This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kedei, N. Articles by Blumberg, P. M. Search for Related Content PubMed PubMed Citation Articles by Kedei, N. Articles by Blumberg, P. M.

Characterization of the Interaction of Ingenol 3-Angelate with Protein Kinase C

¹ Laboratory of Cellular Carcinogenesis and Tumor Promotion, and ² Laboratory of Experimental Carcinogenesis, National Cancer Institute, Bethesda, Maryland, and ³ Peplin Biotech, Fortitude Valley, Queensland, Australia

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Ingenol 3-angelate (I3A) is one of the active ingredients in Euphorbia peplus, which has been used in traditional medicine. Here, we report the initial characterization of I3A as a protein kinase C (PKC) ligand. I3A bound to PKC- in the presence of phosphatidylserine with high affinity; however, under these assay conditions, little PKC isoform selectivity was observed. PKC isoforms did show different sensitivity and selectivity for down-regulation by I3A and phorbol 12-myristate 13-acetate (PMA) in WEHI-231, HOP-92, and Colo-205 cells. In all of the three cell types, I3A inhibited cell proliferation with somewhat lower potency than did PMA. In intact CHO-K1 cells, I3A was able to translocate different green fluorescent protein-tagged PKC isoforms, visualized by confocal microscopy, with equal or higher potency than PMA. PKC- in particular showed a different pattern of translocation in response to I3A and PMA. I3A induced a higher level of secretion of the inflammatory cytokine interleukin 6 compared with PMA in the WEHI-231 cells and displayed a marked biphasic dose-response curve for the induction. I3A was unable to cause the same extent of association of the C1b domain of PKC- with lipids, compared with PMA or the physiological regulator diacylglycerol, and was able to partially block the association induced by these agents, measured by surface plasmon resonance. The in vitro kinase activity of PKC- induced by I3A was lower than that induced by PMA. The novel pattern of behavior of I3A makes it of great interest for further evaluation.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The protein kinase C (PKC) family of serine/threonine kinases plays a central role in mediating the signal transduction of extracellular stimuli that result in production of the second messenger diacylglycerol (DAG). PKC isoforms are differentially involved in the regulation of cell proliferation, differentiation, cell survival, apoptosis, and carcinogenesis (1 , 2) . For example, elevation in PKC-ß_II is an early event in colon carcinogenesis (3) , overexpression of PKC- leads to increased sensitivity for epidermal carcinoma formation (4) , whereas overexpression of PKC- decreases the susceptibility to tumor promotion (5) and results in growth inhibition and differentiation (6) . PKC- is negatively involved in skin tumor promotion through the stimulation of differentiation of keratinocytes, whereas PKC- and - are frequently associated with cell survival and suppression of apoptosis (7, 8, 9) . Moreover, altered patterns of PKC isozyme expression and/or activation were reported in many diseases, including cancer, psoriasis, and diabetic retinopathy (2 , 10 , 11) .

The profound role of PKCs in the regulation of cell proliferation, differentiation, survival, and apoptosis makes them drug targets (12) . Bryostatin 1, a functional antagonist of many PKC-mediated effects, is in clinical trials as a cancer chemotherapeutic agent (13) . The PKC-ß-specific inhibitor LY333531 shows promise for prevention of diabetic retinopathy and is also being evaluated for other indications (11) . Short-chain substituted 12-deoxyphorbol derivatives such as prostratin or 12-deoxyphorbol 13-phenylacetate are antitumor promoting and may be of utility in combination with highly active anti-retroviral therapy for AIDS (14) . Phorbol 12-myristate 13-acetate (PMA) itself has been reported to have activity as a differentiating agent for the treatment of human myeloid leukemia cells (15) .

Consistent with their somewhat different biological functions, PKC isoforms differ in their structure, tissue distribution, subcellular localization, mode of activation, and substrate specificity (16, 17, 18) . The DAG-dependent PKC family members are composed of an NH₂-terminal regulatory domain containing a pseudosubstrate domain, a phospholipid and DAG-binding C1 domain, and a COOH-terminal catalytic domain. The Ca²⁺- and DAG-dependent isoforms also contain a Ca²⁺-binding C2 domain within the regulatory domain. In the inactivated state of the enzyme (usually localized to the cytoplasm), the pseudosubstrate domain of the regulatory region binds to the catalytic domain, inhibiting catalytic activity. PKC is activated endogenously by DAG, which likewise recruits PKC to the membrane (translocation).

Phorbol esters, tumor-promoting diterpene derivatives from plants of the family Euphorbiaceae, bind to the C1 domain in PKCs, as do DAGs, and activate the enzyme in a phospholipid-dependent manner (19, 20, 21, 22) . Although phorbol esters and related natural products appear to be ultrapotent analogs of DAGs, it has long been recognized that there is marked heterogeneity in the patterns of biological responses induced by these agents (23) . The most prominent example for the divergent biological responses is provided by the bryostatins. Bryostatins are more potent activators of PKC than are phorbol esters, but they induce only a subset of biological responses typical for phorbol esters and block in a dominant fashion the responses that they do not induce (24, 25, 26) . The short-chain derivatives of 12-deoxyphorbol-13 monoesters, prostratin and 12-deoxyphorbol 13-phenylacetate, are potent inflammatory agents on mouse skin, but the inflammation they induce is transient and they block tumor promotion induced by PMA (27) . Finally, phorbol esters with unsaturated side chains tend to be inflammatory but nonpromoting (28) .

Recently, Wada et al. (29) have reported that a phorbol ester with a hydrophilic 12-ester inhibited PKC activation by PMA in vitro. Such behavior would be predicted based on the current understanding of phorbol ester action. It is well known from the crystallographic (30) , nuclear magnetic resonance (31) , and modeling studies (32) of phorbol ester-PKC complexes that phorbol esters insert into a hydrophilic cleft in the C1 domain. The phorbol ester thus completes a hydrophobic surface on the upper surface of the C1 domain, promoting membrane interaction. In addition, the ester chains of the ligand further contribute to the hydrophobic surface, enhancing the membrane interaction. A hydrophilic ester or a short-chain ester decreases this latter hydrophobic contribution, resulting in decreased membrane stabilization. This structure activity relationship can explain why phorbol esters with different side chains induce different biological responses.

Ingenol derivatives are structurally closely related to the phorbol esters (33) but have been much less extensively investigated because of their more complicated chemistry. Ingenol 3-angelate (Fig. 1) is one of the active compounds from Euphorbia antiquorum and Euphorbia peplus, extracts of which have been used as traditional medicines in treating a number of conditions, including warts, corns, waxy growths, skin cancer, asthma, and catarrh (34, 35, 36) . Ingenol 3-angelate was first isolated and characterized for inflammatory and tumor-promoting activity by Adolf et al. in the early 1980s (37) . In those studies, ingenol 3-angelate showed unusual biology in mouse skin. It was highly potent in causing mouse ear inflammation, being more toxic than the long-chain substituted derivatives, but the inflammation was transient, unlike that of PMA. This transient inflammatory activity and local toxicity of ingenol 3-angelate was reminiscent of that of short-chain esters of deoxyphorbol (27) . Crude extracts from E. peplus have been shown to be active against a variety of tumor cell lines in vitro, including strains of malignant melanoma that are resistant to conventional therapeutic agents (38) .

View larger version (17K): [in this window] [in a new window] [Download PPT slide]   Fig. 1. Chemical structures of ingenol 3-angelate (I3A) and phorbol 12-myristate 13-acetate (PMA).

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Materials.
[³H]Phorbol 12,13-dibutyrate (PDBu; 21 Ci/mmol) was from Perkin-Elmer (Boston, MA) and [-³³P]ATP (3000 Ci/mmol) was from ICN (Costa Mesa, CA). PDBu was from LC Laboratories (Woburn, MA), and PMA was purchased from Alexis Biochemicals (Pittsburg, PA). Ingenol 3-angelate was from Peplin Biotech. (Fortitude Valley, Queensland, Australia). 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine; 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine; and 1,2-dioleyl-sn-glycerol (DOG) were purchased from Avanti Polar Lipids (Alabaster, AL). Reagents for expression and purification of glutathione S-transferase (GST) fusion proteins and enhanced chemiluminescence were obtained from Amersham Biosciences Corp. (Piscataway, NJ). CM5, L1 chips, and buffers used in surface plasmon resonance experiments were from Biacore Inc. (Piscataway, NJ). Gö-6983 and Gö-6976 were from Calbiochem (La Jolla, CA). The purified human PKC isoforms (PKC-, -ß, -, -, -) were purchased from Panvera (Madison, WI). DMEM and fetal bovine serum were purchased from American Type Culture Collection (Manassas, VA). RPMI, F-12 (Ham) medium, and antibiotics used for cell culture (100 units/ml penicillin and 100 µg/ml streptomycin) were from Invitrogen (Carlsbad, CA). All other chemicals were from Sigma (St. Louis, MO).

Cell Cultures.
CHO-K1 cells (CCL-61) and WEHI-231 cells (CRL-1702) were obtained from the American Type Culture Collection (Manassas, VA). Colo-205 and Hop-92 cells were from the Biological Testing Branch, National Cancer Institute, NIH (Frederick, MD). CHO-K1 cells were grown under usual conditions in F-12 medium (Ham), Colo-205 and Hop-92 in RPMI 1640, all supplemented with 10% fetal bovine serum and antibiotics. WEHI-231 cells were cultured in DMEM supplemented with 10% fetal bovine serum, antibiotics, and 50 µM ß-mercaptoethanol.

XTT Cell Proliferation Assay.
Cells grown on 96-well plates for 24 h were treated with single doses of the drugs (6 wells/dose), and the 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide inner salt (XTT)cell proliferation assay was performed 2 days after the treatment (3 days for Hop-92 cells) according to the manufacturer’s protocol (Roche Molecular Biochemicals, Indianapolis, IN). EC₅₀ values were calculated using Origin 6.0 (OriginLab Corp., Northampton, MA).

IL-6 ELISA.
The concentration of secreted IL-6 was determined in the supernatant of WEHI-231 cells using an IL-6 ELISA kit (Biosource International Inc, Camarillo, CA). One x 10⁵ viable cells/ml were plated in 6-cm dishes, treated after 24 h with the specified concentrations of the drugs [final solvent (DMSO) concentration 0.1%] for the indicated times (3–9 h and 24 h), after which, aliquots of supernatant were removed for IL-6 determination. The cells were pelleted from the supernatants by centrifugation at 1000 x g for 10 min. The IL-6 concentration in the supernatants was measured following the manufacturer’s protocol. During the optimalization of the assay, the IL-6 concentration in the supernatants was determined in triplicates. Later the IL-6 concentration of the supernatants was determined as a single measurement in each experiment or in duplicates for the experiments with the PKC inhibitors.

[³H]PDBu Binding.
The binding affinity of ingenol 3-angelate to partially purified mouse PKC- (39) , to purified C1b domain of PKC- (40) and to purified human PKC-, -ß, -, -, and - (Panvera) isoforms was determined by competition of [³H]PDBu binding. [³H]PDBu binding was measured using the polyethylene glycol precipitation assay as described in detail in Lewin and Blumberg (41) .

For the Schild plot, apparent binding affinities of [³H]PDBu to PKC- were determined in the presence of 0, 0.3, 0.6, and 0.9 nM ingenol 3-angelate. The apparent K_D values were plotted against the concentrations of ingenol 3angelate, and the K_i was calculated using the relationship K_i = K_D/slope, where K_D is the intersection of the straight line with the Y axis, indicating the K_D of [³H]PDBu.

Kinase Assay.
Activation of purified PKC- and PKC- was assayed by measuring the incorporation of ³³P from [-³³P 3000 Ci/mmol]ATP (ICN) into their specific peptide substrate as described previously (42) . PKC- activation was measured using an assay mixture (50 µl) consisting of 20 mM Tris-HCl (pH 7.5), 0.25 mg/ml BSA, 7.5 mM magnesium acetate, 0.1 mM CaCl₂, 100 µg/ml phospholipid (phosphatidylserine:phosphatidylcholine, 20:80 w/w, prepared by sonication), 0.03% Triton X-100, 600 nM PKC Selectide peptide (Calbiochem), ATP (50 µM final concentration, 1 µCi) and ingenol 3-angelate or PMA (20 nM-40 µM) from appropriate DMSO stocks, the final concentration of the diluent not exceeding 0.2%. The assay mixture (50 µl) used for measuring PKC- activity contained 20 mM Tris-HCl (pH 7.5), 0.25 mg/ml BSA, 7.5 mM magnesium acetate, 0.1 mM EGTA, 30 µM PKC- peptide (Calbiochem), 200 µg/ml phospholipid (phosphatidylserine:phosphatidylcholine 40:60 w/w), 0.03% Triton X-100, ATP (50 µM final concentration, 1 µCi), and ingenol 3-angelate or PMA (40 nM–80 µM). The assay tubes were incubated for 10 min at 30°C. The reaction was stopped by chilling on ice. Aliquots (25-µl) were spotted onto DE81 ion exchange chromatography paper (Whatman Ltd., Maidstone, England), followed by washing in 0.5% phosphoric acid three times for 5 min each. The bound radioactivity was measured in a scintillation counter. In each single experiment, each ligand concentration was assayed in triplicate and a dose-response curve was plotted.

Preparation of Lipid Vesicles.
Different large unilamellar vesicles with diameter of 100 nm, referred to as liposomes, were prepared for the surface plasmon resonance measurements. The control vesicles contained 80 mol % 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine and 20 mol % 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine; and the PMA-liposomes were made by the addition of 0.25 mol % of PMA to the control liposomes, and the DOG-liposomes were made by the addition of 4 mol % of DOG to the control liposomes. Aliquots of lipids in chloroform were mixed and dried under a stream of nitrogen, then were resuspended in 170 mM sucrose and 20 mM Tris-HCl (pH 7.4) to a concentration of 20 mM. The samples were vortexed for 30 s, were subjected to four freeze-thaw cycles by placing them in a 42°C water bath and in dry ice alternately, and then were extruded 40 times through two-stacked 0.1-µm pore polycarbonate filters using a LipoFast microextruder (Sigma-Aldrich, St. Louis, MO) to form liposomes. Specified concentrations of liposomes refer to those of the constituent phospholipids.

Biacore Experiments Using the CM5 Chip.
Anti-GST antibody (4500–5000 resonance units) were covalently bound to the surface of the CM5 chip according to the manufacturer’s protocol in HBS-EP buffer [150 mM NaCl, 10 mM HEPES, 3 mM EDTA, 0.005% Surfactant P20 (pH 7.4)]. Later on, HBS-EP or filtered and degassed PBS [100 mM Na₂HPO₄/NaH₂PO₄, 150 mM NaCl (pH 7.4) was used as running buffer at a flow rate of 20 µl/min. One hundred twenty µl of 150 nM GST-fused C1b domain of PKC- were injected over the surface and were incubated for 2 min before the injection of 180 µl of 0.75 mM PMA- or DOG-liposome with 0.1% DMSO (control) or specified concentrations of drugs in 0.1% DMSO final concentration (10 nM-10 µM). In membrane stabilization experiments, 0.5 mM of control liposomes were injected over the GST surface in the presence of specified concentrations of ingenol 3-angelate and PMA (0.1 nM-10 µM). The dissociation was observed for 10 min. The C1b domain and the lipids were washed down after each run with two 10-µl injections of 10 mM glycine (pH 2.0). Each sample was followed by a control when the C1b domain was replaced with running buffer. The sensorgram of the controls (measurement without C1b domain) was subtracted from the sensorgram of samples (with C1b domain) during the evaluation. The amount of bound liposome (resonance unit measured at the end of injection) was plotted against the drug concentration. These experiments were done on a Biacore 1000 instrument.

Biacore Experiments Using the L1 Chip.
A Biacore 2000 instrument (Biacore Inc.) was used to perform these experiments. Control liposomes and DOG-liposomes of the same concentration were immobilized on flow cell 1 and flow cell 2 of an L1 chip using HBS-N buffer [10 mM HEPES, 150 mM NaCl (pH 7.4); Biacore Inc.] as running buffer. Before the injection of liposomes, the flow cells were washed with 25 mM CHAPS for 1 min at a flow rate of 20 µl/min. The lipid surface was prepared by the injection of 1 mM liposome (240 µl) at a flow rate of 5 µl/min, followed by 1-min injection of 10 mM NaOH at a flow rate of 50 µl/min to remove the nonbound liposomes. Two hundred µl of 300 nM purified C1b domain of PKC- (diluted in running buffer, containing 0.1% DMSO for control or specified concentrations of the drugs in 0.1% DMSO final concentration) was injected over the surfaces at a flow rate of 40 µl/min. The dissociation was monitored for 5 min. DMSO or the drugs were added to the C1b domain just before the injection. The lipid surface was regenerated by a pulse of 25 µl of 10 mM glycine (pH 1.5) at a flow rate of 50 µl/min. As the liposome surfaces became contaminated by the drugs, a routine "Desorb" procedure was run to clean the surfaces of the L1 chip and new liposomes were immobilized. During the evaluation, the resonance unit at the very beginning of the dissociation was plotted against the drug concentration.

In the measurements of the direct binding of PMA or ingenol 3-angelate to the liposomes, the liposome surface was prepared as described, and 240 µl of drugs at specified concentrations in 0.1% final DMSO concentration, and control containing 0.1% DMSO, were injected at a flow rate of 40 µl/min; the dissociation was observed for 10 min. The sensorgram of the DMSO control was subtracted from the sensorgram of samples and the resonance unit_max was plotted against the drug concentration.

Down-Regulation of Different PKC Isoforms by Ingenol 3-Angelate and PMA.
The down-regulation experiments were performed as described previously (43) . Briefly, WEHI-231, HOP-92, and Colo-205 cells plated on 6-cm dishes were treated with the specified concentrations of ingenol 3-angelate or PMA (0.1–1000 nM) for 24 h. After treatment, the cells were washed, total cell lysates were prepared and analyzed on 10% SDS-PAGE, followed by electrotransfer and immunostaining. The primary antibodies used for immunostaining were against PKC- (C-20), -ß_I (C-16), -ß_II (C-18), - (C-20), - (C-15), and -µ (C-20), all from Santa Cruz Biotechnology (Santa Cruz, CA). The immunostaining was visualized by ECL (Amersham Biosciences). The membranes were immunostained with anti-actin antibody as a control for sample loading after stripping the membranes in Restore Western blot stripping buffer (Pierce, Rockford, IL) for 15 min at 60°C. Films were scanned, densitometry was performed using Image J (developed at National Institute of Mental Health, NIH), and EC₅₀ values were calculated using Origin 6.0.

Translocation of Mouse PKC-, -ß_I, -, -, -µ and Human PKC- in Living Cells.
The translocation experiments were performed as described in detail previously (44) . Briefly, CHO-K1 cells were seeded onto 40-mm circular glass coverslips, and 24 h later, were transiently transfected with green fluorescent protein (GFP)-tagged mouse PKC-, -ß_I, -, -, -µ and human PKC- subcloned into pEGFP-N1 plasmid. Translocation of PKC isoforms was visualized by confocal microscopy 48 h after transfection.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Binding of Ingenol 3-Angelate to PKC Isoforms.
We determined the binding affinity of ingenol 3-angelate to PKC and its possible in vitro selectivity among PKC isoforms. The PKCs were assayed in the presence of phosphatidylserine (100 µg/ml), and the K_i values were measured by competition of [³H]PDBu binding. Ingenol 3-angelate bound to mouse PKC- with high affinity; its K_i was 0.10 ± 0.02 nM compared with a K_i of 0.3 nM for phorbol 12,13-dibutyrate. This potent in vitro binding affinity is consistent with the impressive in vivo inflammatory potency of ingenol 3-angelate on the mouse ear, for which its EC₅₀ (0.007 nmol/ear) is 2-fold more potent than that for PMA (0.016 nmol/ear), although the different time courses of action prevent exact comparison (37) .

To verify that the inhibition of [³H]PDBu binding by ingenol 3-angelate reflected a competitive mechanism, the apparent dissociation constant of [³H]PDBu was measured in the presence of increasing concentrations (0, 0.3, 0.6, and 0.9 nM) of ingenol 3-angelate. The apparent K_d values of [³H]PDBu were plotted as a function of the concentration of ingenol 3-angelate (Fig. 2B) . Consistent with a competitive mechanism, the Schild plot was linear, yielding a K_i value of 0.162 nM for ingenol 3-angelate, a value very similar to that obtained above by competition at a fixed concentration of [³H]PDBu (K_i = 0.10 ± 0.02 nM).

View larger version (8K): [in this window] [in a new window] [Download PPT slide]   Fig. 2. Binding of ingenol 3-angelate to different protein kinase C (PKC) isoforms. A, binding affinities of ingenol 3-angelate to PKC-, -ß, -, -, and -. Binding affinities of ingenol 3-angelate to PKC-, -ß, -, -, and - were measured in the presence of phosphatidylserine by inhibition of [3H]phorbol 12,13-dibutyrate (PDBu) binding. Values represent the mean ± SE of three independent experiments. B, Schild plot of binding of ingenol 3-angelate to PKC- in the presence of phosphatidylserine. Apparent binding affinities (Kapp) of [3H]PDBu to PKC- were measured in the presence of 0, 0.3, 0.6, and 0.9 nM of ingenol 3-angelate and were plotted against the concentration of the drug. Values represent the average ± SE of five independent experiments.

Down-Regulation of Different PKC Isoforms.
Prolonged treatment with phorbol esters and other natural products is known to result in down-regulation of different PKC isoforms (43 , 45) . We determined whether there was any isoform selectivity in down-regulation caused by 24-h treatment with ingenol 3-angelate compared with PMA in three different cell lines, WEHI-231, Hop-92, and Colo-205 cells. In all of the cell lines, PMA and ingenol 3-angelate were able to cause down-regulation of the tested classical and novel PKCs and to change the mobility of PKC-µ in a dose-dependent manner. There were reproducible differences in the sensitivity and selectivity of the isoforms to the drugs in the different cell lines (Fig. 3 ; Table 1 ). In WEHI-231 cells, ingenol 3-angelate was 10-fold less potent than PMA in down-regulating conventional PKCs present in this cell line (PKC-, PKC-ß_I, and PKC-ß_II), with less (PKC-) or no (PKC-) apparent difference in their potency for novel PKCs (Fig. 3A ; Table 1 ). Similar changes were seen in HOP-92 cells with some exceptions; most of the isoforms (PKC-, -ß_I, -) were about 3–10-fold less sensitive to the drugs than in WEHI-231 cells, and the level of PKC- was not changed with the treatments (Table 1) . A shift in the mobility of PKC-µ was caused by low concentrations of PMA (0.3 nM) or ingenol 3-angelate (1 nM) in both of the cell lines. We assume that this shift is brought about by the phosphorylation of PKC-µ by different PKCs, as reported previously (46) . As a matter of comparison of isoforms, we observed marked differences in the sensitivity of the different PKC isoforms to down-regulation by the treatments in these two cell lines. For ingenol 3-angelate, PKC-ß_I and PKC- were the most sensitive, followed by PKC- and PKC-ß_II. PKC- and PKC-µ were the least sensitive. Phosphorylation of PKC-µ was somewhat more sensitive than was down-regulation of any of the PKC isoforms. A markedly different pattern in isoform sensitivity and selectivity was observed in Colo-205 cells. PKC- was the most sensitive for the ingenol 3-angelate treatment, followed by PKC- and PKC-µ. The classical PKC- and PKC-ß_I were the least sensitive in this cell line (Fig. 3B ; Table 1 ). Furthermore, ingenol 3-angelate was about 10–30-fold more potent in down-regulating PKC- and - than was PMA, having a modest biphasic effect on PKC- (Fig. 3B) .

View larger version (91K): [in this window] [in a new window] [Download PPT slide]   Fig. 3. Down-regulation of different protein kinase C (PKC) isoforms after 24-h treatment with ingenol 3-angelate or phorbol 12-myristate 13-acetate (PMA) in WEHI-231 and Colo-205 cells. Cells were treated with ingenol 3-angelate or PMA for 24 h. After the treatment, total cell lysates were prepared and analyzed on 10% SDS-PAGE, followed by electrotransfer and immunostaining with the indicated anti-PKC antibodies. Anti-actin immunostaining was used as a control for loading. The data shown are from one experiment. Two additional experiments yielded similar results. A, WEHI-231; B, Colo-205.

View larger version (73K): [in this window] [in a new window] [Download PPT slide]   Fig. 4. Translocation of green fluorescent protein (GFP)-tagged protein kinase C (PKC) isoforms in CHO-K1 cells. A, translocation of GFP-tagged PKC-ß, -, -, -µ on ingenol 3-angelate (I3A) treatment. CHO-K1 cells expressing GFP-tagged PKC-ß, -, -, -µ were treated with 10 nM I3A for 45 min, and images were taken every 30 s. Images taken at 0 and 30 min are shown. Arrows, the principal sites of translocation. Data represent confocal images obtained from a representative experiment. Two additional experiments yielded similar results. B, translocation of PKC- and - by I3A. CHO-K1 cells expressing GFP-tagged PKC- or - were treated with 10 nM or 100 nM of I3A for 45 min, and images were taken every 30 s. Images taken at 0 and 30 min are shown. Arrows, the principal site of translocation. Data represent confocal images obtained from a representative experiment. Two additional experiments yielded similar results. C, different translocation pattern of GFP-tagged PKC- treated with 10 nM I3A or 100 nM phorbol 12-myristate 13-acetate (PMA). CHO-K1 cells expressing PKC- were treated with 10 nM I3A or with 100 nM PMA for 45 min and images were taken every 30 s. Images at 0, 5, 10, 15, and 30 min are shown. Arrows, the principal sites of translocation. Data represent confocal images obtained from a representative experiment. Two additional experiments yielded similar results.

View larger version (40K): [in this window] [in a new window] [Download PPT slide]   Fig. 4A. Continued.

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Fig. 5. Effect of phorbol 12-myristate 13-acetate (PMA) and ingenol 3-angelate on the proliferation of WEHI-231, HOP-92, and Colo-205 cells. Cells were treated with the specified concentrations of PMA and ingenol 3-angelate for 48 h (72 h for HOP-92 cells). The cell proliferation was determined by the 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide inner salt (XTT) assay. Results are presented as percentage of the DMSO-treated (0.1% DMSO) control. Values represent the average ± SE of three independent experiments. A, WEHI-231; B, Hop-92; C, Colo-205.

View larger version (49K): [in this window] [in a new window] [Download PPT slide]   Fig. 6. Concentration of interleukin 6 (IL-6) secreted into the supernatant of WEHI-231 cells. A, time course and dose response of IL-6 secretion. WEHI-231 cells were treated with specified concentrations of ingenol 3-angelate and phorbol 12-myristate 13-acetate (PMA) for the indicated periods of time and the IL-6 secretion in the supernatant was measured. The concentration of IL-6 secreted into the supernatant was determined by sandwich ELISA. Results presented are the mean ± SE of three independent experiments. To permit comparison among experiments, results were normalized to the concentration of IL-6 induced by 10 nM ingenol 3-angelate at 6 h. Insert, IL-6 concentration in the supernatant of WEHI-231 cells after 24-h treatment with the indicated concentrations of ingenol 3-angelate or PMA. B, effects of protein kinase C (PKC) inhibitors on the ingenol 3-angelate and PMA induced IL-6 production. WEHI-231 cells were treated with the specified concentrations of ingenol 3-angelate or PMA in the absence or in the presence of PKC inhibitors Gö-6983 and Gö-6976. The concentration of IL-6 secreted into the supernatant was determined by sandwich ELISA. Mean values and range of duplicates in a single experiment are shown. The experiment was repeated two additional times with similar results.

Stabilization of the Interaction of PKC with Membranes.
In elegant studies, Bertolini et al. (51) demonstrated that phorbol esters with hydrophilic patterns of substitution are less able to cause PKC to associate with liposomes. Because of the somewhat hydrophilic nature of the ingenol 3-angelate, we tested the ability of ingenol 3-angelate to stabilize the interaction of C1b domain with phospholipids using surface plasmon resonance. In the first experimental approach, the C1b domain of PKC- was immobilized and liposomes were injected over the surface in the presence of specified concentrations of ingenol 3-angelate or PMA (0.1 nM-10 µM). Ingenol 3-angelate and PMA increased the binding of the liposomes to the C1b domain of PKC- in a biphasic manner (Fig. 7A) . Two differences were apparent. First, ingenol 3-angelate was somewhat less potent for inducing liposome binding to the C1b domain. Second, the maximal level of association of the liposomes with the C1b domain of PKC- was less for ingenol 3-angelate than it was for PMA. We concluded that ingenol 3-angelate was less efficient in inducing stabilization of the membrane interaction of the C1b domain.

View larger version (22K): [in this window] [in a new window] [Download PPT slide]   Fig. 7. Membrane stabilization of the C1b domain caused by ingenol 3-angelate (I3A) or phorbol 12-myristate 13-acetate (PMA); effect of I3A on binding of PMA- or 1,2-dioleyl-sn-glycerol (DOG)-liposomes to the C1b domain of protein kinase C- (PKC-). A, enhancement of liposome binding to the C1b domain by I3A or PMA. Binding of 0.5 mM control liposomes [20% 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine:80% 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPS:POPC) diluted in PBS] to immobilized C1b domain of PKC- was measured in the presence of the indicated concentrations of the drugs by surface plasmon resonance. The maximal value of liposome binding, measured at the end of injection, was plotted against the concentration of the drug. Points represent the mean values ± SE of four independent experiments. For comparison among experiments, values were normalized, the binding of liposomes in the absence of drugs being defined as 0 and the maximal binding being defined as 1. B, binding of PMA-liposome to the immobilized C1b domain of PKC- in the absence or presence of I3A. Seven hundred fifty µM PMA-liposome (0.25 mol % PMA in POPC/POPS liposomes) in the presence or absence of I3A (10 nM-10 µM) was injected over the C1b domain of PKC- immobilized to the -glutathione S-transferase (GST)-CM5 chip. The concentration of DMSO (used as diluent) was 0.1% in all cases. One representative experiment is shown. Four additional experiments gave similar results. RU, resonance units. C, effect of I3A on binding of PMA-liposomes or control liposomes to the immobilized C1b domain of PKC-. PMA liposomes (0.25 mol % PMA in POPC/POPS liposomes) or control liposomes (20% POPS:80% POPC) were injected over the C1b domain-containing surface, and the maximum resonance units (RU) value at the end of injection was plotted against the concentration of I3A as percentage of control. Binding of PMA-liposome in the absence of I3A represents 100%. Mean values and range of duplicates or SE of triplicates run in a single experiment are shown. The experiment was repeated two additional times with similar results. D, effect of I3A on binding of DOG-liposomes or control liposomes to the immobilized C1b domain of PKC-. The experiment was performed as described in C. Binding of DOG-liposome (4 mol % DOG in POPC/POPS liposomes) without treatment was taken as 100%. Mean values and range of duplicates or SE of triplicates run in a single experiment are shown. The experiment was repeated two additional times with similar results.

We also evaluated the partial antagonism of ingenol 3-angelate on the binding of the C1b domain to lipids using the reciprocal technical approach, in which the lipids were immobilized on the surface and the GST-fused C1b domain was in the soluble phase. The C1b domain of PKC- bound both to the control lipid surface and to the DOG-lipid surface in a dose-dependent manner (40–1280 nM tested) with much higher binding to the lipid surface containing DOG (data not shown). GST alone did not bind to any of the lipid surfaces (up to 2.5 µM; data not shown). In the subsequent experiments evaluating the effect of ingenol 3-angelate on the binding, 300 nM C1b domain was used. Consistently with the experiments in which the C1b domain was immobilized, ingenol 3-angelate inhibited the binding of the C1b domain to the DOG-lipid surface and increased the binding to the control lipid surface (Fig. 8, A and B) . Indeed, 1 µM ingenol 3-angelate fully decreased the binding of the C1b domain to the DOG-lipid surface to the level of binding to the lipid surface without DOG (Fig. 8A) . PMA was also able to increase the binding of the C1b domain to the control lipid surface and to decrease the binding to the DOG-lipid surface in a concentration-dependent manner (Fig. 8C) .

View larger version (21K): [in this window] [in a new window] [Download PPT slide]   Fig. 8. Effect of ingenol 3-angelate and phorbol 12-myristate 13-acetate (PMA) on binding of the C1b domain of protein kinase C (PKC)- to immobilized liposomes; Binding of ingenol 3-angelate and PMA to immobilized liposomes. A, binding of the C1b domain of PKC- to immobilized control liposomes and 1,2-dioleyl-sn-glycerol (DOG)-liposomes in the absence or presence of 1 µM ingenol 3-angelate. Control liposomes [(20% 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine:80% 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPS:POPC)] or DOG-liposomes (4 mol % DOG in POPC/POPS liposomes) were immobilized on flow cell 1 and 2 of an L1 chip and 300 nM C1b domain of PKC- was injected over the surfaces. Representative binding curves are shown. The experiment was repeated two additional times with similar results. B, effect of ingenol 3-angelate on binding of the C1b domain to different liposomes; binding of ingenol 3-angelate to liposomes. Control liposomes (20% POPS:80% POPC) or DOG-liposomes (4 mol % DOG in POPC/POPS liposomes) were immobilized on flow cell 1 and 2 of an L1 chip. Three hundred-nM C1b domain of PKC- in the absence or presence of the specified concentrations of ingenol 3-angelate was injected. The same concentrations of ingenol 3-angelate without the protein were also injected over control-liposome and DOG-liposome covered surfaces. The maximal values of binding were plotted against the concentration. The concentration of DMSO used as diluent was 0.1% in all cases. Mean values and range of duplicate or SE of triplicate measurements run in a single experiment are shown. Two additional experiments showed similar results. C, effect of PMA on binding of the C1b domain to different liposomes; binding of PMA to liposomes. Control liposomes (20% POPS: 80% POPC) or DOG-liposomes (4 mol % DOG in POPC/POPS liposomes) were immobilized on flow cell 1 and 2 of an L1 chip. Two hundred µl of 300 nM C1b domain of PKC- in the absence or presence of the specified concentrations of PMA was injected over the surface. The same concentrations of PMA without the protein were also injected over control-liposome and DOG-liposome covered surfaces. The maximal values of binding were plotted against the concentration. The concentration of DMSO used as diluent was 0.1% in all cases. Mean values and range of duplicate or SE of triplicate measurements run in a single experiment are shown. One additional experiment showed similar results.

In Vitro Kinase Assay.
Finally, we compared the effect of ingenol 3-angelate and PMA on the kinase activity of purified human PKC- and PKC- in an in vitro kinase assay. Both compounds increased the basal activity of PKC- and PKC- in a dose-dependent manner (Fig. 9) . Ingenol 3-angelate caused lower increase in kinase activity of PKC- compared with PMA, consistent with its lower potency in stabilizing membrane interaction. No such difference was observed with PKC-, in which ingenol 3-angelate was even slightly more potent in inducing kinase activity, the maximum of activation being similar.

View larger version (18K): [in this window] [in a new window] [Download PPT slide]   Fig. 9. Effect of phorbol 12-myristate 13-acetate (PMA) and ingenol 3-angelate (I3A) on the in vitro kinase activity of protein kinase C (PKC)- and PKC-. A.) Kinase activity of PKC- induced by PMA and I3A. Activation of purified human PKC- was assayed by measuring the incorporation of 33P from ATP into PKC Selectide substrate in the presence of 100 µg/ml lipid [1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine:1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPS:POPC) 20:80], 0.03% Triton X-100, 0.1 mM Ca2+, 7.5 mM magnesium acetate for 10 min at 30°C. Mean values ± SE of triplicate measurements from a single experiment are shown. Two additional experiments showed similar results. B, kinase activity of PKC- induced by PMA and I3A. Activation of purified human PKC- was assayed by measuring the incorporation of 33P from ATP into PKC- peptide substrate in the presence of 200 µg/ml lipid (POPS:POPC 40:60), 0.03% Triton X-100, 0.1 mM EGTA, 7.5 mM magnesium acetate for 10 min at 30°C. Mean values ± SE of triplicate measurements from a single experiment are shown. Two additional experiments showed similar results. Where error bars are not visible, they are less than the size of the symbol.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Ingenol 3-angelate is a rather unusual hydrophilic ligand of PKC. Its calculated Log P value is 3.89, much lower than that of for PMA (6.65, both calculated by ChemDraw). Its limited selectivity for different isoforms in vitro is like that of compounds such as (–)octylindolactam V (39) , or the DAG-lactone HK654 (54) . It is important to note that lack of selectivity in vitro does not necessarily predict lack of specificity in the intact cell. Although the DAG-lactone HK654 did not discriminate between different PKC isoforms in in vitro binding or kinase assays, it showed PKC- selectivity in translocation and apoptosis in LNCaP prostate cells (54) .

Prolonged activation of PKCs may cause down-regulation of different isoforms through calpain, caspases, and the proteosome pathway (55, 56, 57) . Different PKC activators behave differently in this aspect. For example, bryostatin 1 is able to prevent PMA-induced down-regulation of PKC- in some systems (45) . Ingenol 3-angelate caused down-regulation of classical and novel PKC isoforms in three tested cancer cell lines similarly to PMA, but showing 1–30-fold differences in the potency of the drugs. There were also pronounced differences in the selectivity of the isoforms to the drugs in the different cell lines highlighting that ingenol 3-angelate is not just another phorbol ester with different potency. These differences in isoform selectivity should cause, in a dose-dependent manner, differences in the ratios of in situ activities of PKC isoforms. Because different PKC isoforms have somewhat different functions, this could be reflected in biological differences, such as we indeed observed for IL-6 secretion.

One hallmark of PKC activation is translocation to membranes (58 , 59) . Most inactive isozymes are localized to subcellular structures and, on activation, translocate to new distinct intracellular sites depending on the stimulus and the specific anchoring proteins. For example, PKC- was shown to translocate to the plasma membrane or to adhesion foci (for review see Ref. 60 ), PKC- translocated to the plasma membrane, nuclear membrane, and mitochondrial membrane (depending on the system; Refs. 44 , 61 , 62 ), PKC- associated with the plasma membrane and the cytoskeleton (for review see Ref. 63 ). The substrate specificity of PKCs will depend in part on their localization, because the localization will determine which potential substrates are accessible to the enzymes (64) . Different patterns of translocation would thus be expected to lead to somewhat different biological responses, and, indeed, one mechanism for antagonism would be to localize an activated PKC to a cellular compartment in which its normal substrate is not found.

Different PKC ligands drive different patterns of PKC localization, at least in the case of PKC- (62) . Although most of the structural features of ligands determining the different patterns of PKC localization remain uncertain, lipophilicity is clearly one critical element (44) . Ingenol 3-angelate is more hydrophilic than PMA, and the translocation pattern was found to be similar to that caused by more hydrophilic phorbol ester derivatives, such as 12-deoxyphorbol 13-phenylacetate or PDBu in CHO cells expressing different GFP-tagged PKC isoforms. The difference in the translocation pattern of PKC- could contribute to the different pattern of biological response of ingenol 3-angelate compared with the more hydrophobic phorbol esters.

Phorbol esters are known to cause differentiation, inhibition of cell proliferation, and cell cycle arrest in different cancer cells. Ingenol 3-angelate caused inhibition of cell proliferation in the three tested cell lines (B cell lymphoma, colon cancer, and lung cancer) similarly to PMA with somewhat lower potency, behaving as a typical phorbol ester in this aspect. There was a difference in the sensitivity of the cell lines to the drugs, the WEHI-231 cells being the most sensitive, followed by HOP-92 and Colo-205 cells, consistent with the down-regulation experiments.

Inflammation was reported to be one of the prominent biological responses to ingenol 3-angelate in mouse skin (37) . IL-6 is a pleiotropic cytokine that acts on a wide variety of cell types. It has important regulatory functions in the immune system exerting both pro- and anti-inflammatory activity, is a mediator of the acute-phase response, and is involved in the regulation of differentiation, proliferation, and survival of target cells (for review see Refs. 65 , 66 ). The regulation of IL-6 expression at the molecular level is complex (67, 68, 69) . PMA, as well as calcium ionophore and cAMP, was shown to up-regulate IL-6 gene expression in astrocytes (47 , 70 , 71) , as well as to increase IL-6 mRNA stability (72) , showing that PMA can regulate IL-6 secretion at multiple levels. Different PKC isoforms were shown to be involved in the regulation of IL-6 secretion (73, 74, 75, 76) . The results on PMA-induced IL-6 secretion in astrocytes or on PMA-induced cytokine production in the EL4–6.1 thymoma cell line (70 , 77) , as well as our data in WEHI-231 cells, suggest that multiple PKC isoforms are involved in the regulation of IL-6 secretion in these cells, some of them increasing, others diminishing IL-6 secretion. The different relative potencies of the two drugs for the different isoforms and the different sensitivity of isoforms to the drugs could explain the observed biphasic effect and the difference in response. Whether ingenol 3-angelate and PMA behave differently for cytokine induction in mouse skin remains to be determined, but it could plausibly contribute to the toxicity of ingenol 3-angelate that was described.

DAG and phorbol esters bind to the hydrophilic cleft of the C1 domain of PKCs, forming hydrogen bonds, and the 13-ester side-chain completes the hydrophobic surface of the protein (30, 31, 32) . The continuous hydrophobic surface allows the insertion of the protein into membranes and activation of enzyme activity. This membrane stabilization contributes to the stabilization of the activated complex and to substrate specificity, and molecules that bind to the C1 domain but do not stabilize the interaction with the membrane should block binding by other DAG analogs and, at the same time, cause only limited association of PKC-ligand complex with the membranes. The overall result, therefore, could be partial agonism at the level of the in vivo PKCs. The plausibility of this partial association with membranes was previously shown (51) . In addition, Wada et al. reported (29) that a phorbol ester with a hydrophilic 12-ester chain could inhibit the activation of PKC- by PMA. The consistent model that emerges from the surface plasmon resonance (Biacore) experiments is that ingenol 3-angelate is able to bind to PKC but is less able to stabilize its membrane interaction. It is clear, however, that the systems reflect multiple actions of the ligands, because higher concentrations of either of the phorbol esters suppressed the association with the lipids. The difference in membrane stabilization presumably contributes to the decreased ability of ingenol 3-angelate in activating PKC- in an in vitro kinase assay and an altered pattern of PKC- localization.

Natural products have provided an impressive window into the extent of heterogeneity of biological response that can be elicited from the PKC receptor family and the related receptors with DAG-responsive C1 domains. As with the antipromoting 12-deoxyphorbol 13-monoesters such as prostratin, the distinct behavior of ingenol 3angelate in the early characterization by Adolf et al. (37) using the mouse model is reflected in its novel behavior in biochemical and cellular systems, as described here. The further characterization of ingenol 3-angelate in a range of systems will help to define the extent of its differences relative to compounds such as PMA. Ingenol derivatives have been relatively neglected compared with the phorbol esters or teleocidins for structure activity analysis. Evaluation of the contribution of the ingenol scaffold to the special properties of ingenol 3-angelate represents a future challenge for PKC pharmacology. Conversely, the concepts emerging from the behavior of ingenol 3-angelate provide strategies for drug design and screening using other structural templates targeted to the C1 domain such as the constrained DAGs (78) . Finally, ingenol 3-angelate, a constituent of the traditional Thai medicines "Yang Sa-Lad-Dai" and "Yan Nam Radom Pol" (37) , illustrates yet again in this era of combinatorial chemistry the wealth of pharmacological leads afforded by traditional medicines.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Peter M. Blumberg, Laboratory of Cellular Carcinogenesis and Tumor Promotion, Building 37, Room 4048B, National Cancer Institute, 37 Convent Drive, MSC 4255, Bethesda, MD 20892-4255. Phone: (301) 496-3189; Fax: (301) 496-8709; E-mail: blumberp{at}dc37a.nci.nih.gov

⁴ N. Kedei and P. Blumberg, unpublished data.

Received 10/30/03. Revised 1/20/04. Accepted 2/27/04.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Mellor H, Parker PJ The extended protein kinase C superfamily. Biochem J, 332: 281-92, 1998.
2. Dempsey EC, Newton AC, Mochly-Rosen D, et al Protein kinase C isozymes and the regulation of diverse cell responses. Am J Physiol Lung Cell Mol Physiol, 279: L429-38, 2000.[Abstract/Free Full Text]
3. Murray NR, Davidson LA, Chapkin RS, Clay Gustafson W, Schattenberg DG, Fields AP Overexpression of protein kinase CßII induces colonic hyperproliferation and increased sensitivity to colon carcinogenesis. J Cell Biol, 145: 699-711, 1999.[Abstract/Free Full Text]
4. Reddig PJ, Dreckschmidt NE, Zou J, Bourguignon SE, Oberley TD, Verma AK Transgenic mice overexpressing protein kinase C in their epidermis exhibit reduced papilloma burden but enhanced carcinoma formation after tumor promotion. Cancer Res, 60: 595-602, 2000.[Abstract/Free Full Text]
5. Reddig PJ, Dreckschmidt NE, Ahrens H, et al Transgenic mice overexpressing protein kinase C in the epidermis are resistant to skin tumor promotion by 12-O-tetradecanoylphorbol-13-acetate. Cancer Res, 59: 5710-8, 1999.[Abstract/Free Full Text]
6. Mischak H, Pierce JH, Goodnight J, Kazanietz MG, Blumberg PM, Mushinski JF Phorbol ester-induced myeloid differentiation is mediated by protein kinase C- and - and not by protein kinase C-ßII, -, -, and -. J Biol Chem, 268: 20110-5, 1993.[Abstract/Free Full Text]
7. Deacon EM, Pongracz J, Griffiths G, Lord JM Isoenzymes of protein kinase C: differential involvement in apoptosis and pathogenesis. Mol Pathol, 50: 124-31, 1997.[Free Full Text]
8. Yuspa SH The pathogenesis of squamous cell cancer: lessons learned from studies of skin carcinogenesis. J Dermatol Sci, 17: 1-7, 1998.[CrossRef][Medline]
9. Kuroki T, Kashiwagi M, Ishino K, Huh N, Ohba M Adenovirus-mediated gene transfer to keratinocytes—a review. J Investig Dermatol Symp Proc, 4: 153-7, 1999.[Medline]
10. Rasmussen HH, Celis JE Evidence for altered protein kinase C (PKC) signaling pathway in psoriasis. J Investig Dermatol, 101: 560-6, 1993.[CrossRef][Medline]
11. Aiello LP The potential role of PKCß in diabetic retinopathy and macular edema. Surv Ophthalmol, 47 Suppl 2: S263-9, 2002.