Tetra-O-methyl nordihydroguaiaretic acid induces growth arrest and cellular apoptosis by inhibiting Cdc2 and survivin expression

Abstract

We previously reported that Sp1-dependent Cdc2 gene expression is inhibited by tetra-O-methyl nordihydroguaiaretic acid (M₄N) and that M₄N is likely responsible for causing growth arrest in M₄N-treated transformed C3 cells. Here, we show that after M₄N treatment and cell-cycle arrest, expression of the Sp1-dependent survivin gene, a member of the inhibitor of apoptosis family, is also suppressed, and the mitochondrial apoptotic pathway is activated. To confirm that inhibition of Cdc2 and survivin gene expression is necessary for M₄N-induced growth arrest and apoptosis, we tested the effect of adding Cdc2 and survivin back to M₄N-treated cells. Cell division was transiently restored in the presence of M₄N after transfection of an exogenous Cdc2 gene copy under the control of the Sp1-independent cytomegalovirus promoter. Caspase-3 activation was also reduced by 50% and 75% in transiently and stably survivin-transfected C3 cells, respectively. The results suggest that M₄N induces growth arrest and apoptosis by suppressing Cdc2 and survivin expression, which constitutes the cellular basis of its antitumoric action.

Cancers are diverse and heterogeneous, affected by both genetic and epigenetic alterations. Yet, all cancers display deregulated cell proliferation and suppressed apoptosis (1). Development of anticancer drugs that can selectively target defects in key pathways controlling these two processes and eliminate transformed cells without damaging both regenerating and differentiated host tissues has since become the main challenges facing all oncologists today.

Our laboratory has been focusing on finding small organic compounds that can selectively arrest growth and induce apoptosis of cancer cells with minimal toxicity to normal tissues. We found earlier that an anti-HIV proviral transcriptional inhibitor, tetra-O-methyl nordihydroguaiaretic acid (M₄N) (2, 3), was able to cause growth arrest of a variety of transformed human and mouse cells in culture and in mice (4). As we have previously reported that 3-O-methyl-NDGA (Mal.4) inhibited HIV proviral transcription by specifically blocking Sp1 transcriptional factor to bind HIV long-terminal repeats (5), M₄N, a DNA major groove binder for G/C-rich sequence, was found to block cell-cycle progression at G₂/M by inhibiting the transcription of an Sp1-dependent gene coding for cyclin-dependent kinase (Cdc2). Intratumoral and i.p. injection of M₄N was exceedingly effective in eliminating C3 cells in mice (4) and murine melanoma cells (6), respectively. At the 6th International Conference on Head and Neck Cancer (August 2004, Washington, DC), it was presented that clinically evident tumor necrosis was also observed in head and neck cancer patients following intratumoral administration of M₄N. High concentrations of M₄N in long-term feeding experiment also caused no apparent toxicities to mice (7). Reasons for such differential cytocidal effects of M₄N toward tumors vs. normal adult tissues are not clear at the present time. Molecular studies reported in the present communication may provide some insights on this matter.

It is well known that apoptosis is a regulated physiological process leading to cell death. It is characterized by cell shrinkage, membrane blebbing, and nuclear condensation, mediated by the actions of caspases. Two apoptotic pathways have been described: the death receptor and the mitochondrial pathways, each of which leads to the activation of the downstream effector caspases 3 and 7, which dismantle the cell (8).

There are also genes that regulate the apoptotic pathways. These include survivin gene whose product is an inhibitor of apoptosis protein that has been implicated in integrating cell division and apoptosis (9). Survivin is overexpressed in most cancers but undetectable in most terminally differentiated normal adult tissues (10, 11), with the exception of thymocytes (10), CD34⁺ stem cells (12), and basal colonic epithelial cells (13). Means of targeting survivin such as antisense and ribozyme have been exploited to induce apoptosis in cancer cells (14–16). The expression of survivin is G₂/M cell-cycle specific and is highly Sp1-dependent, with transcriptional activity requiring two critical Sp1 sites (17, 18). In the present study, we show that M₄N not only can inhibit Cdc2 gene expression, but is also an effective inhibitor for transcription of Sp1-dependent survivin gene. Apoptosis was induced when survivin production was inhibited in M₄N-treated C3 cells. Such inhibitory effects were found to be remarkably specific for transcriptions of Sp1-regulated endogenous survivin and Cdc2 genes. For example, both cell arrest and caspase-3 activity can be substantially reversed in M₄N-treated C3 cells when transfected with exogenous Cdc2 and survivin gene constructed with an Sp1-independent cytomegalovirus (CMV) promoter, respectively.

The histology of skin, spine, and muscle tissues from mice injected s.c. with M₄N dissolved in DMSO did not differ from those tissues injected with DMSO and uninjected normal tissues, while inducing apoptosis in the targeted tumor cells (19). Because there is no survivin expression, M₄N-induced apoptosis is perhaps inoperative in normal tissues. Yet M₄N functions fully in suppressing growth and causing programmed cell death in transformed cells by its dual inhibitory effect on expression of two cell-cycle related genes, Cdc2 and survivin. Development of M₄N as a tumor-selective, anticancer agent for humans seems to be worthy of consideration.

Materials and Methods

Cell Culture and M₄N Treatment. The C3 cell line (20) was obtained from W. M. Kast (University Hospital, Leiden, The Netherlands) and maintained in Iscove's modified Dulbecco's medium, supplemented with 5% FBS/10 μM 2-mercaptoethanol/100 units/ml penicillin and streptomycin. M₄N was dissolved at 70°C for 3 h in Hybri-max DMSO (Sigma).

Terminal Deoxynucleotidyltransferase-Mediated dUTP Nick End Labeling Assay. C3 cells were grown on coverslips followed by M₄N treatment for 3 days. Labeling of fragmented DNA was carried out with the Cell Death Detection kit (Roche Diagnostics) according to the manufacturer's protocol. The cells were analyzed on a Zeiss Axioplan microscope, and apoptotic cells in three random fields were scored.

Antibodies, Western Blotting, and Kinase Assay. The antibodies used for Western blotting analysis were the antibodies against mouse caspase-3, cyclin B, and survivin (Santa Cruz Biotechnology), mouse caspase-8 (Pharmingen), mouse caspase-9 and cytochrome c (Cell Signaling Technology, Beverly, MA), and mouse Cdc2 (Oncogene Research Products, Cambridge, MA). Cell lysates were prepared in lysis solution (4). For Western blots, proteins were analyzed with Western Light Plus Protein Detection kit (Tropix, Bedford, MA). For Cdc2 kinase assay, C3 cells were lysed with cold lysis buffer (10 mM Tris, pH 7.4/1.0% Triton X-100/0.5% Nonidet P-40/150 mM NaCl/20 mM sodium fluoride, 1× protease inhibitor mixture); 300 μg of protein of cell lysate was incubated with 3 μg of monoclonal cyclin B1 antibody for 1 h at 4°C with gentle agitation. Protein A/G agarose pellets were equilibrated with immunoprecipitation (IP) buffer (10 mM Tris, pH 7.4/1.0% Triton X-100/0.5% Nonidet P-40/150 mM NaCl/20 mM, 1× protease inhibitor mixture) and added to protein extract solutions. Samples were incubated at 4°C overnight, followed by three washes with IP buffer and two washes with kinase buffer (10 mM Tris·HCl, pH 7.4/150 mM NaCl/10 mM MgCl₂/0.5 mM DTT). Pellets were incubated with 50 μl of the kinase buffer containing 25 μM ATP, 2.5 μCi of [³²γ]ATP (1 Ci = 37 GBq), and 50 μg of histone H1 at 37°C for 30 min. Reactions were stopped by the addition of 6× SDS sample buffer and boiling the mixture for 5 min. Proteins were separated on an SDS/12% PAGE gel. The gel was stained with Coomassie brilliant blue, dried, and exposed to x-ray film.

Cytochrome c Release Assay. C3 cells treated with 1% DMSO or 40 μM M₄N in 1% DMSO for 3 days were harvested. Cytosolic and mitochondrial extracts were prepared. For testing cytochrome c release, a high-quality, well established mitochondrial fractionation procedure was used (21). The supernatant (cytosol) and pellet (mitochondria) from this fractionation were analyzed by Western blotting.

Caspase-3 Activity Assay. Cells were lysed in 50 mM Hepes (pH 7.4)/100 mM NaCl/0.1% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS)/1 mM DTT/0.1 mM EDTA, and 30 μg of protein of cell lysate was added to the assay buffer (50 mM Hepes, pH 7.4/100 mM NaCl/0.1% CHAPS/10 mM DTT/0.1 mM EDTA/10% glycerol) in a final volume of 90 μl in a 96-well flat bottom plate. The reaction was initiated by the addition of 10 μl of 2 mM caspase-3 chromogenic tetrapeptide substrate (Ac-DEVD-pNA) (Calbiochem) and kept at 37°C for 6 h. OD at 405 nm was measured with a 96-well microplate reader.

RT-PCR Analysis. Total RNA was isolated from cultured cells by the guanidinium thiocyanate and phenol method (22). A 343-bp RT-PCR product was generated by using survivin-specific sense (5′-GCATCGCCACCTTCAAGAACTGGCCC-3′) and antisense (5′-CGGGTAGTCTTTGCAGTCTCTTCAAACTC-3′) primers, which detected only survivin 140, the survivin species that possesses the inhibitor of apoptosis protein functions. GAPDH sense (5′-GAATCTACTGGCGTCTTCACC-3′) and antisense (5′-GTCATGAGCCCTTCCACGATGC-3′) primers were used to generate a 238-bp RT-PCR product as a control. The RT-PCR products were separated by electrophoresis on a 1.8% agarose gel containing ethidium bromide and photographed. The signal intensities were quantified by scion image (Scion, Frederick, MD).

Plasmids and Transfection. Full-length mouse survivin cDNA (kindly provided by E. M. Conway, University of Leuven, Leuven, Belgium) and full-length Cdc2 (generated by RT-PCR of RNA from C3 cells) were subcloned into pcDNA3.1-myc-His or pIRESpuro2 to generate proper expression clones for Cdc2 and survivin under a CMV promoter. Transient transfections were carried out with FuGENE 6 (Roche Applied Science) according to manufacturer's protocol. For the recovery experiment using pIRESpuro2-Cdc2, cells were then incubated with media containing 40 μM M₄N in 1% DMSO for 4 days. Cell viability was assessed by trypan blue exclusion. For the cotransfection experiment, C3 cells were cotransfected with the various cDNA-containing constructs and lacZ in pcDNA3.1-myc-His for 8 h in serum-free media, which were then replaced by complete media containing 40 μM M₄N. After M₄N treatment, cells were fixed and stained with 5-bromo-4-chloro-3-indolyl β-d-galactoside. Blue cells were scored in 10 random fields for each dish. For the caspase-3 inhibition study, cells were transfected with the Cdc2 and survivin pcDNA3.1-myc-His constructs. After drug treatment with 40 μM M₄N for 3 days, cell lysates were prepared and analyzed for caspase-3 activity as described above. Stable lines were established by transfecting C3 cells with pIRESpuro2-survivin-myc-His and selected in 2.0 μg/ml puromycin. A single clone with maximal expression of survivin was selected to be used in current studies.

Immunofluorescence. Monolayer cultures of C3 cells were fixed with 3% paraformaldehyde in serum-free media for 10 min and then with 3% paraformaldehyde in PBS for 20 min at 4°C, followed by quenching with 50 mM NH₄Cl-PBS. The fixed cells were blocked for 30 min in PBS containing 1% BSA and 0.075% saponin and incubated with primary antibody for 1 h and then with FITC-conjugated goat anti-rabbit secondary antibody. Coverslips were mounted with VECTASHIELD containing 4′,6-diamidino-2-phenylindole (Vector Laboratories) and analyzed on a Zeiss Axioplan microscope.

Results

M₄N Induces Apoptosis. Our previous data on M₄N showed that it inhibited cell proliferation and induced cell death in several mammalian cell lines (4). Cells undergoing programmed death or apoptosis exhibit characteristic morphological changes, such as DNA cleavage, nuclear condensation, and formation of apoptotic bodies. We used 4′,6-diamidino-2-phenylindole staining and the terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling assay to assess these nuclear morphological features in C3 cells treated with 40 μM M₄N for 3 days. In the terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling assay, DNA breaks are labeled by fluorescein nucleotides catalyzed by terminal deoxynucleotidyltransferase (23). In the M₄N-treated C3 cells, a significant number of nuclei was extensively fragmented and appeared bright green or light blue with a 4′,6-diamidino-2-phenylindole counter stain (Fig. 1A). Quantitatively, apoptotic cells and the total number of cells were scored in three random fields, and the percentage of apoptotic cells was obtained. As expected, the percentage of apoptotic cells increased with M₄N concentration (Fig. 1 A).

Fig. 1.

M₄N treatment induces apoptosis in C3 cells. (A) C3 cells were treated with 1% DMSO (Upper Left) or 40 μMM₄N in 1% DMSO (Upper Right) for 72 h and then assayed as described under Materials and Methods. Arrows indicate apoptotic bodies. The percentage of apoptotic cells was obtained from three random fields and expressed as mean ± SD (Lower). (B) C3 cells were treated with the indicated concentrations of M₄N and assayed for caspase-3 activity. The results are expressed as mean ± SD of three independent experiments. (C) C3 cells were treated with indicated concentrations of M₄N and analyzed by Western blotting with antibodies against caspase-3. (D) C3 cells were treated with indicated concentrations of M₄N and analyzed by Western blotting with antibodies against caspases 8 and 9 and cyclin B1. Cyclin B1 was used as a loading control. For cytochrome c release assay, C3 cells were treated with the indicated concentrations of M₄N, and cytosolic and mitochondrial extracts were analyzed by Western blotting with antibodies against cytochrome c.

To biochemically assay the induction of apoptosis by M₄N, we measured the activation of caspase-3. C3 cells were treated with M₄N for 3 days. Caspase-3 activity was quantified using a chromogenic tetrapeptide substrate (Ac-DEVD-pNA). M₄N treatment resulted in dose-dependent increase in caspase-3 activity (Fig. 1B), up to an 8.3-fold increase after treatment with 40 μMM₄N. At the higher dosage of 60 μM, however, caspase-3 activity was reduced to only 2.2 times more than the control, possibly a reflection of caspase-3 degradation as found for many cellular proteins at advanced stages of apoptosis.

The effect of M₄N on caspase-3 activation was also assessed by Western blotting using anticaspase-3 antibodies. Caspase-3 is activated by specific proteolytic processing of the 32-kDa procaspase-3 protein to the 20-kDa caspase-3 polypeptide. Western blot analysis of total protein isolated from the treated cells with anti-caspase-3 antibody demonstrates a good correlation between the dose-dependent increase in caspase-3 activity (Fig. 1B) and the appearance of the 20-kDa caspase-3 cleavage product (Fig. 1C).

We next asked which of the two apoptotic pathways, the death receptor or the mitochondrial pathway, was activated in M₄N-treated C3 cells. A same Western blot was probed sequentially with antibodies against caspases 8 and 9, and cyclin B-1, the loading control. It showed that caspase-8 was not activated, whereas caspase-9 was; this was shown by the decreasing levels of procaspase-9 with increasing concentrations of M₄N, whereas procaspase-8 and cyclin B1 remained constant (Fig. 1D). Activation of caspase-9 requires the release of cytochrome c into the cytosol from mitochondria. We investigated whether M₄N treatment caused mitochondria to expel cytochrome c. Cytosolic and mitochondrial extracts from C3 cells treated with DMSO or 40 μMM₄N were separated and then analyzed by Western blotting. In the DMSO-treated cells, cytochrome c remained in the mitochondrial extract, and only very little was present in the cytosolic extract (Fig. 1D, 0 μM, lanes M and C). However, in the M₄N-treated cells, the mitochondrial pool of cytochrome c was reduced, whereas a significant amount of cytochrome c was present in the cytosol (Fig. 1D, 40 μM, lanes M and C). This was consistent with the activation of caspase-9, suggesting the notion that M₄N induced the mitochondrial apoptotic pathway.

Down-Regulation of Survivin Expression by M₄N Treatment. Our previous studies showed that M₄N prevents Sp1 binding to its cognate binding sites in the promoter regions of Sp1-dependent genes (2, 3). Given that the expression of survivin is Sp1-dependent (17, 18), and our finding that M₄N induces caspase-3 activation, we next investigated the effect of M₄N on survivin gene expression. We performed RT-PCR analysis of survivin mRNA levels in C3 cells. The expression of GAPDH, a non-Sp1-dependent housekeeping gene not affected by M₄N treatment (4), was used for normalization. Treatment with M₄N for 3 days greatly reduced survivin mRNA levels. For cells treated with 60 μM M₄N for 3 days, expression of survivin mRNA was reduced to ≈20% compared with the cells treated with 1% DMSO (Fig. 2A). Treatment with M₄N also inhibited survivin protein levels. Western blots of proteins extracted from untreated C3 cells using anti-survivin antibodies showed a 16.5-kDa survivin protein band, whereas cyclin B1 level remained unaffected by the M₄N treatments (Fig. 2B). Treatment with M₄N for 3 days significantly reduced the level of survivin protein in response to increasing M₄N concentrations, consistent with the activation of caspase-3 and induction of apoptosis in these cells.

Fig. 2.

Effect of M₄N treatment on survivin expression. (A) C3 cells were treated with the indicated concentrations of M₄N for 72 h. Total RNA was analyzed for survivin mRNA levels by RT-PCR. GAPDH expression was used for normalization. (B) C3 cells treated with the indicated concentrations of M₄N were analyzed for survivin and cyclin B1 expression by Western blotting. Cyclin B1 was used as a loading control. (C) C3 cells treated with 1% DMSO (Left) and 40 μM M₄N in 1% DMSO (Right) for 72 h were fixed and incubated with primary antibody against survivin, FITC-labeled secondary antibody, and DAPI.

During interphase, survivin has been localized in the cell in some studies to the cytoplasm and in others to the nucleus (10, 11, 24, 25). By using fluorescent antibody staining with anti-survivin antibodies, we observed that survivin was expressed and localized to both subcellular compartments in C3 cells at interphase, although most was dispersed throughout the cytoplasm (Fig. 2C Left). Consistent with the Western blot results, treatment with 40 μM M₄N for 3 days cleared both nuclear and cytosolic pools of survivin in C3 cells (Fig. 2C Right).

M₄N-Induced Growth Arrest Is Reversible and Correlates with the Levels of Cdc2 and Cdc2/Cyclin B Activity in Short Duration. Many therapeutic approaches that induce cell growth arrest are irreversible and lead to cell death often quickly. To determine the permanence of M₄N induced growth arrest, we treated C3 cells for 3 days with sufficient amounts of drug to stop cell division (40 μM). The cells were then washed twice with PBS and given fresh media of the original type, except that half of the M₄N-treated cells were now switched to media containing only 1% DMSO. Cells viability was assessed by trypan blue exclusion. When treated with these concentrations of M₄N transiently, all cell lines previously arrested with M₄N demonstrated marked recovery of growth within 48 h after the removal of M₄N (Fig. 3A). As the time course continued, the growth rate of these cells continued to rise. Similar growth recovery was seen in a separate experiment in which the human cervical cancer cell line, C33a, was treated with M₄N, as above, and then reseeded at low density in a colony-forming assay (19). All cells retained greater than 95% viability during these experiments. We next examined the Cdc2 protein levels from cells incubated with 20 or 40 μM M₄N for 3 days followed by incubation in fresh media containing either 1% DMSO or M₄N in 1% DMSO for 3 more days. Cells from which the drug was removed (Fig. 3B, 20D and 40D) produced normal levels of Cdc2 protein as compared to cells grown with just 1% DMSO in the media throughout (DMSO). However, cells incubated with M₄N for 6 days (Fig. 3B, lanes 20 and 40) expressed much less Cdc2. In all samples, the levels of cyclin B remained equivalent. Such reduction of Cdc2 expression should also correlate with Cdc2 kinase activity. Indeed, similar recovery of Cdc2 kinase activities from M₄N treatment was observed (Fig. 3B).

Fig. 3.

Reversible Cdc2 expression, kinase activity, localization, and growth. (A) Cells were treated with 1% DMSO or 40 μMM₄N in 1% DMSO (left arrow). After 3 days, the cells were washed two times with PBS, and half of the M₄N-treated cells were switched to 1% DMSO and the remaining cells received media of the original type of M₄N concentrations (right arrow). Viability of representative cell dishes was assessed by trypan blue exclusion. (B) C3 cells were incubated for 6 days in media containing 1% DMSO (DMSO), 20 or 40 μM M₄N in 1% DMSO (20 and 40), or 3 days in media containing 20 or 40 μMM₄N in 1% DMSO followed by 3 additional days in media containing 1% DMSO (20D and 40D). For Cdc2 kinase assay, the same M₄N treatment was used as in B.(C) C3 cells were incubated for 6 days in media containing 1% DMSO (Left), 6 days in 40 μMM₄N in 1% DMSO (Center), or 3 days in 40 μMM₄N in 1% DMSO followed by 3 days in 1% DMSO (Right). Cells were fixed, stained with DAPI, and anaylzed by immunofluorescence with Cdc2 antibodies (yellow). The FITC (Cdc2, yellow) channel was overexposed to clearly view localization.

M₄N Treatment Induced a Reversible Shift of Cdc2 Protein from the Nucleus to the Cytoplasm. Active Cdc2 protein is primarily found in the nucleus of cells, where it performs its role in myriad processes including histone phosphorylation. When the active protein is not needed, it is transferred to the cytoplasm for degradation (26). To examine the impact of M₄N on Cdc2 localization, we incubated C3 cells for 3 days in media containing 1% DMSO or 40 μM M₄N. Following this treatment, the media were removed, and the cells were washed twice with PBS. Half of the drug-exposed cells were then switched to media containing only 1% DMSO, whereas the other half received fresh media containing 40 μM M₄N. As expected, Cdc2 was primarily found in the nucleus of C3 cells treated with 1% DMSO (Fig. 3C Left). Interestingly, residual Cdc2 protein of cells treated with 40 μM M₄N for 6 days was almost entirely located in the cytoplasm (Fig. 3C Center). Moreover, cells that had received 1% DMSO and 40 μM M₄N for 3 days followed by 3 days in media without M₄N showed a full recovery of Cdc2 in the nucleus (Fig. 3C Right). Additional experiments (19) determined that the cytoplasmic localization of residual Cdc2 in M₄N-treated cells was apparent after as little as 48 h, indicating that the Cdc2 localization seen in Fig. 3C Right may result from newly synthesized Cdc2 and from those recovered from the state of Cdc2 localization observed in Fig. 3C Center. Thus, when M₄N-treated cells are returned to normal media, Cdc2 is mainly localized in the nucleus.

Reduction of M₄N-Induced Caspase-3 Activity by Cdc2 and Survivin Overexpression. We next sought to investigate whether an exogenous gene subcloned in the pcDNA3.1 plasmid, a Sp1-unregulated CMV promoter, could be expressed in transfected C3 cells in the presence of M₄N. Survivin cDNA was subcloned into pcDNA3.1-myc-His, and expression of CMV-survivin was examined. In the absence of M₄N, cells not transfected and transfected with the empty vector only expressed endogenous survivin as expected. Cells transfected with survivin-myc-His expressed both the endogenous and the larger tagged survivin. After treatment with 40 μM M₄N for 3 days, the endogenous survivin protein was no longer detectable. However, the same M₄N treatment did not inhibit the overexpression of survivin-myc-His in cells transfected with CMV-survivin (Fig. 4A).

Fig. 4.

M₄N-induced caspase-3 activity is reduced by Cdc2 or survivin overexpression. (A) A cDNA copy of survivin was subcloned under the CMV-driven vector pcDNA3.1-myc-His. C3 cells and C3 cells transfected as indicated were treated with 1% DMSO (–) or 40 μM M₄N (+). Survivin expression was analyzed by Western blotting with survivin antibody. (B) C3 cells were transfected with empty vector, Cdc2, or survivin. Eight hours posttransfection, the cells were treated with 40 μM M₄N. The cells were lysed and assayed for caspase-3 activity. Results are expressed as mean ± SD of three independent experiments. (C) C3 cells or C3 cells stably transfected with survivin-myc-His were analyzed for survivin expression (Left). C3 cells, C3 cells transiently transfected with survivin-myc-His, and C3 cells stably expressing survivin-myc-His were treated with 1% DMSO or 40 μMM₄N for 3 days. The cells were lysed and assayed for caspase-3 activity for 6 h. Caspase-3 activities of the DMSO-treated samples were subtracted from those of the M₄N-treated samples to obtain kinetics of M₄N-induced caspase-3 activities. Data are mean of three independent determinations (Right).

If inhibition of survivin synthesis was indeed to play a central role in enhancing caspase-3 activity in M₄N-treated cells, supplement of survivin-myc-His via overexpression should allow reversion of this trend. We performed experiments to test this. For the control, C3 cells were only treated with 40 μM M₄N, whereas other C3 cell samples were first transfected with the vector, Cdc2, or survivin before the 40 M₄N treatment. Eight hours posttransfection, media were replaced with complete growth media containing the same M₄N concentration. The activity of the control measured at 0.0315 of OD (405 nm) was set as the basal M₄N-induced caspase-3 activity. Compared to the control, the caspase-3 activity in Cdc2-transfected cells and survivin-transfected cells was reduced by 36% and 54%, respectively, whereas transfection with the vector showed no effect on M₄N-induced caspase-3 activity (Fig. 4B).

We next investigated whether a C3 cell line stably expressing survivin-myc-His was even more resistant to M₄N-induced apoptosis relative to C3 cells or C3 cells transiently transfected with survivin-myc-His. The survivin stable line was expanded from a single clone and expressed His-tagged survivin (Fig. 4C). Cells of the stable line were treated with either 1% DMSO or 40 μM of M₄N for 3 days. Caspase-3 activity was measured every 5 min for 6 h to generate caspase-3 activity kinetics of these samples. During the linear portion of the graph, maximal activities occurred around 3 h after incubation of cell extracts with Ac-DEVD-pNA, at which time the caspase-3 activities of C3 cells transfected with survivin-myc-His and C3 cells stably expressing survivin-myc-His were compared. We found that M₄N-induced caspase-3 activity in C3 cells was reduced by 50% and 75% in transiently and stably survivin-myc-His transfected cells, respectively (Fig. 4C).

Effect of M₄N on Cell Arrest and Apoptosis Can Be Separately Reversed by Overexpression of Cdc2 and Survivin. If cell-cycle arrest in M₄N-treated C3 cells is the direct result of reduced levels of Cdc2 protein, then delivering exogenous Cdc2 to the cells should reverse the arrest and allow the cells to enter mitosis and divide. For this study, the pIRESpuro2-Cdc2 construct was transfected into C3 cells, and the transiently transfected cells were then treated with 1% DMSO and 40 μM M₄N in the growth media.

The impact of M₄N treatment on cell growth was not apparent before 24 h of treatment (Fig. 5A). Following 2 days in media containing M₄N, however, only the C3 cells transfected with the Cdc2 gene continued growth. Over the next 3 days, the number of these cells increased significantly, indicating that they were, at least partially, resistant to M₄N effects. In contrast, those cells that were transfected with the pIRESpuro2 vector (pIRES), or only received FuGENE 6, ceased growth.

Fig. 5.

Recovery of M₄N-induced effects by Cdc2 or survivin overexpression. (A) A cDNA copy of Cdc2 was cloned into the pIRESpuro2 vector and transfected into C3 cells. The cells were then incubated with media containing 40 μM M₄N in 1% DMSO for 4 days. Cell viability was assessed by trypan blue exclusion. (B) C3 cells were cotransfected with the lacZ construct and the empty vector pcDNA3.1-myc-His (a and b), Cdc2 (c and d), or survivin construct (e and f) and treated with M₄N and analyzed as described under Materials and Methods. The results are expressed as mean ± SD. (C) C3 cells treated with 40 μM M₄N for 3 days were fixed and then incubated with Cdc2 antibody; transfected cells treated with 1% DMSO or treated with 40 μMM₄N for 3 days were fixed and then incubated with anti-His antibody. Cells were then incubated with FITC-labeled secondary antibody and DAPI.

To specifically identify and quantitate those C3 cells spared from M₄N-induced effects by exogenous supplementation of Cdc2, we characterized the recovery of M₄N-treated C3 cells by cotransfection of a lacZ containing marker plasmid with the test plasmid pcDNA3.1-myc-His, containing Cdc2. Eight hours after transfection, the C3 cells were treated with 40 μM M₄N. After 2 days and 4 days following M₄N treatment, the cells were fixed and stained with 5-bromo-4-chloro-3-indolyl β-d-galactoside for β-galactosidase expression. The cells carrying the transfected genes appeared dark blue under light microscope. In the case of the vector, the number of blue cells continued to decrease from day 2 to day 4 in the presence of M₄N (presumably because both endogenous Cdc2 and survivin expression was down-regulated) (Fig. 5B). In the case of the vector carrying the transfected Cdc2 gene, the number of blue cells increased by 37.1% from day 2 to day 4. This suggests that the added Cdc2 is able to reverse the M₄N-induced cell-cycle arrest. We also tested whether the addition of survivin alone was able to protect M₄N-arrested cells from progressing to cell death. The number of cells transfected with survivin as determined by lacZ staining remained nearly unchanged from day 2 to day 4 (a mere 2.4% increase), suggesting that addition of survivin alone does not reverse cell arrest, although these cells are protected from M₄N-induced apoptosis (Fig. 5B).

To determine whether Cdc2 localization is responsible for the recovery of C3 cell growth, we transiently transfected C3 cells with pcDNA3.1-Cdc2-myc-His. C3 cells treated with 40 μMM₄N were probed with monoclonal Cdc2 antibody; transfected cells treated with 1% DMSO or treated with 40 μMM₄N were probed with anti-His antibody. As expected, M₄N treatment inhibited Cdc2 expression in nontransfected cells, and very little residual Cdc2 was present in the cytoplasm, but absent in the nucleus (Fig. 5C). In the Cdc2-transfected cells, Cdc2 was expressed throughout the M₄N-treated and untreated cell, indicating that CMV-driven Cdc2 overexpression was not affected by M₄N treatment. In nondividing cells, Cdc2 was predominantly localized to the cytoplasm and only some to the nucleus. In dividing cells, brighter FITC staining of Cdc2-His was colocalized with chromosomes, indicating that transfected M₄N-treated C3 cells had precise Cdc2 localization for normal growth (Fig. 5C).

Discussion

In this study, we have shown that M₄N treatment results in nuclear DNA cleavage, the formation of condensed nuclei and apoptotic bodies, and the activation of caspase-3, which are all hallmark features of apoptotic cells. We have shown further that M₄N-induced apoptosis is mediated via the mitochondrial pathway but not the death-receptor pathway. More interestingly, it seems that M₄N affects at least one of the regulators of apoptosis, survivin. We demonstrated in this study that M₄N strongly suppresses survivin gene expression. This phenomenon is time-dependent and occurs as early as 24 h post-M₄N treatment, before the onset of significant caspase-3 activity and apoptosis (unpublished data). This finding suggests that survivin inhibition by M₄N may occur before caspase-3 activation and apoptosis rather than a shutdown of survivin synthesis as a result of apoptosis.

Although the CMV promoter contains three putative Sp1 binding sites (27), it has been shown that deletion of these elements does not have significant impact on CMV transcription and replication at high multiplicities of infection (28). Because the CMV promoter can compensate for the loss of the Sp1 binding sites, the prevention of Sp1 binding to CMV promoter by M₄N is likely inconsequential. However, both mouse and human Cdc2 and survivin endogenous genes require successful binding of Sp1 to their promoters to initiate transcription (17, 18, 29). We observed that C3 cells transfected with Cdc2 and survivin, under the control of the CMV promoter, are resistant to M₄N-mediated cell arrest or death. The discovery conforms to the model that the effect of M₄N is selective to Sp1-dependent genes.

Phosphorylation on Thr-34 of survivin by p34^cdc2 (30) was found to be important for survivin protein stability. Survivin protein is degraded via the ubiquitin-proteasome pathway. In addition, an unphosphorylatable T34A survivin mutant has a faster turnover rate than the wild type (31). This may explain why genetic inactivation (32) or pharmacological inhibition (33) of Cdc2 results in cell apoptosis. Such a cytoprotective role of Cdc2 was also observed in CMV-Cdc2-transfected cells. Phosphorylation of residual survivin protein left over in M₄N-treated cells could have generated a more stable form of survivin, leading to the reduction of M₄N-induced caspase-3 activity as observed (Fig. 4B). It is therefore likely that the overexpressed Cdc2 phosphorylates residual survivin, which can become more stable before survivin transcription is inhibited by M₄N. Further studies on this possibility will be conducted.

In the present study, we have demonstrated what appeared to be the dual effect of M₄N in counteracting both the uncontrolled growth and the inactivated apoptosis in C3 transformed cells. The notion that M₄N may be able to suppress other types of cancer in vivo by the same mechanism is further supported by our recent observation that five human tumor cell lines (MCF-7, breast; Hep3B, liver; HT-29, colorectal; LNCaP, prostate; K-562, leukemia) and their cancer xenografts in Thy^–/Thy^– mice are suppressed by M₄N through transcriptional inhibition of Cdc2 and survivin expression.