Evidence supporting a role for calcium in apoptosis induction by the synthetic triterpenoid 2-cyano-3,12-dioxooleana-1,9-dien-28-oic acid (CDDO).

The synthetic triterpenoid 2-cyano-3,12-dioxooleana-1,9-dien-28-oic acid (CDDO) is a novel anticancer agent that induces apoptosis in tumor cells. The cytotoxic stress underpinning CDDO-induced apoptosis has not been established. This study compared and contrasted the effects of CDDO on COLO 16 human skin cancer cells and their respiration-deficient (rho(0)) clones to elucidate the stress signal responsible for initiating apoptosis. CDDO promoted apoptosis in COLO 16 cells in a dose- and time-dependent manner. The rho(0) clones appeared to be more sensitive to CDDO-induced apoptosis implying that the disruption of mitochondrial respiration was not directly associated with triggering cell death. After a 4-h exposure to CDDO, mitochondrial inner transmembrane potential-sensitive dyes revealed mitochondrial hyperpolarization in the COLO 16 cells and mitochondrial depolarization in the rho(0) clones. Electron microscopy illustrated that this exposure also promoted mitochondrial condensation, endoplasmic reticulum dilation, and chromatin condensation in the COLO 16 cells. Endoplasmic reticulum dilation and chromatin condensation were also observed in the rho(0) clones, but the mitochondria in these cells were markedly swollen implying that the disruption of intracellular Ca(2+) homeostasis was associated with cell death. A Ca(2+)-sensitive dye confirmed that CDDO increased cytoplasmic free Ca(2+) in the COLO 16 cells, their rho(0) clones, as well as in malignant breast and lung epithelial cells. A cell-permeant Ca(2+) chelator reduced the CDDO-induced increase in cytoplasmic free Ca(2+), and inhibited caspase activation, the development of apoptotic morphology, and DNA fragmentation in the COLO 16 cells, implying that Ca(2+) played a pivotal role in signaling the initiation of apoptosis.

Triterpenoids derived from plants are used for medicinal purposes in many Asian countries and have been reported to have anticancer activity (1). Many of the plant-derived triterpenoids are considered weak anti-inflammatory and/or anticancer agents, which has encouraged the identification of novel synthetic analogues with greater potency (2). One such analogue, 2-cyano-3,12-dioxooleana-1,9-dien-28-oic acid (CDDO), 1 has gained recognition as a promising cancer chemopreventive and therapeutic agent (3). In vitro studies have shown that nanomolar concentrations of CDDO can inhibit the proliferation of various human tumor cell types (3,4), suppress the activity of inflammatory cytokines such as interferon-␥, interleukin-1, and tumor necrosis factor (2,3), and inhibit the induction of inflammatory mediators such as cyclooxygenase-2 and nitric oxide synthase (2,3). More recently, low micromolar concentrations (Յ10 M) of CDDO have been shown to induce apoptosis in human myeloid (5,6) and lymphocytic leukemia cells (7), osteosarcoma cells (8), and breast cancer cells (4).
Apoptosis is the mechanism utilized by metazoans to eliminate redundant or potentially deleterious cells and is considered an essential process for regulating tissue homeostasis. Apoptosis induction is arguably the most potent defense against cancer making it a desirable end point for both chemoprevention (9) and chemotherapy (10). Apoptosis is a relatively linear process. It is triggered by an initiation phase that is highly varied depending on cell type and the underlying stress stimulus (e.g. oxidative stress, DNA damage, ion fluctuations, and cytokines). This is followed by an effector phase where the cell undergoes distinct biochemical changes that result in the systematic activation of catabolic enzymes (proteases and nucleases) that ultimately participate in the cleavage of proteins and DNA characteristic of the degradation phase (11,12).
Two effector mechanisms associated with caspase (cysteine proteases involved in apoptosis) activation have been characterized extensively. These include the extrinsic or death receptor-mediated effector mechanism and the intrinsic or mitochondrial-mediated effector mechanism (13). The extrinsic mechanism is mediated by the activation of caspase-8, while the intrinsic mechanism relies on the release of soluble apoptogenic mitochondrial proteins (i.e. cytochrome c) to activate caspase-9 and initiate cellular degradation. These apical caspases can trigger the activation of downstream proteases like caspase-3. Furthermore, cross-talk between the extrinsic and intrinsic effector mechanisms can be established via the activity of pro-apoptotic Bcl-2 family members (e.g. Bid and Bax) to amplify cell death signaling (14). In addition to the mitochondria, growing evidence suggests that other organelles, including the endoplasmic reticulum (ER), Golgi apparatus, and lysosomes are also major points of integration for damage sensing, pro-apoptotic signaling, and caspase activation (15).
CDDO appears to induce apoptosis through the activation of both extrinsic and intrinsic effector mechanisms. The activation of caspase-8 (5)(6)(7)(8), the cleavage of Bid (5,8), the induction of mitochondrial permeability transition (MPT) (4), the release of cytochrome c from the mitochondria (5,6,8), the activation of caspase-9 (7), and the activation of caspase-3 (5-8) all appear to be involved in CDDO-induced apoptosis. However, none of the previously referenced studies investigated the underlying cytotoxic signal responsible for triggering the downstream effector mechanisms associated with CDDO-induced apoptosis. This study compared and contrasted the effects of CDDO on COLO 16 human cutaneous squamous cell carcinoma cells and their respiration-deficient derivatives depleted of mitochondrial DNA ( 0 clones). We demonstrate that CDDO triggers apoptosis in these cells by increasing cytoplasmic free Ca 2ϩ ([Ca 2ϩ ] c ). The increase in [Ca 2ϩ ] c caused by CDDO appeared to encourage caspase activation and chromatin condensation without inducing MPT in the parental cells. The data would also suggest that CDDO could conceivably promote the release of soluble apoptogenic mitochondrial proteins (e.g. cytochrome c, endonuclease G, and apoptosis inducing factor) via Ca 2ϩinduced MPT in the 0 clones, which may be dependent on the Ca 2ϩ buffering capacity of their mitochondria.

EXPERIMENTAL PROCEDURES
Cell Culture and Reagents-The COLO 16 cell line was provided by Dr. Janet Price (University of Texas M. D. Anderson Cancer Center, Houston, TX). The 0 derivatives of COLO 16 cells were isolated and characterized as described previously (16). The MCF-7 human malignant breast carcinoma cell line was purchased from the American Type Culture Collection (Manassas, VA). The H1792 human non-small cell lung carcinoma cell line was provided by Dr. Adi Gazdar (University of Texas Southwestern Medical Center, Dallas, TX). The cells were cultured in Dulbecco's modified Eagle's medium containing 4.5 mg/ml D-glucose (Sigma) supplemented with 110 g/ml pyruvate, 50 g/ml uridine (both from Sigma), and 2% fetal bovine serum (Invitrogen). Cell cultures were incubated at 37°C in humidified air containing 5% CO 2 . Treatment with CDDO and other agents was performed on subconfluent cultures.
Apoptosis Assays-DNA fragmentation was ascertained by using a hypotonic solution of propidium iodide (PI) (17). Cells were treated with the specified concentrations of CDDO, BAPTA acetoxymethyl ester, or an equal volume of the vehicle Me 2 SO for the specified times. Where indicated, BAPTA acetoxymethyl ester was added to the culture medium 1 h before treatment with CDDO. The cells were harvested by trypsinization and combined with their respective culture media that was removed before trypsinization. The cells were pelleted by centrifugation and the pellet was resuspended in 1 ml of hypotonic PI solution (50 g/ml PI, 0.1% sodium citrate, and 0.1% Triton X-100 (all from Sigma) dissolved in Ca 2ϩ -free phosphate-buffered saline (Ca 2ϩ -free PBS at 4°C) and stored overnight at 4°C. The cell suspensions were analyzed the next day for PI fluorescence intensity by flow cytometry.
Caspase activity was detected with PhiPhiLux-G 1 D 2 (Oncoimmunin Inc., Gaithersburg, MD), a cell-permeant fluorogenic caspase substrate (DEVD-rhodamine) that is cleaved in a DEVD-dependent manner to produce fluorescent molecules of rhodamine (17). Cells in 10-cm plastic tissue culture plates were treated for the designated times with 5 M CDDO, 10 M BAPTA acetoxymethyl ester, the combination of 5 M CDDO and 10 M BAPTA acetoxymethyl ester as described above, or an equal volume of the vehicle Me 2 SO. The cells were harvested by trypsinization, pelleted by centrifugation, resuspended in 50 l of 10 M PhiPhiLux-G 1 D 2 reagent, incubated for 1 h at 37°C, and washed with flow cytometry buffer (Oncoimmunin Inc.) according to the manufacturer's recommendations. The resulting cell suspensions were analyzed immediately for rhodamine fluorescence intensity by flow cytometry.
Assays for Reactive Oxygen Species (ROS) Production and Mitochondrial Inner Transmembrane Potential (⌬⌿ m )-Hydroperoxide and superoxide production were determined using 2Ј,7Ј-dichlorofluorescin diacetate and dihydroethidium, respectively (17,18). Briefly, cells in 10-cm plastic tissue culture plates were treated for 2 or 4 h with 20 M 2Ј,7Ј-dichlorofluorescin diacetate and 5 M CDDO, 5 M 4HPR, 5 M CDDO and 5 M 4HPR, or Me 2 SO. The cells were harvested by trypsinization, washed in 5 ml of Ca 2ϩ -free PBS at 37°C, pelleted by centrifugation, resuspended in 1 ml of Ca 2ϩ -free PBS at 37°C, and analyzed immediately for 2Ј,7Јdichlorofluorescein (DCF) fluorescence intensity by flow cytometry. For superoxide detection, cells in 10-cm plastic tissue culture plates were treated for 2 or 4 h with 5 M CDDO or Me 2 SO. Twenty min before the cells were harvested, dihydroethidium was added directly to the culture medium to a final concentration of 5 M. The cells were harvested as described above and analyzed immediately for ethidium fluorescence intensity by flow cytometry.
The cationic dyes DiOC 6 (3) and CMXRos were used to evaluate ⌬⌿ m (16). Cells in 10-cm plastic tissue culture plates were treated with 5 M CDDO or Me 2 SO for 4 h. Twenty min before the cells were harvested, DiOC 6 (3) or CMXRos were added directly to the culture medium to a final concentration of 60 and 150 nM, respectively. The cells were harvested as described above, resuspended in 1 ml of Ca 2ϩ -free PBS at 37°C and analyzed immediately for DiOC 6 (3) or CMXRos fluorescence intensity by flow cytometry.
Assay for Ca 2ϩ Mobilization-[Ca 2ϩ ] c was measured using the cellpermeant Ca 2ϩ -sensitive fluorescent dye Fluo-3 acetoxymethyl ester (17). Where indicated, BAPTA acetoxymethyl ester (10 M) was added to the culture medium of cells in 10-cm plastic tissue culture plates for a 1-h exposure prior to the loading procedure with Fluo-3 acetoxymethyl ester. The medium was removed from the tissue culture plates and replaced with 4 M Fluo-3 acetoxymethyl ester diluted in Krebs-Ringer buffer (KRB) (10 mM D-glucose, 120 mM NaCl, 4.5 mM KCl, 0.7 mM Na 2 HPO 4 , 1.5 mM NaH 2 PO 4 , and 0.5 mM MgCl 2 (pH 7.4 at 37°C)) (Sigma) for 20 min. The dishes were washed once with 5 ml KRB to remove the residual dye. The cultures were then treated for the designated times with the indicated concentrations of CDDO or the vehicle Me 2 SO diluted in KRB alone or including 2 mM CaCl 2 . The cells were harvested by trypsinization, washed in 5 ml of Ca 2ϩ -free PBS at 37°C, pelleted by centrifugation, resuspended in 1 ml of Ca 2ϩ -free PBS at 37°C, and analyzed immediately for Fluo-3 fluorescence intensity by flow cytometry.
Flow Cytometry-All flow cytometric procedures were performed with a Coulter XL flow cytometer, and data analysis was accomplished using System II XL software (Coulter Corp., Miami, FL). The experiments employing flow cytometry were typically conducted using duplicate samples for each treatment, unless otherwise indicated, and each experiment was repeated at least two times. Approximately 10,000 cells were analyzed for each sample. In all cytofluorometric determinations, cell debris and cells clumps were excluded from the analysis of the cell suspensions by gating.
Electron Microscopy-Cells were seeded in single-well tissue culture slides and allowed to attach and proliferate for 24 -48 h. The culture medium was removed and replaced with fresh medium containing 5 M CDDO or an equal volume of the vehicle Me 2 SO. After 4 h, the medium was removed from the wells and the cells were washed twice with 3 ml of Ca 2ϩ -free PBS at 37°C. The last wash was replaced with 2 ml of fixative (3% glutaraldehyde and 2% paraformaldehyde dissolved in 0.1 M sodium cacodylate (pH 7.3 at 37°C)) (all from Sigma). The samples were stored overnight at 4°C, embedded, sectioned, and stained using standard electron microscopy procedures. The samples were viewed using a JEOL 1010 transmission electron microscope (JEOL USA, Inc., Peabody, MA), and images were acquired using an Advanced Microscopy Techniques imaging system (Advanced Microscopy Techniques Corp., Danvers, MA).

CDDO Induces Apoptosis in COLO 16 Cells and Their 0
Clones-Previous studies have demonstrated that low micromolar (Յ10 M) concentrations of CDDO can induce apoptosis in various human tumor cell types (4 -8). We exposed COLO 16 cells to 1, 5, and 10 M CDDO for 24 h and examined them for DNA degradation characteristic of apoptosis (indicated by hypoploid DNA content using a hypotonic PI stain). As shown in Fig. 1A, increasing concentrations of CDDO promoted increasing amounts of hypoploid cells in the treatment populations relative to the Me 2 SO-treated control. The 5 and 10 M concentrations of CDDO appeared to be equally effective in promoting apoptosis in the majority of the treatment populations. Between 12 and 24 h after exposure to CDDO, there was also a time-dependent increase in the percentages of hypoploid cells in the treatment populations (Fig. 1B).
COLO 16 0 cells are more resistant than COLO 16 cells to apoptosis induction following exposure to several putative cancer chemopreventive agents. These agents (i.e. 4HPR (18), 6-[3-(1-adamantyl)-4-hydroxyphenyl]-2-naphthalene carboxylic acid (CD437) (16), capsaicin (17), and resiniferatoxin (17) appear to target mitochondrial respiration to promote enhanced ROS production, MPT, and caspase activation in COLO 16 cells. Furthermore, several studies have demonstrated that 0 cells are more resistant than their parental counterparts to the apoptogenic effects of various cancer chemotherapeutic agents (19,20), ceramide (21), and extrinsic mediators of apoptosislike tumor necrosis factor (22) and TRAIL (23) implying that the disruption of mitochondrial respiration was associated with cell death. We exposed the 0 cells to 5 M CDDO for 12 and 24 h and examined them for hypoploid DNA content to determine whether there was any difference in their sensitivity to CDDO-induced apoptosis relative to the parental cells. Interestingly, the 0 cells appeared to be more sensitive to CDDOinduced apoptosis. The majority of the 0 cells exhibited hypoploid DNA content after a 12-h exposure to 5 M CDDO, while a longer exposure was required for the same treatment to attain a similar response in the parental cells (Fig. 1B). These results suggested that the disruption of mitochondrial respiration was not directly associated with CDDO-induced apoptosis. Short term exposure of the COLO 16 cells to 5 or 10 M CDDO promoted morphological changes indicative of early apoptosis in the majority of the treatment populations. As illustrated in Fig. 2A, a 4-h exposure to 5 M CDDO caused the parental cells to shrink in size and round up from the tissue culture plate relative to the Me 2 SO-treated controls (Fig. 2B). Similar morphological changes were also observed in the 0 clones after a 4-h exposure to 5 M CDDO (not shown). In chronic lymphocytic leukemia cells, CDDO promoted the rapid activation of caspase-8 and caspase-3 that peaked within 2 h after treatment (7). In an attempt to determine whether caspase activity was associated with the early morphological changes observed in the COLO 16 cells, these cells were treated with 5 M CDDO for various times and examined for DEVDaselike caspase activity using a cell permeant fluorogenic caspase substrate (DEVD-rhodamine). We detected a time-dependent increase in rhodamine fluorescence intensity indicative of DEVDase-like caspase activity in COLO 16 cells that appeared to require at least 4 h to peak following exposure to CDDO and subsequently declined after 8 h (Fig. 2C).
CDDO Modulates ROS Production in COLO 16 Cells-Certain cancer chemopreventive agents can trigger the rapid and marked production of hydroperoxides and/or superoxide in COLO 16 cells (16 -18). To determine whether oxidative stress was possibly associated with CDDO-induced apoptosis, we exposed COLO 16 cells to 5 M CDDO for 2 and 4 h and examined them for evidence of enhanced hydroperoxide and/or superoxide production. However, there was only a slight, but discernable and reproducible, increase in both the DCF and ethidium fluorescence intensity at each of the time points examined relative to the Me 2 SO-treated controls (not shown). We repeated these assays using the 0 cells and found that CDDO treatment failed to increase the fluorescence intensity of either DCF or ethidium at each of the time points examined relative to the Me 2 SO-treated controls (not shown). These results implied that enhanced ROS production was not directly associated with CDDO-induced apoptosis. Furthermore, the slight increase in ROS generation observed in the parental cells following short term exposure to CDDO probably originated from the mitochondria.
We have reported previously that the synthetic retinoid 4HPR appears to function as a pro-oxidant in shin cancer cells by undergoing redox cycling at the mitochondrial electron transport chain (18). If the slight ROS production triggered by CDDO reflected an increase in mitochondrial respiration in these cells, we would predict that CDDO would enhance 4HPR-induced ROS production. Indeed, we observed that COLO 16 cells exposed for 2 h to 5 M 4HPR exhibited ϳ3 times the mean DCF fluorescence intensity displayed by the Me 2 SO-and CDDOtreated cells, which was increased an additional 4-fold when 4HPR was combined with an equal molar concentration of CDDO (Fig. 3A). These results suggested that short term exposure to CDDO enhanced mitochondrial respiration in COLO 16 cells.
CDDO Modulates ⌬⌿ m in COLO 16 Cells and Their 0 Clones-Mitochondrial electron transport drives the generation of ⌬⌿ m , which is used to produce ATP (24), regulate intracellular Ca 2ϩ homeostasis (25,26), and facilitate the transport of proteins and metabolic substrates into the matrix (27). Glycolysis is responsible for ATP generation in 0 cells (24). The generation of ⌬⌿ m in 0 mitochondria is mediated by the activity of the mitochondrial F1-ATPase, which hydrolyzes ATP to pump protons out of the matrix, thereby maintaining most aspects of mitochondrial physiology (27). We have observed the dissipation of ⌬⌿ m in breast cancer cells exposed to CDDO implying that MPT was associated with apoptosis induction (4). Therefore, we exposed the parental and 0 cells to 5 M CDDO for 4 h and examined them for ⌬⌿ m using two cationic ⌬⌿ m -sensitive dyes to determine whether CDDO could alter the retention of these dyes as a surrogate indicator of changes in mitochondrial physiology. As shown in Fig. 3B, a 4-h exposure to CDDO appeared to promote mitochondrial hyperpolarization in the COLO 16 cells as evidenced by the slight, but discernable and reproducible, increase in both DiOC 6 (3) and CMXRos fluorescence intensity compared with the Me 2 SO-treated controls. Conversely, a similar exposure to CDDO caused mitochondrial depolarization in the 0 cells as indicated by the marked loss in the fluorescence intensity of both DiOC 6 (3) and CMXRos relative to the Me 2 SOtreated controls (Fig. 3C). The apparent hyperpolarization of ⌬⌿ m observed in the parental cells, in addition to the modulation of mitochondrial ROS generation, further substantiated the notion that short term exposure to CDDO increased mitochondrial respiration. The conspicuous mitochondrial depolarization detected in the 0 cells suggested that CDDO triggered MPT in the 0 mitochondria, which appeared to be in some way related to their respiration deficiency. COLO 16 Cells and Their 0 Clones-Electron microscopy was utilized to examine the parental and 0 cells exposed to Me 2 SO or 5 M CDDO for 4 h to see whether we could detect any physical changes in their mitochondria that would mirror the changes in ⌬⌿ m we had observed and possibly provide additional clues to the underlying cell stress associated with CDDO-induced apoptosis. Cells with representative characteristics were selected for presentation in Fig. 4. Fig. 4A shows a Me 2 SO-treated COLO 16 cell, which displayed a well defined plasma membrane and nuclear envelope containing two nucle- oli. The mitochondria in these cells were electron dense and exhibited an orthodox conformation. The ER could also be identified in close proximity to the mitochondria (Fig. 4B). A 4-h exposure to CDDO promoted cytoplasmic and nuclear shrinkage in these cells as represented by the COLO 16 cell in Fig. 4C. The nucleoli were absent and chromatin condensation was evident along with clusters of vacuolization in the cytoplasm. A higher magnification revealed that the mitochondria in these cells were intact but had all undergone varying de-grees of condensation depending on their location in the cell (not shown). The areas of enhanced cytoplasmic vacuolization appeared to be the ER that was dilated (Fig. 4D). The mitochondria adjacent to these clusters were markedly condensed and clearly more electron dense than the mitochondria observed in the Me 2 SO-treated COLO 16 cell.

CDDO Promotes Distinct Cellular and Subcellular Alterations in
The COLO 16 0 cells were typically larger than their parental counterparts as illustrated by the Me 2 SO-treated 0 cell in Fig.  4E. These cells also displayed a well defined plasma membrane and nuclear envelope containing multiple nucleoli. However, the mitochondria in these cells were morphologically distinct from those observed in the parental cells. Most notably, they were less electron dense and had very sparse, disorganized cristae (Fig.  4F), and their size varied considerably (Fig. 4E). As observed in the parental cells, the mitochondria in the 0 cells were also interlaced with ER. A 4-h exposure to CDDO triggered cytoplasmic and nuclear shrinkage, chromatin condensation, and extensive cytoplasmic vacuolization in the 0 cells (Fig. 4G). A higher magnification confirmed that the larger vacuoles were the remnants of the mitochondria, which appeared abnormally enlarged or swollen compared with those observed in the Me 2 SO-treated 0 cell. The cristae were absent, and the disruption of outer membrane integrity was evident (Fig. 4H). These mitochondrial changes appeared to be consistent with MPT induction. The smaller vacuoles surrounding the mitochondria presumably represented what remained of the ER. The possibility of MPT induction serving as a central and coordinating event in CDDO-induced apoptosis could be ruled out, at least in the parental cells, because the mitochondria in these cells were intact, albeit noticeably condensed, and apparently hyperpolarized at a time when CDDO-induced caspase activity and chromatin condensation were prominent. Furthermore, in addition to chromatin condensation, the dilation of the ER appeared to be a common subcellular feature in both the COLO 16 cells and their 0 derivatives exposed to CDDO implying that ER stress was directly associated with apoptosis induction.

CDDO Increases [Ca 2ϩ ] c in a Dose-and Time-dependent Manner-
The ER comprises approximately half of the intracellular membranes in eukaryotic cells, and its labyrinthine spaces are the primary sites for sequestering Ca 2ϩ in the cell. Consequently, the ER is directly involved in the regulation of intracellular Ca 2ϩ homeostasis. Concentrations of Ca 2ϩ in the ER can exceed ϳ10 times the average Ca 2ϩ concentration in the cytoplasm (28). Using the Ca 2ϩ -sensitive dye Fluo-3 (17), we investigated whether short term exposures to CDDO could increase [Ca 2ϩ ] c as a surrogate indicator of ER stress. COLO 16 cells were exposed for 1 h to 0 (Me 2 SO), 1, 5, or 10 M CDDO diluted in KRB alone or containing 2 mM CaCl 2 . We observed a dose-dependent increase in Fluo-3 fluorescence intensity signifying that a short term exposure to CDDO could promote an increase in [Ca 2ϩ ] c . The addition of 2 mM CaCl 2 to the KRB did not appreciably alter the response of Fluo-3 to CDDO treatment in these cells (Fig. 5A), suggesting that the increase in [Ca 2ϩ ] c originated from a release from an intracellular store, probably the ER. The increase in Fluo-3 fluorescence intensity was also time-dependent indicating that CDDO exposure increased [Ca 2ϩ ] c in a continuous manner (Fig. 5B). Furthermore, the 0 clones, as well as human malignant breast (MCF-7) and lung (H1792) epithelial cells, also displayed increases in Fluo-3 fluorescence intensity after a 1-h exposure to 5 M CDDO that were ϳ3-4-fold greater than their respective Me 2 SO-treated controls. This would imply that the Ca 2ϩ mobilizing effect of CDDO was not unique to cutaneous keratinocytes (Fig. 5C).
BAPTA Inhibits CDDO-induced Apoptosis-Exposure to CDDO progressively increased [Ca 2ϩ ] c in COLO 16 cells imply- Certain cell cultures were pretreated with BAPTA acetoxymethyl ester as described in the legend to B followed by a 12-h co-treatment with CDDO where indicated. Hypoploid DNA content was determined using hypotonic PI staining as described in Fig. 1A. The results in C are expressed as a percentage of the mean hypoploid cells detected in triplicate samples for each treatment Ϯ S.D. (error bars).
ing that the disruption of Ca 2ϩ homeostasis was responsible for triggering apoptosis. To test our assumption, we sought to determine whether buffering the level of [Ca 2ϩ ] c with a cell permeant Ca 2ϩ chelator could inhibit CDDO-induced apoptosis. We pretreated COLO 16 cells with BAPTA acetoxymethyl ester, a cell-permeant agent that is hydrolyzed in the cytoplasm to yield the Ca 2ϩ chelator BAPTA (29), to determine whether we could buffer the increase in [Ca 2ϩ ] c caused by CDDO treatment. As illustrated in Fig. 6A, a 2-h exposure to 5 M CDDO promoted a 6-fold increase in the mean Fluo-3 fluorescence intensity compared with the Me 2 SO-treated control. This was decreased to less than a 1-fold increase when BAPTA was included as a pretreatment, while BAPTA alone promoted a slight decrease in the mean Fluo-3 fluorescence intensity compared with the Me 2 SO-treated control. Pretreating the cells with BAPTA also inhibited CDDO-induced DEVDase-like caspase activity as evidenced by the decrease in the mean rhodamine fluorescence intensity compared with that exhibited by a 4-h exposure to 5 M CDDO alone (Fig. 6B), suggesting that the increase in [Ca 2ϩ ] c was responsible for triggering caspase activity.
We monitored the various treatment populations for up to 12 h and found that the onset of apoptotic morphology was diminished when BAPTA was included as a pretreatment before the addition of CDDO. As shown in Fig. 7A, the cells exposed to 5 M CDDO for 12 h were markedly shrunken, and ϳ30% of the cells in the treatment population were suspended in the culture medium. However, these cells are not seen in the picture because the medium was removed prior to imaging the cultures. A 1-h pretreatment with BAPTA delayed the onset of apoptotic morphology in the COLO 16 cells co-treated an additional 12-h with CDDO (Fig. 7B). Less than 10% of the cells in this treatment population were suspended in the culture medium (not shown). The attached cells, and the cells treated with BAPTA alone (not shown), did not appear to be markedly different from the Me 2 SO-treated control cells (not shown, please see Fig. 2B for a picture of Me 2 SO-treated COLO 16 cells). We stained these cells with a hypotonic solution of PI to determine whether DNA fragmentation had occurred. A 12-h exposure to 5 M CDDO promoted hypoploid DNA content in ϳ44% of the sample populations (Fig. 7C). This was reduced to ϳ17% when the parental cells were pretreated with BAPTA followed by a 12-h co-treatment with CDDO. A comparable inhibition of CDDO-induced apoptosis was also observed in the 0 clones pretreated for 1 h with BAPTA followed by a 12-h co-treatment with CDDO (Fig. 7C). DISCUSSION The results presented in this study strongly support a role for Ca 2ϩ in CDDO-induced caspase activation and apoptosis and provide a novel mechanism for the possible anticancer activity of this agent. It is commonly accepted that sustained elevations in [Ca 2ϩ ] c can mediate apoptotic cell death in various cell systems via the activation of Ca 2ϩ -dependent enzymes like proteases, endonucleases, and phospholipases (30,31). Furthermore, the disruption of intracellular Ca 2ϩ homeostasis and/or ER stress can also promote the activation of caspase-12, which is believed to reside on the cytoplasmic face of the ER (32)(33)(34). Caspase-12 activity has been implicated in the activation of caspase-8, caspase-9, and caspase-3 (35). We have also shown that 0 cells respond to CDDO-induced apoptosis in a manner similar to that exhibited by their parental counterparts. This finding suggests that the Ca 2ϩ -mobilizing effect of CDDO could be exploited to trigger apoptosis in tumor cells irrespective of their metabolic phenotype.
The ER is the primary site in eukaryotic cells that is utilized for Ca 2ϩ storage and the regulation of intracellular Ca 2ϩ ho-meostasis (28). In addition to the ER, the mitochondria can also contribute to the regulation of intracellular Ca 2ϩ homeostasis. Sustained Ca 2ϩ release from the ER can trigger Ca 2ϩ uptake by the mitochondria (26,36). The mitochondria possess an exceptional capacity to sequester Ca 2ϩ and buffer local variations in [Ca 2ϩ ] c levels via the activity of electrophoretic uniporters localized in the inner membrane (26,37). The close proximity of the mitochondria to the ER is believed to facilitate this process, which is possibly mediated via ER-mitochondria contact sites (37,38). Mitochondrial Ca 2ϩ uptake has long been known to stimulate the turnover of matrix dehydrogenases coupled to the tricarboxylic acid cycle. This activity provides substrates for mitochondrial respiration, which can promote an increase in ⌬⌿ m (26,37,39,40).
The sustained increase in [Ca 2ϩ ] c followed by mitochondrial Ca 2ϩ uptake and the stimulation of respiration could explain why CDDO promoted a slight increase in mitochondrial ROS production, greatly enhanced 4HPR-induced ROS generation, and caused mitochondrial hyperpolarization in the COLO 16 cells. We also observed mitochondrial condensation in these cells after a short term exposure to CDDO. Mitochondrial condensation has been reported in cells treated with the Ca 2ϩ ionophore A23187 (41), and mitochondrial condensation and matrix remodeling have also been confirmed in wild-type and Bax/Bak-deficient cells primed for apoptosis with agents that promote ER stress (42). Furthermore, CDDO-induced caspase activation and chromatin condensation occurred concomitant to the changes in mitochondrial physiology and morphology in the COLO 16 cells supporting a role for Ca 2ϩ signaling in all of these processes.
The impairment of bioenergetic processes due to deletions or mutations in mitochondrial DNA can diminish the Ca 2ϩ buffering capacity of the mitochondria (43)(44)(45)(46) and potentially sensitize these organelles to Ca 2ϩ -induced MPT (45). Exposure to CDDO increased [Ca 2ϩ ] c in the 0 cells, which appeared to be required for apoptosis induction. Moreover, a short term exposure to CDDO also promoted mitochondrial depolarization and mitochondrial swelling in these cells implying that Ca 2ϩ -induced MPT was responsible for these effects. Among the various stimuli that cause MPT, Ca 2ϩ has been investigated extensively and, thus, is considered a prototypical inducing agent (47). MPT allows water and solutes up to 1500 Da to infiltrate the mitochondrial matrix. This results in colloidal osmotic swelling of the matrix (47,48), which can cause the physical rupture of the outer mitochondrial membrane (49). These events can disrupt mitochondrial membranes to the extent that soluble apoptogenic mitochondrial proteins like cytochrome c (50), apoptosis-inducing factor (51), and endonuclease G (52) are released to participate in the biochemical reactions involved in cellular degradation. Mitochondrial respiration appeared to protect against MPT in the parental cells following a short term exposure to CDDO. This could have been due to the apparent hyperpolarization of ⌬⌿ m , given that enhanced ⌬⌿ m can decrease the sensitivity of mitochondria to MPT induction (47,53). Together, these findings suggest that the early mitochondrial changes indicative of MPT observed in the 0 cells exposed to CDDO were a by-product of Ca 2ϩ overload, which could conceivably reflect the diminished Ca 2ϩ buffering capacity of their mitochondria.
CDDO reportedly produced a synergistic effect on apoptosis induction in human myeloid leukemia cells when combined with tumor necrosis factor (5). Tumor necrosis factor apparently promotes enhanced mitochondrial ROS generation (54) and inhibits mitochondrial respiration by blocking complex I activity (54,55). In the presence of increased [Ca 2ϩ ] c , the inhibition of mitochondrial respiration could conceivably sensitize the mitochondria to Ca 2ϩ overload and/or oxidative stress resulting in MPT induction (47). Therefore, it is interesting to speculate that the disruption of mitochondrial respiration caused by tumor necrosis factor coupled with an elevation in [Ca 2ϩ ] c triggered by CDDO could amplify stress signaling to enhance cell killing. The results obtained in this study would also suggest that CDDO could have a synergistic effect on cell killing when combined with 4HPR. This may well be achieved by increasing the pro-oxidant quality of 4HPR, which appears to be intimately associated with its ability to trigger apoptosis in various tumor cell types (18, 56 -58).
In conclusion, Ca 2ϩ appears to be an important regulator of CDDO-induced apoptosis in human cutaneous squamous cell carcinoma cells. Determining the integration points for Ca 2ϩ and the effector mechanisms associated with CDDO-induced apoptosis, while outside of the scope of this study, should reveal further clues to the cellular and molecular activity of this agent. Further studies are warranted to examine these integration points to fully understand how CDDO exploits endogenous apoptosis-inducing mechanisms. These studies should provide additional information to support the use of CDDO in the prevention or treatment of cancer. Moreover, the results presented in this study suggest that CDDO, and perhaps its structural analogues, should be considered as potential candidates for the prevention or treatment of skin cancer.