Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

(E)-(1S,4S,10S,21R)-7-[(Z)-ethylidene]-4,21-diisopropyl-2- oxa-12,13-dithia-5,8,20,23-tetraazabicyclo[8,7,6]-tricos-16-ene-3,6,9,19,22-pentanone (FR901228; NSC-630176), a bicyclic depsipeptide isolated from Chromobacterium violaceum cultures, possesses potent antitumor effects for murine and human tumor cells derived from breast, lung, and colon adenocarcinomas (Ueda et al., 1994a,b,c; Wang, 1999), and its clinical value is currently under investigation in clinical trials. We have shown that treatment of MCF7 or MDA-MB231 cultures with FR901228 resulted in increased expression of p21^Cip1 in a p53-independent manner and phosphorylation of Bcl-2, which may contribute to cell growth inhibition and apoptosis (Rajgolikar et al., 1998). Induction of p21^Cip1 expression plays an important role in growth arrest of cells in the G₁ phase of the cell cycle and cell senescence (Chang et al., 2000). Treatment of human B-cell chronic lymphocytic leukemia cells with FR901228 resulted in induction of apoptosis, increased expression of Bax, and a decreased level of p27^Kip1, in a dose-dependent manner (Byrd et al., 1999). FR901228 also showed inhibitory activity to histone deacetylase that results in accumulation of acetylated histone species in cells and may play a critical role in growth arrest of cells and morphological reversion of transformed cells (Nakajima et al., 1998; Zhu et al., 2001). On the other hand, pretreatment of T-cell hybridoma cultures with FR901228 blocked CD3 activation-induced apoptosis as well as the suppression of c-Myc and Fas ligand expression (Wang et al., 1998). Since both c-Myc and Fas ligand expression are important in induction of apoptosis, the suppression of c-Myc and Fas ligand expression may contribute to the activity of FR901228 to block CD3-induced apoptosis of T cells. The discrepancy between the activities of FR901228 on induction and blockage of apoptosis of different cell types remains to be clarified. Treatment of ras-transformed NIH3T3 cells with FR901228 resulted in reversal of cell morphology and reduction of c-Myc expression (Ueda et al., 1994c), indicating that FR901228 may block the Ras-induced signaling pathway. However, the molecular mechanism or target for FR901228 to block the Ras-induced pathway has not been revealed.

Growing evidence has shown that ras-induced cellular transformation results in increased susceptibility of cells to anticancer agents, such as etoposide VP16 (Chen et al., 1997) and lovastatin (Chang et al., 1998). Ras mediates signals initiated from membrane-associated growth factor receptors to downstream signaling pathways, such as the extracellular signal-regulated kinase (ERK) pathway that consists of Raf, Mek, Erk, and Rsk and the phosphoinositide 3-kinase (PI3-K) pathway that consists of PI3-K, phosphoinositide-dependent kinase, Akt, and S6K (Campbell et al., 1998). Activation of the ERK pathway results in induction of cell proliferation or differentiation, and induction of the PI3-K pathway plays a critical role in protein synthesis and suppression of apoptotic signaling elements for cell survival (Jarpe et al., 1998). Activation of Ras may lead to induction of other members of the Ras superfamily including Rho-, Rac-, and cdc42-dependent pathways and may lead to changes in cell morphology (Campbell et al., 1998). Aberrant induction of the Ras-induced pathways may result in cellular transformation (Campbell et al., 1998). Many studies have shown that induction of the ERK or PI3-K pathway protects cells from apoptosis, and suppression of these pathways leads to induction of apoptosis (Jarpe et al., 1998). Besides the ERK pathway, two other mitogen-activated protein kinase pathways that are involved in stress-activated cellular events, JNK and p38, also play roles in mediating signals downstream from Ras (Campbell et al., 1998; Jarpe et al., 1998). Constant activation of the JNK or p38 pathway in cells may result in induction of caspases leading to apoptosis (Jarpe et al., 1998; Nunez et al., 1998). Induction of Ras-involved signaling pathways through increased expression of Ras or mutation of the ras gene has been postulated in development of many human cancers (Campbell et al., 1998). Studying specific agents that may exhibit selectivity on ras-transformed cells is a current approach to identify effective agents for anticancer treatment. It is important to understand the molecular mechanism of each anticancer agent to develop a combined adjuvant for treating cancers involving aberrant induction of Ras activity.

In this communication, we present evidence indicating a discriminating activity of FR901228 to regulate signaling pathways in ras-transformed 10T1/2 cells differently from their counterpart pathways in nontransformed 10T1/2 cells, which may contribute to the selectivity of FR901228 to induce apoptosis of ras-transformed 10T1/2 cells whereas it induces growth arrest of nontransformed counterpart cells in G₀/G₁ phase of the cell cycle.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Reagents. Mouse 10T1/2 fibroblast cell lines transformed by the H-ras gene or transfected with nascent vector (gifts from Drs. S. J. Parsons at University of Virginia and E. J. Taparwosky at Purdue University) were maintained in Basal medium Eagle (Invitrogen, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum, 50 units/ml penicillin, and 5 µg/ml streptomycin and cultivated at 37°C. Stock aqueous solutions of FR901228 (Chemistry and Synthesis Branch, the National Cancer Institute, Bethesda, MD) and staurosporine (Bertrand et al., 1994) (Sigma-Aldrich, St. Louis, MO) were prepared in dimethyl sulfoxide and were diluted in culture medium before treatment of cultured cells.

Cytotoxicity Assay. Cells were grown in 12-well culture plates. After treatment, cells were trypsinized, washed, resuspended in a staining buffer containing Basal medium Eagle and 0.2% trypan blue for 2 min, and counted for live and dead cells in a hemocytometer (Rajgolikar et al., 1998).

Flow Cytometric Analysis of Cell Population. Cultures were rinsed with phosphate-buffered saline; cells were trypsinized in culture dishes, rinsed with Ca²⁺ and Mg²⁺ free phosphate-buffered saline, fixed in 70% cold ethanol, and stained for 30 min with 10 µg/ml propidium iodide (Darzynkiewicz et al., 1994). Flow cytometry analysis was performed on the Coulter EPICS Elite Cytometer (Beckman Coulter, Inc., Fullerton, CA) using 15-mW air-cooled argon laser to produce 488-nm light. Phosphatidylinositol fluorescent light emission was collected with a 610LPDC filter. Extended analysis of DNA content and calculation of the percentage of cells in each phase of the cell cycle were performed on Multicycle software (Phoenix Flow System Inc, San Diego, CA).

Detection of Apoptosis. Morphological changes related to apoptosis were detected microscopically. Internucleosomal DNA degradation in apoptotic cells was detected by the method of terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate biotin nick-end labeling (TUNEL) (Gavrieli et al., 1992). Briefly, cells were seeded and cultivated on an eight-chamber slide and treated with FR901228 for 48 h. Cultured cells were labeled with digoxigenin-conjugated nucleotides by terminal deoxynucleotidyl transferase and complexed with peroxidase-conjugated antidigoxigenin antibody, stained with peroxidase substrate, counterstained with methyl green, and detected microscopically using the ApopTag peroxidase detection system as suggested by the manufacturer (Intergen, Purchase, NY). The percentage of apoptotic cells was determined by counting stained apoptotic cells and entire population of cells in a culture (BD Biosciences, San Jose, CA) microscopically.

Preparation of Cell Lysates. Cell pellets were incubated in lysis buffer (10 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, 5 mM EDTA, 10 mM sodium pyrophosphate, 10% glycerol, 0.1% Na₃VO₄, 50 mM NaF, pH 7.4) on ice for 10 min (Mehta et al., 1998). Cell lysates (S20) were isolated from the supernatants after centrifugation of crude lysates at 20,000g for 20 min. Protein concentration in cell lysates was measured using the BCA assay (Pierce Chemical, Rockford, IL).

Western Immunoblotting. Proteins were resolved by electrophoresis in 10 or 14% SDS-polyacrylamide gels and transferred to a nitrocellulose filter (pore size, 0.4 µ; Invitrogen). Nonspecific protein-binding sites on the filter were saturated by incubation of the filter with 3% nonfat milk or 5% bovine serum albumin in STT buffer (10 mM Tris-HCl, pH 7.2; 150 mM NaCl, 0.05% Tween 20) at ambient temperature for 30 min. Filters were then incubated with the specific primary antibody at 4°C for 15 h. Specific antibodies to Raf-1, Mek, Akt, p38, and active fragment p17 of caspase-3 were purchased from New England BioLabs (Beverly, MA). Specific antibodies to phosphorylated/activated forms of Raf1, Mek1, p44/42^Erk, Akt, and p38 were also purchased from New England BioLabs. Specific antibodies to Ras, p44/42^Erk1/2, p21^Cip1 (Ab-6), p53 (PAb 421), and p32 of procaspase-3 were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The filters were rinsed five times and incubated with horseradish peroxidase-conjugated antibodies at ambient temperature for 30 min. Antigen-antibody complexes on filters were detected by the SuperSignal chemiluminescence kit as indicated by the manufacturer (Pierce Chemical) and visualized by autoradiography.

Ras Activity Assay. Glutathione transferase-fused Ras-binding domain (aa51-131 of Raf1) (GST-RBD) was prepared and complexed with glutathione agarose (de Rooij and Bos, 1997), quantified by staining peptides with Coomassie Brilliant Blue and compared with protein standard of bovine serum albumin after being resolved in SDS-polyacrylamide gel electrophoresis. Twenty micrograms of cell lysates were incubated with 1 µg of GST-RBD at 4°C for 30 min. After washes, active Ras in the complex with GST-RBD-glutathione agarose was detected by Western immunoblotting with specific antibody to Ras (Santa Cruz Biotechnology Inc.).

Immunoprecipitation. Immunoprecipitation of p63^Krs1-related kinases was carried out by incubation of S20 with the specific antibody Ab-KQ (Wang and Fecteau, 2000) at 0°C for 1 h. Immune complexes were adsorbed to Pansorbin (Staphylococcus aureus) (Calbiochem, San Diego, CA) at 0°C for 30 min and washed with STE buffer (50 mM Tris-HCl, pH 7.0, 150 mM NaCl, 1 mM Na₂EDTA) supplemented with 0.1 mM dithiothreitol and 0.1% Nonidet P-40, and ST buffer (10 mM Tris-HCl, pH 7.2, 150 mM NaCl) supplemented with 1 mM dithiothreitol. Washed immune complexes were resuspended in 1× SDS sample buffer and boiled for 2 min.

In-Gel Kinase Assay. Myelin basic protein (MBP) (Invitrogen) was used as a substrate, and the in-gel kinase assay was performed as previously described (Wang and Fecteau, 2000). Briefly, 10% SDS-polyacrylamide gel was copolymerized with 0.4 mg/ml MBP. Cellular proteins were resolved in the MBP-immobilized SDS-polyacrylamide gel electrophoresis, followed by rinsing the gel with 20% isopropanol in buffer (100 mM Tris, pH 8). The gel was denatured in buffer B (100 mM Tris, pH 8, 5 mM -mercaptoethanol) supplemented with 6 M guanidine-HCl and renatured with buffer B supplemented with 0.04% Tween 40. The kinase reaction was carried out by incubating the gel with kinase buffer (20 mM Tris, pH 7.2, 10 mM MgCl₂, 15 mM -glycerolphosphate) supplemented with 50 µM ATP and 50 µCi of [-³²P]ATP at 22°C for 30 min. The gel was washed with 1% sodium pyrophosphate in 5% trichloroacetic acid. Protein kinases that phosphorylated MBP in the gel were detected by autoradiography of the -³²P-labeled MBP.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Selectivity of FR901228 to ras-Transformed 10T1/2 Cells. In the study of cytotoxic activity of FR901228, we detected a dose- and time-dependent activity of FR901228 to induce growth inhibition and apoptosis of ras-transformed and nontransformed counterpart 10T1/2 cells (Fig. 1, A and B). We also detected that treatment of ras-transformed 10T1/2 cells with a discriminating dose of FR901228 at 1 nM concentration induced apoptosis (Fig. 1C), whereas this treatment induced growth arrest of nontransformed counterpart cells in G₀/G₁ phase of the cell cycle (Table 1). Cultures of ras-transformed (Fig. 1A, top right panel) and nontransformed counterpart (Fig. 1A, top left panel) 10T1/2 cells were treated with different concentrations of FR901228 at 0.2, 1, and 5 nM for 24 and 48 h. FR901228 treatment induced growth inhibition and cell death of ras-transformed 10T1/2 cells by 24 and 48 h, respectively (Fig. 1A, top right panel). By 48 h, cultures of FR901228-treated ras-transformed cells exhibited apoptotic morphology including nuclear condensation, cytoplasmic blebbing, and cell shrinkage (Fig. 1B, panels f, g, and h). In contrast, treatment of nontransformed 10T1/2 cells with 0.2 or 1 nM FR901228 resulted in suppression of cell growth (Fig. 1A, top left panel). Furthermore, the TUNEL assay, used to detect internucleosomal DNA degradation in apoptotic cells, revealed a higher population of apoptotic cells in ras-transformed cultures (approximately 25%) (Fig. 1C, panel b) than in nontransformed counterpart cultures (approximately 5%) treated with 1 nM FR901228 (panel a). The extent of cell death detected by the TUNEL assay in untreated growing nontransformed 10T1/2 cultures was similar to 1 nM FR901228-treated nontransformed cultures (data now shown), indicating that FR901228 treatment did not induce significant cell death in the growth-arrested nontransformed 10T1/2 cultures. However, treatment of nontransformed 10T1/2 cells with 5 nM FR901228 induced growth inhibition by 24 h and cell death by 48 h (Fig. 1A, top left panel) and cell shrinkage (Fig. 1B, panel d). Consistent with previous studies using NIH3T3 cells (Wang and Fecteau, 2000), flow cytometric analysis showed an increased proportion of FR901228-treated nontransformed 10T1/2 cell population in G₀/G₁ phase (Table 1), indicating that FR901228 induced growth arrest of cells in G₀/G₁ phase of the cell cycle. In addition, we noted that the ras-transformed cells that survived FR901228 treatment were also mainly arrested in G₀/G₁ phase of the cell cycle (Table 1). It is important to note that results showing sensitivity to FR901228 were similar among multiple lines of ras-transformed 10T1/2 cells. Results obtained using FR901228-treated nontransformed and vector-transfected nontransformed 10T1/2 cultures were also similar to each other. As a comparison, ras-transformed and nontransformed 10T1/2 cultures showed similar degrees of growth inhibition and induction of apoptosis in response to a potent apoptosis-inducing agent staurosporine (Fig. 1A, bottom panels). Accordingly, selectivity of FR901228 to ras-transformed 10T1/2 cells is unlikely to be a cell clone-related event.

View larger version (96K): [in this window] [in a new window]   Fig. 1.   Cytotoxicity of FR901228 for nontransformed and ras-transformed 10T1/2 cells. A, nontransformed (left panels) and ras-transformed (right panels) 10T1/2 cultures were treated with 0.2, 1, and 5 nM FR901228 (FR) (top panels) or 1, 5, and 25 nM staurosporine (STSP) (bottom panels) for 48 h (hrs). At 24 and 48 h, cells were harvested by trypsinization and stained with 0.2% trypan blue. Relative cell survival rate was determined by counting live and dead cells. Each value represents a mean of triplicates, and error bars represent standard deviation. B, cell morphology of growing cultures of nontransformed (a, b, c, d) and ras-transformed (e, f, g, h) 10T1/2 cells treated with 0 (a and e), 0.2 (b and f), 1 (c and g), or 5 nM FR901228 (d and h) for 48 h. C, nontransformed (a) and ras-transformed 10T1/2 (b) were treated with 1 nM FR901228 for 48 h. Internucleosomal DNA degradation in apoptotic cell nuclei was detected by the TUNEL assay using ApopTag peroxidase agents as indicated by arrows.

View this table: [in this window] [in a new window]   TABLE 1 Flow cytometric analysis of 10T1/2 and ras-transformed 10T1/2 cultures treated with FR901228 Proliferating cultures of nontransformed and ras-transformed 10T1/2 cells were maintained in complete Delbecco's modified Eagle's medium (FR901228) or treated with 1 nM FR901228 (+FR901228) for 48 h. Cultures were trypsinized and cells were washed, fixed, and stained for flow cytometry to analyze DNA contents and determine cell population in each phase of the cell cycle. Each value represents a mean of duplicates. Standard error was ±3%.

Induction of p21^Cip1 in Growth-Arrested Cells by FR901228. Increases of p21^Cip1 expression were induced in growth-arrested nontransformed 10T1/2 cultures by FR901228 treatment in a dose-dependent manner (Fig. 2A, lanes 2 and 3). To investigate the ability of FR901228 to induce growth arrest versus apoptosis between nontransformed and ras-transformed 10T1/2 cells, respectively, we studied regulation of p21^Cip1, the Cdk2 inhibitor that is often associated with growth inhibition (Chang et al., 2000). In contrast to induction of p21^Cip1 in nontransformed 10T1/2 cells by FR901228, treatment resulted in a decrease of p21^Cip1 expression in ras-transformed cells (Fig. 2A, lanes 5 and 6). Study of expression kinetics of p21^Cip1 indicated that elevation or suppression of p21^Cip1 was induced in nontransformed or ras-transformed 10T1/2 cells (Fig. 2B, lanes 2 and 5), respectively, by FR901228 in 24 h. In addition, consistent with our previous study that treatment of human mammary adenocarcinoma MCF7 and MDA-MB231 cells with FR901228 resulted in increased expression of p21^Cip1 in a p53-independent manner (Rajgolikar et al., 1998), FR901228 treatment did not result in induction of p53 in either nontransformed or ras-transformed cultures undergoing growth arrest or apoptosis, respectively (data not shown).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Regulation of p21Cip1 expression by FR901228. A, proliferating nontransformed and ras-transformed 10T1/2 cultures (lanes 1 and 4) were treated with 0.2 (lanes 2 and 5) or 1 nM FR901228 (FR) (lanes 3 and 6) for 48 h. B, proliferating nontransformed and ras-transformed 10T1/2 cultures (lanes 1 and 4) were treated with 1 nM FR901228 for 24 (lanes 2 and 5) or 48 (lanes 3 and 6) h. Cell lysates were analyzed to determine expression levels of p21Cip1 by Western immunoblotting using specific antibodies. Bars indicate p21Cip1.

Induction of Caspase-3 in Apoptotic Cultures by FR901228. Induction of active caspase-3 (Fig. 3A) correlated closely with the extent of apoptosis induced in nontransformed and ras-transformed 10T1/2 cultures by FR901228 in a dose-dependent manner. To investigate the differential ability of FR901228 to induce apoptosis between nontransformed and ras-transformed 10T1/2 cells, we studied regulation of caspase-3 activity as an index for induction of apoptosis (Nunez et al., 1998). We detected that the basal levels of procaspase-3 and caspase-3 active fragment were higher in ras-transformed 10T1/2 cells (Fig. 3A, lane 4) than in nontransformed counterpart cells (lane 1). Treatment of nontransformed 10T1/2 cells with the discriminating dose of 1 nM FR901228 that induced growth inhibition also induced a modest level of procaspase-3 but not expression of active caspase-3 (Fig. 3A, lane 2). In contrast, 1 nM FR901228 induced apoptosis of ras-transformed 10T1/2 cells and active caspase-3 in concert with decreased level of procaspase-3 (Fig. 3A, lane 5). Treatment of cultures of ras-transformed or nontransformed 10T1/2 cells with 5 nM FR901228 resulted in induction of apoptosis and significant expression of active caspase-3 (Fig. 3A, lanes 3 and 6). In addition, a higher level of procaspase-3 was also detected in ras-transformed NIH3T3 cells (Fig. 3B, lane 4) than in nontransformed counterpart cells (lane 3).

View larger version (40K): [in this window] [in a new window]   Fig. 3.   Regulation of caspase-3 activation by FR901228. A, proliferating nontransformed and ras-transformed 10T1/2 cultures (lanes 1 and 4) were treated with 1 (lanes 2 and 5) or 5 nM FR901228 (FR) (lanes 3 and 6) for 48 h. Cell lysates were analyzed to determine expression levels of 32 kDa procaspase-3 and active 17 kDa fragment of caspase-3 by Western immunoblotting using specific antibodies to procaspase-3 and caspase-3 active fragment, respectively. B, levels of procaspase-3 in proliferating nontransformed (lanes 1 and 3) and ras-transformed (lanes 2 and 4) 10T1/2 or NIH3T3 cultures were detected by Western immunoblotting using a specific antibody to procaspase-3. Bar indicates procaspase-3. Arrow indicates active from of caspase-3.

Modulation of the ERK Signaling Pathway in Cells by FR901228. Specific antibodies to detect activation-related phosphorylation of kinases were used to determine the kinase activity in comparison with their cognate protein expression in Western immunoblotting. We found that the overall kinase activity of p44/42^Erk1/2 was significantly reduced in ras-transformed 10T1/2 cells by FR901228 treatment. As shown in Fig. 4A, treatment of ras-transformed cells with FR901228 resulted in decreased kinase activity of p44/42^Erk1/2 but did not induce significant changes in protein level of Erk1/2. Accordingly, the specific kinase activity of p44/42^Erk1/2 was suppressed in ras-transformed cells by FR901228 treatment in a dose-dependent manner. It was notable that the discriminating dose of 1 nM FR901228 induced a significant suppression of Erk activity (Fig. 4A, lane 3). Apparently, the ERK pathway in ras-transformed 10T1/2 cells is a target for FR901228 activity. To determine whether oncogenic Ras was targeted by FR901228 activity, we studied the level of active Ras in association with a specific Raf peptide containing the Ras-binding domain (de Rooij and Bos, 1997) and protein expression in ras-transformed 10T1/2 cells treated with FR901228. As shown in Fig. 4B, neither the Ras activity (lane 2) nor the protein expression (lane 4) was affected in ras-transformed cultures undergoing apoptosis induced by FR901228 compared with untreated counterpart cultures (lanes 1 and 3). The result indicated that oncogenic Ras is not a target for FR901228 activity. To investigate the target site(s) of FR901228 activity in the ERK pathway downstream from Ras (Campbell et al., 1998), we studied the regulation kinetics of kinase activity and protein expression of Erk1/2 (Fig. 4E) and upstream regulators Mek (Fig. 4D) and Raf-1 (Fig. 4C) in ras-transformed 10T1/2 cells using the discriminating dose of 1 nM. We detected that the overall kinase activity of Raf was decreased in parallel with decreases of cognate protein expression in a time-dependent manner (Fig. 4C). Suppression of Raf activity appeared to be attributed to the reduction of protein expression. In concert with decreased Raf activity (Fig. 4C), both the overall kinase activities of Mek (Fig. 4D) and Erk (Fig. 4E) were reduced. In concert with reduced Raf expression level, -actin level was also reduced in cells undergoing apoptosis induced by FR901228 (Fig. 4F). However, we did not detect any significant changes in the protein expression of Mek and Erk. The result indicated that the specific kinase activities of Mek and Erk were down-regulated in concert with the decreased overall Raf activity in ras-transformed 10T1/2 cells undergoing FR901228 treatment.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Regulation of ERK pathway in ras-transformed 10T1/2 cells by FR901228. A, proliferating cultures of ras-transformed 10T1/2 cells were treated with 0.2, 1, or 5 nM FR901228 (FR) for 48 h. B, proliferating cultures of ras-transformed 10T1/2 cells were treated with 1 nM FR901228 for 48 h. Cell lysates were analyzed by Western immunoblotting to detect active Ras in association with RBD peptide (lanes 1 and 2) or Ras protein in cell lysates (lanes 3 and 4). C, D, E, and F, proliferating cultures of ras-transformed 10T1/2 cells were treated with 1 nM FR901228 for 24 or 48 h. Cell lysates were analyzed by Western immunoblotting with specific antibodies to detect levels of (C) phosphorylated/activated Raf-1 (p-Raf-1) and Raf-1 protein, (D) phosphorylated/activated Mek (p-Mek) and Mek protein, (A and E) phosphorylated/activated Erk1/2 (p-Erk1/2) and Erk1/2 protein, and (F) -actin. Arrows indicate phosphorylated kinases p-Raf-1, p-Mek, and p-Erk1/2. Bars indicate cognate protein expression of Ras, Raf, Mek, Erk1/2, and -actin.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Regulation of ERK pathway in nontransformed 10T1/2 cells by FR901228. A, proliferating cultures of nontransformed 10T1/2 cells were treated with 0.2, 1, or 5 nM FR901228 (FR) for 48 h. B, C, D, and E, proliferating cultures of nontransformed 10T1/2 cells were treated with 1 nM FR901228 for 24 or 48 h. Cell lysates were analyzed by Western immunoblotting with specific antibodies to detect levels of (B) phosphorylated/activated Raf-1 (p-Raf1) and Raf-1 protein, (C) phosphorylated/activated Mek (p-Mek) and Mek protein, (A and D) phosphorylated/activated Erk1/2 (p-Erk1/2) and Erk1/2 protein, and (E) -actin. Arrows indicate phosphorylated kinases p-Raf-1, p-Mek, and p-Erk1/2. Bars indicate cognate protein expression of Raf-1, Mek, Erk1/2, and -actin.

Modulation of Akt in Cells by FR901228. Treatment of ras-transformed or nontransformed 10T1/2 cells by FR901228 resulted in decreases in protein expression of Akt in a dose- and time-dependent manner (Fig. 6, A and B). We used specific antibodies to detect phosphorylation of specific residues in Akt for measuring the kinase activity in comparison with its cognate protein expression in Western immunoblotting. Higher activity of Akt was detected in ras-transformed 10T1/2 cells (Fig. 6A, lane 1) than in counterpart nontransformed cells (lane 5). Suppression of the overall Akt activity was in concert with decreased levels of protein expression in ras-transformed 10T1/2 cells by FR901228 treatment (Fig. 6A, lanes 2 to 4) compared with basal level of Akt activity and protein level in untreated ras-transformed cultures (lane 1). Study of regulation kinetics of Akt activity and protein expression indicated that suppression of Akt was induced in ras-transformed 10T1/2 cell cultures by 48 h of treatment with 1 nM FR901228 (Fig. 6B, lane 3) when cultures underwent apoptosis.

View larger version (48K): [in this window] [in a new window]   Fig. 6.   Regulation of Akt in 10T1/2 cells by FR901228. A, proliferating cultures of ras-transformed 10T1/2 (lanes 1 to 4) and nontransformed counterpart cells (lanes 5 to 8) were treated with 0.2, 1, or 5 nM FR901228 (FR) for 48 h. B, proliferating cultures of ras-transformed 10T1/2 (lanes 1 to 3) and nontransformed cells (lanes 4 to 6) were treated with 1 nM FR901228 for 24 or 48 h. Cell lysates were analyzed by Western immunoblotting using specific antibodies to detect levels of phosphorylated/activated Akt (p-Akt) and Akt protein. Arrow indicates phosphorylated Akt (p-Akt), and bar indicates cognate protein expression of Akt.

Modulation of Stress-Related Pathways. To investigate whether stress-activated protein kinase signaling pathways (Campbell et al., 1998; Jarpe et al., 1998) were involved in the induction of apoptosis of cells by FR901228 treatment, we studied regulation of p38 and JNK as indexes for the pathways. We detected a substantial level of p38 activity in ras-transformed 10T1/2 cells (Fig. 7A, lane 1). FR901228 treatment of ras-transformed 10T1/2 cells resulted in suppression of p38 activity in a dose-dependent manner (Fig. 7A, lanes 2-4). However, the protein level of p38 was not changed significantly in ras-transformed 10T1/2 cultures undergoing FR901228-induced apoptosis, indicating that the specific kinase activity of p38 was down-regulated in ras-transformed 10T1/2 cultures by FR901228 treatment. On the other hand, nontransformed counterpart cultures treated with FR901228 exhibited increased expression of p38 (Fig. 7B, lanes 2-4) compared with untreated cells (lane 1). Changes in the kinase activity of p38 were not detectable, regardless of cultures undergoing growth arrest or apoptosis induced by FR901228. In contrast, UV irradiation of nontransformed cultures induced apoptosis and profound activation of p38 (Fig. 7B, lane 5). Therefore, FR901228-induced apoptosis of ras-transformed or nontransformed counterpart 10T1/2 cells was unlikely to involve activation of p38-involved stress signaling pathway.

View larger version (48K): [in this window] [in a new window]   Fig. 7.   Regulation of stress-activated pathway by FR901228. Proliferating cultures of (A) ras-transformed 10T1/2 cells and (B) nontransformed 10T1/2 counterpart cells were treated with 0.2, 1, or 5 nM FR901228 (FR) for 48 h. UV-irradiated cultures received 5 mJ/cm2. Cell lysates were analyzed by Western immunoblotting using specific antibodies to detect levels of phosphorylated/activated p38 (p-p38) and p38 protein. Arrow indicates phosphorylated p38 (p-p38), and bar indicates cognate protein expression of p38.

Activation of Krs1-Related p33 Kinase by FR901228. Induction of Krs1- or Krs2-related kinase products p33 or p36 by anticancer agents or stress shock have been reported to correlate with induction of apoptosis (Graves et al., 1998; Kakeya et al., 1998; Lee et al., 1998; Watabe et al., 1999). We have reported that induction of apoptosis of human breast cancer MCF7 and MDA-MB231 cells correlated with activation of p33 (Rajgolikar et al., 1998). During investigation of the involvement of proteolytic products of Krs1 in FR901228-induced apoptosis of ras-transformed 10T1/2 cells or growth arrest of nontransformed counterpart cells, we detected that the kinase activity of p33, which is produced through proteolytic modification of p63^Krs1 (Wang and Fecteau, 2000), was induced in growth-arrested or apoptotic cultures. As shown in Fig. 8A, the kinase activity of p33 was induced to a similar level in nontransformed 10T1/2 cultures that were undergoing growth arrest or apoptosis by 1 or 5 nM FR901228 (lanes 3 and 4), respectively. However, treatment with 1 or 5 nM FR901228 induced different degrees of apoptosis in cultures of ras-transformed 10T1/2 cells and different degrees of p33 activation (Fig. 8A, lanes 7 and 8). Increases of p33 activity were in harmony with decreases of p63^Krs1 activity in both ras-transformed 10T1/2 (Fig. 8A, lanes 7 and 8) and nontransformed counterpart cultures (lanes 3 and 4).

View larger version (59K): [in this window] [in a new window]   Fig. 8.   Regulation of p63Krs1-related kinases by FR901228. A, proliferating cultures of nontransformed (lanes 1 to 4) or ras-transformed 10T1/2 cells (lanes 5 to 8) were treated with 0.2, 1, or 5 nM FR901228 (FR) for 48 h. Kinase activities of p63Krs1 and p33 in cell lysates (S20) were determined. B, cultures of ras-transformed 10T1/2 cells were treated with 1 nM FR901228 or 5 nM staurosporine (SP) for 48 h. The immune complexes of p63Krs1, p36, and p33 were prepared from 10 µg of S20 (lanes 5 and 6) incubated with 1 µl of a Krs1-specific antibody Ab-KQ (IP) (lanes 1 and 2) in the presence of 1 µg of antigen peptides (+ Ag) (lanes 3 and 4). The kinase activities of p63Krs1, p36, and p33 in S20 or IP were analyzed by the in-gel kinase assay using MBP as a substrate. Arrows indicate p63Krs1, p36, and p33.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Investigation of differential regulation of intracellular signaling pathways in cancerous cells and their normal counterpart cells treated with anticancer agents provides cellular and molecular basis for developing strategies to selectively induce apoptosis of cancerous cells while bypassing normal counterpart cells. Malignant transformation of cells is accompanied by molecular and cellular changes, such as changes in chromosomal integrity, gene expression, intracellular signal transduction, subcellular structure, cellular interaction, and/or cell motility. These changes contribute to different susceptibilities of malignantly transformed cells from their normal counterpart cells in response to various forms of stress. Our efforts on studying selectivity of FR901228 to particular types of malignantly transformed cells and identification of molecular targets for the agent may provide a basis for developing new strategies using combined anticancer agents to control cancers. Results from this study indicated a dose-dependent activity of FR901228 to induce apoptosis of ras-transformed fibroblasts, whereas the treatment induced growth arrest of nontransformed counterpart cells in G₀/G₁ phase of the cell cycle. Investigation of signaling pathways in ras-transformed and nontransformed counterpart cells also produced important information for the development of an anticancer agent that may be capable of selectively inducing apoptosis of malignantly transformed cells involving Ras activation, bypassing normal counterpart cells.

To eliminate misrepresentation of results from any one of available assays, we used three independent parameters to determine the induction of apoptosis in cultured cells by FR901228, including decreased cell growth survival rate, apoptotic morphology, and increased internucleosomal DNA degradation (TUNEL assay). Consistent with the observation of others that apoptotic 10T1/2 cells detach from culture plates (Tomei et al., 1993), we detected that a significant population of apoptotic 10T1/2 or ras-transformed counterpart cells with extended internucleosomal DNA degradation were floating with cell debris in cultures. Accordingly, study of relative survival rate of anchored cells in culture revealed cell proliferation, growth inhibition, and cell death semiquantitatively. Detection of apoptotic morphology and internucleosomal DNA degradation in anchored population of cells in culture was able to determine apoptotic-like cell death qualitatively. Additionally, flow cytometric analysis of anchored population of cells determined growth arrest of cells in G₀/G₁ phase of the cell cycle by FR901228 treatment. Results from these assays taken together are competent to determine the biological activity of FR901228 to induce growth arrest or apoptosis.

Cellular transformation by the ras gene of cultured 10T1/2 fibroblast cells resulted in induction of the ERK, PI3-K, and p38 pathways and increased expression of procaspase-3. Using 1 nM, a discriminating dose that induced apoptosis of ras-transformed 10T1/2 and caused growth arrest of nontransformed counterpart cells, we detected differential regulation of the ERK, PI3-K, and p38 signaling pathways in ras-transformed 10T1/2 and nontransformed counterpart cells. Inhibition of the ERK pathway in ras-transformed 10T1/2 cells by FR901228 appears to be due to decreased Raf activity that results from decreased protein content, leading to down-regulation of downstream kinases Mek and Erk. Since suppression of the ERK pathway in cells has been shown to potentiate cell apoptosis (Campbell et al., 1998; Jarpe et al., 1998), suppression of the ERK pathway may play a role in the induction of apoptosis of ras-transformed 10T1/2 cells by FR901228. Raf protein content was also reduced significantly in nontransformed 10T1/2 cells; however, FR901228 treatment resulted in increased specific kinase activity of Raf that appeared to moderately induce the downstream kinases Mek and Erk. This observation is consistent with reports that a prolonged increase in ERK activity results in growth arrest of NIH3T3 cells with concomitant induction of p21^Cip1 (Pumiglia and Decker, 1997). Modest induction of the ERK pathway contributes to growth arrest of some human breast cancer cells and p53-independent induction of p21^Cip1 (Lee et al., 1999). We have also detected increased expression of p21^Cip1 in nontransformed 10T1/2 cells by FR901228 treatment. It has been shown that p21^Cip1-mediated growth arrest of a human fibrosarcoma cell line is accompanied by induction of several genes encoding secreted proteins with mitogenic or antiapoptotic activity (Chang et al., 2000). Up-regulation of p21^Cip1 in human breast adenocarcinoma or sarcoma cells decreases apoptosis induced by anticancer agent paclitaxel through inhibition of Cdc2, growth arrest of cells, and exiting from abnormal mitosis (Yu et al., 1998; Li et al., 1999). It is possible that up-regulation of p21^Cip1 and the ERK pathway contributed to the survivability of nontransformed 10T1/2 cells undergoing FR901228 treatment. In contrast, lack of up-regulation of p21^Cip1 in ras-transformed 10T1/2 cells contributed to the increased susceptibility to FR901228-induced cell death. However, the role of increased Erk1/2 activity accompanied with increased expression of p21^Cip1 in the survival of FR901228-treated nontransformed 10T1/2 cells still remains to be determined.

Since activity of the PI3-K pathway plays a critical role in cell survival (Campbell et al., 1998; Jarpe et al., 1998), the discrepancy in regulation of the specific kinase activity of Akt between ras-transformed 10T1/2 and nontransformed counterpart cells by FR901228 may contribute to its selectivity to induce apoptosis of ras-transformed 10T1/2 cells. The differential regulation of Akt appears to be similar to the regulation of Raf in FR901228-treated ras-transformed 10T1/2 and nontransformed counterpart cells. Both protein contents of Akt and Raf were significantly reduced in ras-transformed and nontransformed cells. The overall kinase activities of Akt and Raf were reduced in concert with decreased protein contents in ras-transformed 10T1/2 cells. However, the overall kinase activities of Akt and Raf were not reduced in nontransformed 10T1/2 cells treated with FR901228. Accordingly, the specific kinase activities of Akt and Raf were elevated in nontransformed 10T1/2 cells undergoing growth arrest by FR901228 treatment. However, the mechanism that enhanced the specific kinase activities of Akt and Raf in growth-arrested nontransformed 10T1/2 cells remains to be addressed.

FR901228-induced apoptosis or growth inhibition was unlikely to involve induction of the stress-activated pathways in ras-transformed 10T1/2 or nontransformed cells, respectively. Instead, deactivation of p38 in ras-transformed 10T1/2 cells appeared to correlate with the induction of apoptosis by FR901228 treatment. A recent study indicated that activation of p38 is crucial in expression of collagenases for the invasive phenotype of transformed cells (Johansson et al., 2000). However, whether deactivation of p38 contributes to induction of apoptosis of ras-transformed cells by FR901228 is unclear.

Induction of apoptosis in cultures of ras-transformed or nontransformed 10T1/2 cells by FR901228 treatment was accompanied by production of active caspase-3. Apparently, induction of caspase-3 activity is involved in FR901228-induced apoptosis. We also detected higher expression levels of procaspase-3 in ras-transformed 10T1/2 and NIH3T3 cells than in nontransformed counterpart cells. Whether the increased basal levels of procaspase-3 and active caspase-3 in ras-transformed 10T1/2 cells may have potentiated the caspase cascade's susceptibility to specific agents such as FR901228 to execute apoptosis remains to be determined.

We have reported a Krs1-related p33 kinase, the activation of which is correlated highly with growth arrest of nontransformed NIH3T3 cells in the quiescent state and the deactivation of which is correlated closely with entry of quiescent cells into the cell cycle (Wang and Fecteau, 2000). Furthermore, increase in p33 activity is correlated with induction of apoptosis in quiescent cultures undergoing stress shock (Wang and Fecteau, 2000). In this study, we noted that different degrees of p33 activity induced in ras-transformed 10T1/2 cultures correlated with degrees of cell death induced by different doses of FR901228. However, p33 activity did not change in nontransformed 10T1/2 cultures regardless of undergoing growth arrest or apoptosis induced by FR901228. The role p33 may play in the process of apoptosis remains to be characterized.

In summary, our study indicates the potential value of FR901228 in the treatment of human cancers involving aberrant activation of Ras-induced signaling pathways through its preferential induction of the caspase cascade and suppression of the ERK, PI3-K, and p38 pathways. On the other hand, induction of p21^Cip1 expression and moderate Erk1/2 activity by FR901228 treatment may help in growth arrest and cell survival of nontransformed 10T1/2 cells. The actions of FR901228 may be attributed to a discriminating dose-related activity that selectively induces apoptosis of ras-transformed 10T1/2 cells. Although the anticancer activity of FR901228 still requires further effort to reveal other molecular targets for its selectivity, our study has provided evidence for the differential activity of FR901228 in inducing cell death of ras-transformed cells, whereas nontransformed counterpart cells are growth-arrested in the cell cycle.