Carboxyamido-triazole Induces Apoptosis in Bovine Aortic Endothelial and Human Glioma Cells¹

Apoptosis or programmed cell death (1) is characterized by cell membrane blebbing, chromatin condensation, and genomic DNA fragmentation (2). In contrast to necrotic cell death, which triggers an inflammatory response, apoptotic cell death results in membrane-bound apoptotic bodies that are eliminated via phagocytosis either by the endothelial reticular system or by macrophages without invoking an inflammatory cascade (3). As such, apoptosis is a vital process in normal development used to eliminate cells during organ maturation, such as in the brain, limbs,and digits, and in T-cell maturation in the thymus (4).

Apoptosis also plays an important role in disease pathology. For example, suppression of apoptosis results in the lack of negative selection of self-reacting T lymphocytes during thymus development,resulting in autoimmune disease (4). In addition, Bcl-2 was originally identified as a result of the t(14:18) translocation in B-cell follicular lymphoma whereby its juxtaposition with the IgH enhancer leads to dysregulated overexpression. This overexpression inhibits B-cell apoptosis and results in lymphoproliferative disease (5).

Traditionally, it is believed that cancer develops due to uncontrolled cell proliferation secondary to loss of tumor suppressor genes or gain of oncogenes. Thus, cancer chemotherapy had focused on antiproliferative strategies. Now it is generally recognized that tumor growth occurs when cells lose the normal balance between cell proliferation and apoptosis. We also know that apoptosis contributes to tumor cell death induced by many cancer chemotherapeutic agents (6); thus, delineating the biological pathways of candidate anticancer agents is imperative.

Initially developed as a coccidiostatic agent,CAI,³ an inhibitor of receptor-operated calcium channel-mediated calcium influx, was shown to have antiproliferative and anti-invasive functions in several human cancer cell lines, including human glioblastoma cells (7, 8, 9). Its mechanism of action has been proposed to inhibit calcium-sensitive signal transduction pathways and modulate the production of secondary messengers such as calcium, inositol phosphate,and arachidonic acid (7). Despite its well-documented cytostatic effect, CAI has not previously been shown to have any cytocidal function. In this report, we demonstrate that CAI induces apoptosis in BAECs and human glioma cells in a dose and time dependent manner.

CAI was a gift from Dr. Elise C. Kohn (Laboratory of Pathology,National Cancer Institute, Bethesda, MD). Stock solution was made in DMSO and kept at −20°C. Poly-l-lysine, DAPI, and PI were purchased from Sigma. Texas Red-X phalloidin was purchased from Molecular Probes. The APO-Direct TUNEL direct immunofluorescence kit was purchased from PharMingen.

BAECs (Clonetics) were a gift from Dr. Steven Brown (Henry Ford Hospital, Detroit, MI). BAECs and human glioblastoma cell line U251N (10) were cultured in MEM (10% FCS and 1%nonessential amino acids) and DMEM (10% FCS), respectively, at 37°C with 5% CO₂. Cells were cultured to approximately 50% confluence at 37°C with 5%CO₂ overnight to insure complete attachment of cells to the culture matrix. The next day, cells were treated with prewarmed fresh medium, with or without CAI, for the various time periods indicated in each of the experiments.

Approximately 2 -3 × 10⁵ cells were plated onto poly-l-lysine-coated coverslips and cultured overnight to ensure complete attachment of cells to coverslips. The next day,cells were changed to fresh medium containing 0, 5, 10, 20, and 40μ m concentrations of CAI for 24 h. After treatment,cells were washed once in PBS and then fixed in 3.7% paraformaldehyde in PBS for 30 min at room temperature, followed by two washes in PBS. Cells were then permeabilized in 1% Triton X-100 in PBS for 3 min and washed twice in PBS. Cells were stained with the nuclear staining dye DAPI (10 μg/ml in PBS) for 20 min at room temperature, followed by two washes in PBS. Cells on coverslips were mounted onto glass slides in antifade medium and sealed with nail polish. Nuclear staining was visualized using a Zeiss Axiophot 2 epi-fluorescent microscope (×400 magnification). The number of cells with apoptotic bodies and total nuclei from three to six high-power fields were counted, and the percentage of apoptosis was calculated as the mean ± SD.

Cells were cultured and treated as described above and stained according to the manufacturer’s recommendation, except that F-actin was also stained with Texas Red-conjugated phalloidin (10 units/ml) to visualize the cytoskeleton. Cells were viewed with either an Axiophot 2 epi-fluorescent microscope (Zeiss) or an Axiovert 100 (Zeiss)microscope attached to a MRC 1024 krypton/argon laser confocal imaging system (Bio-Rad). Images were acquired and composed with LaserSharp 2.0 software (Bio-Rad).

Approximately 2 × 10⁶ cells were collected and washed once in PBS. Cell pellets were lysed in 0.5 ml of lysis buffer [10 mm Tris (pH 7.6) and 0.6% SDS], followed by the addition of 4.0 m NaCl to a final concentration of 1.0 m, and mixed well. Lysates were centrifuged at 12,000 rpm for 30 min at 4°C. Supernatants were collected and incubated at 37°C for 60 min in the presence of 50 μg/ml RNase A, followed by a single phenol-chloroform (v:v, 1:1) extraction. Genomic DNA was precipitated with 2 volumes of 100% ethanol in the presence of 0.01 m MgCl₂ and 25 μg/ml tRNA at−80°C for at least 2 h. DNA was pelleted by centrifugation at 12,000 rpm for 30 min at 4°C. Pellets were washed twice with 70%ethanol and then air-dried before being resuspended and stored in 25μl of TE [10 mm Tris and 1 mm EDTA (pH 8.0)] buffer. The amount of DNA was determined by A_{260 nm}value, and its purity was determined by A_{260 nm}:A_{280 nm}ratio. All DNA preparations had an A_{260 nm}:A_{280 nm}ratio of >1.8. Equal amounts of DNA were loaded onto a 0.8% agarose gel and electrophoresed in 0.5× TBE buffer [45 mm Tris-borate and 1 mmEDTA (pH 8.0)] for 16 h at 2 V/cm. DNA fragmentation was visualized by ethidium bromide (0.5 μg/ml) staining under UV light. Images were captured using the UVP Video Copy System (Mitsubishi) and processed using Adobe Photoshop.

Alternatively, 2 × 10⁶ cells were collected and washed in PBS. Cell pellets were suspended in 40 μl of PC buffer([192 parts of 0.2 mNa₂HPO₄ (pH 7.4) and 8 parts of 0.1 m citric acid (pH 7.8)] for 30 min at room temperature and then centrifuged at 1000 × g for 5 min (11). The supernatants were transferred into a new tube,concentrated in a Speed Vac just until they were dry, resuspended in 0.25% NP40 (3 μl) and 1 mg/ml RNase A (3 μl), and incubated at 37°C for 30 min. After the addition of 1 mg/ml proteinase K (3 μl),the samples were incubated for 30 min at 37°C, followed by the addition of 12 μl of loading buffer (0.25% bromphenol blue and 30%glycerol). Total contents of the tubes were loaded onto a 0.8% agarose gel in 0.5× TBE buffer and electrophoresed for 16 h at 2 V/cm.

Cells (2 × 10⁶) were treated and collected as described above and fixed in 3 ml of ice-cold 70% ethanol for 30 min on ice. Ethanol was removed by centrifugation at 450 × g for 5 min. Cell pellets were washed twice in PBS containing 1% bovine serum and then incubated with 3 ml of PC buffer for 10–15 min on ice. Cells were pelleted, washed once in PBS containing 1% bovine serum, and then incubated in 1 ml of PI-RNase (25μg/ml PI, 0.5 mg/ml RNase A in PBS) solution for 30 min at room temperature. Samples were analyzed using a Coulter Epics XL/XL-MCL flow cytometer. Apoptotic cells with DNA contents less than the G₁-G₀ content appeared as the sub-G₀-G₁ peak on the DNA histogram (25,000 events were counted). The percentage of apoptosis reflects the percentage of total cells that were in sub-G₀-G₁.

Multiple linear regression analysis was used to model the percentage of cells that were apoptotic. The CAI concentration, time, and the product of CAI concentration and time were used as predictor variables. Linear correlation coefficients and Ps from the regression models were computed.

To determine whether CAI induces apoptosis, we treated BAECs with increasing concentrations of CAI as described in “Materials and Methods.” After 24 h of CAI treatment, apoptotic nuclei were visualized using the DNA intercalating dye DAPI. The bright blue apoptotic nuclei were readily identified by their condensed chromatin and apoptotic bodies (Fig. 1,A). As reported for most other cultured cells, a few BAECs were undergoing apoptosis under normal culture conditions (0 μm). However, we observed a concentration-dependent increase in the number of apoptotic cells 24 h after CAI treatment (Fig. 1,A). CAI concentrations of 0, 5, 10, 20, and 40 μmresulted in 1.9%, 3.5%, 6.6%, 13.1%, and 27.2% apoptosis,respectively (Fig. 1 B). The relationship between CAI concentration and the percentage of apoptosis was statistically significant (r = 0.99; P < 0.001).

We further investigated the CAI-induced DNA strand breaks in BAECs by TUNEL, as described in “Materials and Methods.” As shown in Fig. 1 C, TUNEL-positive signals demonstrating DNA strand breaks were observed in cells that also demonstrated characteristic nuclear chromatin condensation and apoptotic body formation.

To determine whether CAI induces apoptotic DNA fragmentation, another hallmark of apoptosis, total genomic DNA was isolated from CAI-treated BAECs and resolved by agarose gel electrophoresis (see Fig. 2,A). CAI induced DNA fragmentation at concentrations of 20 and 40μ m. To further characterize the dose- and time-dependent responses of CAI-induced apoptosis, we treated BAECs with 0, 20, and 40 μm CAI for 1, 2, and 5 days. After 1 day and 2 days of treatment, DNA fragmentation was present in the 40 μm CAI-treated groups, but not in the 20μ m CAI-treated groups. However, when the treatment was prolonged to 5 days, both 20 μmCAI- and 40 μm CAI-treated BAECs underwent apoptosis by this criterion (Fig. 2 B).

Although we observed a dose-dependent response in CAI-induced apoptosis in BAECs by morphological analysis (Fig. 1, A and B), we could not quantify the percentage of apoptosis of the total cell population because a significant number of cells detached at 40 μm CAI treatment (Fig. 1,A). To better quantitate the CAI-induced apoptosis, we performed a flow cytometric analysis (Fig. 2 C). Increased apoptosis was observed in cells after 1 day and 2 days of treatment with 20μ m CAI (2.2% and 14.4%, respectively);however, 44.6% of the cells underwent apoptosis at 5 days of treatment. Consistent with the DNA fragmentation results, 40μ m CAI induced increasing levels of apoptosis at days 2 (59.5%) and 5 (86.9%) of treatment. The time by CAI concentration interaction term was significant (P =0.034), suggesting that the relationship between CAI concentration and apoptosis differs by day.

To determine whether CAI also induces DNA fragmentation in a human glioma cell line, U251N cells were treated with CAI (Fig. 3, A and B) as described previously for BAECs. Fig. 3 A demonstrates the dose-dependent response of U251N cells to CAI treatment. At a concentration of 40 μm, CAI induced DNA fragmentation. No apparent DNA fragmentation was observed in cells treated with CAI concentrations lower than 20μ m.

Dose- and time-dependent induction of DNA fragmentation in U251N cells is illustrated in Fig. 3, A and B. A CAI concentration of 40 μm induced DNA fragmentation as early as 1 day, with increasing fragmentation occurring by days 2 and 5. A CAI concentration of 20μ m did not induce DNA fragmentation until 5 days after treatment.

Quantitative analysis of apoptosis in the U251N cells was evaluated with flow cytometry (Fig. 3 C) and showed dose- and time-dependent patterns similar to those demonstrated by the DNA fragmentation gels. At 40 μm concentration,18.7%, 36.8%, and 48.4% of U251N cells underwent apoptosis after 1,2, and 5 days treatment with CAI, respectively. For the group treated with 20 μm CAI, there was a less prominent but time-dependent response (6%, 9.9%, and 21.8% apoptosis at days 1, 2,and 5, respectively). Statistically, there was a significant(P = 0.044) interaction between time and concentration of CAI, suggesting that the relationship between CAI and apoptosis differs by day. In particular, CAI was significantly (P = 0.05) related to apoptosis at day 5 only, and not at day 2(P = 0.21) or day 1 (P = 0.21). Also,there was a significant day effect at CAI concentrations of 20μ m (P = 0.002).

CAI, a calcium channel blocker, was shown to have antitumor effects, including antimigratory, antiproliferative, and antimetastatic effects (7, 8, 9). Furthermore, in vivo studies demonstrate no significant end organ toxicities to endothelium (12) while targeting angiogenesis (13). Whereas CAI is considered to be cytostatic, its cytocidal effects (14) have not been fully explored. We report here that CAI at concentrations of ≥20 μm, induces apoptosis in BAECs and U251N, a human glioma cell line. This apoptotic effect is both dose and time dependent, as demonstrated by morphological and biochemical parameters. Thus, in addition to its cytostatic effect as reported by others, CAI may also exert a cytotoxic effect via induction of apoptosis in both vascular endothelial cells and tumor cells.

During the process of apoptosis, cells undergo serial structural and molecular changes. These include the plasma membrane blebbing, nuclear chromatin condensation, apoptotic body formation, and cleavage of genomic DNA into high molecular weight (300 and 50 kb) DNA or 150–200-bp nucleosomal DNA (15). Therefore, to fully evaluate CAI-induced apoptosis, we used several detection methods:(a) cell morphology and DAPI nuclear staining to illustrate chromatin condensation and apoptotic body formation; (b)TUNEL to demonstrate DNA strand breaks; (c) gel electrophoresis to illustrate DNA fragmentation; and (d)flow cytometry to assess the sub-G₀-G₁ fraction of DNA content. We observed apoptosis using all of these analyses. Therefore,by four independent methods, our data support the conclusion that CAI induces apoptosis in BAECs and tumor cells.

In our current report, we demonstrate that CAI induces apoptosis in both BAECs and human glioma cells in both a dose- and time-dependent manner. The 40-μm concentration of CAI induced apoptosis within 24 h after treatment, with increasing levels observed at days 2 and 5. However, the 20-μm concentration of CAI did not induce apoptosis until 5 days after treatment. This is significant in that it implies that relatively low doses of CAI can achieve the cytocidal effects observed with high doses of CAI by prolonging the treatment. In clinical settings, doses of CAI are limited by adverse side effects, but achievable concentrations of CAI for more prolonged periods may provide adequate doses for effect. The observation that CAI induces apoptosis in proliferating endothelial cells as effectively as in tumor cells suggests that it may be equally effective as an antiangiogenic and antitumor agent.

The antiproliferative and anti-invasive effects of CAI have been well documented (7, 8, 9, 13, 14). In this study, we used concentrations of CAI that were previously shown to negatively effect tumor cell proliferation and invasion. The CAI concentrations that demonstrate apoptosis were those that produced the maximum antiproliferative and anti-invasive effects (20 and 40μ m). Concentrations below 10 μm did not induce apoptosis, although these concentrations did affect both proliferation and invasion. These results support the reported data that CAI concentrations above 10 μm are cytotoxic (14), and our data further demonstrate that the cytotoxicity is mediated by apoptosis.

As a key process in signal transduction pathways, calcium influx has been reported both to induce and protect cells from apoptosis,depending on the cell type and experimental systems (15, 16, 17, 18, 19, 20). Although the mechanism of calcium-induced apoptosis remains elusive, it has been reported that calcium influx directly activates calcium-dependent protease calpain and caspase as well (19, 20). Furthermore, specific calpain or caspase inhibitors block this type of apoptosis. In contrast, blocking of calcium influx also induces apoptosis in certain cell types, including neuronal cells (17) and hematopoietic cells (18).

The mechanism of action of CAI by inhibiting non-voltage-gated calcium uptake and the downstream calcium-sensitive signal transduction pathways is important in cancer cell growth and invasion. These signaling pathways include the release of arachidonic acid and its metabolites, the generation of inositol phosphates, and tyrosine phosphorylation (7, 13, 14). Our data do not permit us to define the pathway(s) involved with CAI-induced apoptosis. As an end-stage process, DNA fragmentation has been used extensively for the detection of apoptosis. DNA fragmentation factor is responsible for the cleavage of genomic DNA (21). In our study, DNA fragmentation was detected by all means used. Therefore, it is reasonable to speculate that DNA fragmentation factor activation is involved in CAI-induced apoptosis. Although CAI-mediated effects in migration and proliferation are a result of calcium influx inhibition (8, 9, 14), additional studies are required to determine whether a similar mechanisms or other mechanisms are involved in CAI-induced apoptosis. Our ongoing investigations are focused on elucidating the mechanisms involved.

We thank Dr. Elise Kohn for CAI and Dr. Steven Brown for the BAECs.