INTRODUCTION

ara-C¹ was developed as an inhibitor of DNA synthesis, and incorporation of ara-C into DNA is associated with its toxicity. However, it is also evident that ara-C has additional effects. For example the rapidity of cell killing observed within 24-36 h of drug administration is inconsistent with the long duration of the DNA synthesis phase in human leukemia cells (>48 h) (1). Thus, although ara-C is the most effective single agent in the treatment of acute myelogenous leukemia, the mechanism of leukemic cell death in response to ara-C is not well defined. We have previously shown that ara-C stimulates the synthesis of two lipid second messengers, diglyceride and ceramide (2). Kharbanda et al. (3) demonstrated that ara-C induces a protein kinase C-like activity in HL-60 cells. Thus, the ara-C-induced DAG could exert its biological activities through the activation of PKC. Ceramide is also recognized as a second messenger, and there is evidence that PKC activation is antagonistic to ceramide-induced apoptosis (4-6). Therefore, ara-C induces two lipid mediators that are potentially involved in opposing pathways in HL-60 cells with respect to induction of cell death by apoptosis. To test this idea we studied DAG-activated signaling in response to ara-C using ET-18-OCH₃, a member of a class of membrane-interactive drugs selective for growth inhibition of cancer cells (7-9). ET-18-OCH₃ directly inhibits PKC activity in vitro by acting as a competitive inhibitor with respect to phosphatidylserine, a lipid cofactor required by PKC (7, 10-12). Our results demonstrate that inhibition of PKCII by either ET-18-OCH₃ or antisense oligodeoxynucleotides to PKCII enhances ara-C-induced apoptosis. This suggests that the diacylglycerol formed in response to ara-C is activating a signaling pathway that is antagonistic to ara-C-stimulated apoptosis.

EXPERIMENTAL PROCEDURES

ara-C (Upjohn) was dissolved in sterile deionized water. TPA was purchased from LC Services and stored in dimethyl sulfoxide. ET-18-OCH₃ (Medmark Pharma, GmbH, Grunwald, Germany) and C₂-ceramide (Matreya) were stored in 100% ethanol. Stock preparations of all reagents were stored at 20 °C under light-free conditions. Vehicle controls were included and were consistently found to be without effect on the parameters studied. Monoclonal PKCII and PKC antibodies (Dr. David Burns, Sphinx Pharmaceuticals), polyclonal PKC (UBI), and monoclonal antibody to Bcl-2 (Pharmingen) were stored at 20 °C and diluted in phosphate-buffered saline, supplemented with 3% bovine serum albumin prior to use. In some experiments, monoclonal PKC antibodies (Signal Transduction Laboratories) were used since HL-60 cells do not have PKCI (13).

HL-60 cells were purchased from ATCC and used between passage numbers 29 and 65. Cells were grown in RPMI 1640 supplemented with L-glutamine, penicillin, and streptomycin (Life Technologies, Inc.), and 10% heat-inactivated fetal bovine serum. All cultures were maintained at 37 °C under an atmosphere of 95% air, 5% CO₂, were passaged three times per week at 2 × 10⁵ cells/ml, and exhibited a doubling time of approximately 30 h. Cell number and viability were determined using a hemocytometer and trypan blue exclusion assay. HL-60 cells were suspended at a density of 2 × 10⁵ cells/ml and incubated 18-24 h at 37 °C prior to the addition of exogenous compounds.

DNA fragmentation was analyzed by agarose gel electrophoresis as reported previously (14), with some modifications. Briefly, HL-60 cells (0.5-1 × 10⁶) were treated with various agents, and following treatment, cells were washed twice in serum-free media and once with ice-cold phosphate-buffered saline. The final pellet was resuspended in 50 µl of lysis buffer (10 mM Tris-HCl (pH 7.4), 40 mM EDTA, 100 mM NaCl, and 0.5% SDS) containing 200 ng/µl RNase A and incubated for 1 h at 37 °C. The lysate was diluted to 300 µl in the same buffer but containing 225 ng/µl Proteinase K (Sigma) and incubated for 16 h at 50 °C. Total genomic DNA was isolated by phenol/chloroform extraction and ethanol precipitation. The DNA was resuspended in TE buffer, pH 8.0. Equal amounts of DNA were loaded into wells of a 1.5% (Seakem GTG, FMC) agarose gel (3 mm thick), and DNA fragments were resolved by electrophoresis at 115 V for 90 min in Tris:borate:EDTA buffer. A 123-base pair DNA molecular weight reference (Promega) was run in parallel. After electrophoresis, the DNA was visualized by staining with ethidium bromide and exposure to UV light.

For quantitation of small molecular weight DNA, HL-60 cells (5 × 10⁶) were exposed for 6 h to drugs and then centrifuged at 2500 rpm for 15 min at 4 °C. Aliquots of the supernatant were removed, adjusted to 5 mM Tris, 15 mM EGTA, 15 mM EDTA, and assayed for released DNA fragments as described below. The remaining medium was aspirated, and the cell pellets were resuspended in 5 mM Tris, 20 mM EDTA, pH 8.0, and 0.1% Triton X-100 lysis buffer (100 µl/10⁶ cells). Cell lysates and medium samples were centrifuged at 30,000 × g for 45 min at 4 °C to separate fragmented DNA (supernatant) from intact DNA (pellet). The lysate and medium samples were diluted in TNE (10 mM Tris, 3 mM NaCl, 1 mM EDTA, pH 7.4) containing 1 µg/ml Hoescht reagent 33528 (Sigma). Relative DNA amounts were determined by spectrofluorophotometry with excitation at 365 (100-nm bandwidth) and filtered emission at 460 nm (10-nm bandwidth).

A sucrose gradient (11:15:40 (w/v) in a 2:1:1 ratio) was used to fractionate cytosol and membranes (15). Cells were washed twice in phosphate-buffered saline, and cell pellets (10⁸ cells) were resuspended in 800 µl of an 11% sucrose solution containing 100 mM phenylmethylsulfonyl fluoride, 1 µg/ml each of aprotinin and leupeptin, and sodium vanadate. Sonicates were layered over 15% and 40% gradients (supplemented with phenylmethylsulfonyl fluoride, aprotinin, leupeptin, and vanadate) and centrifuged at 100,000 × g for 30 min at 4 °C. The top 700 µl contained cytosolic proteins, and the interface between the 15% and 40% sucrose layers contained the membrane proteins.

Protein concentration was determined using Pierce BCA protein assay reagent, and the cytosolic and membrane fractions were diluted to equal protein concentrations with sample buffer (2% (w/v) SDS, 0.05 M dithiothreitol, 62.5 mM Tris, pH 6.8). After boiling and dilution in glycerol/bromphenol blue solution, equal amounts of protein were loaded into the wells of 9% (for PKC) or 12% (for Bcl-2) SDS-polyacrylamide gels, electrophoresed, and transferred to nitrocellulose membranes (Schleicher & Schuell). Membranes were stained with Ponceau S (Sigma) to confirm equal loading of protein. Immunoblotting was performed and proteins were visualized by an enhanced chemiluminescent assay (NEN Life Science Products). Scanning densitometry was performed, and relative protein amounts were measured by the Macintosh software program IPLab Gel.

Antisense and Sense Oligodeoxynucleotides to PKCII

Unprotected antisense and sense oligodeoxynucleotides to PKCII were synthesized by the DNA core laboratory of the Comprehensive Cancer Center of Bowman Gray School of Medicine of Wake Forest University. The 5 region of human PKCII was targeted and had the following 25-mer sequence: 5-CCCCGCAGCCGGGTCAGCCATCTTG-3. The complementary sequence over the same region was used to synthesize sense oligonucleotides to PKCII: 5-CAAGATGGCTGACCCGGCTGCGGGG-3. The purified oligonucleotides were stored at 20 °C. To minimize the action of nucleases, the oligonucleotides were added at a final concentration of 15 µM to cells in serum-free RPMI (supplemented with 5 µg/ml insulin/transferrin). The cultures (2 × 10⁵ cells/ml) were allowed to equilibrate at 37 °C for 24 h prior to the addition of drugs.

RESULTS AND DISCUSSION

Isoforms of PKC appear to have different targets in signaling pathways of specific cell types, and this can lead to diverse cellular responses (16-22). Calcium- and phospholipid-dependent PKCII has been demonstrated to be important in a signaling pathway resulting in TPA-induced differentiation of HL-60 cells to monocytes and adherent macrophages (23-26). In in vitro assays, ET-18-OCH₃ inhibits total PKC activity induced by TPA, as well as TPA-stimulated activity of the partially purified PKCII isoform.² To determine the effect of ET-18-OCH₃ on specific PKC isoforms, we used immunoblot analysis to evaluate translocation from cytosol to membranes, an indicator of PKC activation by TPA. Fig. 1 shows that the majority of PKCII is in the cytosol of unstimulated HL-60 cells. A 1-h treatment with 100 nM TPA induced translocation of this isoform. Simultaneously, we examined the effect of ET-18-OCH₃ on PKCII in unstimulated and TPA-treated cells. The TPA-stimulated translocation of PKCII is inhibited by pretreatment of HL-60 cells for 16 h with 4 µM ET-18-OCH₃. Under these experimental conditions, ET-18-OCH₃ also caused an observable decrease of PKCII in the membranes of non-stimulated cells treated while in log phase growth.

Inhibition of TPA-induced PKCII translocation by ET-18-OCH₃.

Kharbanda et al. (27) demonstrated a time-dependent increase in a serine/threonine kinase activity in HL-60 cells treated with 10 µM ara-C. Therefore, the 2-fold increase in the DAG subclass in response to ara-C could activate one or more DAG-binding PKC isoforms (28). We found that PKC was not significantly translocated during ara-C (10 µM) exposure (0-9 h) of HL-60 cells (data not shown). Cytosolic and membrane levels of Ca²⁺-independent PKC (data not shown) were also not affected by ara-C. However, PKCII levels increased in the membranes isolated from cells treated with 10 µM ara-C (Fig. 2A). Examination of total PKCII by immunoblot analysis demonstrated that ara-C stimulates PKCII up-regulation (28). A maximal increase was consistently observed at approximately 3-5 h, but levels were above the unstimulated control for up to 9 h. ET-18-OCH₃ (2 µM) inhibited the effect of ara-C on membrane-associated levels of PKCII (Fig. 2) over the time period studied.

The loss of viability in the cells treated with the combination of drugs was greater than that observed with ET-18-OCH₃ or ara-C alone. This was observed in cells treated for up to 24 h with 10 µM ara-C and 0.5-2.0 µM ET-18-OCH₃. Inhibition of colony formation by ara-C was enhanced by co-treatment with ET-18-OCH₃ (data not shown). These data prompted an investigation of the effect of ET-18-OCH₃ on ara-C-induced apoptosis in HL-60 cells. One marker for apoptosis is the presence of internucleosomal DNA fragmentation observed as a DNA ladder in multiples of 180-200 base pairs in an ethidium bromide-stained agarose gel. We observed a dose-dependent increase in the levels of internucleosomal DNA cleavage in HL-60 cells treated with ara-C (0.1-10 µM for 8 h). ET-18-OCH₃ at the concentrations we used (0.1-2.0 µM) did not result in significant DNA cleavage (Fig. 3). However, at higher concentrations ET-18-OCH₃ alone can cause apoptosis (29-32). HL-60 cells treated with 10 µM ara-C and ET-18-OCH₃ (0.5-2 µM) showed a marked dose-dependent increase in DNA cleavage compared with ara-C alone (Fig. 3). Enhancement of DNA cleavage was also observed with low concentrations of ara-C (0.1 µM) in the presence of 0.1 µM ET-18-OCH₃ at 16 h (28). These data support the hypothesis that ara-C-induced apoptosis is suppressed by the activation of PKCII. C₂-ceramide causes apoptosis without activation of PKCII, and ET-18-OCH₃ did not further stimulate C₂-ceramide-induced apoptosis (28).

Down-regulation of Bcl-2 by ara-C Is Enhanced by ET-18-OCH₃

Bcl-2 is a member of a recently discovered family of proteins that are regulators of apoptosis (33-37). Bcl-2, a 25-26-kDa phosphoprotein, functions to suppress apoptosis. ara-C treatment induces transcriptional down-regulation of Bcl-2 in HL-60 cells undergoing apoptosis (34). Furthermore, inhibition of Bcl-2 with antisense oligonucleotides induces apoptosis and sensitizes leukemic cells to ara-C (38). Recently, it has been demonstrated that cell-permeable ceramide analogs transcriptionally down-regulate Bcl-2 (39). Therefore, we determined the effect of ET-18-OCH₃ on Bcl-2 protein levels in HL-60 cells treated with ara-C, TPA, or C₂-ceramide. Fig. 4 illustrates the effects of TPA (10 nM), ara-C (10 µM), and C₂-ceramide (100 µM) on Bcl-2 levels in HL-60 cells. TPA caused a time-dependent increase in Bcl-2 while C₂-ceramide caused a marked reduction in Bcl-2. ara-C caused only a moderate decrease in Bcl-2. Therefore, the failure of ara-C to cause a marked reduction in Bcl-2 could be due to the simultaneous stimulation of PKC. Pretreatment of HL-60 cells with ET-18-OCH₃ caused a reduction in Bcl-2 levels in both ara-C-treated (Fig. 5) and TPA-treated cells (28). ET-18-OCH₃ alone did not significantly affect Bcl-2 levels (28). Pretreating HL-60 cells with ET-18-OCH₃ did not have an effect on the reduction of Bcl-2 levels in response to ceramide which did not cause PKCII translocation (28).

Comparison of Bcl-2 protein levels in response to TPA, ara-C, and C₂-ceramide treatment.

ET-18-OCH₃ effect on Bcl-2 levels in response to ara-C in HL-60 cells.

Effect of Antisense Oligonucleotide to PKCII on ara-C Induction of DNA Fragmentation in HL-60 Cells

In addition to PKC, ET-18-OCH₃ can inhibit other enzymes (29, 40-44). Therefore, to provide further evidence that inhibition of PKCII activity is an important event we examined the effect of oligonucleotides to PKCII mRNA on ara-C-induced internucleosomal DNA fragmentation. HL-60 cells in serum-free media were pretreated with or without antisense or sense oligonucleotides to PKCII for 24 h, followed by addition of 10 µM ara-C. Serum-free conditions were used to minimize nuclease activity. This pretreatment was without significant effect on the growth of control, HL-60 cells not treated with ara-C (data not shown). Pretreatment with the antisense (but not sense) oligonucleotides inhibited ara-C-stimulated increases in PKCII (Fig. 6A). Inhibiting PKCII with antisense oligonucleotides also enhanced ara-C-induced DNA fragmentation (Fig. 6B). Treatment with either antisense or sense oligonucleotides for a total of 30 h did not result in significant DNA fragmentation above the level observed in untreated HL-60 cells (Fig. 6B).

Pretreatment with antisense oligonucleotides inhibits PKC activation by ara-C and enhances internucleosomal DNA cleavage induced by ara-C.

In summary, the data presented indicate that the ara-C-stimulated formation of DAG and ceramide lead to activation of two antagonistic signaling pathways. The protective effects of the PKCII-dependent pathway are due to effects on the levels of Bcl-2. Inhibition of the PKCII-dependent pathway results in an enhancement of apoptosis in response to ara-C. Overall, these data indicate that inhibition of PKC may enhance the clinical effectiveness of ara-C and point to strategies for the use of chemotherapeutic agents in new combinations.