Overexpression of the Atypical Protein Kinase C ζ Reduces Topoisomerase II Catalytic Activity, Cleavable Complexes Formation, and Drug-induced Cytotoxicity in Monocytic U937 Leukemia Cells*

In this study, we evaluated the influence of protein kinase Cζ (PKCζ) on topoisomerase II inhibitor-induced cytotoxicity in monocytic U937 cells. In U937-ζJ and U937-ζB cells, enforced PKCζ expression, conferred by stable transfection of PKCζ cDNA, resulted in total inhibition of VP-16- and mitoxantrone-induced apoptosis and decreased drug-induced cytotoxicity, compared with U937-neo control cells. In PKCζ-overexpressing cells, drug resistance correlated with decreased VP-16-induced DNA strand breaks and DNA protein cross-links measured by alkaline elution. Kinetoplast decatenation assay revealed that PKCζ overexpression resulted in reduced global topoisomerase II activity. Moreover, in PKCζ-overexpressing cells, we found that PKCζ interacted with both α and β isoforms of topoisomerase II, and these two enzymes were constitutively phosphorylated. However, when human recombinant PKCζ (rH-PKCζ) was incubated with purified topoisomerase II isoforms, rH-PKCζ interacted with topoisomerase IIβ but not with topoisomerase IIα. PKCζ/topoisomerase IIβ interaction resulted in phosphorylation of this enzyme and in decrease of its catalytic activity. Finally, this report shows for the first time that topoisomerase IIβ is a substrate for PKCζ, and that PKCζ may significantly influence topoisomerase II inhibitor-induced cytotoxicity by altering topoisomerase IIβ activity through its kinase function.

DNA topoisomerases II are nuclear enzymes that modify DNA topology by their ability to break and reseal both strands in concert. Topoisomerases II have important functions in DNA replication and can serve as a cancer chemotherapy target. Indeed, drugs such as etoposide (VP- 16) or mitoxantrone, form drug-topoisomerase II-DNA ternary complexes referred to as "cleavable complex." The primary cytotoxic effect of these socalled "topoisomerase II inhibitors" is not by inhibition of topoisomerase II activity but rather by stabilizing topoisomerase II cleavable complexes. This interaction prevents the DNAresealing step normally catalyzed by topoisomerase II. The ternary complex constitutes a latent DNA-damaging state, which is ultimately converted to an irreversible DNA doublestrand break (DSB). 1 Although the mechanism by which complex formation mediates cell death is still poorly understood, it has been largely documented with few exceptions that the amount of cleavable complexes and the subsequent number of DNA breaks correlates with cytotoxicity (1). These observations suggest that abnormal intracellular distribution or a decrease in expression level, activity, and sensitivity of the inhibited topoisomerase may have major impacts on topoisomerase inhibitor clinical efficacy. This has been confirmed by the molecular characterization of the so-called atypical multidrug resistant phenotype (at-MDR) resulting from selection by topoisomerase II inhibitors. Indeed, at-MDR cells display crossresistance to other topoisomerase II inhibitors and have been associated with a number of functional and/or structural topoisomerase II alterations, including decreased catalytic activity, abnormal interaction between topoisomerase II and nuclear matrix, reduced expression, point mutation and, finally, altered phosphorylation (2).
The role of phosphorylation on topoisomerase II function has been debated and remains controversial. Indeed, previous studies have shown that topoisomerase II contains potential serine phosphorylation sites and that this enzyme is a substrate for various serine kinases, including casein kinase II, p34 cdc2 kinase, and classic protein kinase C (PKC). In a cell-free system, PKC-induced phosphorylation of topoisomerase II results in an increase in its catalytic activity by enhancing ATP hydrolysis (3,4). In the absence of antineoplastic drugs, phosphorylation has a negligible effect on other steps of topoisomerase II catalytic cycle, including DNA binding or DNA cleavage/religation equilibrium. However, in the presence of VP-16 or amsacrine, phosphorylation decreases the ability of drugs to stabilize DNA-topoisomerase II complexes, apparently by increasing the rates of religation of DNA by the enzyme (5). Other studies have provided indirect evidences that PKC might also influence topoisomerase II function in vivo. For example, PKC inhibitors, such as suramin or staurosporine, decrease topoisomerase II phosphorylation and catalytic activity in intact cells as well as drug-induced topoisomerase II-mediated cleavage (6,7). However, the role of topoisomerase II phosphorylation in drug resistance has been minimized on the basis of independent studies that have shown that, in at-MDR cells, topoisomerase II could be either hyperphosphorylated or hypophosphorylated (8 -10).
At least 12 different isoforms of PKC have been characterized so far and have been separated into three categories based on the Ca 2ϩ requirement for activation and phorbol ester binding activity. Conventional PKCs (␣, ␤I, ␤II, and ␥) are Ca 2ϩdependent phorbol ester receptor kinases; novel PKCs (␦, ⑀, , and ) are Ca 2ϩ -independent phorbol ester receptor kinases; and atypical PKCs (, , , and ) are independent of both Ca 2ϩ and phorbol ester. Previous studies have shown that topoisomerase II is phosphorylated in vitro by each of the conventional PKC isoforms (11). However, the influence of these PKC isozymes on cellular topoisomerase function in vivo is still largely unknown. Moreover, to the best of our knowledge, the influence of atypical PKC isozymes on topoisomerase II phosphorylation and function has not been investigated.
PKC is an atypical PKC isoform, which is activated directly or indirectly by a variety of important signaling molecules, including ceramide (12,13), phosphatidic acid (14), and diacylglycerol generated from phosphatidylcholine hydrolysis (15), phosphoinositide 3-kinase lipid products (16), and p21Ras (17). PKC has emerged as a critical regulator of a number of cellular functions, including proliferation, differentiation, and apoptosis inhibition (18). Despite the critical role of this enzyme in cellular signaling, its implication in the regulation of topoisomerase II function has never been examined. This study was aimed to evaluate the effect of PKC overexpression on the formation of cleavable complexes and cytotoxicity induced by VP-16 in the human leukemic U937 cells.  ing to full-length rat PKC) or 20 g of the vector without the PKC insert using a Bio-Rad Gene Pulser as previously described (19). For this study, two clones, U937-J and U937-B, were selected and were compared with control U937-neo cells. Cells were cultured in RPMI complemented with 10% fetal calf serum. U937-J, U937-B, and U937neo cells, displayed similar growth kinetics with a doubling time of about 25 h. PKC overexpression resulted in a 2.5-fold increase in PKC activity as measured by MBP phosphorylation after immunoprecipitation with anti-PKC antibody.

Materials-Recombinant
MTT Assay-This assay is based on the ability of viable mitochondria to convert MTT, a soluble tetrazolium salt (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) into an insoluble formazan precipitate, which is dissolved in dimethyl sulfoxide and quantified by spectrophotometry. Cells (30,000) were seeded in 96-well plates and treated with cytotoxic agents for 48 h. Absorbance corresponding to MTT conversion was read at two wavelengths, 540 and 690 nm.
DNA Filter Elution Assays-Exponential growing cells were labeled with [ 3 H]thymidine (0.02 Ci/ml) for 48 h, chased for 2 h in isotope-free medium, and exposed to VP-16 for the indicated time. Equal numbers of cells (5 ϫ 10 5 ) were loaded onto polycarbonate or PVC filters, lysed, and subjected to elution (20). Radioactivity in the DNA fractions was counted, and the fraction of the DNA retained on the filter was calculated as follows: fraction retained/(filter plus lysis plus fraction retained). Elution of DNA through polycarbonate filters reflects the presence of DSB, and elution of DNA through PVC filters reflects the presence of DNA-protein cross-link (DPC). DPC frequency (in radequivalents) was computed according to the formula: Protein Analysis-For cytoplasmic protein, 1 ϫ 10 7 cells were washed twice in phosphate-buffered saline and lysed by resuspension in lysis buffer containing 10 mM HEPES (pH 7.8), 100 mM EDTA, 100 mM EGTA, 1 mM PMSF, 2 M pepstatin A, 0.6 g/ml aprotinin on ice for 10 min. Nonidet P-40 (0.3%) was then added for 2 min, and the cytoplasmic lysate (supernatant) was collected after centrifugation at 10,000 ϫ g for 2 min at 4°C. For nuclear lysate, 1 ϫ 10 7 cells were washed twice in phosphate-buffered saline and lysed by resuspension in lysis buffer containing 10 mM HEPES (pH 7.8), 100 mM EDTA, 100 mM EGTA, 1 mM PMSF, 2 M pepstatin A, 0.6 g/ml aprotinin on ice for 10 min. Nonidet P-40 was then added at 0.3% final for 5 min, and the nuclear pellet was resuspended in 20 mM HEPES (pH 7.8), 400 mM NaCl, 1 mM EDTA, 1 mM EGTA. Aliquots were sonicated and centrifuged at 20,000 ϫ g for 10 min at 4°C, and supernatants containing nuclear proteins were collected. Nuclear or total cell lysates were resuspended in a denaturing loading buffer, and proteins were loaded in SDS-PAGE (7.5 or 10%), transferred onto nitrocellulose, and probed with anti-topoisomerase II ␣ and/or ␤ or anti-PKC antibodies. Immune complexes were detected by using the chemiluminescent detection system.
Preparation of Nuclear Extracts-Cells (1 ϫ 10 7 ) were washed once with phosphate-buffered saline and twice with buffer A (1 mM KH 2 PO 4 , 5 mM MgCl 2 , 150 mM NaCl, 1 mM EGTA, 1 mM DTT, 1 mM PMSF, 2 g/ml leupeptin, 2 g/ml aprotinin) and resuspended for 10 min with buffer C containing 0.3% Triton X-100. Cells were then centrifuged and resuspended in 100 l of buffer A. One-hundred microliters of buffer A containing 0.55 M NaCl was then added. After mixing and gentle rotation for 30 min at 4°C, samples were centrifuged for 10 min at 14,000 rpm. Supernatants were used as nuclear extracts.
Topoisomerase II Decatenation Assay-Decatenation assays were carried out by incubating 0.25 g of kinetoplast DNA with nuclear extracts or recombinant proteins in buffer B containing 10 mM Tris-HCl, pH 7.9, 50 mM NaCl, 50 mM KCl, 5 mM MgCl 2 , 0.1 mM EDTA, 15 g/ml bovine serum albumin, 1 mM ATP. After 30 min at 30°C, reactions were quenched by addition of 1% SDS, 0.5% bromphenol blue, and 30% glycerol. DNA products were resolved on 1% agarose-Tris borate-EDTA gels at 16 V overnight. Agarose gels were stained with ethidium bromide, and fluorescence was quantified by UV imager.
Co-immunoprecipitation of PKC and Topoisomerase II-Extracts were prepared by lysing cells (15 ϫ 10 6 ) in buffer C. Cells extracts (1.5 mg) were sonicated, clarified, and immunoprecipitated with 3 g of anti-PKC antibody overnight at 4°C. Immune complexes were collected by incubation with protein G-Sepharose beads for 60 min at 4°C. The beads were then extensively washed with buffer D. For in vitro interaction, recombinant PKC was preincubated with topoisomerase II␤ for 1 h at 32°C in buffer B and immunoprecipitated with 3 g of anti-PKC antibody overnight at 4°C. Immune complexes were collected with protein A-Sepharose beads for 60 min at 4°C. The beads were then washed with buffer B. Denaturing loading buffer was added to immune complexes from in vivo or in vitro experiments. Samples were boiled for 5 min, run in SDS-PAGE (7.5%), transferred onto nitrocellulose membrane, and probed with anti-topoisomerase II antibodies. Proteins were detected by chemiluminescence.
Influence of PKC Overexpression on VP-16-induced DNA Strand Breaks in U937 Cells-U937-neo and U937-J cells were prelabeled with [ 3 H]thymidine for 48 h, chased with fresh medium, and treated with VP-16 for 1 h and DSB were determined using alkaline elution (20,21). As shown in Fig. 2, VP-16 produced significantly less DNA DSB in U937-J cells than in U937-neo cells. VP-16-induced DPC were also compared in U937-neo and U937-J cells. As shown in Fig. 3, whereas VP-16 induced DPC in a dose-dependent manner in both U937-neo and U937-J cells, the levels of DPC were significantly lower in U937-J. These results showed that PKC overexpression resulted in reduced VP-16-induced DNA damage. To rule out the possible influence of PKC on drug transport, we measured VP-16-induced DPC in isolated nuclei from U937-neo and U937-J cells. As shown in Fig. 4, VP-16 produced a dosedependent increase in DPC in U937-neo nuclei, whereas there was no detectable DPC formation in U937-J nuclei. This result confirmed that PKC overexpression resulted in a significant reduction in VP-16-induced DNA damage, which could not be explained by altered drug transport.
Influence of PKC Overexpression on Topoisomerase II Expression in U937 Cells-To investigate the possible influence of PKC on topoisomerase II expression, U937-neo, U937-J, and U937-B cell extracts were analyzed by Western blotting with anti-topoisomerase II␣ and anti-topoisomerase II␤ antibodies. As shown in Fig. 5, topoisomerase II␣ and topoisomerase II␤ expression levels in U937-neo, U937-J, and U937-B cells were comparable. This result suggests that reduced VP-16induced DNA damage in PKC-overexpressing cells was not due to decreased topoisomerase II expression. For this reason, we hypothesized that PKC overexpression may result in reduced topoisomerase II activity.
Influence of PKC Overexpression on Topoisomerase II Activity in U937 Cells-Decatenation of kinetoplast DNA was used as a specific assay to evaluate topoisomerase II activity of nuclear extracts prepared from U937-neo, U937-J, and U937-B cells (22). Nuclear extracts from U937-neo cells exhibited topoisomerase II activity in a dose range comprised between 500 and 1000 ng of total nuclear protein, whereas topoisomerase II activities contained in U937-J and U937-B preparations were dramatically reduced (Fig. 6A). Based on Western blot analysis of nuclear cell extracts, it appeared that nuclear PKC expression was inversely correlated with topoisomerase II activity (Fig. 6B). For this reason, we hypothesized FIG. 9. Influence of rH-PKC on purified topoisomerase II activity in a cell-free system. rH-PKC (1 or 3 g) (A) or PKC␣, ␤, ␥ (50 ng) (B) were incubated with topoisomerase II (50 ng) for 1 h at 32°C, and topoisomerase II activity was measured by the decatenation of the kinetoplast DNA as a specific assay for topoisomerase II activity. Data are from one experiment representative of three experiments.
FIG. 10. Interaction between rH-PKC and purified topoisomerase II␤ isoform in a cell-free system. rH-PKC (1 g) was incubated with topoisomerase II␤ (1 g) for 1 h at 32°C and PKC/ topoisomerase II␤ complexes were subjected to immunoprecipitation using anti-PKC antibody following by immunoblotting with either anti-PKC antibody (A) or anti-topoisomerase II␤ (B). PKC kinase activity was checked using MBP as a substrate as described under "Experimental Procedures" (C). Topoisomerase II␤ (1 g) was incubated with rH-PKC (1 g) for 1 h at 32°C in the presence of [␥-32 P]ATP and phosphatidylserine (4 g), and phosphorylated topoisomerase II␤ was revealed by autoradiography after separation in a SDS-PAGE (7.5%) (D). Data are from one experiment representative of three experiments. that PKC inhibited topoisomerase II activity by influencing topoisomerase II phosphorylation.
Influence of PKC Overexpression on Serine Phosphorylation of Topoisomerase II in U937 Cells-U937-neo, U937-J, and U937-B cell extracts were immunoprecipitated with antiphosphoserine antibody, and topoisomerase II was immunoblotted with anti-topoisomerase II␣ and anti-topoisomerase II␤ antibodies. As shown in Fig. 7, U937-J and U937-B cells exhibited constitutive topoisomerase II␣ and ␤ serine hyperphosphorylation, compared with U937-neo cells, whereas antiphosphothreonine antibody, used as control, was not reactive. This result shows that PKC increased phosphorylation of both isoforms of topoisomerase II.
Interaction between PKC and Topoisomerase II in U937-neo and PKC Overexpressing U937 Cells-PKC/topoisomerase II interaction was assessed by immunoprecipitation. However, because anti-topoisomerase II␣-and ␤-specific antibodies used in this study were not suitable for immunoprecipitation, cellular extracts of U937-neo, U937-J, and U937-B were immunoprecipitated with anti-PKC antibody and topoisomerase II␣ and ␤ proteins were immunoblotted with relevant antibodies. In U937-neo cell extracts, we found that PKC interacted neither with topoisomerase II␣ nor with topoisomerase II␤, although a significant PKC amount was detected in the immunoprecipitates (Fig. 8). However, in U937-J and U937-B cellular extracts, both topoisomerase II␣ and ␤ co-immunoprecipitated with PKC (Fig. 8). These results suggest that, in PKC-overexpressing U937 cells, the enzyme may directly or indirectly interact with both topoisomerase II␣ and ␤ isoforms and that these interactions seriously interfere with topoisomerase II activity. To investigate this hypothesis, we evaluated in a cell-free system the influence of recombinant PKC on the activity of purified topoisomerase II preparations containing both topoisomerase II␣ and ␤.
Influence of PKC on Purified Topoisomerase II Activity in a Cell-free System-In these experiments, recombinant human PKC (rH-PKC) (1 or 3 g) was incubated with topoisomerase II (50 ng), and topoisomerase II activity was measured by the decatenation assay. As shown in Fig. 9A, PKC was found to inhibit topoisomerase II activity in a dose-dependent manner.
In contrast, a mixture containing PKC␣, PKC␤, and PKC␥ was found to stimulate topoisomerase II activity as previously described (Fig. 9B) (11). This result suggests that PKC does interfere with either topoisomerase II␣ or ␤ activities. To further investigate this finding, we evaluate the capacity of PKC to interact with purified topoisomerase II␣ or ␤.
Interaction between rH-PKC and Purified Topoisomerase II Isoforms in a Cell-free System-rH-PKC was co-incubated with either topoisomerase II␤ or topoisomerase II␣ in a molar ratio of 2:1. PKC/topoisomerase II complexes were immunoprecipitated using anti-PKC antibody, and topoisomerase isoforms were revealed by immunoblotting with specific antitopoisomerase antibodies. Controls were provided by immunoblotting immunoextracts with anti-PKC (Fig. 10A). As shown in Fig. 10B, a small amount of topoisomerase II␤ nonspecifically bound to protein A-Sepharose. However, the level of topoisomerase II␤ detected following immunoprecipitation with anti-PKC antibody was significantly increased, suggesting that most of topoisomerase II␤ specifically bound to PKC. In contrast, despite many efforts, we were unable to detect topoisomerase II␣ in anti-PKC immunoextracts (data not shown). These results showed that PKC was able to interact with topoisomerase II␤ but not topoisomerase II␣. The consequence of this interaction on topoisomerase II phosphorylation was also investigated. Topoisomerase II␤ (1 g) was incubated with rH-PKC (1 g) in the presence of [␥-32 P]ATP and phosphatidylserine, and phosphorylated topoisomerase II␤ was revealed by autoradiography. PKC kinase activity was checked using MBP as a substrate (Fig. 10C). Topoisomerase II␤ was found to be constitutively phosphorylated. However, incubation with PKC resulted in a 2-fold increase in topoisomerase II␤ phosphorylation (Fig. 10D). Altogether, these results demonstrated that PKC did interact with topoisomerase II␤ and phosphorylated this enzyme.
Influence of rH-PKC on Topoisomerase II␤ Activity-In these experiments, rH-PKC was incubated with topoisomerase II␤ at a molar ratio of 2:1, and topoisomerase II␤ activity was measured by the decatenation assay. As shown in Fig.  11, PKC was found to inhibit topoisomerase II␤ activity. FIG. 11. Influence of rH-PKC on purified topoisomerase II␤ activity. rH-PKC (530 ng) was incubated with topoisomerase II␤ (650 ng) for 1 h at 32°C, and topoisomerase II␤ activity was measured by the decatenation assay. Data are from one experiment representative of three experiments.

DISCUSSION
This study shows that PKC overexpression in U937 cells resulted in inhibition of apoptosis and increased survival of U937 cells treated with VP-16 and mitoxantrone, two topoisomerase II inhibitors. Enforced PKC expression resulted in a marked decrease in VP-16-induced DPC and DNA DSB, whereas the level of topoisomerase II␣ and topoisomerase II␤ expression was unchanged compared with control cells. These results suggest that PKC can interfere with topoisomerase II function. In fact, we found that PKCoverexpressing cells exhibited reduced topoisomerase II catalytic function as measured by the decatenation assay. Altered topoisomerase II catalytic cycle may explain reduced drug-induced DNA damage and cytotoxicity. Thus, this study shows for the first time that a specific PKC isozyme may inhibit topoisomerase II catalytic activity and VP-16-induced apoptosis and cytotoxicity by interfering with drug-induced DNA damage.
Based on the kinase function of PKC, we hypothesized that PKC overexpression might result in abnormal topoisomerase II phosphorylation. In fact, we found that, in PKC-overexpressing cells, PKC was not only found to interact with topoisomerase II␣ and topoisomerase II␤ but also that these two topoisomerase II isoforms were heavily phosphorylated on serine residues. These results suggest that, in PKC-overexpressing cells, PKC not only directly or indirectly interacts with the two topoisomerase II isoforms but also phosphorylates these enzymes. However, using a cell-free system, we described that only topoisomerase II␤ is a substrate for PKC and that PKC inhibits topoisomerase II␤ activity. This result suggests that, in PKC-overexpressing cells, PKC interacts directly with topoisomerase II␤ and inhibits topoisomerase II␤ catalytic activity. This hypothesis is consistent with the role of this topoisomerase II␤ form in the cytotoxicity of topoisomerase II inhibitors (23,24).
With regard to topoisomerase II␣, the fact that this enzyme was found to interact in vivo, but not in vitro, with PKC, suggests that, in PKC-overexpressing cells, PKC/topoisomerase II␣ interaction involves one or several other proteins required for the constitution of this complex. Moreover, the fact that, in PKC-overexpressing cells, topoisomerase II␣ was found to be constitutively phosphorylated whereas, in vitro, PKC was unable to phosphorylate this enzyme, suggests that topoisomerase II␣ is phosphorylated by another PKC-regulated kinase. In this perspective, it is interesting to note that in a recent study topoisomerase II␣ was found to be phosphorylated in intact cells by ERK2, the effector serine kinase of the classic MAPK module (25). Based on previous studies, which have documented that PKC is a downstream target of MAPK (26,27), topoisomerase II␣ phosphorylation could result from PKC-mediated ERK2 activation in PKC-overexpressing cells. The fact that, in these cells, ERK2 was found to be constitutively activated and accumulated in the nucleus (data not shown) supports this hypothesis.
The role of atypical PKC isoforms, including PKC, in cell survival has been previously documented. Indeed, it has been described that the blockade of PKC or PKC/ with dominantnegative mutants or antisense oligonucleotides is sufficient to promote apoptosis (28,29). The inactivation of PKC by caspasedependent proteolysis during apoptosis induced by UV (30) or by cisplatin (31) strengthens the role of PKC in the cellular protection against genotoxic stress. The mechanism by which atypical PKC isoforms exert their anti-apoptotic effect has received a great deal of attention. These studies strongly suggested that NF-B signaling pathways could play an important role in PKC-induced inhibition of apoptosis (32). Indeed, NF-B is a negative regulator of apoptosis induced by genotoxic agents, including topoisomerase II inhibitors (33,34). Therefore, we cannot rule out that PKC overexpression may result in the activation of anti-apoptotic signals that interfere with the post-damage apoptotic response and, therefore, contribute to drug resistance.
To conclude, we propose a model in which, upon PKC accumulation in the nucleus, this enzyme interacts with and phosphorylates nuclear topoisomerase II␤. Topoisomerase II␤ hyperphosphorylation reduces catalytic function and decreases formation of ternary complexes and drug-induced cytotoxicity. If so, nuclear PKC accumulation might function to regulate topoisomerase II function. Although very little is known about expression and subcellular localization of PKC in tumor cells, PKC may translocate to the nucleus upon stimulation by differentiating agents (35), growth factors (36,37), cytokines (38), or hypoxia (39). Whether PKC alters topoisomerase II function in these conditions will be the subject of further investigations.