Activation of NF-kappa B by Antineoplastic Agents
ROLE OF PROTEIN KINASE C*

(Received for publication, October 2, 1996, and in revised form, February 19, 1997)

Kumuda C. Das and Carl W. White Dagger

From the Department of Pediatrics, National Jewish Medical and Research Center, Denver, Colorado 80206

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
REFERENCES


ABSTRACT

Paclitaxel can induce tumor necrosis factor (TNF) and interleukin-1 gene expression, similar to lipopolysaccharides. Since lipopolysaccharide-induced expression of TNF is related to activation of NF-kappa B, we determined whether NF-kappa B could be activated by paclitaxel. In the human lung adenocarcinoma cell line A549, paclitaxel activated NF-kappa B in a dose-dependent manner with maximal activation after 2-4 h. Since paclitaxel could up-regulate TNF and interleukin-1 secretion and subsequent NF-kappa B activation could be caused by these cytokines, the effect of two other groups of anticancer drugs including vinca alkaloids (vinblastine and vincristine) and anthracyclines (daunomycin and doxorubicin), neither of which induce TNF or interleukin-1 gene expression, were examined. Like paclitaxel, vinblastine, vincristine, daunomycin, and doxorubicin each caused activation of NF-kappa B. Therefore, it is unlikely that activation of NF-kappa B caused by these agents or by paclitaxel is mediated via cytokine up-regulation. Furthermore, actinomycin D and cycloheximide, inhibitors of transcription and translation, respectively, did not inhibit paclitaxel-induced NF-kappa B activation. Several other transcription factors such as AP-1, AP-2, CREB, SP-1, or TFIID were not activated by antineoplastic agents demonstrating specificity of NF-kappa B activation. The involvement of both subunits in the NF-kappa B DNA binding complex was demonstrated by its abrogation by anti-p65 and by supershift by anti-p50 antibodies. Since protein phosphorylation is implicated in the activation of NF-kappa B, the effect of anticancer drugs on protein kinase C activity was measured. Vincristine, daunomycin, and paclitaxel significantly increased protein kinase C activity, and vinblastine and doxorubicin caused similar trends. Following treatment with antineoplastics (1-4 h), cytoplasmic Ikappa Balpha degradation occurred concomitantly with translocation of p65 to the nucleus. Specific protein kinase C inhibitors (bisindolylmaleimide (GF109203X) and calphostin C) blocked the activation of NF-kappa B by each compound. Hence, protein kinase C activation may contribute to NF-kappa B activation by antineoplastic agents.


INTRODUCTION

Paclitaxel, a diterpene compound, was originally isolated from the stem bark of Taxus brevifolia and shown to have antiproliferative activity against various cultured cells as well as antineoplastic activity in tumor patients (1). These effects of paclitaxel appear to be related to its ability to bind to tubulin, to promote microtubule assembly, and to stabilize microtubules by bundle formation (2-4). Recently, Ding et al. (5). have found paclitaxel to exhibit cell cycle-independent, endotoxin-like effects on murine macrophages. Paclitaxel, like endotoxin (lipopolysaccharide; LPS),1 causes murine macrophages to down-regulate TNF receptors and initiate synthesis and secretion of TNF. In a similar fashion, paclitaxel can induce expression of both IL-1alpha and IL-1beta (6). In addition, paclitaxel also induces tyrosine phosphorylation of microtubule associated protein kinases (7). Likewise, paclitaxel enhances gamma -interferon induction of nitric oxide synthase and secretion of nitric oxide (6), a macrophage tumoricidal factor. The pathways linking these responses to paclitaxel are believed to be similar to those causing such responses to LPS (5). Thus, an intracellular target affected by paclitaxel might be involved in actions of LPS in macrophages and other cells. Determination of which intracellular molecule paclitaxel and LPS affect in common could provide further insight into the actions of LPS on mammalian cells and participation of the cytoskeleton in responses of cells to their environment. One potential intracellular target of these compounds is the transcription factor NF-kappa B. NF-kappa B, named for its ability to recognize a kappa  light chain immunoglobulin gene regulatory element, can participate in the regulation of numerous genes (reviewed in Baeuerle (8)). Under basal conditions, it usually exists as a heterodimer of 50- and 65-kDa subunits bound to an inhibitor protein Ikappa B in the cytoplasm. Various stimuli cause Ikappa B to dissociate from the complex allowing the heterodimer to migrate to the nucleus and activate gene expression. Because NF-kappa B is involved in several gene expression phenomena caused by LPS (9), we sought to determine if NF-kappa B is activated by paclitaxel in the pulmonary adenocarcinoma cell line A549.

Herein, we report that paclitaxel can cause NF-kappa B activation. Because paclitaxel can induce cytokines such as TNF and IL-1, which could then activate NF-kappa B (8), we assessed the effect of additional classes of antitumor drugs, vinca alkaloids, and anthracyclines, which do not induce TNF or IL-1 secretion. The mechanism of action of the vinca alkaloids is related to the inhibition of microtubule formation in the mitotic spindle resulting in an arrest of the dividing cells at metaphase (10). By contrast, the mechanism of action of anthracyclines is related to their intercalation into DNA (11) and inhibition of RNA transcription (12). This interaction with DNA confers the cytotoxic effect of these agents. These drugs also cause DNA strand breaks that appear to be mediated by DNA topoisomerase II (13). Thus, the mode of action, structure, and function of paclitaxel is entirely different than that of vinca alkaloids and anthracyclines. Nonetheless, each of these classes of antineoplastics shared the property of NF-kappa B activation.

Because protein kinases are involved in NF-kappa B activation by a variety of agents (14, 15), we investigated a potential role for these proteins in NF-kappa B activation by antineoplastics. Protein kinase C (PKC) is a calcium and phospholipid-dependent protein kinase that plays an important role in signal transduction pathways regulating cell growth (16). In some drug-resistant cell lines, elevation of PKC activity occurs in response to antitumor drugs (17). Therefore, the effect of antitumor drugs on the activity of protein kinase C was examined. In addition, the effects of calphostin C, a specific inhibitor of PKC, which binds to its regulatory region (18), and the bisindolylmaleimide GF109203X, which specifically inhibits by binding to the catalytic region of PKC, were assessed relative to the activation of NF-kappa B by anti-tumor drugs.


EXPERIMENTAL PROCEDURES

Materials

Vinblastine (Velban®, Lilly) and vincristine (Oncovin®, Lilly) were obtained commercially. Paclitaxel (Taxol®) was obtained from Bristol-Meyers Squibb Company. Doxorubicin hydrochloride, daunomycin hydrochloride, and cremophor EL were obtained from Sigma. Anti-p50 and anti-p65 antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). All other materials were obtained in the highest available grade.

Cell Culture and Treatments

A549 cells (adenocarcinoma cells) were grown in F12-K (Kaigan's modified) supplemented with 10% fetal calf serum and 100 units of penicillin/streptomycin. Confluent monolayers were treated with various concentration of drugs for various time periods as indicated in the figure legends. Cell viability was determined by the trypan blue exclusion method. Cell viability was unchanged by any of the agents studied over the time courses (up to 4 h) and at the dosages studied. Paclitaxel was dissolved in Me2SO.

Nuclear Extract Preparation

Nuclear extracts were prepared by the method of Staal et al. (20) with the following modifications. Approximately 107 cells were washed in 10 ml of phosphate-buffered saline and centrifuged (1,500 × g for 5 min). The pellet was resuspended in phosphate-buffered saline (1 ml), transferred into an Eppendorf tube, and centrifuged again (16,000 × g; 15 s). Phosphate-buffered saline was removed, and the cell pellet was resuspended in 400 µl of buffer A (10 mM HEPES, pH 7.8, 10 mM KCl, 0.1 mM EDTA, 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, leupeptin (0.5 mg/ml), antipain (0.3 mg/ml)) by gentle pipetting. The cells were allowed to swell on ice (15 min), after which 25 µl of 10% Nonidet P-40 (Sigma) were added, and the tube was vortexed vigorously for 10 s. The homogenate was centrifuged for 30 s in a microcentrifuge. The nuclear pellet was resuspended in buffer C (20 mM HEPES, pH 7.8, 0.42 M NaCl, 5 mM EDTA, 5 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 10% (v/v) glycerol), and the tube was rocked gently at 4 °C for 30 min on a shaking platform. The nuclear extract was centrifuged for 10 min in a microcentrifuge at 4 °C, and the supernatant was frozen at -70 °C in aliquots until electrophoretic mobility shift assay (EMSA) was done. Protein was quantified by Bradford assay (Bio-Rad) (21).

Electrophoretic Mobility Shift Assay

For the EMSA the following specific oligonucleotides were obtained from Promega and end-labeled using T4 polynucleotide kinase (Life Technologies, Inc.) and [gamma -32]P ATP (ICN) in 10 × kinase buffer (0.5 M Tris-HCl, pH 7.5, 0.1 M MgCl2, 50 mM dithiothreitol, 1 mM spermidine, and 1 mM EDTA). For competition studies, 3.5 pmol of unlabeled oligonucleotides were used. Oliognucleotides were: NF-kappa B, 5' AGT TGA GGG GAC TTT CCC AGG C 3'; AP-1, 5' CGC TTG ATG AGT CAG CCG GAA 3'; SP1, 5' ATT CGA TCG GGG CGG GGC GAG C 3'; AP-2, 5' GAT CGA ACT GAC CGC CCG CGG CCC GT 3'; CREB, 5' AGA GAT TGC CTG ACG TCA GAG AGC TAG 3'; TFIID, 5' GCA GAG CAT ATA AGG TGA GGT AGG A 3'. 1.75 pmol was end-labeled with [gamma -32P]ATP (ICN) using T4 polynucleotide kinase (Life Technologies, Inc.) in 10 × kinase buffer. The labeled double-stranded oligonucleotide was separated from free [gamma -32P]ATP using a G-50 Sephadex spin column (5 Prime right-arrow 3 Prime, Inc., Boulder, CO). The nuclear protein-32P-labeled oligonucleotide complex was separated from free 32P-labeled oligonucleotide by electrophoresis through a 6% native polyacrylamide gel in a running buffer of 0.25 × TBE (5 × TBE = 500 mM Tris, pH 8.0, 450 mM borate, 5 mM EDTA). The binding reaction was performed according to a modified method of Garner and Revzin (22). The reaction mixture without labeled oligonucleotide was preincubated for 15 min at 4 °C followed by a 20-min incubation after addition of labeled oligonucleotide at room temperature. The binding reaction contained 10 µg of sample protein, 5 µl of 5 × incubation buffer (20% glycerol, 5 mM MgCl2, 5 mM EDTA, 5 mM dithiothreitol, 500 mM NaCl, 50 mM Tris-HCl, pH 7.5, 0.4 mg/ml calf thymus DNA). The binding reaction also contained poly(dI-dC) (2 µg; Pharmacia Biotech Inc.) in a total reaction volume of 25 µl.

Supershift Assay

For supershift assay some of the binding reactions contained 150 ng of anti-p50 or anti-p65 (Santa Cruz Biotechnology) along with 2 µg of poly(dI-dC) (Pharmacia). Binding reactions were incubated at room temperature for 15 min followed by additional incubation for 3 h at 4 °C.

Western Blot Detection of Ikappa Balpha and p65

Nuclear extract was prepared as described earlier. Post nuclear supernatant was treated as the cytoplasmic extract. Nuclear or cytoplasmic extracts (25 µg) were separated by electrophoresis in a 12% polyacrylamide gel under denaturing and reducing conditions. Protein was transferred to a nitrocellulose membrane (Hybond-ECL, Amersham Corp.) and immunoblotted with monoclonal anti-Ikappa Balpha or anti-p65 (Santa Cruz Biotechnology) antibodies, and visualized by the ECL-system (Amersham) using anti-rabbit horseradish peroxidase IgG (Sigma).

Protein Kinase C Activity Assay

PKC activity was determined using PKC assay kit of Life Technologies, Inc. based on the measurement of the phosphorylation of acetylated myelin basic protein as described by Yasuda et al. (23). Briefly, 5 × 106 cells were washed in phosphate-buffered saline followed by scraping of the monolayers in 0.5 ml of buffer A (20 mM Tris, pH 7.5, 0.5 mM EDTA, 0.5 mM EGTA, 0.5% Triton X-100, 25 µg/ml each of aprotinin and leupeptin). Cells were homogenized with 10-15 strokes of a precooled Dounce homogenizer with a B-type pestle at 4 °C. Following incubation of the cell homogenate with buffer A for 30 min, the lysate was centrifuged, and the supernatant was applied to a DEAE ion exchange column (Bio-Rad). Partially purified PKC was eluted with 5 ml of buffer C (20 mM Tris, pH 7.5, 0.5 mM EDTA, 0.5 mM EGTA, 10 mM 2-mercaptoethanol, 0.2 M NaCl). The DEAE eluate (25 µl) was assayed for PKC activity using 50 µM acetylated myelin basic protein as a substrate. Activity of PKC was expressed as pmol/min/5 × 106 cells.


RESULTS

Activation of NF-kappa B by Paclitaxel

Dose Response

To investigate whether paclitaxel induces NF-kappa B activation, we treated A549 cells with various concentration of paclitaxel. Nuclear extracts were examined for NF-kappa B binding activity by EMSA using consensus NF-kappa B binding sequences. As shown (Fig. 1A), paclitaxel increased NF-kappa B activation in a dose-dependent manner (lanes 6-9). As little as 7 µM paclitaxel activated NF-kappa B.


Fig. 1. Activation of NF-kappa B by paclitaxel (Taxol). Time course: A549 cells were incubated with paclitaxel (98 µM) for 1-4 h. Following incubation, nuclear extract was prepared and EMSA performed as described in the text. Lane 1, unstimulated cells; lane 2, cells treated with paclitaxel (98 µM) for 1 h; lane 3, 2 h; lane 4, 3 h; lane 5, 4 h of paclitaxel treatment. Throughout the gel, the uppermost band represents NF-kappa B. Dose response: A549 cells were incubated with paclitaxel (7, 28, 56, and 98 µM) for 2 h. After incubation, nuclear extract was prepared and EMSA was performed as described under "Experimental Procedures." Lanes 6-9, paclitaxel; lane 10, competition with cold NF-kappa B oligonucleotide with nuclear extract of paclitaxel-stimulated cells.
[View Larger Version of this Image (62K GIF file)]

Time Course

Paclitaxel treatment resulted in a rapid activation of NF-kappa B binding activity. NF-kappa B binding activity was detectable after 1 h, and continued to increase through 4 h (Fig. 1B). Thus, induction of NF-kappa B by paclitaxel is a relatively rapid response.

Effect of Actinomycin D and Cycloheximide on NF-kappa B Activation by Paclitaxel

To determine whether secretion of TNF-alpha or IL-1 by paclitaxel (5, 6) is responsible for its activation of NF-kappa B, cells were pretreated with indicated concentrations of actinomycin D or cycloheximide (10 min) followed by exposure to paclitaxel (2 h). As demonstrated (Fig. 2), neither actinomycin D nor cycloheximide inhibited the activation of NF-kappa B. Instead there was a potentiation of activation of NF-kappa B by paclitaxel following actinomycin D or cycloheximide pretreatment. In addition, both cycloheximide and actinomycin D alone activated NF-kappa B. Thus, activation of NF-kappa B by paclitaxel persisted despite inhibition of transcription or of new protein synthesis.


Fig. 2. Effect of actinomycin D and cycloheximide on the activation of NF-kappa B by paclitaxel (Taxol). A, A549 cells were preincubated with actinomycin D (1 µg/mL) or cycloheximide (10 µg/mL) for 10 min followed by incubation with paclitaxel for an additional 2 h. Nuclear extract was prepared and EMSA was done as described in the text. Lane 1, unstimulated cells; lane 2, paclitaxel (113 µM); lane 3, actinomycin D (1 µg/ml); lane 4, paclitaxel (113 µM) + actinomycin D (1 µg/ml); lane 5, cycloheximide (10 µg/ml); lane 6, paclitaxel (113 µM) + cycloheximide (10 µg/ml). B, effect of dexamethasone on NF-kappa B activation by paclitaxel. A549 cells were preincubated for 16 h with dexamethasone (1 µM) followed by incubation with paclitaxel (2 h; 113 µM). Nuclear extracts were prepared, and EMSA was done as described under "Experimental Procedures." Lane 1, unstimulated cells; lane 2, dexamethasone (1 µM); lane 3, paclitaxel (113 µM); lane 4, dexamethasone (1 µM) + paclitaxel (113 µM).
[View Larger Version of this Image (39K GIF file)]

Effect of Dexamethasone on NF-kappa B Activation by Paclitaxel

In macrophages, paclitaxel has been reported to induce TNF-alpha and IL-1 production (5, 6). Production of these cytokines is inhibitable by dexamethasone. Therefore, to rigorously evaluate a potential role for TNF-alpha and IL-1 in paclitaxel-induced activation of NF-kappa B, we pretreated the cells with dexamethasone (1 µM) for 16 h followed by paclitaxel treatment (2 h). By EMSA, there was no inhibition of NF-kappa B activation by paclitaxel due to dexamethasone pretreatment (Fig. 2, lanes 7-10). This indicates that TNF-alpha or IL-1 could not be responsible for initial activation of NF-kappa B by paclitaxel.

Activation of NF-kappa B by Vinblastine and Vincristine

Dose Response

Vinblastine and vincristine treatment resulted in activation of NF-kappa B in a dose-dependent manner (Fig. 3A). This activation was specific as demonstrated by competition reaction with unlabeled oligonucleotide. Activation of NF-kappa B by vinblastine was strikingly dose-dependent. Similarly, vincristine induced activation of NF-kappa B and demonstrated a dose dependence up to 50 µM. However, an increase in the vincristine concentration to 100 µM (Fig. 3A, lane 9) did not cause a further increase in the activation of NF-kappa B. Control cells were treated with the diluent for both agents (mannitol dissolved in distilled water; 100 mg/ml).


Fig. 3. Activation of NF-kappa B by vinblastine and vincristine. A, dose response: A549 cells were incubated with indicated amount of vinblastine or vincristine for 4 h. Following incubation nuclear extract was prepared and EMSA was performed as described under "Experimental Procedures." Lane 1, untreated cells; lane 2, vinblastine (10 µM); lane 3, vinblastine (30 µM); lane 4, vinblastine (50 µM); lane 5, vinblastine (100 µM); lane 6, vincristine (10 µM); lane 7, vincristine (20 µM); lane 8, vincristine (50 µM); lane 9, vincristine (100 µM); lane 10, competition (Comp.) of nuclear extracts stimulated with vinblastine (50 µM) with cold NF-kappa B consensus oligonucleotides; lane 11, competition of nuclear extracts stimulated with vincristine (50 µM) with cold NF-kappa B consensus oligonucleotides. B, time course: A549 cells were incubated with vinblastine (50 µM) or vincristine (50 µM) for various time periods. After incubation, nuclear extract was prepared and EMSA was performed as described in the text. Lane 1, control; lanes 2-6, NF-kappa B binding after 30 min to 4 h of stimulation with vinblastine. Lanes 7-11, NF-kappa B binding after 30 min to 4 h of stimulation with vincristine; lane 12, competition of nuclear extracts stimulated with vinblastine (50 µM) for 4 h with cold NF-kappa B consensus oligonucleotide; lane 13, competition of nuclear extracts stimulated with vincristine (50 µM) for 4 h with cold NF-kappa B consensus oligonucleotide.
[View Larger Version of this Image (76K GIF file)]

Time Course

A549 cells were incubated with vinblastine (50 µM) or vincristine (50 µM) for various intervals (30 min, and 1, 2, 3, or 4 h). As demonstrated in Fig. 3B, NF-kappa B activation was increased in a time-dependent manner. Vinblastine and vincristine both activated NF-kappa B after 30 min through 4 h. Vinblastine-induced activation was time-dependent. Vincristine-induced activation differed only in that it was of relatively delayed onset (Fig. 3B, lanes 9-11).

Activation of NF-kappa B by Doxorubicin and Daunomycin

Dose Response

Daunomycin activated NF-kappa B in a dose-dependent manner. Maximal activation occurred at a concentration of 5 µM (Fig. 4A, lane 4). Further increases in the concentration of daunomycin caused decreased NF-kappa B activation (Fig. 4B, lanes 3-5). This is perhaps not surprising since the toxic effect of these drugs at higher concentrations has been well documented (24, 25). Doxorubicin showed a similar effect on NF-kappa B activation. Maximal activation occurred at 30 µM (Fig. 4B, lane 7). Additional increases in concentration did not potentiate NF-kappa B activation further. Thus, daunomycin-induced NF-kappa B activation is dose-dependent up to 5 µM and that due to doxorubicin is dose-dependent up to 30 µM.


Fig. 4. Activation of NF-kappa B by daunomycin and doxorubicin. A, dose response: A549 cells were incubated with daunomycin (0, 2, 4, or 5 µM) or doxorubicin (1, 3, 5, 10, 15, or 30 µM) for 3 h. After incubation, nuclear extract was prepared and EMSA was performed as described under "Experimental Procedures." Lane 1, unstimulated cells; lanes 2-4, daunomycin at indicated concentrations; lanes 5-10, doxorubicin at indicated concentrations; lanes 11 and 12, competition (Comp.) with cold NF-kappa B oligonucleotide with nuclear extract of daunomycin (Dauno) (5 µM) or doxorubicin (Doxo) (15 µM) stimulated cells. B, lane 1, unstimulated cells; lanes 2-5, daunomycin (Daun.) 5, 10, 15, and 20 µM; lanes 6-9, doxorubicin (Doxo.) 15, 30, 50, and 100 µM. C, time course: A549 cells were incubated with daunomycin (5 µM) or doxorubicin (15 µM) for various time periods. After incubation, nuclear extract was prepared and EMSA was performed as described under "Experimental Procedures." Lanes 1 and 7, unstimulated cells; lanes 2-5, NF-kappa B binding after 1-4 h of stimulation with daunomycin (5 µM); lanes 8-11, NF-kappa B binding after 1-4 h of stimulation with doxorubicin (15 µM); lane 6, competition of nuclear extract of cells treated with daunomycin (5 µM) for 4 h, and lane 12, competition of nuclear extract of cells treated with daunomycin (15 µM) for 4 h with cold NF-kappa B consensus oligonucleotide.
[View Larger Version of this Image (48K GIF file)]

Time Course

In addition, the drugs increased NF-kappa B activation in a time-dependent manner (Fig. 4C), with continued activation through 4 h. Specifity of NF-kappa B activation was confirmed by competition studies with unlabeled oligonucleotides.

Effect of Anti-p50 and Anti-p65 Antibodies on Activation of NF-kappa B by Antineoplastic Agents

Each of the antineoplastic agents activated NF-kappa B, and anti-p65 eliminated DNA binding of NF-kappa B induced by each drug (Fig. 5). Anti-p50 supershifted NF-kappa B binding to a considerable extent. Thus, antineoplastic agents caused the translocation of p50-p65 heterodimer to the nucleus, and both subunits appear to be present in the kappa B-binding complex. However, p65 appears to be primarily responsible for binding to DNA since anti-p65 abolished DNA binding of NF-kappa B.


Fig. 5. Effect of anti-p50 and anti-p65 on the DNA binding of NF-kappa B induced by antineoplastic agents. Nuclear extracts of cells treated with vinblastine (Vinb.) (50 µM), vincristine (Vinc.) (50 µM), daunomycin (Daun.) (5 µM), doxorubicin (Doxo.) (15 µM), and paclitaxel (Taxol) (56 µM) were incubated with 150 ng of anti-p50 or anti-p65 antibody in a 25-µl binding reaction with poly(dI-dC) (2 µg) as described under "Experimental Procedures." Lane 1, cells stimulated with vinblastine (50 µM); lanes 2 and 3, vinblastine-stimulated nuclear extract incubated with anti-p50 or anti-p65 antibody; lane 4, cells stimulated with 50 µM vincristine; lane 5 and 6, vincristine-stimulated nuclear extract incubated with anti-p50 or anti-p65 antibody; lane 7, cells stimulated with daunomycin (5 µM); lanes 8 and 9, daunomycin-stimulated nuclear extract incubated with anti-p50 or anti-p65 antibody; lane 10, cells stimulated with 15 µM doxorubicin; lanes 11 and 12, doxorubicin-stimulated nuclear extract incubated with anti-p50 or anti-p65 antibody; lane 13, cells stimulated with paclitaxel (56 µM); lanes 14 and 15, paclitaxel-stimulated nuclear extract incubated with anti-p50 or anti-p65 antibody.
[View Larger Version of this Image (55K GIF file)]

Activation of NF-kappa B by Antineoplastic Agents Is Specific

To determine specificity of activation of NF-kappa B by antineoplastic agents, EMSA was performed with probes for AP-1, AP-2, CREB, TFIID, and SP-1 binding factors. As shown (Figs. 6, 7, 8), none of these transcription factors was activated. Thus, activation of NF-kappa B by anticancer drugs appears to be a specific response.


Fig. 6. Effect of vincristine and daunomycin on other transcription factors. A549 cells were incubated with vincristine (Vinc.) (50 µM) or daunomycin (Daun.) (15 µM) for 3 h. After incubation, nuclear extract was prepared and EMSA was performed as described under "Experimental Procedures." Labeled probes for AP-1, AP-2, CREB, TFIID, and SP-1 were incubated with 10 µg of nuclear extract of vincristine- or daunomycin-treated cells. Lanes 1-15, vincristine-stimulated extract probed with AP-1, AP-2, CREB, TFIID, and SP-1 binding oligonucleotides. Lanes 16-30, daunomycin-stimulated extract probed with these same labeled oligonucleotides.
[View Larger Version of this Image (41K GIF file)]


Fig. 7. Effect of doxorubicin and paclitaxel on other transcription factors. A549 cells were incubated with doxorubicin (Doxo.) (15 µM) for 3 h or paclitaxel (Taxol) (56 µM) for 2 h. After incubation, nuclear extract was prepared, and EMSA was performed as described under "Experimental Procedures." Labeled probes for AP-1, AP-2, CREB, TFIID, and SP-1 were incubated with 10 µg of nuclear extract of doxorubicin- or paclitaxel-treated cells. Lane 1-10, doxorubicin-stimulated extract probed with AP-1, AP-2, CREB, TFIID, and SP-1 binding oligonucleotides. Lanes 11-20, paclitaxel-stimulated extract probed with these same oligonucleotides.
[View Larger Version of this Image (94K GIF file)]


Fig. 8. Effect of vinblastine on other transcription factors. A549 cells were incubated with vinblastine (50 µM) for 3 h. After incubation, nuclear extract was prepared, and EMSA was performed as described under "Experimental Procedures." Labeled probes for AP-1, AP-2, CREB, and SP-1 were incubated with nuclear extract (10 µg) of vinblastine-treated cells.
[View Larger Version of this Image (78K GIF file)]

Antineoplastic Agents Cause Ikappa Balpha Degradation and p65 Translocation

Among the events that occur during activation of NF-kappa B are the degradation of the Ikappa B inhibitory unit and migration of the p65 subunit to the nucleus. To investigate whether the anticancer drugs cause the degradation of Ikappa B, we prepared cytoplasmic extracts from the anticancer drug-treated cells and examined the extract using gel electrophoresis followed by Western blot using Ikappa B-specific polyclonal antibodies. Each of the anticancer drugs caused degradation of Ikappa B (Fig. 9). Paclitaxel treatment for 1 h resulted in Ikappa B degradation, which was complete after 2-3 h. Although there were minor variations in the kinetics with which Ikappa B disappeared from the cytoplasm and was later restored following treatment with the other anti-cancer drugs, the effect of each was similar.


Fig. 9. Degradation of Ikappa B-alpha induced by anti-cancer drugs. A549 cells were incubated with indicated concentrations of anticancer drugs for various intervals. Following incubation, cytoplasmic extract was prepared as described under "Experimental Procedures." Cytoplasmic extract (25 µg protein) was resolved in a 12% Tris-glycine polyacrylamide gel. Protein was transferred to nitrocellulose, and Western blot was performed using anti-Ikappa B antibody. Bands were visualized using ECL procedure (Amersham). Row 1, unstimulated cells and cells stimulated with paclitaxel (Taxol) (56 µM) for 15 min to 3 h; row 2, unstimulated cells (Con) and cells stimulated with vinblastine (50 µM) for 1 h to 4 h; row 3, unstimulated cells and cells stimulated with vincristine (50 µM) for 1-4 h; row 4, unstimulated cells and cells stimulated with daunomycin (5 µM) for 1-4 h; row 5, unstimulated cells and cells stimulated with doxorubicin (15 µM) for 1-4 h.
[View Larger Version of this Image (42K GIF file)]

When we examined the appearance of p65 in the nuclear extracts of cells stimulated with anti-cancer drugs or untreated controls, p65 was not detected in nuclear extracts of control cells (Fig. 10). Paclitaxel treatment caused appearance of p65 after 2 h which continued through 4 h. Again, minor differences in the rate of appearance of p65 in the nucleus occurred, but each agent caused this effect in a manner which paralleled the associated activation of NFkappa B measured by EMSA.


Fig. 10. Translocation of p65 to nucleus induced by anti-cancer drugs. Nuclear extracts of A459 cells stimulated with indicated concentrations of anti-cancer drugs for various intervals (1-4 h) were examined for the presence of NFkappa B p65 subunit. Nuclear protein extract (25 µg) was resolved in a 12% Tris-glycine polyacrylamide gel. Protein was transferred to nitrocellulose, and Western blot was performed using anti-p65 antibody. Bands were visualized using the ECL procedure (Amersham).
[View Larger Version of this Image (34K GIF file)]

Effect of Antineoplastic Drugs on Protein Kinase C Activity

Since protein kinase C is implicated in activation of NF-kappa B by a variety of stimuli, we evaluated the effect of anticancer drugs on PKC activity. A549 cells were treated with indicated concentrations of anti-tumor drugs for 30 min. Following incubation, cells were harvested and assayed for PKC activity as described. Vincristine-, daunomycin-, and paclitaxel-treated cells demonstrated significant increases in PKC activity (Table I). Vinblastine and doxorubicin increased PKC activity by 62 and 90%, although these changes were not statistically significant.

Table I. Effect of anticancer drugs on protein kinase C activity

A549 cells were incubated with indicated concentration of anticancer drugs for 30 min. After incubation, cells were washed twice with HBSS, PKC was partially purified by DEAE ion exchange chromatography, and extracts were assayed as described in the text. Total cellular PKC was expressed as picomoles of PKC/min/5 × 106 cells.

Stimulus PKC/min/5 × 106 cells

µM pmol
None 9.8  ± 1.5
Vinblastine 50 15.94  ± 1.08
Vincristine 50 31.64  ± 7.09a
Daunomycin 4 30.30  ± 2.75a
Doxorubicin 15 17.94  ± 2.25
Paclitaxel 56 22.46  ± 3.09a

a Significantly different from control at p < 0.05.

Effect of Bisindolylmaleimide and Calphostin C, Specific Inhibitors of Protein Kinase C, on NF-kappa B Activation by Antineoplastic Agents

GF109203X, a bisindolylmaleimide derivative, is a specific inhibitor of PKC (19). GF109203X (15 or 25 µM) inhibited activation of NF-kappa B by each of the antineoplastic agents (Fig. 11A). This effect was potent and consistent. Calphostin C, a specific inhibitor of PKC, which binds its regulatory region, also inhibited NF-kappa B activation by each of the antineoplastic agents (Fig. 11B).


Fig. 11. Inhibition of NF-kappa B activation by specific PKC inhibitors. A, inhibition by bisindolylmaleimide. A549 cells were preincubated with GF109203X (15 or 25 µM) for 1 h followed by incubation with the indicated concentration of anti-cancer drugs. Nuclear extract was prepared, and EMSA was performed. Lane 1, unstimulated cells; lane 2, cells treated with vincristine (Vinc.) (50 µM); lanes 3 and 4, cells pretreated with GF109203X (15 or 25 µM) followed by incubation with vincristine; lane 5, unstimulated cells; lane 6, cells treated with daunomycin (Daun.) (15 µM); lane 7, cells treated with doxorubicin (Doxo.) (15 µM); lanes 8 and 9, cells pretreated with GF109203X (15 or 25 µM) followed by incubation with daunomycin; lane 10, cells pretreated with GF109203X (15 µM) followed by incubation with doxorubicin; lane 11, unstimulated cells; lane 12, cells treated with vinblastine (Vinb.) (50 µM); lanes 13 and 14, cells pretreated with GF109203X (15 or 25 µM) followed by incubation with vinblastine; lane 15, cells treated with paclitaxel (Taxol) (56 µM); lanes 16 and 17, cells pretreated with GF109203X (15 or 25 µM) followed by incubation with paclitaxel. B, inhibition by calphostin C. Lane 1, unstimulated cells; lane 2, cells treated with calphostin C (1 µM); lanes 3 and 4, cells treated with daunomycin (5 µM) alone or daunomycin (5 µM) plus calphostin C; lanes 5 and 6, cells treated with doxorubicin alone or doxorubicin + calphostin C; lanes 7 and 8, cells treated with vinblastine alone or vinblastine plus calphostin C (800 nM); lanes 9 and 10, cells treated with vincristine (50 µM) alone or vincristine (50 µM) plus calphostin C (800 nM); lanes 11 and 12, cells treated with paclitaxel alone or paclitaxel (113 µM) plus calphostin C (400 nM).
[View Larger Version of this Image (79K GIF file)]


DISCUSSION

In the present investigation, we demonstrated that paclitaxel induces NF-kappa B activation in the pulmonary adenocarcinoma cell line A549 in a dose- and time-dependent manner. Extension of these studies revealed that compounds from two additional classes of antineoplastic agents also caused NF-kappa B activation. Thus, this appears to be a generalized effect of antiproliferative drugs and suggests that a number of perturbations of cellular homeostasis can activate NF-kappa B. This is not altogether surprising in view of the wide variety of noxious stimuli now recognized to cause NF-kappa B activation.

In macrophages, paclitaxel can induce TNF-alpha or IL-1 gene expression (5, 6). Thus, it is possible that paclitaxel-induced TNF-alpha could then activate NF-kappa B. To evaluate this possibility we pretreated cells with the transcriptional inhibitor actinomycin D or the translational inhibitor cycloheximide. Neither of these inhibited activation of NF-kappa B by paclitaxel. Instead, there was superinduction of NF-kappa B by paclitaxel in the presence of actinomycin D or cycloheximide. Hence, this activation of NF-kappa B by paclitaxel is not dependent on new synthesis of mRNA or protein, which excludes the possibility that the activation of NF-kappa B by paclitaxel could have been cytokine-mediated. To further evaluate a potential role of cytokines in paclitaxel-induced NF-kappa B activation, we examined the effect of dexamethasone which inhibits induction of TNF and IL-1 by paclitaxel. As shown under "Results," dexamethasone pretreatment did not inhibit NF-kappa B activation by paclitaxel. We also evaluated the effect of other antitumor drugs that do not induce cytokine gene expression and which are different in structure and mode of action than paclitaxel. These included vinblastine, vincristine, daunomycin, and doxorubicin. Each of these drugs induced NF-kappa B activation in a dose- and time-dependent manner. Thus, although autocrine mechanisms cannot be altogether excluded, the data do not support a role for cytokine up-regulation in the initial activation of NF-kappa B by anti-tumor drugs. Nonetheless, it remains quite plausible that autocrine mechanisms such as induction of cytokines could amplify NF-kappa B activation in vivo once the process is initiated by these agents.

As demonstrated by supershift assay (Fig. 5), p50/p65 heterodimer is involved in the activation of NF-kappa B by anti-tumor drugs. When activated nuclear extract was probed with consensus oligonucleotides for transcription factors such as AP-1, CREB, SP-1, or TFIID, no DNA binding was observed (Figs. 6, 7, 8), suggesting that the activation of NF-kappa B by anti-tumor drugs is specific. Activation of NF-kappa B is dependent on the detachment and degradation of inhibitor protein Ikappa B. To further investigate the mechanism of NF-kappa B activation, we probed the cytoplasmic extracts of antitumor drug-stimulated cells with antibody to Ikappa B in a Western blot analysis (Fig. 9). As demonstrated, Ikappa B was rapidly detached from the complex and subsequently degraded. This phenomenon occurred concomitantly with appearance of p65 in the nuclear extract (Fig. 10).

Since removal of Ikappa B from the NF-kappa B complex could be achieved by phosphorylation of Ikappa B, we suspected a role for a protein kinase in the activation of NF-kappa B by antitumor drugs. Protein kinase C has been shown to be involved in the activation of NF-kappa B by a variety of agents (15, 16). Thus, when we measured PKC activity after 30 min of stimulation with antitumor drugs, an increased activity was observed. This activity was significantly higher in paclitaxel-, vincristine-, and daunomycin-stimulated cells than in unstimulated cells. Similar trends in cell PKC activity occurred following treatment with vinblastine and doxorubicin (Table I). To further ascertain the role of PKC in the activation process, we incubated cells with specific inhibitors of PKC, including calphostin C or bisindolylmaleimide GF103209X. Calphostin C is a specific inhibitor which binds to the regulatory region of PKC (18). On the other hand, GF109203X is a specific inhibitor which binds to the catalytic domain of PKC (19). As demonstrated, both compounds inhibited activation of NF-kappa B by antineoplastic agents (Fig. 11). Thus, a role for PKC in the activation of NF-kappa B by antitumor drugs is demonstrated.

It has been reported that NF-kappa B can be activated by microtubule-depolymerizing agents (26). In that study, such activation could be abrogated by microtubule-stabilizing agents (26), suggesting a possible role for cytoskeletal disorganization as a signaling event for the activation of NF-kappa B. However, in the present report we have shown that paclitaxel, which is a microtubule-stabilizing anti-tumor agent, vinblastine and vincristine, which are microtubule-depolymerizing agents, and daunomycin and doxorubicin, which are drugs that intercalate into DNA and inhibit RNA synthesis, all could induce NF-kappa B. Although these drugs are structurally and functionally dissimilar, they share a common property, which is activation of protein kinase C. Thus, since these drugs have effects other than interfering with microtubule assembly, it is possible that they induce NF-kappa B by mechanisms distinct from microtubule-associated mechanisms. As demonstrated, one such potential mechanism is through the increased activity of protein kinase C.


FOOTNOTES

*   This work was supported in part by a Fellowship Grant from the American Lung Association (to K. C. D.) and National Institutes of Health Grants HL 46481 (to C. W. W.) and 1RO1 HL 52732 (to C. W. W.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Dagger    Recipient of an American Heart Association Established Investigator Award with partial funding support of the American Heart Association of Colorado. To whom correspondence should be addressed: Dept. of Pediatrics, National Jewish Center, J-103, 1400 Jackson St., Denver, CO 80206. Tel.: 303-398-1617; Fax: 303-270-2189.
1   The abbreviations used are: LPS, lipopolysaccharide; TNF, tumor necrosis factor; PKC, protein kinase C; EMSA, electrophoretic mobility shift assay; IL, interleukin.

REFERENCES

  1. Wani, M. C., Taylor, H. L., Wall, M. E., Coggon, P., and McPhail, A. T. (1971) J. Am. Chem. Sci. 93, 2325-2327
  2. Kumar, N. (1981) J. Biol. Chem. 256, 10435-10441 [Abstract/Free Full Text]
  3. Schiff, P. B., Fant, J., and Horwitz, S. B. (1979) Nature 277, 665-667 [Medline] [Order article via Infotrieve]
  4. Schiff, P. B., and Horwitz, S. B. (1980) Proc. Natl. Acad. Sci. U. S. A. 77, 1561-1565 [Abstract]
  5. Ding, A. H., Porteu, F., Sanchez, E., and Nathan, C. F. (1990) Science 248, 370-372 [Medline] [Order article via Infotrieve]
  6. Bogdan, C., and Ding, A. (1992) J. Leukocyte Biol. 52, 119-121 [Abstract]
  7. Manthey, C. L., Branses, M. B., Perra, P. Y., and Vogel, S. N. (1992) J. Immunol. 149, 2459-2465 [Abstract/Free Full Text]
  8. Baeuerle, P. A. (1991) Biochim. Biophys. Acta 1072, 63-80 [CrossRef][Medline] [Order article via Infotrieve]
  9. Shakhov, A. N., Collart, M. A., Vassalli, P., Nedospasov, S. A., and Jongeneel, C. V. (1990) J. Exp. Med. 171, 35-47 [Abstract]
  10. Watanabe, K., and West, W. L. (1982) Fed. Proc. 41, 2292-2299 [Medline] [Order article via Infotrieve]
  11. Pigram, W. J., Fuller, W., and Amilton, L. D. H. (1971) Nat. New Biol. 235, 17-22
  12. Murphree, S. A., Tritton, T. R., Smith, P. L., and Sartorelli, A. C. (1981) Biochim. Biophys. Acta 649, 317-324 [Medline] [Order article via Infotrieve]
  13. Tewey, K. M., Rowe, T. C., Yang, L., Halligan, B. D., and Liu, L. F. (1984) Science 226, 466-468 [Medline] [Order article via Infotrieve]
  14. Muroi, M., and Suzuki, T. (1993) Cell. Signalling 5, 289-298 [CrossRef][Medline] [Order article via Infotrieve]
  15. Dominguez, I., Sanz, L., Arenzana-Seisdedos, F., Diazmeco, M. T., Virelizier, J. L., and Moscat, J. (1993) Mol. Cell. Biol. 13, 1290-1295 [Abstract]
  16. Martelly, I., and Castagna, M. (1989) Curr. Opin. Cell. Biol. 1, 206-210 [Medline] [Order article via Infotrieve]
  17. Ido, M., Asao, T., Sakurai, M., Inagaki, M., Saito, M., and Hidaka, H. (1986) Leuk. Res. 10, 1063-1069 [Medline] [Order article via Infotrieve]
  18. Tamaoki, T., Nomoto, H., Takahashi, I., Kato, Y., Morimoto, M., and Tomita, F. (1992) Biochem. Biophys. Res. Commun. 135, 397-402
  19. Toullec, D., Pianetti, P., Coste, H., Bellevergue, P., Grand-Perret, T., Ajakane, M., Baudet, V., Boissin, P., Boursier, E., Loriolle, F., Duhamel, L., Charon, D., and Kirilovsky, J. (1991) J. Biol. Chem. 266, 15771-15781 [Abstract/Free Full Text]
  20. Staal, F. J. T., Roederer, M., Herzenberg, L., and Herzenberg, L. A. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 9943-9947 [Abstract]
  21. Bradford, M. M. (1976) Anal. Biochem. 72, 248-251 [CrossRef][Medline] [Order article via Infotrieve]
  22. Garner, M. M., and Revzin, A. (1981) Nucleic Acids Res. 9, 3047-3059 [Abstract]
  23. Yasuda, I, Kishimoto, A., Tanaka, S., Tominaga, M., Sakurai, A., and Nishizuka, Y. (1990) Biochem. Biophys. Res. Comm. 166, 1220-1227 [Medline] [Order article via Infotrieve]
  24. Tritton, T. R., and Yee, G. (1982) Science 217, 248-250 [Medline] [Order article via Infotrieve]
  25. Siegfried, J. A., Kennedy, K. A., Sartorelli, A. C., and Tritton, T. R. (1983) J. Biol. Chem. 258, 339-343 [Abstract/Free Full Text]
  26. Rosette, C., and Karin, M. (1995) J. Cell. Biol. 128, 1111-1119 [Abstract]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.