Angoline and Chelerythrine, Benzophenanthridine Alkaloids That Do Not Inhibit Protein Kinase C*

Starting with an extract derived from the stem ofMacleaya cordata (Papaveraceae) that was active in the process of inhibiting phorbol 12,13-dibutyrate binding to partially purified protein kinase C (PKC), the benzophenanthridine alkaloid angoline was isolated and identified. This discovery appeared in context, as a related benzophenanthridine alkaloid, chelerythrine, has been reported to mediate a variety of biological activities, including potent and selective inhibition of protein kinase C (PKC). However, in our studies, angoline was not observed to function as a potent inhibitor of PKC. Moreover, we were unable to confirm the reported inhibitory activity of chelerythrine. In a comprehensive series of studies performed with various PKC isozymes derived from a variety of mammalian species, neither chelerythrine nor angoline inhibited activity with high potency. To the contrary, chelerythrine stimulated PKC activity in the cytosolic fractions of rat and mouse brain in concentrations up to 100 μm. In addition, chelerythrine and angoline did not inhibit [3H]phorbol 12,13-dibutyrate binding to the regulatory domain of PKC at concentrations up to 40 μg/ml, and no significant alteration of PKC-α, -β, or -γ translocation was observed with human leukemia (HL-60) cells in culture. Further, chelerythrine did not inhibit 12-O-tetradecanoylphorbol 13-acetate-induced ornithine decarboxylase activity with cultured mouse 308 cells, but angoline was active in this capacity with an IC50 value of 1.0 μg/ml. A relatively large number of biological responses have been reported in studies conducted with chelerythrine, and alteration of PKC activity has been considered as a potential mechanism of action. In light of the current report, mechanisms independent of PKC inhibition should be considered as responsible for these effects.

mitters of secondary messengers have been shown to mediate their effects through PKC. PKC subtypes have been grouped as (i) conventional (␣, ␤I, ␤II, and ␥), which are activated by calcium, phospholipids, diacylglycerol, and 12-O-tetradecanoylphorbol 13-acetate (TPA); (ii) novel (␦, ⑀, , , and ), which do not require calcium for activation; and (iii) atypical ( and ), which are activated by phospholipids but not by calcium, diacylglycerol, or TPA (2,3). It is likely that these isozymes have distinct and distinguishable functions (4 -7). Since PKC has been elaborated as a major intracellular phorbol ester receptor (8,9), it has become clear that this protein plays an important role in the process of tumor promotion (10). Accordingly, inhibition of PKC can be viewed as a rational method of blocking or inhibiting tumor promotion, and several natural product inhibitors have been identified. Examples include staurosporine (11), UCN-1028C (12), and isoquinolinesulfonamide H-7 (13). In addition, the benzophenanthridine alkaloid, chelerythrine, was described as a potent and selective inhibitor of PKC, with an IC 50 value of 0.66 M illustrated with enzyme derived from rat brain (14). Biological responses mediated by chelerythrine, such as cytotoxic activity with L-1210 tumor cells (14) and antiplatelet activity (15), have been ascribed to inhibition of PKC. Additional activities mediated by chelerythrine include inhibition of alanine aminotransferase (16), inhibition of Na ϩ ,K ϩ -ATPase (17), and antibacterial effects (18).
In searching for novel natural product cancer chemopreventive agents, the benzophenanthridine alkaloid angoline was obtained from an extract of Macleaya cordata (Papaveraceae); this plant extract had previously been shown to antagonize the interaction of [ 3 H]phorbol 12,13-dibutyrate (PDBu) with PKC receptor. Due to the structural similarity of angoline and chelerythrine, we undertook a series of studies to investigate the activity of angoline, especially in the context of affecting PKCmediated responses. As reported herein, neither angoline nor chelerythrine were effective inhibitors of PKC. Alternate mechanisms should be taken into account when considering the biological responses mediated by these compounds.

PKC Preparations and Analysis of Activity
The purification of PKC-␣ from calf brain and the PKC assay were performed essentially as described by Da Silva et al. (23). Briefly, fresh calf brain homogenates were centrifuged at 100,000 ϫ g for 60 min. The supernatant was concentrated using an ultrafiltration cell (Amicon Division, Beverly, MA) equipped with a XM50 membrane and subjected to passage over a DEAE-cellulose column equilibrated with buffer A (20 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 0.5 mM EGTA, 1 mM dithiothreitol, 10% glycerol). The sample was eluted with a linear gradient of KCl (0 -0.3 M) in buffer A. Fractions were collected and enzyme activities were monitored by [ 3 H]PDBu binding (24). The major fractions showing binding activity were combined and concentrated, and further purification of PKC was conducted by sequential column chromatography over phenyl-Sepharose, Sephacryl S-200, poly-L-lysine-agarose, and hydroxyapatite. Additional samples of brain used for the preparation of cytosolic fractions (12,000 ϫ g) bearing PKC activity were from mouse, rat, rabbit, chicken, and dog (Pel-Freez ® Biologicals, Rogers, AR).
PKC catalytic activity was assayed by quantifying transfer of the ␥-phosphate group of [␥-32 P]ATP into histone IIIs. The incubation mixture (80 l) consisted of 20 mM Tris-HCl buffer, pH 7.5, 200 g/ml histone (type IIIs), 10 mM MgCl 2 , 0.5 mM CaCl 2 , 500 g/ml phosphatidylserine in 3% Triton X-100, 10 M ATP containing [␥-32 P]ATP, and test compounds. After a 10-min preincubation at 30°C, reactions were initiated by the addition of 20 l of enzyme solution (4 g). After an additional incubation period of 10 min, 85 l of the reaction mixtures were spotted onto Whatman P-81 cation exchange paper (Whatman Paper Ltd., Maidstone, Kent, United Kingdom). The papers were washed three times with 0.2% aqueous phosphoric acid solution, airdried, and subjected to liquid scintillation counting. Test compounds were evaluated for potential to affect basal activity (no activator added), or in the presence of PDBu (0.4 M) or 1,2-diolein (180 M). Results were expressed as a percentage, relative to solvent-treated controls, and dose-response relationships were obtained by evaluating serial dilutions of test compounds in duplicate.

Assessment of PDBu Binding to the Regulatory Site of PKC
PDBu binding assays were performed in 96-well microtiter plates essentially as described by De Vries et al. (24), using calf brain homogenate or other sources of soluble PKC as receptor. Each reaction mixture (200 l final volume) contained homogenate (25 g of protein) as a source of PKC, bovine serum albumin (200 g), 20 mM Tris-HCl buffer, pH 7.4, and [ 3 H]PDBu (5 nM, 20 Ci/mmol). Test substances (10 l) were added to the reaction mixtures (in duplicate) and incubated for 1 h at 37°C. Unbound [ 3 H]PDBu was removed by filtration with 50 mM Tris-HCl, pH 7.4, through glass fiber filter mats (Type B; Wallac Inc., Gaithersburg, MD) using a Tomtec Mach III ® 96-well plate harvester (Tomtec, Orange, CT). Radioactivity was determined by scintillation counting using a Wallac 1450 MicroBeta ® liquid scintillation counter (Wallac, Inc.). The amount of [ 3 H]PDBu bound in the presence of nonradioactive PDBu (4 M) was used to assess nonspecific binding. Specific binding was defined as the difference between total and nonspecific binding, and results were expressed as a percentage, relative to the solvent-treated control group. IC 50 values were calculated using semilog plots.
Preparation and Fractionation of HL-60 Cells-HL-60 cells in logarithmic phase (1 ϫ 10 7 cells) were treated with test compounds for 3 or 24 h. Cells were harvested by centrifugation (500 ϫ g, 10 min), washed twice with 10 ml of ice-cold phosphate-buffered saline, pH 7.4, and homogenized (4°C) in 1.0 ml of buffer (20 mM Tris-HCl, pH 7.4, 5 mM NaCl, 1 mM EDTA, 5 mM MgCl 2 , 2 mM dithiothreitol, 200 M phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 10 g/ml leupeptin) for 30 s using a Brinkmann Polytron ® PT-3000 homogenizer (Littau, Switzerland). The homogenates were centrifuged at 100,000 ϫ g for 1 h (4°C), and the resulting supernatant fractions (used as a source of cytosolic PKC) were stored at Ϫ80°C. The pellets were solubilized in buffer containing 1% Triton X-100 by vortexing and incubating on ice for 2 h, and the suspensions were centrifuged at 100,000 ϫ g for 1 h (4°C) to obtain membrane/microsomal PKC. The resulting supernatant fractions were used as a source of membrane-associated particulate PKC, and stored at Ϫ80°C. All protein contents were quantified using the bicinchoninic acid (BCA) method and bovine serum albumin as a standard.

Assessment of TPA-induced Ornithine Decarboxylase (ODC) Activity with ME 308 Cells
Mouse epidermal 308 cells were cultured in S-minimal essential medium containing non-essential amino acids (1ϫ), dialyzed fetal bovine serum (5%), Ca 2ϩ (0.05 mM), and antibiotic-antimycotics (1ϫ) at 37°C in a 5% CO 2 atmosphere. For determination of TPA-induced ODC activity, cells were distributed to 24-well plates at an initial density of 2 ϫ 10 5 cells/ml/well. After an 18-h preincubation, test materials dissolved in Me 2 SO were added in duplicate (5 l, 0.5% final Me 2 SO concentration) before the induction of ODC activity with TPA (200 nM). After an additional incubation period of 6 h, plates were washed twice with phosphate-buffered saline and stored at Ϫ80°C until tested. ODC activity was assayed directly in the 24-well plates as described previously (25). In brief, frozen cells were lysed by quickly thawing the bottom of the culture plates in a warm water bath (37°C, 2 min). A substrate and cofactor mixture (200 l containing 2 l of L-[1-14 C]ornithine (200 nCi, 56 mCi/mmol, 100 Ci/ml), 50 l of sodium phosphate buffer (0.2 M, pH 7.2), 16 l of EDTA (12.5 mM), 10 l of dithiothreitol (50 mM), 4 l of pyridoxal phosphate (5 mM in 10 mM NaOH) and 118 l of unlabeled L-ornithine (78 g/ml)) were added to each well. Released [ 14 C]CO 2 was captured by paper discs, which were moistened with 30 l of 1 N NaOH during incubation of plates at 37°C for 1 h while shaking. The amount of radioactivity captured on NaOH-impregnated filter disks was determined by scintillation counting using a Wallac 1450 MicroBeta ® liquid scintillation counter. Protein was determined by the Bio-Rad method using bovine serum albumin as a standard. Results were calculated as nanomoles of [ 14 C]CO 2 /h/mg of protein and expressed as a percentage in comparison with a control treated with Me 2 SO and TPA.

Effects of Chelerythrine and Angoline on PKC Activity-
Based on the structural similarity of angoline and chelerythrine (Fig. 1), initial studies were performed to assess the potential of angoline, isolated from M. cordata, to inhibit the catalytic activity of PKC. Surprisingly, angoline demonstrated little inhibitory activity with calf brain PKC; IC 50 values for basal and PDBu-stimulated activity were approximately 100 g/ml (Fig. 2). We then examined chelerythrine in this assay system. In contrast to the results of Herbert et al. (14), chelerythrine did not mediate potent inhibitory activity with calf brain PKC. IC 50 values were approximately 60 and 80 g/ml with basal and PDBu-stimulated activity, respectively (Fig. 2). The lack of efficacious inhibition by chelerythrine was not due to the nature of the stimulator, since similar dose-response patterns were observed in the presence of either PDBu or diolein (Fig. 3). Using the same experimental conditions, staurosporine (133 ng/ml) inhibited activity over 90%, irrespective of which stimulator was employed (data not shown). Chelerythrine was also ineffective as an inhibitor using mouse and rat brain cytosol fractions as sources of PKC. In fact, with these preparations, chelerythrine enhanced PKC activity in concentrations ranging up to 100 M (Fig. 4).
Effects of Chelerythrine and Angoline on PDBu Binding-In order to investigate the effect of chelerythrine and angoline on the regulatory domain of PKC, PDBu binding activity was assessed with crude calf brain particulate as a source of PKC. At concentrations of 2 g/ml, chelerythrine and angoline enhanced [ 3 H]PDBu binding. This was followed by diminution of binding to the levels of control. When tested at a concentration 200 g/ml, chelerythrine and angoline reduced PDBu binding to levels corresponding to 60 and 45% of control, respectively (Fig. 5). Using other sources of soluble PKC, including mouse, rat, dog, chicken, and rabbit brain cytosol, chelerythrine and angoline did not show inhibitory effects at concentrations ranging up to 20 g/ml (data not shown). In each experiment, unlabeled PDBu (0.4 M) inhibited [ 3 H]PDBu binding to PKC by over 95% (data not shown). Therefore, chelerythrine and angoline did not affect binding to the regulatory domain of PKC with strong potency.
Effects of Test Compounds on PKC Translocation-In order to evaluate the potential of chelerythrine and angoline to effect PKC translocation, Western blot analyses were performed with preparations derived from cultured HL-60 cells treated with test compounds. Mouse anti-PKC-␣, -␤, or -␥ antibodies were used as probes (Fig. 6). As expected, treatment with TPA (200 nM) for 3 h induced translocation of PKCs from the cytosol to particulate fractions (Fig. 6, A and B), and increasing exposure time to 24 h resulted in down-regulation (Fig. 6, C and D). When tested at concentrations as high as 1.0 g/ml, however, alterations in PKC levels or distribution patterns were not facilitated by chelerythrine or angoline (Fig. 6). When co-incubated with TPA for 24 h, down-regulation of PKCs was slightly potentiated by the test compounds.
Effects of Chelerythrine and Angoline on TPA-induced ODC Activity-ODC is the key enzyme in the biosynthesis of polyamines and is inducible by stimuli such as growth factors, hormones, and tumor promoters, including phorbol esters (26). Activity is controlled by various factors, including expression, stability and transcription rate of ODC mRNA, stability and translation rate of the ODC enzyme, and post-translational modifications (27,28). Using mouse 308 cells as a model system, chelerythrine did not inhibit TPA-induced ODC activity within the non-cytotoxic range (IC 50 Ͼ1.0 g/ml), but promoted a slight enhancement of activity at lower concentrations. Angoline did not alter the TPA-induced effect at lower concentrations, but reduced activity by approximately 50% at a test concentration of 1.0 g/ml (Fig. 7). DISCUSSION Since PKC plays an important role in signal transduction and phorbol ester-promoted tumorigenesis, various inhibitors have been studied in detail. One such substance, chelerythrine, has been reported as a potent and specific inhibitor. The phosphorylation reaction mediated by PKC was reduced with an IC 50 value of 0.66 M. Inhibition was competitive with respect to phosphate acceptor (histone IIIs), and noncompetitive with respect to ATP. No substantive competition was observed with respective to [ 3 H]PDBu binding, thus suggesting interaction does not occur at the regulatory site of PKC. Other kinases, such as protein kinase A, tyrosine protein kinase, and Ca 2ϩ / calmodulin-protein kinase, were not inhibited with great efficacy, indicating specificity for PKC (14).
Accordingly, when we obtained angoline from a plant extract demonstrating antagonism of PDBu binding with PKC, it was suspected that this substance would function in a mode similar to chelerythrine. However, angoline was found ineffective as an inhibitor of PKC activity. Thus, as a positive control, the inhibitory potential of chelerythrine was investigated, obtaining this substance from a commercial source (Sigma). However, in repeated attempts, chelerythrine was found to be practically inactive as an inhibitor of partially purified calf brain PKC. Although the compound appeared pure and stable (HPLC anal- ysis), additional samples were procured (LC Service Corp., Extrasynthese) and evaluated, and also found to be inactive. This lack of inhibitory activity was observed with PKC obtained from a variety of species, either in the presence or absence of activators (PDBu or diolein). In fact, to the contrary, when evaluated with cytosol fractions derived from rat or mouse brain, chelerythrine enhanced PKC activity, and, with calf brain preparations, [ 3 H]PDBu binding was elevated. These types of activities are consistent with recent reports by Lombardini (29,30) who demonstrated chelerythrine-stimulated phosphorylation of a ϳ20-kDa protein present in the mitochondrial fraction of rat retina. Further, using intact cells, Gopalakrishna et al. (31) noted that chelerythrine and sanguinarine could induce an initial activation of membrane translocation from the cytosol, and subsequent down-regulation of PKC. These responses parallel those of tumor promoters, such as phorbol esters, and would not be anticipated by an inhibitor of PKC. Along these lines, we evaluated angoline and chelerythrine for potential to facilitate translocation of PKC-␣, -␤, and -␥ with cultured HL-60 cells, but no significant alterations were observed. Interestingly, however, with cultured ME 308 cell system, chelerythrine stimulated TPA-induced ODC activity at lower concentrations, whereas angoline was inactive at lower concentrations but reduced activity approximately 50% at a test concentration of 1.0 g/ml. Polyamines play essential roles in normal cell proliferation and differentiation, but are overexpressed in various cancer cells. Since ODC is a key component that regulates intracellular polyamines, inhibitors of ODC activity are of interest. Accordingly, it should be of value to explore the mechanism by which angoline functions in this capacity.
Chelerythrine is known to mediate a variety of biological responses. For example, rat liver alanine aminotransferase is inhibited by adduct formation between thiol groups of the enzyme and the iminium bond of chelerythrine (16). Chelerythrine also inhibits taxol-mediated polymerization of rat brain tubulin (32), histone kinase activity using partially purified PKC from the rat lacrimal gland (33), pyrogen-induced expres-sion of tissue factor in endothelial cells, histamine release induced by aggregation of IgE-receptor on human basophils (34,35), Na ϩ ,K ϩ ,2Cl Ϫ cotransporter in Ehrlich mouse ascites tumor cells (36), and invasion of metastatic human follicular thyroid cancer (37). We have recently found that angoline and chelerythrine inhibit 7,12-dimethylbenz(a)anthracene-induced preneoplastic lesion formation in mouse mammary gland organ culture by approximately 60% at a test concentration of 10 M (data not shown). Based on the report of Herbert et al. (14), some of these effects have been attributed to PKC inhibition. As indicated by the current report, however, chelerythrine is not a general and specific inhibitor of PKC. Accordingly, PKCindependent pathways should be considered when exploring the biological potential of chelerythrine and related compounds. For example, as noted above, interaction of the iminium bond of chelerythrine and active sites in various target molecules can play an important role in biological responses that are mediated by the compound. An interaction of this type, of course, would not be specific for or limited to PKC, and could be more broadly related to the mechanism by which chelerythrine functions.