Enzastaurin, a Protein Kinase Cβ Inhibitor, Suppresses Signaling through the Ribosomal S6 Kinase and Bad Pathways and Induces Apoptosis in Human Gastric Cancer Cells

Although the incidence of gastric adenocarcinoma has decreased significantly in Western countries, it is still among the most common malignancies in South America, in many former Eastern European countries, and across Asia. In Korea, according to statistics reported in 2005, gastric cancer was the most prevalent cancer in men (1). Although chemotherapy is commonly used in clinical practice, the prognosis of advanced gastric cancer is still poor and treatment is usually unsuccessful. In breast, lung, and colorectal cancers, the identification of novel molecular targets and the development of therapies specific to these targets have provided successful new approaches to treatment in clinical practice. However, targeted therapy has not been successful in the treatment of gastric cancer, and attempts to develop and apply new agents are urgently needed.

The protein kinase C (PKC) signaling pathway, which functions through serine/threonine kinase activity, plays a key role in tumor-induced angiogenesis, tumor growth, differentiation, cytokine secretion, migration, and apoptosis and is a possible target for anticancer therapy (2, 3). PKC activation contributes to tumor cell survival and proliferation and, as such, has been implicated repeatedly in the malignant progression of human cancers, such as B-cell lymphoma (4), malignant gliomas (2), and colorectal carcinomas (5–7). PKC is stimulated by vascular endothelial growth factor (VEGF) receptor activation and is an important mediator of VEGF, the most potent angiogenic factor identified in a variety of solid tumors (8).

Gastric cancer tumor tissue possesses a higher level of PKC activity than normal tissue (9). Two highly specific PKC inhibitors (RO-31-8220 and chelerythrine) have been shown to suppress the growth of gastric cancer cells through apoptosis induction and cell cycle quiescence (10). PKC has been implicated as having an important role in the development of invasive activity in gastric cancer (11). Recently, it was reported that the antisense targeting PKCα and PKCβ1 markedly inhibited gastric cancer cell growth and made gastric cancer cells more responsive to mitomycin C–induced or 5-fluorouracil (5-FU)–induced apoptosis (12). Given its proposed key role in tumorigenesis, PKC may be a promising target for gastric cancer growth inhibition.

Enzastaurin (LY317615.HCl), an acyclinc bisindolylmaleimide, was initially developed as an ATP-competitive selective inhibitor of PKCβ (13). In addition to its major target PKCβ, enzastaurin also potently inhibits other PKC isoforms, including PKCδ, PKCε, PKCγ, and PKCα (14). Although enzastaurin was initially developed for antiangiogenic cancer therapy (15), recent preclinical studies have shown that enzastaurin has a direct effect on a variety of human cancer cells, inducing apoptosis (14, 16–18). Enzastaurin was also shown to target the phosphoinositide 3-kinase (PI3K)/Akt pathway and to inhibit glycogen synthase kinase 3β (GSK3β) phosphorylation (14). Based on these promising preclinical findings, a phase I clinical study that found that enzastaurin at 525 mg once daily was the recommended dose for further clinical trials was completed. At this 525 mg daily dose, a steady-state plasma concentration of enzastaurin and its analytes reached up to 8 μmol/L, with the a mean plasma exposure of 2.2 μmol/L (3). Currently, enzastaurin is being evaluated in several phase II or phase III clinical trials.

However, the role of enzastaurin in gastric cancer remains unknown. The results of our study show that enzastaurin is cytotoxic to gastric cancer cells. These findings strongly support further clinical evaluation of enzastaurin. Furthermore, our data show that enzastaurin induces gastric cancer cell apoptosis by a mechanism not previously reported through the p90 ribosomal S6 kinase (Rsk) and Bad-mediated pathways, in addition to inhibiting the Akt signaling cascade.

Enzastaurin was provided by Eli Lilly Company. Other chemotherapeutic drugs were obtained as follows: 5-FU and cisplatin from Choongwae Pharma Company, paclitaxel from BMS Korea, and irinotecan from CJ Pharmaceutical Company. Enzastaurin was initially dissolved in DMSO (Sigma Chemical Co.) at a concentration of 5 mmol/L and stored in small aliquots at −20°C.

Antibodies to caspase-3, caspase-9, Erk, phosphorylated Erk, Akt, phosphorylated Akt, p90 Rsk, phosphorylated Rsk (Thr³⁵⁹/Ser³⁶³), GSK3β, phosphorylated GSK3β, Bad, phosphorylated Bad (Ser¹¹²), and phosphorylated Bad (Ser¹³⁶) were purchased from Cell Signaling Technology. Antibodies to caspase-8 and poly(ADP-ribose) polymerase (PARP) were provided by BD PharMingen. Antibodies to Bax, Bak, Bcl-X_L, Mcl-1, and cytochrome c were obtained from Santa Cruz Biotechnology. The antibody to Bcl-2 was from DakoCytomation.

Human gastric cancer cell lines (SNU-1, SNU-5, SNU-16, SNU-216, SNU-484, SNU-601, SNU-620, SNU-638, SNU-668, and SNU-719) were obtained from the Korea Cell Line Bank (19) and grown in RPMI 1640 supplemented with 10% fetal bovine serum and gentamicin (10 μg/mL). All cell lines were incubated under standard culture conditions (20% O₂ and 5% CO₂; 37°C).

Tetrazolium dye [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, MTT; Sigma] assays were used to evaluate the growth inhibitory effects of enzastaurin, as described in prior studies (20, 21). Cells were seeded on 96-well plates, incubated for 24 h, and then treated with drugs for 72 h at 37°C. After drug treatment, MTT solution was added to each well and incubated for 4 h at 37°C before the medium was removed. DMSO was then added and shaken for 30 min at room temperature. Cell viability was determined by measuring absorbance at 550 nm in a microplate reader (Spectra Classic, Tecan Co.). Experiments were performed in groups of six for each drug concentration and were repeated thrice.

To determine the in vivo activity of enzastaurin, 6-wk-old female athymic nude mice (Harlan Sprague-Dawley, Inc.) were injected s.c. in the back with SNU-16 or SNU-484 cells in 100 μL PBS. When the tumor reached 100 mm³, enzastaurin was suspended in 10% acacia (Fisher Scientific) in water and given at a dose of 75 mg/kg by oral gavage twice daily. Control groups were treated only with vehicle. Tumor burden was measured twice a week with a caliper (calculated volume = length × width × height × π / 6).

Cells were washed twice in PBS, fixed in 70% ethanol, and stored at −20°C until required for analysis. Before analysis, cell suspensions were washed with PBS, digested with RNase A (50 μg/mL) for 15 min at 37°C, and then stained with propidium iodide (50 μg/mL). The cell DNA contents (10,000 cells per experimental group) were determined using a FACSCalibur flow cytometer (Becton Dickinson Biosciences) equipped with a ModFit LT program (Verity Software House, Inc.), as previously described (20).

After the cells were exposed to enzastaurin for 72 h, the degree of apoptosis was assessed by the Annexin V binding assay as instructed by the manufacturer (BD PharMingen). The harvested cell suspension was incubated with Annexin V for 15 min at room temperature in the dark and then analyzed by flow cytometry, as described previously (20). Single-variable analysis, which used Annexin V–FITC only, was performed due to the self-fluorescence of enzastaurin within the propidium iodide spectrum.

Three core tissue biopsies (4 mm in diameter) were taken from each individual paraffin-embedded tissue sample (donor blocks) and arranged in a new recipient paraffin block (tissue array block) using a trephine apparatus (Superbiochips Laboratories). Each tissue array block contained samples of all animals. Sections of 4 μm were cut from each triplicate tissue array blocks, deparaffinized, and dehydrated. Immunohistochemical detection of apoptosis was carried out using an Apoptag in situ Apoptosis Detection kit (Chemicon International) following the procedures provided by the manufacturer.

Cultured cells were washed with ice-cold PBS and suspended in an extraction buffer [20 mmol/L Tris-Cl (pH 7.4), 100 mmol/L NaCl, 1% NP40, 0.5% sodium deoxycholate, 5 mmol/L MgCl₂, 0.1 mmol/L phenylmethylsulfonyl fluoride, 0.1 mmol/L pepstatin A, 0.1 mmol/L antipain, 0.1 mmol/L chymostatin, 0.2 mmol/L leupeptin, 10 μg/mL aprotinin, 0.5 mg/mL soybean trypsin inhibitor, and 1 mmol/L benzamidine] on ice for 15 min. Samples containing equal amounts of total protein were resolved in a SDS-polyacrylamide denaturing gel, transferred to nitrocellulose membranes, and probed with antibodies. Detection was performed using an enhanced chemiluminescence system (Amersham Pharmacia Biotech).

For total protein extraction from frozen tissue, mouse tissues frozen in liquid nitrogen after excision were powdered with mortar and pestle. Radioimmunoprecipitation assay buffer [50 mmol/L Tris-Cl (pH 8.0), 150 mmol/L NaCl, 1% NP40, 0.5% sodium deoxycholate, 0.1% sodum dodecyl sulfate, 1 mmol/L EDTA, 1 mmol/L phenylmethylsulfonyl fluoride, 1 mmol/L Na₃VO₄, 1 mmol/L NaF, 1 μg/mL aprotinin, 1 μg/mL leupeptin, and 1 μg/mL pepstatin] was added to powdered tissue. Samples were vortexed and then incubated for 45 min on ice. Samples were homogenized and centrifuged at 14,000 × g for 10 min. Then, Western blot analysis was performed using the same method.

For detecting the mitochondrial cytochrome c release into the cytosol, cells were harvested at each experimental time point, resuspended with isotonic isolation buffer [10 mmol/L HEPES, 1 mmol/L EDTA, 250 mmol/L sucrose (pH 7.6)], and collected by centrifugation. Cells were then suspended in hypotonic isolation buffer [10 mmol/L HEPES, 1 mmol/L EDTA, 50 mmol/L sucrose (pH 7.6)] and disrupted by passing through a 27-gauge needle 5 to 10 times. After adding hypertonic isolation buffer [10 mmol/L HEPES, 1 mmol/L EDTA, and 450 mmol/L sucrose (pH 7.6)] to balance the tonicity of the buffer, cells were centrifuged at 16,000 × g for 20 min, and the supernatant was used for the cytosol protein extraction.

Two independent small interfering RNA (siRNA) vectors that target the DNA sequence of Bad (Bad1 AAGAAGGGACTTCCTCGCCCG, Bad 2 GACGAGTTTGTGGACTCCTTT; Qiagen Co.) were used in this experiment, and a nonspecific siRNA (nonhomologous to any known gene sequence) was used as a negative control. SNU-620 cells were transfected with these siRNAs for 4 h using the Lipofectamine Plus reagent (Invitrogen) according to the manufacturer's protocol and recovered in fresh medium containing 10% fetal bovine serum for 12 h. Cells were then treated with enzastaurin or DMSO, and the proportion of apoptotic cells (sub-G₁ fraction) was determined by flow cytometry.

Analyses of drug interactions were performed by constructing an envelope of additivity using the isobologram method of Steel and Peckham, as described previously (21, 22). Based on available dose-response curves, the combined effects of enzastaurin and other cytotoxic chemotherapeutic agents were analyzed at IC₅₀. Three isoeffect curves were drawn as follows.

Mode I line. When the dose of drug A is chosen, there remains an increment of effect to be produced by drug B. If two drugs were to act independently, the addition is performed by taking the increase in doses, starting from 0, that give log survivals which add up to IC₅₀ (heteroaddition).

Mode II(A) line. When the dose of drug A is chosen, an isoeffect curve can also be calculated by taking the dose increment of drug B that gives the required contribution to the total effect up to the limit, in this case, IC₅₀ (isoaddition).

Mode II(B) line. Similarly, when the dose of drug B is chosen, an isoeffect curve can be calculated by taking the dose increment of drug A that gives the required contribution to IC₅₀ (isoaddition).

With combination of graded doses of drug A and a chosen dose of drug B, a single dose-response curve can be drawn. When the experimental IC₅₀ concentration in this drug combination falls left of the envelope (Supplementary Fig. S1, point P1), the two drugs have supraadditive (synergistic) interaction. When the experimental data point is within the envelope, the combination is considered to be noninteractive (additive; Supplementary Fig. S1, point P2). Finally, when the data point is in the area to the right of the envelope, the combination is considered to be antagonistic (Supplementary Fig. S1, point P3 or point P4).

Actual IC₅₀ values were obtained from growth inhibition curves after cancer cells were exposed to a variety of concentrations of enzastaurin or other cytotoxic agents, alone or in combination. When combined, two drugs were applied simultaneously for 72 h.

Statistical analysis. Comparison of quantification of terminal deoxyribonucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) assay or percentage of surviving cells were analyzed by two-tailed Mann-Whitney U test or Student's t test. Statistical analysis to compare tumor sizes in xenograft-bearing mice was performed with Student's two-tailed t test. Differences between groups were considered statistically significant if P < 0.05.

Enzastaurin suppresses gastric cancer cell proliferation in vitro and in a xenograft mouse model. Although enzastaurin was initially introduced for clinical development due to its antiangiogenic activity, it was also shown to have a direct antiproliferative effect on human tumor cells (14). We therefore evaluated the ability of enzastaurin to suppress gastric cancer cell proliferation in culture. Indeed, enzastaurin suppressed the proliferation of gastric cancer cells and showed a wide variety of IC₅₀ values. SNU-620 cells were the most sensitive to enzastaurin (IC₅₀, 3.56 μmol/L) and SNU-1, SNU-5, SNU-16, and SNU-484 cells had a low micromolar range of IC₅₀ values (3.78–6.35 μmol/L). However, enzastaurin was not very cytotoxic with SNU-216 and SNU-719 with IC₅₀ values of >250 μmol/L after 72 hours of exposure (Fig. 1A).

We next sought to assess the in vivo efficacy of enzastaurin using a mouse model of human gastric cancer. The phase I clinical trials for enzastaurin have shown that p.o. administration at 525 mg/d yields ∼2 μmol/L mean steady-state plasma exposure of enzastaurin and its analytes (3). In the tumor xenograft-bearing mice, plasma enzastaurin concentration is ∼2 μmol/L at a dose of 75 mg/kg given p.o. by gavage twice daily (14). At this dose of enzastaurin, tumor growth in the enzastaurin-treated group was significantly suppressed compared with the control group in the SNU-484 xenograft-bearing mice (P < 0.05; Fig. 1B). Enzastaurin treatment also significantly suppressed the growth of SNU-16 gastric cancer xenograft (P < 0.05; data not shown).

Enzastaurin inhibits gastric cancer cell proliferation without cell cycle specificity and induces direct apoptosis. After demonstrating that enzastaurin had antiproliferative effects on gastric cancer cells, we next sought to examine the effects of enzastaurin on cell cycle progression. The cell cycle profiles of the gastric cancer cells were analyzed, and enzastaurin was found to have little effect on the cell cycle progression of SNU-620 and SNU-484. However, the sub-G₁ population increased with the passage of time and in a dose-dependent manner after enzastaurin treatment; this result suggests that enzastaurin induced apoptosis (Fig. 2A). Protein profiling of enzastaurin-treated gastric cancer cells showed cleavage of caspase-3 and PARP. The amount of cleaved forms of these proteins increased in a time-dependent and dose-dependent manner after enzastaurin treatment (Fig. 2B,, top). Consistent with these data, the Annexin V binding assay confirmed a dose-dependent enzastaurin-induced apoptosis in SNU-620 and SNU-484 cells (Fig. 2B , bottom left). Taken together, these analyses show that enzastaurin induced direct apoptosis in human gastric cancer cells in the low micromolar range (2.5–10 μmol/L) without cell cycle–specific inhibition.

The effects of enzastaurin therapy on tumor cell apoptosis were also examined in a xenograft mouse model (SNU-484). In Western blot analysis using protein extraction from frozen xenograft tumors, the induction of caspase-3 activation was shown in enzastaurin-treated SNU-484 xenograft-bearing mice (Fig. 2B,, bottom right). We also performed TUNEL assay on paraffin-embedded xenograft tumors and used tissue arrays, which allowed direct comparison between tissues from different animal groups. The results of TUNEL assay indicated increased levels of apoptosis in enzastaurin-treated animals compared with control animals (P < 0.05; Fig. 2C).

Enzastaurin activates the mitochondrial pathway during apoptosis in gastric cancer cells. Two major apoptotic pathways, the death receptor pathway (extrinsic pathway) and the mitochondrial pathway (intrinsic pathway), have been well characterized in mammalian cells. Over the course of these pathways, activation of the death receptor first triggers caspase-8 activation, whereas the release of mitochondrial cytochrome c activates caspase-9 as an initial caspase, all of which subsequently induce the activation of effector caspases, such as caspase-3, caspase-6, and caspase-7 (23). As shown in Fig. 2B, caspase-3 was activated in gastric cancer cells, and, thus, we next examined the activation of initiator caspases (caspase-8 and caspase-9) in enzastaurin-treated SNU-620 cells. Although active fragments of caspase-8 or caspase-9 were not detected in the Western blotting, the amount of the caspase-9 proform decreased with time. This result suggests that enzastaurin-induced apoptosis is associated with the activation of caspase-9. However, the amount of the caspase-8 proform did not change after enzastaurin treatment (Fig. 2D,, left). In addition, enzastaurin induced the cytochrome c release from the mitochondria to the cytosol in SNU-620 cells (Fig. 2D , right). These results show that enzastaurin treatment led gastric cancer cells to undergo apoptosis through a mitochondrial pathway.

The proapoptotic Bcl-2 family proteins (Bax and Bak) have been shown to be required for the disruption of mitochondria and intrinsic death of cancer cells, whereas the antiapoptotic Bcl-2 family proteins (Bcl-2, Bcl-X_L, and Mcl-1) can prevent cell death by interfering with the action of Bax and Bak. Therefore, the change in the expression of the Bcl-2 family proteins was examined after treating the SNU-620 or SNU-484 cells with DMSO or enzastaurin. The expression levels of Bax and Bak were not changed after enzastaurin treatment in both cell lines. Enzastaurin suppressed both Bcl-2 and Bcl-X_L expression at 2.5 μmol/L in the SNU-620 cells but had no effect on these proteins in the SNU-484 cells. The levels of Mcl-1 were not significantly changed at 2.5 to 5 μmol/L of enzastaurin, but were suppressed at a higher dose of enzastaurin (10 μmol/L) in both cell lines (Supplementary Fig. S2).

Enzastaurin blocks phosphorylation of Akt and p90 Rsk. We sought to examine whether pathways known to be influenced by PKC activity might be affected in gastric cancer cells by enzastaurin treatment. PKC activity has been connected to many intracellular signaling pathways, including the ras-Erk signaling axis and the PI3K/Akt pathway (14). Although Rsk is activated by Erk, Rsk can also be influenced by a PKC-dependent and Erk-independent pathway (24–26). In a previous report, treatment of HCT116 colon cancer cells and U87MG glioblastoma cells with enzastaurin failed to inhibit Erk activity, but enzastaurin inhibited Akt phosphorylation in these cell lines (14). Similarly, treatment of SNU-620 and SNU-484 cells with enzastaurin did not suppress Erk activity in the Western blot analysis. By contrast, enzastaurin showed a clear concentration-dependent reduction of Akt^Ser473 and Rsk^{Thr359/Ser363} phosphorylation (Fig. 3A). Enzastaurin also suppressed the expression of phosphorylated GSK3β, a downstream target of the Akt pathway, in both gastric cancer cell lines (Fig. 3B).

Enzastaurin induces Bad-mediated apoptosis through the Rsk pathway in gastric cancer cells. It is well known that the activation of both ras-Erk and PI3K/Akt signaling pathways promotes cancer cell survival and, along with the cell death machinery, leads to the phosphorylation and inactivation of Bad, a proapoptotic member of the Bcl-2 family proteins. Rsk can also phosphorylate Bad via a PKC-dependent and Erk-independent pathway (24–26). Survival-promoting cytokines suppress the activity of the Bad protein by inducing the phosphorylation of Bad at two critical sites, Ser¹¹² and Ser¹³⁶, which leads to the dissociation of Bad from prosurvival Bcl-2 proteins and the association of Bad with members of the 14-3-3 family of proteins (27). The regulation of Bad by these phosphorylation events suggests that Bad is a point of convergence for multiple signaling pathways that cooperate in cell death or survival. Akt phosphorylates Bad on Ser¹³⁶, whereas Erk or Rsk phosphorylate Bad on Ser¹¹², respectively (24–26). Because enzastaurin suppressed the phosphorylation of Akt and Rsk, we next evaluated whether Bad activity was affected in enzastaurin-treated gastric cancer cells using Western blot analysis. As shown in Fig. 4A, Ser¹¹² phosphorylation of Bad was readily detectable and decreased in a concentration-dependent manner after enzastaurin treatment. Because Erk phosphorylation was not decreased in enzastaurin-treated cells (Fig. 3A), enzastaurin-induced Rsk dephosphorylation is thought to activate Bad via dephosphorylation of Bad at Ser¹¹². These data suggest that activation of Bad through its dephosphorylation at Ser¹¹² is a crucial mechanism for enzastaurin-mediated gastric cancer cell apoptosis.

To confirm the direct involvement of Bad in enzastaurin-induced apoptotic cell death, we analyzed the effect of silencing endogenous Bad protein by siRNA in SNU-620 cells. Transfections of two independent siRNAs targeting Bad (Bad1 and Bad2) strongly silenced endogenous Bad, compared with transfections of nonspecific siRNA (Fig. 4B). When SNU-620 cells were transfected with Bad siRNAs, the extent of enzastaurin-induced apoptosis and cytotoxicity were markedly reduced compared with those cells transfected with the control siRNA (Fig. 4C and D). Taken together, these results suggest that enzastaurin-induced apoptosis and cytotoxicity in gastric cancer cells were mediated through Bad activation via the Rsk pathway.

Enzastaurin is also known to exert its anticancer activity by inhibiting signals through the Akt pathway (14, 17, 18); our data also showed decreased phosphorylation of Akt in enzastaurin-treated gastric cancer cells (Fig. 3A). However, although the phosphorylated Bad (Ser¹³⁶) antibody readily detected the phosphorylated Bad control protein (from Cell Signaling Technology, Inc.), the Bad protein phosphorylated at Ser¹³⁶ from the extracts of gastric cancer cells was not detected (data not shown). Thus, the association between the Akt pathway and the Bad activity could not be confirmed.

The combination of enzastaurin and cytotoxic chemotherapeutic drugs shows additive or synergistic activity in gastric cancer cells. The combined effects of enzastaurin and cytotoxic chemotherapeutic agents (5-FU, cisplatin, paclitaxel, and irinotecan) were tested using the isobologram method in SNU-620 and SNU-484 cells (Fig. 5). The simultaneous exposure of enzastaurin and these additional cytotoxic agents over a period of 72 hours produced additive or synergistic interactions.

PKC is involved in signal transduction pathways that regulate growth factor response, tumor cell proliferation, and apoptosis. PKC therefore represents an attractive and promising target for cancer treatment. Although implicated in gastric cancer pathogenesis, the therapeutic value of targeting the PKC signaling pathway in gastric cancer is to date unknown. We therefore investigated the cytotoxicity and antitumor effects of enzastaurin in gastric cancer. Although SNU-216 and SNU-719 were resistant to enzastaurin, growth of other gastric cancer cells was effectively inhibited by enzastaurin. Specially, SNU-1, SNU-5, SNU-16, SNU-484, and SNU-620 had a low micromolar range of IC₅₀ values (3.78–6.35 μmol/L; Fig. 1A); this range was similar to that of colon carcinoma, glioblastoma, prostate cancer, cutaneous T-cell lymphoma, and myeloma cell lines (14, 17, 18). When combined with cytotoxic agents commonly used in the treatment of advanced gastric cancer, enzastaurin showed synergistic or additive activities at clinically significant concentrations (1–3 μmol/L), as shown in Fig. 5. Oral dosing with enzastaurin to yield plasma concentrations similar to those achieved in clinical trials also significantly suppressed the growth of human gastric cancer cell xenografts (Fig. 1B).

To date, there is no report on the effects of enzastaurin on cancer cell cycle progression. In a previous report, stauroporine and its analogues with high specificity for PKC had differential effects on cell cycle progression in A549 lung cancer cells. Staurosporine and UCN-01 retarded A549 cells in G₀-G₁. However, CGP 41251 had no effect on any phase of the A549 cell cycle, as enzastaurin inhibited SNU-620 and SNU-484 cell growth in a noncycle-specific manner (Fig. 2A). These findings show that PKC inhibitors differ with respect to the mechanisms by which they interfere with the cell cycle (28). Our data suggest that enzastaurin induces in vitro cell growth inhibition in gastric cancer through direct induction of apoptosis without affecting cell cycle progression. This apoptotic induction by enzastaurin was mediated through the activation of caspases in the gastric cancer cells. The increased level of the cytosolic cytochrome c protein and the decreased level of caspase-9 proform, after enzastaurin treatment, reveal that a mitochondrial apoptotic pathway was involved in enzastaurin-induced gastric cancer cell death (Fig. 2D). Enzastaurin-induced apoptosis was also shown in tumor xenograft-bearing mice (Fig. 2B and C). In MM.1S myeloma cells, down-regulation of Mcl-1, but not Bcl-2 and Bcl-X_L, was observed after enzastaurin treatment (2.5–5 μmol/L; ref. 16). However, Mcl-1 expression levels were not changed significantly at 2.5 to 5 μmol/L of enzastaurin in SNU-620 and SNU-484 cells, but its expression was suppressed at a higher dose (10 μmol/L). Both Bcl-2 and Bcl-X_L expression in SNU-620 cells were decreased at lower doses of enzastaurin (2.5 μmol/L; Supplementary Fig. S2). These results show that enzastaurin affects the expression of the antiapoptotic Bcl-2 family proteins differently according to the specific cell type evaluated.

The mechanisms associated with enzastaurin-induced apoptosis in cancer cells are not clear. Recent studies have emphasized the importance of the PI3K/Akt signaling pathway in enzastaurin-induced apoptosis in human cancer cell lines. These studies showed that the expression of phosphorylated Akt and GSK3β, one of the Akt downstream targets, was decreased by enzastaurin treatment. They also suggested that GSK3β is a reliable pharmacodynamic marker for enzastaurin activity (14, 17, 18). Akt activation, which is mediated through phosphorylation, maintains a survival signal that protects cells from apoptosis by phosphorylating proapoptotic proteins, such as caspase-9, Bad, and the cell regulatory protein GSK3β; in addition to indirectly modulating p53 and nuclear factor-κB, it also mediates growth factor–induced cell proliferation (18). Therefore, suppression of Akt phosphorylation may be an important mechanism associated with enzastaurin-induced tumor cell death. However, experiments with in vitro kinase assays showed that enzastaurin caused virtually no inhibition of Akt (14). Thus, it was suggested that enzastaurin may indirectly suppress the Akt signaling pathway through its inhibitory effect on PKC proteins (14). However, the mechanisms underlying enzastaurin activity, linking the PKC and Akt pathways, are still unknown and remain to be elucidated.

In our study, the expression of both phosphorylated Akt and GSK3β was also decreased by enzastaurin treatment; this finding is consistent with previous reports. Additionally, our experiments showed a new mechanism involved in enzastaurin-induced apoptosis in gastric cancer cells—the Bad pathway. Bad is one of the proapoptotic Bcl-2 family member proteins and is inactivated by phosphorylation at two critical sites, Ser¹¹² and Ser¹³⁶. When phosphorylated at Ser¹¹² or Ser¹³⁶, Bad is complexed to the cytosolic 14-3-3 protein and fails to interact with the antiapoptotic Bcl-X_L protein, thus favoring cell survival (27). In our experiments, Bad activation (dephosphorylation) was confirmed in enzastaurin-treated gastric cancer cells (Fig. 4A). In addition, the depletion of Bad conferred resistance to enzastaurin-induced apoptosis and cytotoxicity in the gastric cancer cells (Fig. 4C and D). Because Erk phosphorylation was not decreased, dephosphorylation of Bad at Ser¹¹² is thought to be mediated by Rsk inactivation in enzastaurin-treated gastric cancer cells (24–26). However, a phosphorylated Bad at Ser¹³⁶, which is mediated by the Akt pathway, was not detected in our experiments. One explanation for this finding is that the expression of phosphorylated Bad at Ser¹³⁶ in gastric cancer cells is below the limit of detection with the antibody we have used (from Cell Signaling Technology, Inc.). However, weak or absent phosphorylation on Ser¹³⁶ might be a common phenomenon shared by diverse tumor cells. One study showed efficient Bad phosphorylation at Ser¹³⁶ in normal melanocytes, whereas it was only minimally detected in melanoma cell lines, although the levels of Bad protein were similar in the two cell types (29). Another study reported a similar phenomenon (30). Thus, the association between Akt and Bad cannot be confirmed in our study; further investigation of the mechanisms underlying Akt-associated apoptotic induction in enzastaurin-treated cancer cells need to be performed. Suggesting mechanisms for enzastaurin-induced apoptosis in gastric cancer, including the newly identified Rsk-mediated and Bad-mediated pathways, are illustrated in Fig. 6.

In summary, enzastaurin induced gastric cancer cell death via mitochondria and caspase-mediated pathways without cell cycle specificity. We showed that enzastaurin induced apoptosis in gastric cancer cells via a newly identified mechanism through Rsk-mediated and Bad-mediated pathways, in addition to the Akt signaling cascade. Furthermore, enzastaurin had synergistic or additive activity with other cytotoxic agents in gastric cancer cells. Our data provide a platform for further evaluation of the novel and orally available PKCβ inhibitor, enzastaurin, as a potential therapeutic agent for the treatment of gastric cancer.

Grant support: Seoul National University Bundang Hospital Research grant 02-2007-030 and Eli Lilly Company.

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.