK/ERK) and strongly induces apoptotic signaling pathways (JNK). These actions could explain the rapid and effective induction of apoptosis exerted by this ether phospholipid (Diomede et al., 1993, 1994; Mollinedo et al., 1993a, 1997).

In conclusion, the current results indicate an association between persistent JNK activation and induction of apoptosis in response to ET-18-OCH₃, suggesting that JNK activation is involved in the initiation of programmed cell death by ET-18-OCH₃. Furthermore, the present data indicate that c-Jun is required for the rapid induction of apoptosis by ET-18-OCH₃, suggesting that activation of c-Jun plays a critical role in this apoptotic response.