Differential Effect of Bryostatin 1 and Phorbol 12-Myristate 13-Acetate on HOP-92 Cell Proliferation Is Mediated by Down-regulation of Protein Kinase Cδ

The protein kinase C (PKC) family of serine/threonine kinases plays a central role in mediating the action of growth factors, neurotransmitters, and hormones in cellular growth control and in carcinogenesis (1). The multiple isoforms of PKC differ in their regulation and in their biological functions. The classic PKCs (α, βI, βII, and γ) are Ca²⁺ dependent and diacylglycerol (DAG) responsive; the novel PKCs (δ, ϵ, η, and 𝛉) are Ca²⁺ independent but DAG responsive (2–4). Different PKC isoforms have different biological functions; indeed, one isoform may antagonize the function of another. Thus, in mouse NIH3T3 cells, overexpression of PKCδ is antiproliferative, whereas overexpression of PKCα or PKCϵ stimulates proliferation (5). The function of individual PKC isoforms is also very much context dependent. Thus, PKCδ is proapoptotic in C6 glioma cells in the presence of etoposide (6) but is antiapoptotic in the same cells in response to Sindbis virus infection (7) or in A172 glioma cells in response to tumor necrosis factor–related apoptosis-inducing ligand (TRAIL; ref. 8).

Phorbol esters, such as phorbol 12-myristate 13-acetate (PMA), induce the translocation of PKC from the cytosolic to the particulate fraction (9). By controlling access to potential substrates, differential translocation of PKC isoforms can further contribute to differential activity (10). Using green fluorescent protein (GFP), the translocation of PKC on different stimuli can be monitored in living cells and in real time. Several reports have shown that translocation of PKC is isoform, cell type, and activator specific and is tightly regulated by various cofactors (11, 12).

Chronic treatment of PKC with phorbol esters or other activators may lead to the time-dependent decrease in the level of PKC, termed down-regulation. Down-regulation provides a mechanism by which an activator of PKC can, at least for longer-term responses, inhibit PKC function. It is clear that down-regulation is a complicated process that can reflect contributions of the calpain, caspase, and proteasome pathways (6, 13, 14). The extent of down-regulation depends on the PKC isoform, on the specific PKC ligand, and on the cellular context for PKC, among other variables (15–17).

Bryostatin 1 is currently undergoing several clinical studies against malignancies, including advanced melanoma, relapsed multiple myeloma, metastatic renal cell carcinoma, and refractory B-cell chronic lymphocytic leukemia (18–20). The bryostatins are macrocyclic lactones isolated from the marine organism Bugula neritina. Together with the phorbol esters, the indole alkaloids, the polyacetates, and the iridals, they represent potent ligands and activators of PKC (21). Uniquely, however, the bryostatins induce only a subset of the responses induced by the phorbol esters and functionally antagonize those responses that they themselves do not induce. Thus, bryostatin 1 blocks the induction by phorbol ester of differentiation in primary mouse keratinocytes 1 (22), HL-60 cells (23), and Friend erythroleukemia cells (24).

The mechanism(s) accounting for the functional antagonism of phorbol ester responses by bryostatin 1 remains largely unresolved, although a variety of mechanistic differences between the action of the phorbol esters and bryostatin 1 have been described. Bryostatin 1 causes a different pattern of translocation of PKCδ than do the typical phorbol esters (11). This differential localization, by providing differential access to substrates, could drive different biology. Bryostatin 1 has also been shown to produce more rapid down-regulation of PKCα in LL-MK2 epithelial cells (25) and of PKCα and PKCβ in human T cells (26) than do the phorbol esters. In addition, bryostatin 1 induced a unique biphasic pattern of down-regulation of PKCδ, with high nanomolar concentrations of bryostatin 1 failing to induce down-regulation of PKCδ in NIH3T3 cells (27), in primary mouse keratinocytes (22), in B16/F10 melanocytes (28), and in HeLa cells (29). In the NIH3T3 cells (27) and primary mouse keratinocytes (22), moreover, bryostatin 1 was shown to protect PKCδ from down-regulation by phorbol esters. The protection of PKCδ from down-regulation seems to involve multiple structural elements within PKC, including the catalytic domain (30) and the C1a domain (31).

Understanding of the mechanisms underlying the unique behavior of bryostatin 1 should help to identify the systems in which it has a rational therapeutic application. In the present study, we have tested the effect of the typical phorbol ester, PMA, and of bryostatin 1 on HOP-92 cellular growth and have been able to relate the differential effect of bryostatin 1 to its down-regulation of PKCδ.

Materials. Bryostatin 1 was from the Drug Chemistry and Synthesis Branch, National Cancer Institute (NCI). PMA was purchased from LC Laboratories (Woburn, MA) and dissolved in DMSO. Affinity-purified polyclonal antibodies directed against the various PKC isoforms were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The PKC inhibitor Gö-698, the calpain inhibitor calpeptin, the calpain inhibitor E64d, the proteasome inhibitor lactacystin, and the caspase inhibitors caspase inhibitor 1 and caspase-3 inhibitor 1 were obtained from Calbiochem (La Jolla, CA). Protease inhibitor cocktail tablets were purchased from Roche Molecular Biochemicals (Indianapolis, IN). Glutamine, RPMI 1640, fetal bovine serum (FBS), and LipofectAMINE Plus reagent were obtained from Invitrogen (Gaithersburg, MD). Enhanced chemiluminescence (ECL) reagents were from Amersham Life Sciences (Arlington Heights, IL). SDS-PAGE precast gels and horseradish peroxidase (HRP)–conjugated anti-rabbit IgG were obtained from Bio-Rad (Hercules, CA).

Construction of GFP-PKCδ fusion protein. A plasmid containing the GFP cDNA (pEGFP-N1) was purchased from Clontech Laboratories, Inc. (Palo Alto, CA). A MluI restriction site was generated by inserting a MluI linker into the plasmid digested with SmaI. A mouse PKCδ cDNA fragment with XhoI and MluI sites was subcloned into the expression vector pEGFP-N1 with the GFP attached to the 3′ end of PKCδ. The junction of PKCδ and GFP in the constructed plasmid was verified by sequencing, which was done by the DNA Minicore, Division of Basic Sciences, NCI, NIH. Plasmid DNA was prepared using the Endo-Free Qiagen kit (Qiagen GmbH, Hilden, Germany) to remove the bacterial endotoxins.

Expression of GFP-PKCδ in HOP-92 cells. HOP-92 cells were seeded onto 40-mm round glass coverslips. Twenty-four hours later, the cells were transfected with the GFP-PKCδ construct using LipofectAMINE Plus reagent according to the manufacturer's instruction. Forty-eight hours later, the translocation of GFP-PKCδ was evaluated using confocal microscopy.

Confocal fluorescent images were collected with a Bio-Rad MRC 1024 confocal scan head mounted on a Nikon Optishot microscope (Melville, NY) with a ×60 planapochromat lens. Excitation at 488 nm was provided with a 522/32 emission filter for green fluorescence. For live cell imaging, a Bioptechs Focht Chamber System (Bioptechs, Butler, PA) was inverted and attached to the microscope stage with a custom stage adapter.

Cell cultures. The HOP-92 (human lung carcinoma) cell line was obtained from the Biological Testing Branch, NCI, NIH (Frederick, MD). The cells were cultured in RPMI 1640 supplemented with 10% FBS and glutamine (2 mg/mL) at 37°C in a humidified atmosphere containing 5% CO₂.

RNA isolation. Cells were seeded in six-well plates at a 2 × 10⁵ per well. After treatment, medium was removed and the wells were washed twice with PBS. Trizol reagent (1 mL; Molecular Research Center, Cincinnati, OH) was added to each well. After 5 minutes at room temperature, the supernatants were transferred to Eppendorf tubes and 0.2 mL chloroform/mL was added. Following centrifugation at 10,000 rpm for 10 minutes, the colorless upper aqueous phase was transferred to a fresh tube. RNA was precipitated by mixing with 0.5 mL isopropanol. After centrifugation, the RNA was dissolved in DEPC-treated water. To assure the same concentration of RNA in each sample, the absorbance of the RNA was measured at 260 nm, and the samples were diluted with DEPC-treated water to a final concentration of 1 μg/μL.

Reverse transcription-PCR analysis. PKCδ and β-actin mRNA transcripts were detected by reverse transcription-PCR (RT-PCR). Two kinds of duplex RT-PCR were carried out using the selective primers for PKCδ and β-actin. RT-PCR was carried out using the Takara RNA-PCR kit (Takara Mirus Bio, Madison, WI). For the first-strand cDNA synthesis, 1 μg total RNA was incubated at 42°C for 30 minutes in reverse transcriptase buffer. The Takara PCR kit containing PCR buffer, 0.2 mmol/L deoxynucleotide triphosphate mixture, 0.2 μmol/L of each primer, and 2.5 units Taq polymerase was employed and the reaction was carried out under the following conditions in a MJ Research thermal cycler (MJ Research, Waltham, MA): denaturation (95°C for 30 seconds), annealing (56°C for 30 seconds), and extension (68°C for 1 minute) for 20 or 30 cycles. PCR products were analyzed by agarose gel electrophoresis in the presence of ethidium bromide. The products of amplification were visualized with a UV transilluminator.

Western blotting. Levels of cell cycle regulatory proteins and PKC isozymes from bryostatin 1– or PMA-treated HOP-92 cells were determined by Western blot analysis. Confluent cultures of HOP-92 cells on 60-mm tissue culture plates were treated with bryostatin 1 or PMA (0.1, 1, 10, 100, and 1,000 nmol/L) for 24 hours at 37°C. After treatment, cells were harvested and lysed by incubation in lysis buffer [150 mmol/L NaCl, 20 mmol/L Tris-HCl (pH 7.5), 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L Na₃VO₄, 25 mmol/L NaF, 1% aprotinin, 10 μg/mL leupeptin] on ice for 30 minutes followed by sonication. Protein concentrations of supernatants were determined by the Bio-Rad DC Microprotein Assay using bovine serum albumin as a protein standard. Aliquots (10 μg) of proteins were fractionated by SDS-PAGE on 10% polyacrylamide gels and then transferred to nitrocellulose membranes. Nonspecific binding of antibody was prevented by incubating the membranes for 1 hour in blocking buffer (PBS, 5% nonfat milk) followed by incubation with buffer containing the β-actin or anti-PKC isozyme antibodies [PKCα (C-20), PKCβ1 (C-16), PKCδ (C-20), PKCε (C-15), and PKCμ (C-20); 1:1,000 dilution]. The membranes were washed 3 × 5 minutes in PBS-Tween 20 (0.05% Tween 20 in PBS)] and incubated for 1 hour at room temperature with the HRP-conjugated secondary antibodies. After washing the membranes for 3 × 5 minutes in PBS-Tween 20 (0.05%), the immunostaining was visualized by ECL. Films were scanned and quantitation was done using Image J software (developed at the National Institute of Mental Health, NIH).

Cellular growth assay. HOP-92 cells were seeded in six-well plates at a density of 0.9 × 10⁴ per well. Twenty-four hours later, the cells were treated with bryostatin 1 (1, 10, 100, and 1,000 nmol/L) or PMA (1, 10, 100, and 1,000 nmol/L) for 1, 2, 4, and 6 days. The number of the cells was determined with a Coulter counter (Beckman Coulter, Fullerton, CA). Growth assays were done at least four times. In other experiments, as indicated in the figure legends, cell proliferation was quantitated using the CyQuant cell proliferation assay (Invitrogen, Carlsbad, CA).

Fluorescence-activated cell sorting analysis. HOP-92 cells were treated with various doses of bryostatin 1 (1-1,000 nmol/L) and PMA (1-1,000 nmol/L), collected 24 hours after treatment, washed with PBS, and fixed in 70% ethanol overnight at 4°C. The fixed cells were pelleted and washed in PBS; 100 μL RNase was then added and the cells were incubated at 37°C for 30 minutes. Propidium iodide (20 μg/mL) in Dulbecco's PBS was added and incubated for 30 minutes at 4°C in the dark. The DNA content in stained nuclei was measured with a flow cytometer (FACSCalibur, BD Biosciences, San Jose, CA) and the data were analyzed with CellQuest software (Becton Dickinson, Mountain View, CA).

Small interfering RNA transfection. RNAs (21-nucleotide) were synthesized and purified by high-performance liquid chromatography (Qiagen). For design of appropriate small interfering RNA (siRNA) sequences targeting PKCδ, DNA sequences of the type AA (N19) were selected. The siRNA sequences of siRNA-1, siRNA-2, and siRNA-3 corresponded to the coding regions 202 to 220, 355 to 373, and 186 to 204 nucleotides, respectively, from the first nucleotide of the start codon (siRNA-1: AAGCCGACCATGTATCCTGAG, siRNA-2: AATGGCAAGGCTGAGTTCTGG, and siRNA-3: AACACTGGTGCAGAAGAAGCC). The sequences were subjected to a BLAST search against the human genome to ensure that the sequences were not present in any other known mRNA. For annealing of the siRNAs, complementary single-stranded RNAs were incubated in annealing buffer [20 mmol/L Tris-HCl (pH 7.5), 10 mmol/L MgCl₂, 50 mmol/L NaCl] for 1 minute at 90°C followed by 1 hour at 37°C. Transfection of duplex siRNA was done using LipofectAMINE 2000. After transfection for 1, 2, and 3 days, the efficacy of the siRNA treatment for suppressing PKCδ expression was assessed by Western blotting.

2,3-Bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide inner salt cell proliferation assay. Cells grown on 96-well plates for 24 hours were transfected with siRNA, the control pCMV vector, and pCMV-PKCδ. Cell proliferation at 1, 2, and 3 days after transfection was assessed using the 2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-carboxanilide inner salt (XTT) assay according to the manufacturer's protocol (Roche Molecular Biochemicals).

Bryostatin 1 and PMA differentially regulate proliferation of HOP-92 cells. We investigated the effect of various doses of bryostatin 1 or PMA on proliferation of HOP-92 cells, a non–small cell lung cancer (NSCLC) cell line (Fig. 1A). The HOP-92 cells were treated with bryostatin 1 (1-1,000 nmol/L) or PMA (1-1,000 nmol/L) for 24 hours and cellular growth was determined. Bryostatin 1 yielded a biphasic dose-response curve for stimulation of cell growth. Low doses of bryostatin 1 (1 and 10 nmol/L) caused an increase in the number of cells (P < 0.05), whereas high doses of bryostatin 1 (1,000 nmol/L) had no measurable effect. In contrast to treatment with bryostatin 1, treatment with PMA failed to stimulate at 24 hours at any dose.

The effect of bryostatin 1 treatment on growth of HOP-92 cells as a function of time was determined for bryostatin 1 at 10 nmol/L, the dose that gave the maximum proliferative response at 24 hours, and for 10 nmol/L PMA (Fig. 1B). After bryostatin 1 treatment, proliferation was significantly increased compared with control as determined by counting in a Coulter counter (P < 0.001, day 1; P < 0.001, day 2; P < 0.05, day 4; P < 0.01, day 6). Conversely, the modest suppression of cellular growth by PMA at 24 hours was much more evident at 2 to 6 days (P < 0.005, day 2; P < 0.04, day 4; P < 0.0001, day 6). Qualitatively similar results were observed when proliferation was measured using the CyQuant proliferation assay (for stimulation of growth by bryostatin 1: P < 0.05, day 1; P < 0.0001, day 2; P < 0.001, day 4; P < 0.0001, day 6).

The inhibition of cell growth by PMA is reminiscent of the effect of PMA on other human NSCLC cell lines, where it is associated with induction of differentiation (32). Furthermore, the known sensitivity of HOP-92 cells to growth inhibition by phorbol esters and related compounds (33) provided an initial motivation for the present study of this cell line.

The effect of bryostatin 1 and PMA treatment on cell cycle variables was determined by flow cytometry. HOP-92 cells were treated with bryostatin 1 (1-1,000 nmol/L) or PMA (1-1,000 nmol/L), and cell cycle variables were determined after 24, 48, and 96 hours of treatment (Fig. 1C). Bryostatin 1 caused a biphasic reduction in the proportion of cells in G₀-G₁ phase and an increase in the proportion of cells in S phase. The maximal decrease in G₀-G₁ phase and increase in S phase measured at 24 hours was achieved at 10 nmol/L bryostatin 1, with G₀-G₁ decreasing from 48.7% to 28.8% (P < 0.01) and S phase increasing from 37.5% to 57.8% (P < 0.05) compared with untreated cells. Conversely, PMA treatment for 24 hours decreased the S phase and increased the G₀-G₁ phase in a dose-dependent manner. At 48 and 96 hours, the relative proliferative effect of bryostatin 1 was somewhat diminished, as also seen in the growth curves, and the inhibitory effect of PMA remained evident.

Differential down-regulation of PKC isozymes by bryostatin 1 and PMA. We compared the dose-response curves for down-regulation by bryostatin 1 and PMA of the PKC isoforms present in HOP-92 to determine if any of these isoforms showed differential response to these two ligands. Confluent cultures of HOP-92 cells were treated for 24 hours with increasing concentrations of bryostatin 1 (0.1-1,000 nmol/L) or PMA (0.1-1,000 nmol/L) and the levels of the various PKC isoforms (α, β1, ϵ, μ, and δ) were determined by Western blotting (Fig. 2A). PMA and bryostatin 1 both down-regulated the classic PKCs (PKCα and PKCβ1), with bryostatin 1 being ∼10-fold more potent than PMA in each case. PKCϵ was down-regulated only partially by both ligands. Again, bryostatin 1 was somewhat more potent. PKCμ showed little change.

Unlike the other PKC isoforms, PKCδ showed a very marked difference in response to bryostatin 1 and to PMA. Whereas PMA at any concentration (0.1-1,000 nmol/L) had little effect on PKCδ, bryostatin 1 potently caused down-regulation of PKCδ and did so in a biphasic fashion. Low doses (0.1-10 nmol/L) of bryostatin 1 caused marked loss of PKCδ, whereas the extent of down-regulation was reduced at higher concentrations of bryostatin 1 (>100 nmol/L). Because this pattern of response of PKCδ to bryostatin 1 treatment mirrored that for induction of proliferation of the HOP-92 cells by bryostatin 1, assuming that PKCδ inhibited proliferation, we investigated the effect of bryostatin 1 on PKCδ in further detail.

Dose- and time-dependent effects of bryostatin 1 and PMA on PKCδ expression. The biphasic pattern of down-regulation of PKCδ by bryostatin 1 was confirmed in additional experiments and quantitated by densitometry (Fig. 2B). Maximal down-regulation was achieved at 1 to 10 nmol/L bryostatin 1. No effect was observed at 1 to 10 pmol/L bryostatin 1. Likewise, at 1,000 nmol/L bryostatin 1, little down-regulation was seen. When different concentrations of bryostatin 1 (0.001 nmol/L-1 μmol/L) were coapplied with 100 nmol/L PMA, the suppression of PKCδ by lower bryostatin 1 concentrations (0.1-10 nmol/L) was blocked. Higher concentrations of bryostatin 1 (10-100 nmol/L) still caused some PKCδ suppression, although not to the level observed in the absence of PMA.

We determined the time course for the down-regulation of PKCδ in response to 10 nmol/L bryostatin 1 or 10 nmol/L PMA in the HOP-92 cells (Fig. 2C). The level of PKCδ had decreased by 8 hours of treatment with 10 nmol/L bryostatin 1 and was markedly reduced by 16 and 24 hours. PMA had little effect on PKCδ at any time point tested. To determine whether the reduction in the level of PKCδ protein was a direct effect of the bryostatin 1 treatment or whether it was a consequence of an effect of bryostatin 1 on PKCδ mRNA expression, the HOP-92 cells were treated with bryostatin 1 (10 nmol/L) or PMA (10 nmol/L), total RNA was extracted, and the PKCδ mRNA expression was analyzed by RT-PCR analysis with PKCδ-specific primers. The level of expression of the PKCδ mRNA was not changed by either treatment (Fig. 2D, suggesting that the reduction in PKCδ protein reflected events at the post-transcriptional level.

If the enhanced proliferation of the HOP-92 cells on treatment with 10 nmol/L bryostatin 1 reflected the down-regulation of PKCδ, then the ability of 100 nmol/L PMA to largely block this down-regulation of PKCδ should be reflected in suppression of the bryostatin 1–induced proliferation. This was indeed the case (Fig. 2E).

Bryostatin 1 and PMA induced translocation of PKCδ. Translocation of PKC in response to ligand binding provides another measure of ligand response. We therefore examined the ability of bryostatin 1 (10 nmol/L) and PMA (10 nmol/L) to induce translocation of PKCδ in the HOP-92 cells. Treatment with 10 nmol/L bryostatin 1 caused a rapid shift of PKCδ from the soluble fraction to the Triton X-100 soluble particulate fraction. Most of the PKCδ was lost from the soluble fraction within 10 minutes of treatment and the translocation was complete by 4 hours after treatment (Fig. 3A). Although 10 nmol/L PMA did not induce the down-regulation of PKCδ, PKCδ rapidly translocated from the soluble fraction to the Triton X-100 soluble particulate fraction in response to PMA. Partial translocation was evident at 10 minutes, but some of the PKCδ remained in the soluble fraction over the duration of the experiment.

To explore the possibility that a different pattern of translocation of the PKCδ by PMA and bryostatin 1 might determine the difference in down-regulation, GFP-tagged PKCδ was transiently transfected into HOP-92 cells, and its translocation was monitored in real-time using confocal microscopy (Fig. 3B). In the absence of stimuli, the GFP-PKCδ was present in the cytoplasm. When 10 nmol/L PMA was added, GFP-PKCδ slowly translocated to the plasma and nuclear membrane. With 10 nmol/L bryostatin 1 treatment, GFP-PKCδ also translocated to the cytoplasm and nuclear membrane by 20 minutes.

Overexpression of PKCδ-induced inhibition of cellular proliferation. To explore a possible mechanistic link between the effects of bryostatin 1 on PKCδ and on cellular proliferation in the HOP-92 cells, we examined the effect of overexpression of PKCδ. Cells were transfected with control pCMV vector or with pCMV-PKCδ for 48 hours, and cell lysates were then analyzed by Western blotting for the level of PKCδ expression (Fig. 4A). A ∼5-fold increase in PKCδ expression was detected in the PKCδ-transfected cells, whereas the vector alone had no effect. To determine the involvement of PKCδ in the regulation of cell proliferation in the HOP-92 cells, the cells were transfected with increasing amounts of pCMV-PKCδ or of the control vector and the effects on the cellular proliferation were determined by the XTT assay. Cells overexpressing PKCδ exhibited in a dose-dependent manner a lower rate of cell proliferation compared with control cells (Fig. 4B), whereas the cells transfected with the control vector showed no significant change compared with the control. We next examined the effect of bryostatin 1 treatment (Fig. 4C). In untransfected cells or in cells transfected with the empty vector, cellular growth was increased by bryostatin 1 (10 nmol/L) treatment as expected (P < 0.001 and 0.001, respectively). The minor apparent decrease in proliferation in the empty vector transfected cells compared with the controls was not statistically significant (P = 0.27). Overexpression of PKCδ decreased proliferation (P < 0.001) as likewise expected. In the PKCδ overexpressing cells, however, bryostatin 1 was unable to stimulate proliferation (P = 0.45; Fig. 4C). Consistent with the lack of effect of the bryostatin 1 treatment on proliferation of the HOP-92 cells overexpressing PKCδ, bryostatin 1 was likewise unable to induce down-regulation of PKCδ in the overexpressing cells (Fig. 4D). This result argues that the degradation pathway for PKCδ in response to bryostatin 1 in the HOP-92 cells is of low capacity and subject to saturation. Similar behavior in which overexpression of a PKC isoform renders it insensitive to down-regulation by bryostatin 1 was described for PKCα in LNCaP cells (34).

PKCδ knockdown stimulates cellular proliferation. The inhibitory role of PKCδ in cell proliferation of the HOP-92 cells was further addressed using a PKCδ siRNA strategy to reduce the expression of the endogenous PKCδ in the HOP-92 cells. Three different sites of PKCδ were selected for design of siRNA constructs and transfected into the HOP-92 cells. Forty-eight hours after transfection, cell lysates were prepared and the expression of PKCδ was determined by Western blot analysis using anti-PKCδ antibody. Among the three siRNA constructs, siRNA-3 produced strong suppression of PKCδ expression (Fig. 5A). We therefore selected siRNA-3 for further analysis. HOP-92 cells were seeded onto 60-mm plates. Twenty-four hours later, the cells were transfected with PKCδ siRNA-3 at three different doses (1, 2, and 4 μg) and the expression of PKCδ was determined by Western blotting 1, 2, and 4 days after transfection. Control siRNA (1, 2, and 4 μg) had no effect on PKCδ expression (Fig. 5B). Partial suppression of PKCδ by the siRNA was apparent by 1 day and almost complete suppression was observed with 4 μg siRNA at 3 days (Fig. 5B). To check the selectivity of the PKCδ siRNA, Western blotting was done with anti-PKCα antibody. The level of PKCα was not changed. To determine the effect of suppression of PKCδ expression by siRNA on HOP-92 cell proliferation, the cells were treated with either the PKCδ siRNA or the control siRNA and cellular growth was measured 1 and 3 days later by the XTT assay. Control siRNA did not affect cellular proliferation. PKCδ siRNA significantly increased the cellular growth compared with the control (P < 0.01, at both 1 and 3 days; Fig. 5C), and the magnitude of the stimulation was similar to that observed on treatment with 10 nmol/L bryostatin. Although slightly less stimulation was seen at 3 days of treatment with the PKCδ siRNA than with 1 day of treatment, this difference was not statistically significant (P = 0.4). Because the extent of knockdown of the PKCδ by the siRNA treatment was more pronounced at 3 days than at 1 day, a greater effect at this later time might have been expected.

We conducted limited experiments to determine the proteolytic pathway responsible for the loss of PKCδ in response to treatment with 10 nmol/L bryostatin. Inhibition of the calpain pathway either with the calcium chelator BAPTA (Fig. 6A) or with the calpain inhibitors calpeptin (Fig. 6B) or E64d (Fig. 6C) caused substantial protection of PKCδ, with less effect on PKCα. The proteasome inhibitor lactacystin, conversely, protected PKCα but not PKCδ (Fig. 6D), although some other proteasome inhibitors gave variable protection of PKCδ (data not shown). The caspase-1 and caspase-3 inhibitors under these conditions protected neither PKCδ nor PKCα (Fig. 6E). The ability of treatment with calpeptin or BAPTA to inhibit PKCδ down-regulation by bryostatin 1 predicted that these treatments would prevent the enhanced proliferation induced by bryostatin 1 treatment. This prediction was confirmed, although these inhibitors by themselves did not inhibit proliferation of the control cells (Fig. 6F).

We conclude that bryostatin 1 differentially affects cell proliferation of HOP-92 cells compared with the typical phorbol ester PMA. Our results indicate that the selective loss of PKCδ induced by bryostatin 1 treatment is sufficient to explain this differential behavior. The calpain degradation pathway is implicated in this differential loss of PKCδ.

Bryostatin 1 has attracted considerable attention as a potential therapeutic agent for cancer. Its high potency for PKC and the other signaling proteins containing a phorbol ester/DAG–responsive C1 domain identifies its general site of action (35). Extensive studies have documented the critical role of PKC in cellular signaling by growth factors and other receptors (36, 37) and the rationale for PKC as a therapeutic target in cancer (38, 39). Among high-affinity ligands for PKC, bryostatin 1 is unique in its ability to functionally antagonize many typical phorbol ester responses while activating PKC (21, 24). Critical among the differences in biological response induced by bryostatin 1 and the phorbol esters, such as PMA, is that PMA is a potent tumor promoter in the mouse skin system, whereas bryostatin 1 is not (40).

We were interested in the response of HOP-92 cells because these are one of the more responsive cell lines to PKC activators in the NCI-60 cell line compound screen as indicated by COMPARE (41). Previously, we and others have described that bryostatin 1 can give a biphasic pattern of down-regulation of PKCδ, with reduced down-regulation at higher bryostatin 1 concentrations, whereas PMA causes monophasic down-regulation. This pattern of behavior has been seen in NIH3T3 cells (27), primary mouse keratinocytes (22), B16/F10 melanocytes (28), HeLa cells (29), etc. The HOP-92 cell line characterized here revealed a different and previously undescribed pattern of behavior. In this system, whereas bryostatin 1 caused a biphasic pattern of down-regulation of PKCδ, PMA treatment failed to cause down-regulation at any concentration.

We were able to link this unique pattern of PKCδ down-regulation by bryostatin 1 in the HOP-92 cells to the effect on cell growth of bryostatin 1 in this system. The down-regulation of PKCδ was associated with enhanced proliferation/resistance to inhibition of proliferation in these cells. The biphasic dose-response curve for down-regulation was mirrored in the dose-response curve for proliferation. Overexpression of PKCδ reduced the proliferation in untreated HOP-92 cells. The overexpression further prevented the down-regulation of PKCδ by bryostatin 1 and blocked the enhanced proliferation in response to bryostatin 1. Likewise, the calpain inhibitors calpeptin and BAPTA blocked the PKCδ down-regulation by bryostatin 1 and blocked its induction of proliferation. Finally, reduction in the level of PKCδ through treatment with siRNA enhanced the proliferation of the HOP-92 cells.

Inhibition of proliferation by PKCδ is a generally typical response and has been described on PKCδ overexpression in Chinese hamster ovary cells (42), NIH3T3 cells, vascular smooth muscle cells, human glioma cells, and capillary endothelial cells (5, 43–45). In vascular smooth muscle, PKCδ inhibited cell proliferation by suppressing G₁ cyclin expression (43); in capillary endothelial cells, PKCδ inhibited the S-phase transition (46). PKCδ has likewise been found to be proapoptotic in many systems (47). On the other hand, it is also clear that PKCδ can be proliferative in some cell types (48–50) and is antiapoptotic in some systems [e.g., A172 glioma cells treated with TRAIL (8), C6 glioma cells infected with Sindbis virus (7), or non–small lung cancer lines A549, H1703, H157, H1355, and H1155 (51)].

Given the diversity of roles of PKC isoforms, an issue for the clinical application of bryostatin 1 has been to find those settings for which there is a solid rationale. Its ability to down-regulate PKCδ and to affect cell response in a fashion consistent with this down-regulation of PKCδ predicts that it might be useful in those systems in which PKCδ plays an antiapoptotic role. The factors determining whether PKCδ is proapoptotic or antiapoptotic are not understood, but it is clear that phosphorylation of PKCδ on specific tyrosine residues plays a critical role in its functioning (47, 52–54).

As yet, relatively few isoform selective ligands for PKC isoforms are available. LY333531 is in clinical trials as a selective enzymatic inhibitor of PKCβ (39). ISIS3521 is an antisense oligonucleotide in clinical trials for suppressing the expression of PKCα (55). We have recently reported the identification of a lead structure with selectivity for the Ras-GRP family of related receptors (56). Our results here fit with the concept that one action of bryostatin 1, at least in appropriate cell types, may be as a functional PKCδ selective inhibitor, working through enhanced PKCδ degradation. Of course, bryostatin 1 simultaneously will be having effects on other PKC isoforms, the significance of which will depend on the cell type, but systems in which PKCδ plays a role are candidates for the therapeutic application of bryostatin 1.

Grant support: Intramural Research Program of the Center for Cancer Research, NIH, NCI.

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.