Drugs that target protein kinase C (PKC) are now being evaluated in patients with non-small cell lung cancer (NSCLC), but the role of PKC in NSCLC cells remains unclear. We report here that NSCLC cell lines show enhanced phosphorylation and altered expression of specific PKC isoforms compared with normal lung epithelial cells. PKC inhibition variably increased apoptosis, with rottlerin, a PKCδ inhibitor, being most effective and potentiating chemotherapy-induced apoptosis, especially with trastuzumab. Consistent with PKCδ being anti-apoptotic in NSCLC cells, transient transfection of a kinase-dead mutant of PKCδ increased apoptosis and potentiated chemotherapy-induced apoptosis. Our studies provide a rationale for targeting PKC isoforms in NSCLC cells, especially PKCδ.

The inadequacy of standard chemotherapy in advanced NSCLC2 has driven the development of novel, targeted therapeutic approaches such as those directed against the epidermal growth factor receptor. Recently, our laboratory identified two signaling pathways downstream of growth factor receptors such as epidermal growth factor receptor that are constitutively active in the majority of NSCLC cell lines and promote cellular survival and resistance to chemotherapy and radiation. These pathways are the PI3K/Akt pathway and the MEK/ERK pathway (1, 2). Although we described in a subset of NSCLC cell lines an inverse relationship between these pathways that was possibly related to c-Raf phosphorylation (2), regulatory mechanisms between these pathways are likely to be complex and are likely to involve other kinases.

In the present report, we characterize PKC isoform expression and function in NSCLC cells. One motivation was the known role of PKC isoforms in regulation of the PI3K/Akt and MEK/ERK pathways. PKC isoforms such as PKCα, β, δ, and ζ may be activated in concert with Akt because PDK1, which phosphorylates and activates Akt, also phosphorylates and activates PKCα, β, δ, and ζ (3). Once activated, PKC isoforms may regulate the PI3K/Akt pathway positively (4) or negatively (5). A PKC isoform of particular interest with regard to Akt is PKCδ. PKCδ is proapoptotic in most systems (reviewed in Ref. 6). PKCδ is activated by both PDK1 and mammalian target of rayamycin (mTOR) (7), a downstream substrate of Akt. PKCδ and Akt are activated by cellular stress (8) as well as by the products of PI3K, phosphatidylinositol 4,5-bisphosphate, and phosphatidylinositol 3,4,5-trisphosphate (9). Akt can physically associate with the regulatory domain of PKCδ (10), and Akt may phosphorylate PKCδ under certain conditions (11). PKCδ can also mediate activation of the MEK/ERK pathway caused by phorbol esters or growth factors.

A second motivation for analysis of PKC isoforms in NSCLC cells is that therapeutic approaches targeting PKC are being evaluated in clinical trials. These trials are investigating drugs such as bryostatin or ISIS 3521, that through discovery or design inhibit PKC (12, 13), as well as drugs such as staurosporin or UCN-01 that incidentally inhibit PKC. Thus far, however, little is known about PKC isoforms in vitro or in NSCLC tumor specimens to rationally support studies in patients with NSCLC.

In the present studies, we demonstrate that at least six PKC isoforms are expressed in a panel of NSCLC cell lines that we previously characterized for Akt and ERK activation. Compared with normal lung epithelial cells, NSCLC cells exhibit selective expression of certain PKC isoforms in an active state and exhibit selective sensitivity to PKC inhibitors. A PKCδ inhibitor, rottlerin, was most effective at increasing apoptosis and dramatically sensitized NSCLC cells to the effects of trastuzumab, as well as other chemotherapeutic agents. These effects were recapitulated when NSCLC cells were transfected with a PKCδ kinase-dead mutant but not with wild-type PKCδ. Our biochemical and genetic data suggest that PKCδ promotes cellular survival and chemotherapeutic resistance in NSCLC cells, in contrast to its proapoptotic role in many systems.

Materials

NSCLC cells were obtained and grown as described previously (1). NHBE cells and materials were purchased from BioWhittaker Inc. (Walkersville, MD) and grown according to the manufacturer’s instructions. BEAS-2B cells were a generous gift from Dr. C. Harris (Bethesda, MD) and grown as described previously (14). LY294002 was obtained from Alexis Biochemicals (San Diego, CA). Protease inhibitor mixture, cisplatin, and α-tubulin antibodies were from Sigma Chemical Co. (St. Louis, MO). Protein assay kits were purchased from Bio Rad (Hercules, CA). Protran pure nitrocellulose membranes were from Schleicher & Schuell (Keene, NH). Rottlerin, calphostin C, chelerythrine chloride, and paclitaxel were purchased from Calbiochem (La Jolla, CA). Trastuzumab was from Genentech (San Francisco, CA). Gemcitabine was obtained from Eli Lilly (Indianapolis, IN). Docetaxel was from Aventis Pharmaceuticals Products Inc. (Collegeville, PA). All native PKC isoform antibodies were purchased from Santa Cruz Biotechnologies (Santa Cruz, CA). All phospho-specific antibodies, horseradish peroxidase-conjugated secondary antibodies, and the PARP antibodies were purchased from Cell Signaling (Beverly, MA). P-PKCβII (S660) and P-PKCα/βII (T638/T641) antibodies are directed against autophosphorylation sites that are necessary for PKC activity. P-PKCζ/λ (T410/403), P-PKCδ (T505), and P-PDK-1 (S241) antibodies are directed against residues in the activation loop that are necessary for PKCζ, PKCδ, and PDK-1 activity, respectively. PKCζ (T410) and P-PKCδ (T505) are sites for PDK-1 phosphorylation. Antibodies against GFP were purchased from Zymed Laboratories, Inc. (South San Francisco, CA). Construction of wild-type PKCδ-GFP and K376R (kinase-dead) PKCδ-GFP was described previously.

Methods

Pharmacological Treatments.

To determine the effects of LY294002 on PKC phosphorylation, cells were plated at 9 × 105 to 1.5 × 106 cells/25-cm2 flask to achieve ∼70% confluence. After attachment, medium was changed to DMEM with 0.1% FBS overnight. LY294002 (25 μm) was added 2 h before harvesting. To study the effect of PKC inhibitors on apoptosis, cells were plated at 1 × 105 or 1.5 × 105 in 24- or 12-well plates, respectively, and allowed to attach. Medium was changed to DMEM with 0.1% FBS with or without rottlerin (10 μm), Gö6976 (100 nm), calphostin C (1 μm), or chelerythrine chloride (10 μm). Cells were incubated for 48 h and harvested for measurement of apoptosis. To assess the effects of rottlerin on chemotherapy-induced apoptosis, rottlerin (10 μm) was added simultaneously to the indicated agents in DMEM with 0.1% FBS for 48 h.

Immunoblotting.

To examine PKC isoform expression in NSCLC cells and NHBE and BEAS2B cells, confluent 75-cm2 flasks were utilized. Preparation of cell extracts, protein assays, and immunoblotting were performed as described previously (1). Equivalent loading was confirmed by staining membranes with fast green as described previously. All PKC isoform and phospho-specific PKC antibodies were used at 1:1,000 dilutions. Cleaved PARP antibody was used at 1:1,500, and GFP monoclonal antibody was used at 1:1,000. Secondary antibodies were used at 1:2,000 to 1:10,000 dilutions. All immunoblotting experiments were repeated at least three times.

Flow Cytometry.

Floating cells were collected, and adherent cells were harvested by trypsinization and centrifuged for 5 min. Cells were fixed in ice-cold 70% methanol added dropwise and then incubated at −20°C for 30 min. Cells were centrifuged and incubated with propidium iodide (25 μg/ml) supplemented with RNase A (30 mg/ml) for 30 min at room temperature. Quantification of sub-2N DNA was determined by flow cytometry analysis using a Becton-Dickinson FACSort and by manual gating using CellQuest software. Gating was performed on blinded triplicate samples.

Transient Transfections.

Cells were plated at 3 × 105 cells/well in 6-well dishes (approximately 70% confluence) and transfected in triplicate with plasmids encoding wtPKCδ or K376R PKCδ using the Superfect Transfection Reagent protocol (Qiagen, Valencia, CA). Cells were allowed to recover from transfection for 24 h, and then medium was changed to DMEM with 0.1% FBS or 10% FBS ± chemotherapeutic agents as indicated for 48 h. Whole cell lysates from ungated samples were harvested for immunoblotting to assess GFP levels and PARP cleavage. Samples were also fixed, stained with propidium iodide, and prepared for flow cytometry. Apoptosis was quantified by gating on GFP-positive cells and performing cell cycle analysis. Experiments were repeated at least three times.

Statistical Analysis.

Statistical comparison of mean values was performed using Student’s t test. All Ps are two tailed.

PKC family members are grouped based on structural similarities that determine dependence on activation by Ca2+ and DAG (classical isoforms), DAG alone (novel isoforms), or neither (atypical isoforms; reviewed in Ref. 15). To determine the extent of PKC isoform expression in NSCLC cells, we performed immunoblotting with isoform-specific antibodies. Multiple PKC isoforms within each PKC family were expressed in each of five NSCLC cell lines previously analyzed for Akt and ERK activity, including α and β in the classical family, δ and ε in the novel family, and μ and ζ in the atypical family (Fig. 1 A). PKCλ and η were not examined in these studies. PKCγ was barely detectable in all cell lines. H1155 cells, which have the highest levels of Akt activity, had the lowest expression of PKCα, β, and ε PKCδ levels were lowest in the H157 cells, and the Mr 40,000 catalytic fragment (CF) of PKCδ was only detected in A549 and H1355 cells, which have the highest levels of ERK activity.

Because multiple PKC isoforms were expressed in each NSCLC cell line and because PKC activation can contribute to cellular transformation, we compared PKC isoform expression and phosphorylation in primary human lung epithelial cells (NHBE), NHBE cells immortalized with SV40 large T antigen (BEAS-2B; Ref. 14), and a NSCLC cell line (H1355). Compared with NHBE, H1355 cells had increased expression of PKCα, β, ε, and μ (Fig. 1 B, left panels). PKCδ expression was decreased in H1355 cells, but expression of the PKCδ CF was increased. Immortalized BEAS-2B cells expressed levels of PKCα, β, ε, and μ comparable with those of H1355 cells but had levels of PKCδ comparable with NHBE. Levels of PKCζ did not differ among these three cell types. Although these three cell types are not isogenic, these studies suggest that transformation of lung epithelial cells may be accompanied by changes in PKC isoform expression. Because the levels of PKCα, β, ε, and μ were increased in H1355 cells compared with NHBE, but were similar among the five NSCLC cell lines, increased PKC isoform expression may be a general feature of NSCLC cells.

To assess whether altered protein expression of PKC isoforms in H1355 cells was associated with increased phosphorylation of PKC isoforms, we performed immunoblotting with phospho-specific antibodies that recognize PKC isoforms in their active states (Fig. 1 B, right panels). Phosphorylation of PKCα, β, δ, and ζ was greatest in H1355, intermediate in BEAS-2B, and low or absent in NHBE. Increased PKCα and β phosphorylation mirrored increased protein expression, but increased PKCδ and ζ phosphorylation occurred independently of increased protein expression. Notably, increased PKCδ phosphorylation was observed despite a decrease in protein level. The fact that the highest levels of PKC isoform phosphorylation were observed with H1355 cells prompted us to evaluate PKC phosphorylation in the other NSCLC cell lines.

To determine whether PKC isoforms were maintained in an active state with serum starvation and to determine whether PKC phosphorylation was PI3K dependent (because PKCα, β, δ, and ζ are all substrates of the PI3K-dependent kinase, PDK-1), NSCLC cells were serum starved in the absence or presence of a PI3K inhibitor, and phosphorylation of PKCα, β, δ, and ζ; PDK-1; and Akt was measured. Fig. 1 C shows that in NSCLC cell lines with high levels of Akt activity, Akt phosphorylation was maintained with serum deprivation but was inhibited by a PI3K inhibitor, LY294002. PDK-1 phosphorylation was also maintained with serum deprivation, but it was not affected by LY294002. Likewise, phosphorylation of PKCα, β, ζ, and δ at the PDK-1 sites was maintained under conditions of serum deprivation, but no inhibition was observed with LY294002. In fact, phosphorylation of PKCα, β, and ζ was slightly increased in the H157 cells by LY294002 treatment. Together, these data showed that Akt, PDK-1, and multiple PKC isoforms were maintained in an active state when NSCLC cells were serum deprived but that only Akt phosphorylation was PI3K dependent.

Maintenance of four PKC isoforms in active states under conditions of serum starvation suggested that these PKC isoforms might contribute to NSCLC cell survival. To test this hypothesis, we treated NSCLC cell lines with four widely used PKC inhibitors (16): (a) rottlerin, a selective PKCδ inhibitor; (b) calphostin C, which inhibits binding of DAG to both classical and novel isoforms; (c) chelerythrine, which also inhibits classical and novel isoforms; and (d) Gö6976, which selectively inhibits classical isoforms. As shown in Fig. 2, each NSCLC cell line responded to at least one PKC inhibitor with at least a 6-fold induction of apoptosis. Apoptosis was confirmed by assessing morphological changes in the NSCLC cells and by assessing the binding of annexin V to apoptotic cells via flow cytometry (data not shown). The NSCLC cell lines differed slightly in sensitivity to the PKC inhibitors, but rottlerin, in particular, increased apoptosis in all five NSCLC cell lines and induced the most apoptosis in four of five NSCLC cell lines. Chelerythrine also increased apoptosis in all five NSCLC cell lines, and calphostin C increased apoptosis in all but the H157 cells. Gö6976 was ineffective, even at higher doses (data not shown). The induction of apoptosis by chelerythrine and calphostin C, but not by Gö6976, suggested that a novel (but not classical) PKC isoform might be responsible for promoting NSCLC cell survival. Moreover, the induction of NSCLC cell apoptosis by rottlerin suggested that PKCδ might be the novel isoform promoting NSCLC cell survival. Genetic evidence supporting this conclusion will be presented below.

Because of the robust apoptotic response induced by rottlerin and the fact that other modes of PKC inhibition are being combined with standard chemotherapy in ongoing clinical trials, we combined rottlerin with chemotherapeutic agents used in the treatment of NSCLC and measured apoptosis. In H1703 cells, combining rottlerin with gemcitabine, cisplatin, docetaxel, or trastuzumab resulted in greater-than-additive induction of apoptosis (Fig. 3,A). Similar results were observed when gemcitabine, docetaxel, or trastuzumab was combined with rottlerin in the H1155 cells (Fig. 3,B). In both cell lines, combining rottlerin with paclitaxel was ineffective, but combining rottlerin with a different taxane, docetaxel, resulted in greater-than-additive increases in apoptosis. Rottlerin alone was quite effective in H1355 (Fig. 3,C) and H157 (Fig. 3 D) cells, but combining chemotherapeutic agents with rottlerin in these cells did not increase apoptosis, with the exception of trastuzumab, where greater than additive effects were seen. In H1355 cells, combining cisplatin and rottlerin or docetaxel and rottlerin was antagonistic to apoptosis caused by rottlerin alone. This may be due to cell cycle conflict because both cisplatin and docetaxel alone induced a G2-M-phase arrest, but in combination with rottlerin, H1355 cells underwent a G0-G1 arrest (data not shown).

To confirm the proapoptotic effects of rottlerin and the putative role of PKCδ, we transiently transfected three NSCLC cell lines with plasmids encoding GFP-tagged wild-type PKCδ or a GFP-tagged PKCδ kinase-dead mutant (K376R). After recovery from transfection, cells were serum starved in the absence or presence of chemotherapy, and apoptosis was assessed in the GFP-positive cells using flow cytometry. (In these experiments, basal levels of apoptosis were higher than those in Fig. 3 because of the nonspecific toxic effects of the transfection reagent.) In H1355 cells transfected with wild-type PKCδ, trastuzumab did not increase apoptosis, as assessed by either flow cytometry or PARP cleavage (Fig. 4,A). In contrast, transfecting H1355 cells with K376R decreased PKCδ phosphorylation, doubled serum starvation-induced apoptosis, and increased trastuzumab-induced apoptosis. To address whether the apoptosis resulting from transfection of K376R was an artifact due to cellular fragility caused by serum starvation and exposure to the transfection reagent, we repeated transfection of H1355 cells with wild-type PKCδ or K376R and kept the serum concentration at 10% throughout the experiment. We chose to test cisplatin with wild-type PKCδ or K376R in these experiments because rottlerin did not potentiate trastuzumab-induced apoptosis in 10% FBS (see below). Consistent with the data obtained under conditions of 0.1% FBS, transfecting K376R decreased PKCδ phosphorylation and increased apoptosis with and without chemotherapy (Fig. 4 B).

Similar results were observed when we transfected H1155 cells or H157 cells, even when we chose chemotherapeutic agents that were not effective in combinations with rottlerin. In H1155 cells (Fig. 4,C), cisplatin slightly increased apoptosis in cells transfected with wild-type PKCδ, but transfection of K376R decreased PKCδ phosphorylation and increased apoptosis caused by serum starvation or serum starvation and cisplatin. In H157 cells (Fig. 4 D), paclitaxel did not significantly increase apoptosis in cells transfected with wild-type PKCδ, but in cells transfected with K376R, decreased PKCδ phosphorylation and increased basal and paclitaxel-induced apoptosis were observed. Consistent with these data, we also observed a pro-apoptotic function of K376R when we attempted to generate NSCLC cells stably transfected with wild-type PKCδ or K376R, and we found that we were unable to expand any clones expressing K376R (data not shown). Together, these results show that pharmacological or genetic inhibition of PKCδ increases NSCLC apoptosis caused by serum starvation and administration of chemotherapy, and suggest that PKCδ promotes NSCLC cell survival.

Our study is the first to evaluate PKC isoform expression in cell lines derived from NSCLC, the most lethal human cancer. Despite differences in status of p53, K-ras, Rb, tumor histology, Akt activity, and ERK activity (1, 2), a panel of five NSCLC cell lines expressed similar levels of at least six isoforms of PKC (α, β, δ, ε, μ, and ζ). PKCγ was not expressed in any of these NSCLC cell lines, which is consistent with a previously noted lack of PKCγ expression in lung tissue (17). PKCλ and η were not examined. Multiple PKC isoforms have also been described in other tumor cell systems derived from small cell lung cancer (18), breast (19), colon (20, 21), and brain (22) tumors. Differences in PKC isoform expression and phosphorylation were observed when we compared normal human lung epithelial cells (NHBE), immortalized but nontumorigenic human lung epithelial cells (BEAS-2B), and a NSCLC cell line (H1355). When compared with NHBE, H1355 cells expressed higher levels of PKCα, β, ε, and μ and the CF of PKCδ. BEAS-2B also expressed increased levels of PKCα, β, ε, and μ, suggesting that immortalization with SV40 large T antigen is associated with increased expression of these isoforms and that increased PKC expression might be associated with lung tumorigenesis. Of note, Webb et al. (17) have observed expression patterns of PKC isoform expression in normal lung tissues that are closer to what we observed in H1355 cells rather than NHBE. Unfortunately, the tissues used by Webb et al. (17) were either denuded of epithelium (trachea) or a mixture of multiple cell types (lung), so direct comparison is not possible. Similar data correlating increased PKC isoform expression with increased tumor grade or degree of transformation have been observed by Mandil et al. (22), who showed increased levels of PKCα and ε expression in high-grade brain tumors (glioblastomas) compared with low-grade tumors (astrocytomas).

In addition to increased protein expression, we observed progressively increased phosphorylation of PKCα, β, δ, and ζ in NHBE, BEAS-2B, and H1355 cells, respectively. The phosphorylation sites we assessed on PKCδ and ζ are sites for PDK-1 phosphorylation, and the sites we assessed on PKCα and β become phosphorylated only after phosphorylation by PDK-1. Surprisingly, under conditions where we hypothesized phosphorylation would be decreased (namely, serum starvation ± the addition of a PI3K inhibitor), maintenance of phosphorylation was observed in all cell lines. LY294002, which has previously been shown to decrease phosphorylation of PDK-1-dependent PKC isoforms in many studies, did not decrease phosphorylation in our studies. Although one report has shown that PKCα and β can be phosphorylated by PDK-1 in a PI3K-independent manner in mouse or monkey cells (23), this is the first report in any tumor cell system to show that LY294002 decreases Akt phosphorylation without affecting PDK-1 or PDK-1-dependent PKC isoform phosphorylation. Our studies suggest that PDK-1 and PKC function may be PI3K independent in these NSCLC cells, which is a subject of current investigation. Moreover, they raise the possibility that increased phosphorylation of PKC isoforms may accompany increased PKC expression in the process of lung tumorigenesis.

Our observation that PKC inhibitors increased NSCLC cell apoptosis suggests that PKC isoforms contribute to NSCLC survival and support the concept of targeting PKC in NSCLC. Currently, the only isoform-specific approach against PKC in lung cancer clinical trials is ISIS 3521, which is an antisense approach against PKCα. ISIS 3521 has not been reported to decrease PKCα levels in patients’ tissues, which raises the possibility that its mechanism of action may be unexpected. In our studies, the classical PKC inhibitor, Gö6976, which has been demonstrated to inhibit PKCα in other studies, was ineffective compared with calphostin C, chelerythrine, or rottlerin. Rottlerin increased apoptosis at least 4-fold in all NSCLC cell lines and was the most effective PKC inhibitor tested in four of five NSCLC cell lines. Rottlerin potentiated chemotherapy-induced apoptosis in a cell line- and drug-specific manner, although antagonism was observed in H1355 cells when rottlerin was combined with cisplatin or docetaxel. The most consistent potentiation occurred with trastuzumab, where quantitatively similar, greater-than-additive effects were observed in all NSCLC cell lines tested. Because H1703 cells have much higher levels of erbB2 protein expression compared with the other NSCLC cell lines (data not shown), potentiation of trastuzumab-induced apoptosis by rottlerin does not appear to correlate with erbB2 expression, and the mechanism responsible for this potentiation is unclear.

Interestingly, the potentiation of chemotherapy-induced apoptosis by rottlerin was observed when experiments were performed in 0.1% FBS, but not 10% FBS. These differences may have been due to serum-derived, growth factor-induced changes in cell cycle distribution under the different culture conditions. In 0.1% FBS, approximately 5% of cells were in S phase at 48 h, as assessed by flow cytometry, which closely agrees with the observed growth fraction of solid tumor cells in vivo(24). In 10% FBS, approximately 20% of the NSCLC cells from each cell line were in S phase. Because a 20% growth fraction is rarely observed in NSCLC tumors, we believe data derived from NSCLC cells grown in 0.1% FBS may better reflect conditions in vivo. In spite of the potentially decreased relevance of performing experiments with NSCLC cells in 10% FBS, however, we assessed potentiation of chemotherapy-induced apoptosis by rottlerin in the H1355 or H1155 cells in 10% FBS (data not shown). Although potentiation of chemotherapy-induced apoptosis was not observed, rottlerin was equally toxic in 10% FBS, and cisplatin was more toxic in 10% FBS.

The fact that rottlerin effectively killed NSCLC cells independent of serum concentration suggests that rottlerin is an interesting agent. Evaluation of rottlerin in the National Cancer Institute 60 cell line screen showed that rottlerin inhibited the growth of 57 of 60 cancer cell lines at a concentration of 10 μm.3 Whether inhibition of PKCδ is responsible for this widespread cytotoxicity is unknown, however, because rottlerin can also promote apoptosis in a PKCδ-independent manner through potentiation of tumor necrosis factor α-induced apoptosis (25). Cytotoxicity across multiple cancer cell lines raises the possibility that rottlerin may have utility in many types of tumor cell systems and that combining rottlerin with trastuzumab in cancer cells where trastuzumab has been shown to be effective (such as breast cancer) might yield interesting results.

Because rottlerin is similar to other kinase inhibitors in that it has nonspecific effects and its activity may depend upon the cellular system in which it is evaluated (26), we confirmed the putative role for PKCδ in promoting NSCLC survival by transfecting NSCLC cells with wild-type or a kinase-dead mutant of PKCδ. Compared with wild-type PKCδ, K376R PKCδ decreased PKCδ phosphorylation and increased apoptosis in three NSCLC cell lines. K376R increased apoptosis irrespective of serum concentration. K376R PKCδ also potentiated chemotherapy-induced apoptosis, but the level of potentiation was less than that observed with rottlerin combined with chemotherapy, possibly because of effects of rottlerin unrelated to PKCδ or a previously noted artifact of transient transfection experiments due to gating on GFP-positive cells, which underestimates apoptosis (1).

In many systems, PKCδ promotes rather than inhibits apoptosis (reviewed in Ref. 6). However, it is clear that the role of PKCδ is tissue and system dependent. For example, in glial cells, overexpression of PKCδ inhibits apoptosis induced by Sindbis virus (27), although it enhances apoptosis in the same cells in response to etoposide (28). Likewise, PKCδ is involved in the antiapoptotic action of basic fibroblast growth factor in PC12 cells (29). Further evidence that supports PKCδ as a suppressor of apoptosis includes the dependence of insulin-like growth factor I receptor-mediated transformation on active PKCδ (30), the activation of PKCδ in response to platelet-derived growth factor-mediated transformation (31), and the induction of anchorage-independent growth and increased metastatic potential of breast cancer cells that overexpress PKCδ (32, 33).

Specificity of PKCδ function may be related to different mechanisms of activation of PKCδ in different systems or differential availability of downstream substrates that mediate PKCδ activity in different cell types. Other factors that might play a role in determining the apoptotic potential of PKCδ are the generation of ceramide and the translocation of PKCδ to various intracellular organelles (34). Ceramide, a sphingolipid second messenger, is synthesized in response to cellular stresses such as chemotherapy. Generation of ceramide may stimulate PKCδ activation or may occur as a consequence of PKCδ activation. Translocation of PKCδ to the plasma membrane, mitochondria, or nucleus after exposure to cytotoxic stresses has been reported and interference with translocation may inhibit the induction of apoptosis by PKCδ (35). Translocation of PKCδ to different cellular compartments might contribute to the different cellular effects of PKCδ because PKCδ may interact with and/or phosphorylate different proteins in these specific compartments. The contribution of ceramide generation and/or cellular translocation to the response of NSCLC cells to inhibition of PKCδ, with or without chemotherapy, is unclear.

The results of our studies on the role of PKCδ in NSCLC cells are consistent with a recent evaluation of PKCδ in lung tissue performed by Lounsbury et al. (36). These investigators showed that PKCδ expression and phosphorylation of a downstream substrate were increased in murine lung tissues exposed to asbestos, a lung carcinogen. Increased levels of PKCδ expression and downstream substrate phosphorylation were associated with increased asbestos-stimulated growth of murine lung epithelial cells. This study raises the possibility that increases in PKCδ expression or activity may be involved in asbestos-related lung tumorigenesis.

In conclusion, our studies support the ongoing efforts targeting PKC isoforms in NSCLC therapy but suggest that current approaches may be limited in two ways. First, if increases in PKC isoform expression and/or phosphorylation can be detected in NSCLC tumor specimens compared with precursor lesions, then targeting PKC isoforms may have utility in NSCLC prevention as well as treatment. Secondly, current clinical approaches targeting PKC isoforms are not specifically directed at PKCδ. Our biochemical and genetic data suggest that the role of PKCδ in NSCLC should be reconsidered.

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.

2

The abbreviations used are: NSCLC, non-small cell lung cancer; PKC, protein kinase C; PI3K, phosphatidylinositol 3′-kinase; ERK, extracellular signal-regulated kinase; MEK, mitogen-activated protein kinase/ERK kinase; PDK, phosphoinositide-dependent kinase; GFP, green fluorescent protein; FBS, fetal bovine serum; DAG, diacylglycerol; PARP, poly (ADP-ribose) polymerase; CF, catalytic fragment.

3

http://dtp.nci.nih.gov/docs/cancer/cancer_data.html

Fig. 1.

Analysis of PKC isoforms. A, expression of PKC isoforms in NSCLC cells. Isoform-specific antibodies were used in immunoblotting experiments as described. The CF of PKCδ (PKCδ CF) was detected with the PKCδ-specific antibody. α-Tubulin is shown as a loading control. B, comparison of PKC expression and phosphorylation in normal lung epithelial cells (NHBE), immortalized lung epithelial cells (BEAS-2B), and a lung cancer cell line (H1355). Immunoblotting with isoform-specific antibodies is shown in the left column. Immunoblotting with phospho-specific PKC isoform antibodies is shown in the right column. α-Tubulin is shown as a loading control. C, maintenance of PKC phosphorylation with serum starvation is PI3K independent. NSCLC cells were serum starved in the absence or presence of LY294002 (25 μm). Immunoblotting with phospho-specific PKC, PDK-1, Akt, and α-tubulin antibodies was performed as described. All figures are representative of three or more independent experiments.

Fig. 1.

Analysis of PKC isoforms. A, expression of PKC isoforms in NSCLC cells. Isoform-specific antibodies were used in immunoblotting experiments as described. The CF of PKCδ (PKCδ CF) was detected with the PKCδ-specific antibody. α-Tubulin is shown as a loading control. B, comparison of PKC expression and phosphorylation in normal lung epithelial cells (NHBE), immortalized lung epithelial cells (BEAS-2B), and a lung cancer cell line (H1355). Immunoblotting with isoform-specific antibodies is shown in the left column. Immunoblotting with phospho-specific PKC isoform antibodies is shown in the right column. α-Tubulin is shown as a loading control. C, maintenance of PKC phosphorylation with serum starvation is PI3K independent. NSCLC cells were serum starved in the absence or presence of LY294002 (25 μm). Immunoblotting with phospho-specific PKC, PDK-1, Akt, and α-tubulin antibodies was performed as described. All figures are representative of three or more independent experiments.

Close modal
Fig. 2.

Effect of PKC inhibitors on NSCLC apoptosis. NSCLC cells were plated and serum starved in the absence (black bars) or presence of PKC inhibitors [rottlerin (10 μm; horizontal hatched bars), Gö6976 (100 nm; white bars), calphostin C (1 μm; diagonally hatched bars), or chelerythrine (10 μm; gray bars)] for 48 h. Cells were harvested, and apoptosis was quantified via flow cytometry as described. Bars are the means ± SE from three independently performed experiments. Ps ≤ 0.05 are shown with ∗.

Fig. 2.

Effect of PKC inhibitors on NSCLC apoptosis. NSCLC cells were plated and serum starved in the absence (black bars) or presence of PKC inhibitors [rottlerin (10 μm; horizontal hatched bars), Gö6976 (100 nm; white bars), calphostin C (1 μm; diagonally hatched bars), or chelerythrine (10 μm; gray bars)] for 48 h. Cells were harvested, and apoptosis was quantified via flow cytometry as described. Bars are the means ± SE from three independently performed experiments. Ps ≤ 0.05 are shown with ∗.

Close modal
Fig. 3.

Effect of rottlerin on chemotherapy-induced apoptosis. H1703 (A), H1155 (B), H1355 (C), or H157 (D) cells were plated at 1 × 105 cells/well and placed in 0.1% FBS. Rottlerin (10 μm) and/or chemotherapeutic agents were added simultaneously, and cells were incubated for 48 h. Apoptosis was quantified as described. Bars: C, control (no chemotherapy); T, paclitaxel (1 μm); G, gemcitabine (100 μm); P, cisplatin (20 μm), D, docetaxel (1.16 μm), and H, trastuzumab (10.5 μm). Bars are the means ± SD of a representative experiment done in triplicate from five independently performed experiments for each cell line. P ≤ 0.05 comparing chemotherapy + rottlerin with rottlerin alone are shown with ∗.

Fig. 3.

Effect of rottlerin on chemotherapy-induced apoptosis. H1703 (A), H1155 (B), H1355 (C), or H157 (D) cells were plated at 1 × 105 cells/well and placed in 0.1% FBS. Rottlerin (10 μm) and/or chemotherapeutic agents were added simultaneously, and cells were incubated for 48 h. Apoptosis was quantified as described. Bars: C, control (no chemotherapy); T, paclitaxel (1 μm); G, gemcitabine (100 μm); P, cisplatin (20 μm), D, docetaxel (1.16 μm), and H, trastuzumab (10.5 μm). Bars are the means ± SD of a representative experiment done in triplicate from five independently performed experiments for each cell line. P ≤ 0.05 comparing chemotherapy + rottlerin with rottlerin alone are shown with ∗.

Close modal
Fig. 4.

Dominant negative PKCδ increases NSCLC apoptosis. NSCLC cells (H1355, A and B; H1155, C; H157, D) were plated at 2 × 105 cells/well and transiently transfected with plasmids encoding wild-type PKCδ (wt PKCδ; gray bars) or dominant negative PKCδ (K376R; black bars) as described. Cells were allowed to recover from transfection for 24 h and then exposed to serum starvation ± chemotherapy for 48 h (except B, where 10% FBS was substituted for 0.1% FBS). Parallel samples from each transfection were harvested for immunoblotting to evaluate GFP, PARP, phospho-PKCδ, and α-tubulin expression. Samples were also fixed, stained with propidium iodide, and prepared for flow cytometry. Apoptosis was quantified by gating on GFP-positive cells and performing cell cycle analysis. Ps for comparing apoptosis under various experimental conditions are shown. A representative experiment from three independent experiments performed with each cell line is shown.

Fig. 4.

Dominant negative PKCδ increases NSCLC apoptosis. NSCLC cells (H1355, A and B; H1155, C; H157, D) were plated at 2 × 105 cells/well and transiently transfected with plasmids encoding wild-type PKCδ (wt PKCδ; gray bars) or dominant negative PKCδ (K376R; black bars) as described. Cells were allowed to recover from transfection for 24 h and then exposed to serum starvation ± chemotherapy for 48 h (except B, where 10% FBS was substituted for 0.1% FBS). Parallel samples from each transfection were harvested for immunoblotting to evaluate GFP, PARP, phospho-PKCδ, and α-tubulin expression. Samples were also fixed, stained with propidium iodide, and prepared for flow cytometry. Apoptosis was quantified by gating on GFP-positive cells and performing cell cycle analysis. Ps for comparing apoptosis under various experimental conditions are shown. A representative experiment from three independent experiments performed with each cell line is shown.

Close modal
1
Brognard J., Clark A. S., Ni Y., Dennis P. A. Akt/protein kinase B is constitutively active in non-small cell lung cancer cells and promotes cellular survival and resistance to chemotherapy and radiation.
Cancer Res.
,
61
:
3986
-3997,  
2001
.
2
Brognard J., Dennis P. A. Variable apoptotic response of NSCLC cells to inhibition of the MEK/ERK pathway by small molecules or dominant negative mutants.
Cell Death Differ.
,
9
:
893
-904,  
2002
.
3
Dutil E. M., Toker A., Newton A. C. Regulation of conventional protein kinase C isozymes by phosphoinositide-dependent kinase 1 (PDK-1).
Curr. Biol.
,
8
:
1366
-1375,  
1998
.
4
Li W., Zhang J., Flechner L., Hyun T., Yam A., Franke T. F., Pierce J. H. Protein kinase C-α overexpression stimulates Akt activity and suppresses apoptosis induced by interleukin 3 withdrawal.
Oncogene
,
18
:
6564
-6572,  
1999
.
5
Doornbos R. P., Theelen M., van der Hoeven P. C., van Blitterswijk W. J., Verkleij A. J., van Bergen en Henegouwen P. M. Protein kinase Cζ is a negative regulator of protein kinase B activity.
J. Biol. Chem.
,
274
:
8589
-8596,  
1999
.
6
Gschwendt M. Protein kinase C δ.
Eur. J. Biochem.
,
259
:
555
-564,  
1999
.
7
Parekh D., Ziegler W., Yonezawa K., Hara K., Parker P. J. Mammalian TOR controls one of two kinase pathways acting upon nPKCδ and nPKCε.
J. Biol. Chem.
,
274
:
34758
-34764,  
1999
.
8
Konishi H., Tanaka M., Takemura Y., Matsuzaki H., Ono Y., Kikkawa U., Nishizuka Y. Activation of protein kinase C by tyrosine phosphorylation in response to H2O2.
Proc. Natl. Acad. Sci. USA
,
94
:
11233
-11237,  
1997
.
9
Toker A., Meyer M., Reddy K. K., Falck J. R., Aneja R., Aneja S., Parra A., Burns D. J., Ballas L. M., Cantley L. C. Activation of protein kinase C family members by the novel polyphosphoinositides PtdIns-3,4-P2 and PtdIns-3,4,5-P3.
J. Biol. Chem.
,
269
:
32358
-32367,  
1994
.
10
Konishi H., Kuroda S., Kikkawa U. The pleckstrin homology domain of RAC protein kinase associates with the regulatory domain of protein kinase C ζ.
Biochem. Biophys. Res. Commun.
,
205
:
1770
-1775,  
1994
.
11
Konishi H., Fujiyoshi T., Fukui Y., Matsuzaki H., Yamamoto T., Ono Y., Andjelkovic M., Hemmings B. A., Kikkawa U. Activation of protein kinase B induced by H2O2 and heat shock through distinct mechanisms dependent and independent of phosphatidylinositol 3-kinase.
J. Biochem. (Tokyo)
,
126
:
1136
-1143,  
1999
.
12
Cragg L. H., Andreeff M., Feldman E., Roberts J., Murgo A., Winning M., Tombes M. B., Roboz G., Kramer L., Grant S. Phase I trial and correlative laboratory studies of bryostatin 1 (NSC 339555) and high-dose 1-β-d-arabinofuranosylcytosine in patients with refractory acute leukemia.
Clin. Cancer Res.
,
8
:
2123
-2133,  
2002
.
13
Clamp A., Jayson G. C. The clinical development of the bryostatins.
Anticancer Drugs
,
13
:
673
-683,  
2002
.
14
Reddel R. R., Ke Y., Gerwin B. I., McMenamin M. G., Lechner J. F., Su R. T., Brash D. E., Park J. B., Rhim J. S., Harris C. C. Transformation of human bronchial epithelial cells by infection with SV40 or adenovirus-12 SV40 hybrid virus, or transfection via strontium phosphate coprecipitation with a plasmid containing SV40 early region genes.
Cancer Res.
,
48
:
1904
-1909,  
1988
.
15
Newton A. C. Protein kinase C: structural and spatial regulation by phosphorylation, cofactors, and macromolecular interactions.
Chem. Rev.
,
101
:
2353
-2364,  
2001
.
16
Hofmann J. The potential for isoenzyme-selective modulation of protein kinase C.
FASEB J.
,
11
:
649
-669,  
1997
.
17
Webb B. L., Lindsay M. A., Seybold J., Brand N. J., Yacoub M. H., Haddad E. B., Barnes P. J., Adcock I. M., Giembycz M. A. Identification of the protein kinase C isoenzymes in human lung and airways smooth muscle at the protein and mRNA level.
Biochem. Pharmacol.
,
54
:
199
-205,  
1997
.
18
Basu A., Weixel K., Saijo N. Characterization of the protein kinase C signal transduction pathway in cisplatin-sensitive and -resistant human small cell lung carcinoma cells.
Cell Growth Differ.
,
7
:
1507
-1512,  
1996
.
19
Mao M., Fang X., Lu Y., Lapushin R., Bast R. C., Jr., Mills G. B. Inhibition of growth-factor-induced phosphorylation and activation of protein kinase B/Akt by atypical protein kinase C in breast cancer cells.
Biochem J.
,
352
:
475
-482,  
2000
.
20
Gokmen-Polar Y., Murray N. R., Velasco M. A., Gatalica Z., Fields A. P. Elevated protein kinase C βII is an early promotive event in colon carcinogenesis.
Cancer Res.
,
61
:
1375
-1381,  
2001
.
21
Weller S. G., Klein I. K., Penington R. C., Karnes W. E., Jr. Distinct protein kinase C isozymes signal mitogenesis and apoptosis in human colon cancer cells.
Gastroenterology
,
117
:
848
-857,  
1999
.
22
Mandil R., Ashkenazi E., Blass M., Kronfeld I., Kazimirsky G., Rosenthal G., Umansky F., Lorenzo P. S., Blumberg P. M., Brodie C. Protein kinase Cα and protein kinase Cδ play opposite roles in the proliferation and apoptosis of glioma cells.
Cancer Res.
,
61
:
4612
-4619,  
2001
.
23
Sonnenburg E. D., Gao T., Newton A. C. The phosphoinositide-dependent kinase, PDK-1, phosphorylates conventional protein kinase C isozymes by a mechanism that is independent of phosphoinositide 3-kinase.
J. Biol. Chem.
,
276
:
45289
-45297,  
2001
.
24
Young R. C., DeVita V. T. Cell cycle characteristics of human solid tumors in vivo.
Cell Tissue Kinet.
,
3
:
285
-290,  
1970
.
25
Basu A., Johnson D. E., Woolard M. D. Potentiation of tumor necrosis factor-α-induced cell death by rottlerin through a cytochrome-c-independent pathway.
Exp. Cell Res.
,
278
:
209
-214,  
2002
.
26
Davies S. P., Reddy H., Caivano M., Cohen P. Specificity and mechanism of action of some commonly used protein kinase inhibitors.
Biochem. J.
,
351
:
95
-105,  
2000
.
27
Zrachia A., Dobroslav M., Blass M., Kazimirsky G., Kronfeld I., Blumberg P. M., Kobiler D., Lustig S., Brodie C. Infection of glioma cells with Sindbis virus induces selective activation and tyrosine phosphorylation of protein kinase C δ. Implications for Sindbis virus-induced apoptosis.
J. Biol. Chem.
,
277
:
23693
-23701,  
2002
.
28
Blass M., Kronfeld I., Kazimirsky G., Blumberg P. M., Brodie C. Tyrosine phosphorylation of protein kinase Cδ is essential for its apoptotic effect in response to etoposide.
Mol. Cell. Biol.
,
22
:
182
-195,  
2002
.
29
Wert M. M., Palfrey H. C. Divergence in the anti-apoptotic signaling pathways used by nerve growth factor and basic fibroblast growth factor (bFGF) in PC12 cells: rescue by bFGF involves protein kinase Cδ.
Biochem. J.
,
352
:
175
-182,  
2000
.
30
Li W., Jiang Y. X., Zhang J., Soon L., Flechner L., Kapoor V., Pierce J. H., Wang L. H. Protein kinase C-δ is an important signaling molecule in insulin-like growth factor I receptor-mediated cell transformation.
Mol. Cell. Biol.
,
18
:
5888
-5898,  
1998
.
31
Li W., Yu J. C., Michieli P., Beeler J. F., Ellmore N., Heidaran M. A., Pierce J. H. Stimulation of the platelet-derived growth factor β receptor signaling pathway activates protein kinase C-δ.
Mol. Cell. Biol.
,
14
:
6727
-6735,  
1994
.
32
Kiley S. C., Clark K. J., Duddy S. K., Welch D. R., Jaken S. Increased protein kinase C δ in mammary tumor cells: relationship to transformation and metastatic progression.
Oncogene
,
18
:
6748
-6757,  
1999
.
33
Kiley S. C., Clark K. J., Goodnough M., Welch D. R., Jaken S. Protein kinase C δ involvement in mammary tumor cell metastasis.
Cancer Res.
,
59
:
3230
-3238,  
1999
.
34
Sumitomo M., Ohba M., Asakuma J., Asano T., Kuroki T., Hayakawa M. Protein kinase Cδ amplifies ceramide formation via mitochondrial signaling in prostate cancer cells.
J. Clin. Investig.
,
109
:
827
-836,  
2002
.
35
Majumder P. K., Pandey P., Sun X., Cheng K., Datta R., Saxena S., Kharbanda S., Kufe D. Mitochondrial translocation of protein kinase C δ in phorbol ester-induced cytochrome c release and apoptosis.
J. Biol. Chem.
,
275
:
21793
-21796,  
2000
.
36
Lounsbury K. M., Stern M., Taatjes D., Jaken S., Mossman B. T. Increased localization and substrate activation of protein kinase Cδ in lung epithelial cells following exposure to asbestos.
Am. J. Pathol.
,
160
:
1991
-2000,  
2002
.