Glutathione Oxidation and Mitochondrial Depolarization as Mechanisms of Nordihydroguaiaretic Acid-Induced Apoptosis in Lipoxygenase-Deficient FL5.12 Cells

Shyam S. Biswal1, Kaushik Datta1, Stephanie D. Shaw, Xiang Feng, John D. Robertson and James P. Kehrer2

Division of Pharmacology and Toxicology, College of Pharmacy, The University of Texas at Austin, Austin, Texas 78712-1074

Received June 1, 1999; accepted August 23, 1999


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Nordihydroguaiaretic acid (NDGA) induces apoptosis in a variety of cell lines. The mechanism(s) of this effect is not known, although the focus has been on the ability of NDGA to inhibit lipoxygenase (LOX) activities. In the present study, NDGA-induced apoptosis was studied in a murine hematopoietic cell line, FL5.12. Although this cell line lacks detectable LOX protein or activities, NDGA (10 µM) was able to induce apoptosis. There was a massive loss of mitochondrial membrane potential by 4 h after the addition of NDGA, suggesting that this organelle might be targeted by NDGA. A pro-oxidant NDGA effect has been suggested as playing a role in apoptosis. This was supported by the findings that glutathione disulfide levels were increased by 4 h following treatment with 10 µM NDGA, that pretreatment with N-acetylcysteine completely blocked the NDGA-induced loss of membrane potential and apoptosis, and that lipid peroxidation was enhanced in cells treated with NDGA. However, no evidence of increased levels of reactive oxygen could be seen in NDGA-treated cells loaded with dichlorofluorescin diacetate or dihydrorhodamine and analyzed by flow cytometry. Bcl-xL protein levels were unaffected by NDGA treatment. Caspase-3 was rapidly activated with a peak at 8 h after FL5.12 cells were treated with NDGA. Ac-DEVD-CHO (25 µM) and boc-asp-FMK (20 µM) both inhibited caspase-3 enzyme activity by 97% 8 h after NDGA treatment. Boc-asp-FMK, a more general caspase inhibitor, delayed NDGA-induced apoptosis while Ac-DEVD-CHO, a more specific inhibitor of caspase-3, had no effect. These results suggest that NDGA-induced apoptosis happens through reactions that depolarize mitochondria, oxidize glutathione and lipids, but do not generate significant amounts of free reactive oxygen species.

Key Words: caspase-3; nordihydroguaiaretic acid; apoptosis; lipoxygenase; bcl-xL; mitochondrial potential; N-acetylcysteine; glutathione.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Apoptosis is a form of cell death that occurs as an integral part of tissue homeostasis and in association with various diseases and disorders. Apoptosis can be triggered after receiving any one of a variety of death signals. These can include growth factor withdrawal, stimulation of cell surface receptors, and cellular stresses such as DNA damage, ionizing or UV radiation, viral infection, heat shock, or hypoxia (Hale et al., 1996Go). The apoptosis cascade can be conveniently divided into 3 phases. The induction phase is characterized by multiple signaling pathways that initiate the death program. During the subsequent effector phase, numerous apoptosis-inducing stimuli converge into a few central pathways that irreversibly commit the cell to death. During the degradation phase, vital structures and functions are destroyed, giving rise to the full-blown phenotype of apoptosis.

Lipoxygenase (LOX) products, including hydroperoxy (HPETEs) or hydroxy (HETEs) fatty acids, have been implicated as important elements in the regulation of tumor cell growth by modulating cell proliferation and apoptosis (Tang et al., 1996Go). Blocking the activity of LOXs will induce apoptosis in some cell lines. For example, nordihydroguaiaretic acid (NDGA), a general LOX inhibitor, induces apoptosis in some, but not all cell lines tested, with W256 cells, a monocytic cell line (Tang et al., 1996Go; 1997), being the most sensitive. NDGA also potently induces apoptosis in FL5.12 cells, a pro-B lymphocytic cell line (Datta et al., 1998Go) but blocks apoptosis in nerve, glioma, neuroblastoma, melanoma, and erythroleukemia cells (Li et al., 1997Go; Maccarrone et al., 1997Go; Wagenknecht et al., 1997Go).

Because of the discrepant effects of NDGA in terms of apoptosis, it is possible that the inhibition of LOX by this compound is not related to its ability to induce this form of cell death. NDGA-induced apoptosis is associated with lipid peroxidation and the depletion of glutathione (GSH) (Tang and Honn, 1997Go). Although NDGA is usually considered an antioxidant, this suggests some pro-oxidant activity in at least some biological systems. Since oxidants are capable of inducing apoptosis (Payne et al., 1995Go), this effect could explain the pro-apoptotic activity of NDGA.

The concept that NDGA-induced apoptosis is unrelated to LOX activity is also supported by our finding that NDGA induces apoptosis in murine FL5.12 cells that lack LOX protein and activity (Datta et al., 1998Go). The current study documents the pro-oxidant activity of NDGA by showing increases in cellular levels of glutathione disulfide (GSSG) as well as lipid peroxidation products. Pretreatment with N-acetylcysteine (NAC) blocked apoptosis indicating an oxidative mechanism related to thiols. Interestingly, there was no evidence of an increase in free levels of reactive oxygen species (ROS). There was, however, a complete loss of mitochondrial membrane potential that could be blocked by NAC in NDGA-treated cells. Compared to the dramatic decrease of bcl-xL expression after treatment of these cells with the 5-lipoxygenase-activating protein inhibitor, MK886, which also induces apoptosis (Datta et al., 1998Go), the expression of this anti-apoptotic protein did not change after treatment with NDGA. Although activation of caspase-3 was evident after treatment with NDGA, blocking this activity with only the most general caspase inhibitor affected apoptosis.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cell culture and treatment.
An IL-3-dependent murine pro-B lymphocytic cell line (FL5.12) was used in this study. The cell line, grown in suspension culture, was maintained in RPMI-1640 media (Sigma, St. Louis, MO) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (Atlanta Biologicals, Northcross, GA), 1% glutamine, penicillin, and streptomycin (100 U/ml and 100 µg/ml, respectively, Gibco, Gaithersberg, MD) and 10% (v/v) IL-3-conditioned medium in a 5% CO2:95% air atmosphere at 37°C. IL-3-conditioned medium was obtained from the WEHI-3B cell line grown in suspension culture under similar conditions (Lee et al., 1982Go). Cultures were passaged on alternate days with fresh medium, and cell counts were determined with a T-890 Coulter counter (Coulter, Miami, FL). In most of the experiments, cells were treated with either ethanol (0.1% final concentration) as vehicle control, NDGA (Sigma), or MK886 (Biomol Research, Plymouth Meeting, PA). In some experiments, cells were treated with either boc-asp.FMK (20 µM) or Ac-DEVD.CHO (25 µM) to inhibit caspase activity (Enzyme Systems, Livermore, CA) for 1.5 h before treatment with NDGA. In experiments involving NAC, cells were pretreated for 1.5 h at a concentration of 10 mM.

Measurements of apoptosis.
A characteristic event in apoptosis is DNA fragmentation and release of nucleosomes into the cytoplasm. These can be detected by an ELISA assay that is quite sensitive and specific for apoptosis relative to necrosis (Salgame et al., 1997Go). Cells were gently lysed releasing mono- and oligonucleosomes from apoptotic cells. A 1-step ELISA kit (Roche Molecular, Indianapolis, IN) using streptavidin-coated microplates, biotinylated antihistone, and peroxidase conjugated anti-DNA antibodies completes the assay. Results are expressed as the {Delta}A405–490, which indicates the enrichment of nucleosomes in the cytoplasm.

Apoptosis was quantitated by visualizing distinctive nuclear and cytoplasmic fluorescence, chromatin condensation and formation of multiple apoptotic bodies following staining of cells with acridine orange and ethidium bromide (Duke and Cohen, 1992Go). The validity of the acridine orange/ethidium bromide assay for detecting apoptosis in this cell line has been previously documented (Bojes et al., 1997Go). Treated and control cells were pelleted at 300 x g for 5 min at 4°C. The cells were then resuspended in 20 µl of fresh medium containing 4 µl of ethidium bromide/acridine orange mixture (1:1 v/v, stock solution 100 µg/ml each). This suspension (10 µl) was examined under a fluorescent microscope in order to differentiate early from late apoptotic cells and normal and necrotic cells. A minimum of 400 cells per sample was counted (magnification, 40x).

Additional confirmation of apoptosis was by quantitation of viable, early and late apoptotic (secondary necrotic) cells using a Coulter EPICS-XL flow cytometer with the Annexin V FITC kit (Coulter, Miami, FL), which utilizes phosphatidylserine externalization as a marker of apoptosis.

Glutathione assay.
Total glutathione (GSH + GSSG) and GSSG were determined using a modified method of Neuschwander-Tetri and Roll (1989). Cells (5 x 106/ml) were collected following incubation with NDGA and centrifuged for 10 min at 300 x g at 4°C. The pellet was resuspended in 1 ml EDTA (2 mM) and sonicated for 2 min. Total glutathione was determined in the cell lysate by adding 83 µl of 25 mM NaH2PO4 (pH 7.0) to 250 µl of lysate. An additional 250 µl of cell lysate was mixed with 83 µl of N-ethylmaleimide for GSSG measurements. All samples (200 µl) were then mixed with 200 µl 25 mM dithiothreitol (in 25 mM NaH2PO4, pH 7.0) followed by 100 µl Tris buffer (pH 8.5), and were incubated on ice for 30 min. Following addition of 0.5 ml of 2.5% (w/v) sulfosalicylic acid, the samples were centrifuged for 10 min at 600 x g at 4°C. An aliquot (200 µl) of supernatant was mixed with 200 µl of o-phthalaldehyde (5 mg/ml in 0.4 M potassium borate solution, pH 9.9) and incubated for 2 min at 25°C. The mixture was neutralized and samples were either kept on ice in the dark and used immediately or were stored at –80°C. Derivatized samples were analyzed by HPLC as described previously (Liu and Kehrer, 1996Go).

Lipid peroxidation assay.
Lipid peroxidation was assessed by the thiobarbituric acid assay (Ohkawa et al., 1979Go). Cells were washed in phosphate-buffered saline before undergoing 3 cycles of freeze-thawing in 200 µl of water. A 20 µl aliquot was then removed for protein determination (Bradford, 1976Go). The assay mix (0.4% (w/v) thiobarbituric acid, 0.5% (w/v) SDS, 9.4% (v/v) acetic acid, pH 3.5; 980 µl) was added to the remaining sample. Samples were incubated for 60 min at 95°C, cooled to room temperature, and centrifuged at 14,000 x g for 10 min. A532 of the supernatants was determined.

Detection of ROS.
The generation of ROS was assessed using 2 different cell-permeable flurogenic probes, dihydorhodamine 123 (DHR123) and dichlorofluorescin diacetate (DCHFDA). Cells were preloaded with either DHR123 (1 µM) or DCHFDA (10 µM) for 30 min, respectively. Cells were then treated with 10 µM NDGA and the samples analyzed after 30 min, 1 h, and 3 h in an Epics XL flow cytometer. H2O2 (0.01%) was used as a positive control.

Measurement of mitochondrial membrane potential.
Changes in mitochondrial membrane potential were detected using the fluorescence-based ApoAlertTM mitochondrial membrane sensor kit (Clontech Laboratories, Palo Alto, CA). Cells (106/ml) were treated with 10 µM NDGA for 2 and 4 h. After treatment, cells were pelleted, washed with PBS, and resuspended in 1 ml MitoSensorTM reagent. Following incubation at 37°C in 5% CO2 for 20 min, cells were diluted with 1 ml of Clontech incubation buffer and then pelleted. The final cell pellet was resuspended in 30 µl fresh incubation buffer and cells were observed by fluorescence microscopy, using a Nikon E800 microscope equipped with a FITC/Texas Red, dual-band filter set (#51006, Chroma Technology Corp., Brattleboro VT; wavelengths in nm: excitation 497, 571; dichroic mirror 375–410, 513–552; barrier filter 531, 627). The MitoSensorTM reagent, cationic by nature, is taken up by normal mitochondria where it forms aggregates that exhibit red fluorescence. A loss in mitochondrial membrane potential prevents this aggregation and MitoSensor remains in the cytoplasm as a monomer where it exhibits green fluorescence.

Immunoblotting.
Cells were pelleted and lysed by repeated pipetting in 15 µl/106 cells ice-cold buffer containing 10 mM Tris (pH 7.4), 10 mM NaCl, 3 mM MgCl2, 1 mM EDTA, 0.1% (v/v) NP-40, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, and 30 µl/ml aprotinin. Cell lysates were centrifuged at 16,000 x g for 10 min at 4°C and total protein content in the supernatant was measured (Bradford, 1976Go). Supernatants were run on 15% reducing SDS–polyacrylamide gels. Protein was transferred to PVDF membranes (Millipore, Bedford, MA) and blocked for at least 1 h in 5% (w/v) non-fat dry milk (BioRad, Hercules, CA). The membrane was then incubated for 1 h with anti-bcl-xL rabbit polyclonal antibody (1:1500), or a goat polyclonal antibody (1:1500) that recognizes both the pro-caspase-3 (32-KD) and the p20 cleavage product subunit (both antibodies were from Santa Cruz Biotechnology, Santa Cruz, CA). The proteins were visualized using an enhanced chemiluminescence detection kit (Amersham Pharmacia Biotech, Arlington Heights, IL).

Measurement of caspase-3 activity.
Ac-DEVD-7-amino-4-methyl-coumarin (Ac-DEVD.AMC) was used as a fluorometric substrate for determining caspase-3 activity (Thornberry et al., 1992Go). Cells (0.7 x 106) were pelleted and lysed in 200 µl of 50 mM Tris (pH 7.5) containing 150 mM NaCl, 0.5 mM EDTA, and 0.5% v/v NP-40. Aliquots (50 µl) of the lysate were incubated with tetrapeptide substrate (40 µM), 10 mM HEPES (pH 7.5), 50 mM NaCl and 2.5 mM DTT in a final volume of 200 µl for 2 h at 37°C. The fluorescence of the released 7-amino-4-methyl coumarin was measured at excitation and emission wavelengths of 360 nm and 460 nm.

Statistical analyses.
Data are expressed as the means ± SE. Comparisons between groups were carried out using ANOVA followed by post hoc analyses with Student Newman-Keul's test. A p value of less than 0.05 was considered significant.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
NDGA-Induced Apoptosis
A preliminary dose-response study assessing apoptosis in FL5.12 cells treated with NDGA showed that at least 10 µM was required to induce significant apoptosis at 12 h. The time course of apoptosis following treatment with 10 µM NDGA is shown in Table 1Go. Apoptosis was evident as early as 4 h, increasing to 45% at 18 h when significant necrosis was also seen. Similar results were observed using flow cytometry with Annexin V–FITC (data not shown).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Time Course of Apoptosis after Treatment of FL5.12 Cells with NDGA
 
Oxidative Processes after NDGA
There were no significant changes in total glutathione following treatment with NDGA (Table 2Go). However, GSSG levels did significantly increase, and the total glutathione:GSSG ratio was decreased by as early as 4 h (Table 2Go). A 1.5-h pretreatment with NAC conferred complete protection against the induction of apoptosis by NDGA at 8 h, as seen by flow cytometric analysis of phosphatidylserine externalization (Table 3Go). Further evidence of oxidative processes was obtained using the thiobarbituric acid assay, which revealed a 6-fold increase one hour after treatment with 10 µM NDGA. In contrast, there was no evidence of an increased production of ROS when cells were loaded with either DHR123 or DCFHDA (data not shown).


View this table:
[in this window]
[in a new window]
 
TABLE 2 Cellular Glutathione after NDGA Treatment
 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Protection by N-acetylcysteine against NDGA-Induced Apoptosis
 
Loss of Mitochondrial Membrane Potential
Control cells loaded with MitoSensorTM reagent exhibited a consistent red fluorescence when incubated for up to 4 h (Fig. 1AGo). Two h after 10 µM NDGA, there was no change in this fluorescence (Fig. 1BGo). However, at 4 h after NDGA, there was a massive loss of mitochondrial membrane potential as shown by the complete conversion to green fluorescence (Fig. 1CGo). This effect was completely prevented in cells treated with 10 mM NAC along with NDGA (Fig. 1DGo).



View larger version (15K):
[in this window]
[in a new window]
 
FIG. 1. Mitochondrial potential in FL5.12 cells treated with 10 µM NDGA. MitoSensor was loaded into these cells as described in Materials and Methods. A shows control cells which exhibit the red fluorescence characteristic of mitochondrial accumulation of this dye. B shows cells treated for 2 h with NDGA. These cells appeared identical to control cells. C shows cells treated for 2 h with NDGA. The green fluorescence demonstrates the loss of membrane potential. D shows cells 4 h after treatment with 10 µM NDGA plus 10 mM NAC. The red fluorescence indicates that NAC prevented the NDGA-induced loss in mitochondrial potential. Original magnification, 40x.

 
Bcl-xL Expression
Previous work showed that treatment of FL5.12 cells with MK886, an inhibitor of 5-lipoxygenase activating protein, leads to a rapid loss of the anti-apoptotic protein bcl-xL, possibly contributing to the observed apoptosis (Datta et al., 1998Go). In contrast to that finding, 8 h after a 10-µM NDGA treatment, there was no change in the expression of bcl-xL as measured by Western blot (Fig. 2Go).



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 2. Western blot analysis of bcl-xL expression upon NDGA treatment. Cells (106) were plated and treated for 8 h with either ethanol as a vehicle control or 10 µM NDGA. SDS–PAGE was performed as described in Materials and Methods; 20 µg total protein was loaded per lane.

 
NDGA-Induced Changes in Caspase-3 Activity
Caspase-3-like activity, as measured by enzymatic cleavage of Ac-DEVD-AMC, was 25-fold higher than control cells 8 h after NDGA treatment (Table 4Go). By 18 h after NDGA, caspase-3-like activity had decreased to ~4-fold above control. Western blot analyses, using an antibody that recognizes the holoenzyme as well as the p20 cleavage product of caspase-3, were performed to confirm that enzymatic processing occurred following NDGA treatment. At time zero, immunoblots showed the presence of the 32-kDa precursor of caspase-3, as well as relatively small amounts of the 20-kDa cleavage product (Fig. 3Go). Eight h after treatment with 10 µM NDGA, the p20 subunit of the mature caspase-3 enzyme was increased while the p32 holoenzyme was decreased (Fig. 3Go). By 18 h, the majority of the 32 kDa precursor had disappeared, resulting in maximal amounts of the p20 fragment (Fig 3Go).


View this table:
[in this window]
[in a new window]
 
TABLE 4 Caspase-3 Activity after NDGA and Caspase-Inhibitor Treatment
 


View larger version (16K):
[in this window]
[in a new window]
 
FIG. 3. This shows Western blot analysis of caspase-3 cleavage upon NDGA treatment. Cells (106) were plated and treated with 10 µM NDGA for 0, 4, 8, or 18 h. Cells were lysed and the lysate was used to perform SDS–PAGE as described in Materials and Methods; 20 µg total protein was loaded per lane.

 
Effect of Caspase Inhibitors
Boc-asp.FMK (20 µM), a general caspase inhibitor, and Ac-DEVD.CHO (25 µM), a more specific reversible caspase-3 inhibitor, prevented any increase in caspase-3-like activity at 8 h in cells treated with NDGA (Table 4Go). With boc-asp.FMK, this inhibitory effect persisted for at least 18 h while, with Ac-DEVD.CHO, there was a small increase in caspase-3-like activity at this time (Table 4Go).

Boc-asp.FMK, but not Ac-DEVD.CHO, was able to inhibit NDGA-induced cell death. This was evident using flow cytometry to assess changes in phosphatidylserine distribution (Table 5Go), acridine orange/ethidium bromide staining to assess nuclear morphology and membrane permeability, and an ELISA assay to assess DNA fragmentation (Table 6Go). Protection by boc-asp.FMK was virtually complete at 8 h. At 18 h, the protection provided by boc-asp.FMK remained complete as assessed by ELISA (Table 7Go), but was no longer complete in terms of phosphatidylserine externalization as measured by flow cytometry (Table 5Go) and was absent using acridine orange/ethidium bromide staining (63% apoptosis with NDGA and 61% with NDGA + boc-asp.FMK). These data suggest that boc-asp.FMK delays, but does not prevent NDGA-induced apoptosis. However, since NDGA was present throughout the time period examined, the delay may simply reflect the gradual loss of boc-asp.FMK caspase inhibition activity in the presence of continued levels of NDGA.


View this table:
[in this window]
[in a new window]
 
TABLE 5 Effect of Caspase Inhibitors on NDGA-Induced Apoptosis
 

View this table:
[in this window]
[in a new window]
 
TABLE 6 Effect of Caspase Inhibitors on NDGA-Induced Apoptosis
 

View this table:
[in this window]
[in a new window]
 
TABLE 7 Effect of Caspase Inhibitors on NDGA-Induced Apoptosis as Shown by ELISA
 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A role for LOX in mediating apoptosis in some tumor cell lines has been suggested by several studies. Some of this work has directly explored LOX by demonstrating that W256 cells transfected with a 12-LOX-specific antisense oligonucleotide or antisense oligonucleotides directed to conserved regions of LOXs undergo time- and dose-dependent apoptosis (Tang et al., 1997Go). Other work has been more indirect, employing various inhibitors of this enzyme. One of the most common LOX inhibitors used has been NDGA. However, NDGA actually prevents apoptosis in some systems and is able to induce apoptosis in a cell line that lacks measurable LOX activity or protein (current data and Datta et al., 1998Go). This demonstrates that the induction of apoptosis by NDGA is mediated by pathways independent of LOX.

A report that NDGA leads to lipid peroxidation and a loss of GSH, but that 12-LOX was involved (Tang and Honn, 1997Go), made it of interest to determine whether oxidative processes could be detected following induction of apoptosis with NDGA in FL5.12 cells, which lack LOX. A substantial increase in GSSG was observed in FL5.12 cells, confirming and extending this prior report. The occurrence of lipid peroxidation in FL5.12 cells is also consistent with the conclusion that NDGA stimulates oxidative reactions. The failure to find changes in DHR123 or DCFH fluorescence after treatment with NDGA was, therefore, surprising. A previous study in W256 cells treated with NDGA and loaded with DHR123 reported the same result (Tang and Honn, 1997Go). These authors concluded, based on this finding and other data, that NDGA causes oxidative stress through non-conventional mechanisms. The current data are consistent with this conclusion and suggest that rather than stimulating the production of ROS, NDGA is perhaps directly oxidizing GSH leading to mitochondrial depolarization and apoptosis.

Support for the idea that NDGA acts through a thiol oxidation mechanism comes both from prior work showing that the apoptosis induced by NDGA can be effectively inhibited by GSH-elevating thiol agents and lipid peroxidation inhibitors (Tang and Honn, 1997Go) and the current study demonstrating that NAC protects against the NDGA-induced loss of mitochondrial potential and apoptosis in FL5.12 cells. Numerous studies have implicated mitochondria in apoptosis (Green and Reed, 1998Go), and NDGA is known to have some mitochondrial effects. For example, NDGA inhibits succinate cytochrome c reductase in EMT6 mouse mammary carcinoma cells while sulfhydryl compounds, including GSH and cysteine, prevent the inhibition of succinoxidase activity by NDGA (Shi and Pardini, 1995Go). This suggests that NDGA exerts its biological effects by oxidation of sulfhydryl groups in key mitochondrial systems. The mechanism of such oxidation is unclear, but it could be a direct effect of NDGA or occur indirectly following the formation of GSSG, perhaps as a result of some sort of NDGA-dependent redox cycle.

The disruption of the glutathione redox state through the formation of GSSG may be the critical factor in the induction of apoptosis, since the simple depletion of GSH by BSO is unable to cause apoptosis in this (Bojes et al., 1999Go) or other (Tang and Honn, 1997Go) cell lines. The oxidation of mitochondrial GSH may explain the profound loss in mitochondrial membrane potential by 2 to 4 h after treating cells with NDGA. Although this effect cannot yet be directly linked to the apoptosis observed in this system, the temporal data are suggestive. Specifically, NDGA causes lipid peroxidation by 1 h and GSSG oxidation at 4 h, while no mitochondrial depolarization was evident at 2 h but was complete at 4 h. Since NAC prevents both mitochondrial depolarization and apoptosis, and since other data demonstrate that changes in mitochondrial permeability and thus mitochondrial potential are critical steps in apoptosis (Susin et al., 1998Go), the theory that a relationship between NDGA-induced thiol oxidation, mitochondrial depolarization, and apoptosis is at least consistent with the current data. It is also possible that the NDGA-catalyzed formation of lipid peroxides represents an important apoptosis-signaling mechanism since such species are known to induce apoptosis (Sandstrom et al., 1994Go).

Multiple lines of evidence indicate that caspases are important effectors of apoptosis (Thornberry and Lazebnik, 1998Go). Based on structural homology, caspases can be broken into three subfamilies: (1) Interleukin-converting enzyme (ICE) subfamily, (caspase-1,-4 and –5); (2) CED-3/CPP32 subfamily, (caspases-3, -6, -7, -8, -9 and -10); and (3) ICH-1/Nedd2 subfamily (Cohen, 1997Go). The activation of caspase-3 in particular is considered essential for the progression of apoptosis (Kuida et al., 1998Go). The current study provides the first report that NDGA treatment activates caspase-3. Blocking general caspase activity with boc-asp.FMK largely prevented NDGA-induced apoptosis. In contrast, a more specific caspase-3 inhibitor, Ac-DEVD.CHO, was ineffective. These results suggest that while apoptosis induced by NDGA may require caspase activity, a specific requirement for caspase-3 is unclear. Since not all apoptotic cell deaths are inhibited by caspase inhibitors (Bojes et al., 1999Go; Borner and Monney, 1999Go; Xiang et al., 1996Go), and since the specificity of these inhibitors is not absolute, it is likely that other caspase family members, particularly caspase-9, are involved.

In summary, a low dose of NDGA induced rapid and extensive apoptosis in FL5.12 cells. This effect was accompanied by an increase in GSSG and lipid peroxidation, and could be blocked by NAC. The massive loss of mitochondrial membrane potential evident at the time apoptosis could first be detected could also be blocked with NAC. A significant increase in caspase-3-like activity was seen after NDGA. However, the specific inhibition of caspase-3 did not protect cells from NDGA-induced apoptosis, while the more general inhibition of caspase activities delayed this effect. Thus, caspases other than caspase-3 may be important. The toxicity of NDGA appears to be caused through its pro-oxidant effects on thiols that generate pro-apoptotic signals and lead to a loss of mitochondrial membrane potential.


    ACKNOWLEDGMENTS
 
This work was supported by NIH grants HL48035 and HL51005. Support was also received from NIEHS Center Grant ES07784. JPK is the Gustavus and Louise Pfeiffer Professor of Toxicology.


    NOTES
 
1 These authors contributed equally to this work. Back

2 To whom correspondence should be addressed. Fax: (512) 471-5002. E-mail: KehrerJim{at}mail.utexas.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bojes, H. K., Datta, K., Xu, J., Chin, A., Simonian, P., Nuñez, G., and Kehrer, J. P. (1997). Bcl-xL overexpression attenuates glutathione depletion in FL5. 12 cells following IL-3 withdrawal. Biochem. J. 325, 315–319.[ISI][Medline]

Bojes, H. K., Feng, X., Kehrer, J. P., and Cohen, G. M. (1999). Apoptosis in hematopoietic cells (FL5.12) caused by interleukin-3 withdrawal: Relationship to caspase activity and the loss of glutathione. Cell Death Differ. 6, 61–70.[ISI][Medline]

Borner, C., and Monney, L. (1999). Apoptosis without caspases: An inefficient molecular guillotine? Cell Death Differ. 6, 497–507.[ISI][Medline]

Bradford, M. M. (1976).A rapid and sensitive method for the quantitation microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 136, 248–254.

Cohen, G. M. (1997). Caspases: the executioners of apoptosis. Biochem. J. 326, 1–16.[ISI][Medline]

Datta, K., Biswal, S. S., Xu, J., Towndrow, K. M., Feng, X., and Kehrer, J. P. (1998). A relationship between 5-lipoxygenase-activating protein (FLAP) and bcl-xL expression in murine pro B-lymphocytic FL5.12 cells. J. Biol. Chem. 273, 28163–28169.[Abstract/Free Full Text]

Duke, R. C., and Cohen, J. J. (1992). Morphological and biochemical assays of apoptosis. In Current Protocols in Immunology (J. E. Coligan and A. M. Kruisbeak, Eds.), pp. 3.17–3.17.16. John Wiley & Sons, New York.

Green, D. R., and Reed, J. C. (1998). Mitochondria and apoptosis. Science 281, 1309–1312.[Abstract/Free Full Text]

Hale, A. J., Smith, C. A., Sutherland, L. C., Stoneman, V. E. A., Longthorne, V. L., Culhane, A. C., and Williams, G. T. (1996). Apoptosis: Molecular regulation of cell death. Eur. J. Biochem. 236, 1–26.[Abstract]

Kuida, K., Haydar, T. F., Kuan, C.-Y., Gu, Y., Taya, C., Karasuyama, H., Su, M.-S., Rakic, P., and Flavell, R. A. (1998). Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9. Cell 94, 325–337.

Lee, J. C., Hapel, A. J., and Ihle, J. N. (1982). Constitutive production of a unique lymphokine (IL 3) by the WEHI-3 cell line. J. Immunol. 128, 2393–2398.[Abstract/Free Full Text]

Li, Y., Maher, P., and Schubert, D. (1997). A role for 12-lipoxygenase in nerve cell death caused by glutathione depletion. Neuron 19, 453–463.[ISI][Medline]

Liu, H., and Kehrer, J. P. (1996). The reduction of glutathione disulfide produced by t-butyl hydroperoxide in respiring mitochondria. Free Rad. Biol. Med. 20, 433–442.[ISI][Medline]

Maccarrone, M., Catani, M., Agro, A., and Melino, G. (1997). Involvement of 5-lipoxygenase in programmed cell death of cancer cells. Cell Death and Differentiation 4, 396–402.[ISI]

Neuschwander-Tetri, B. A., and Roll, J. (1989). Glutathione measurement by high-performance liquid chromatography separation and fluorometric detection of the glutathione-orthophthalaldehyde adduct. Anal. Biochem. 179, 236–241.[ISI][Medline]

Ohkawa, H., Ohishi, N., and Yagi, K. (1979). Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal. Biochem. 95, 351–358.[ISI][Medline]

Payne, C. M., Bernstein, C., and Bernstein, H. (1995). Apoptosis overview emphasizing the role of oxidative stress, DNA damage, and signal-transduction pathways. Leuk. Lymphoma 19, 43–93.[ISI][Medline]

Salgame, P., Varadhachary, A. S., Primiano, L. L., Fincke, J. E., Muller, S., and Monestier, M. (1997). An ELISA for detection of apoptosis. Nucleic Acids Res. 25, 680–681.[Abstract/Free Full Text]

Sandstrom, P. A., Tebbey, P. W., Van Cleave, S., and Buttke, T. M. (1994). Lipid hydroperoxides induce apoptosis in T cells displaying an HIV-associated glutathione peroxidase deficiency. J. Biol. Chem. 269, 798–801.[Abstract/Free Full Text]

Shi, L., and Pardini R. S. (1995). Effect of NDGA on beef heart mitochondria and EMT6 mouse mammary carcinoma cells. Res. Commun. Mol. Pathol. Pharmacol. 90, 235–254.[ISI][Medline]

Susin, S. A., Zamzami, N., and Kroemer, G. (1998). Mitochondria as regulators of apoptosis: D oubt no more. Biochim. Biophys. Acta 1366, 151–165.[ISI][Medline]

Tang, D. G., Chen, Y. Q., and Honn, K. V. (1996). Arachidonate lipoxygenases as essential regulators of cell survival and apoptosis. Proc. Natl. Acad. Sci. U S A 93, 5241–5246.[Abstract/Free Full Text]

Tang, D. G., and Honn, K. V. (1997). Apoptosis of W256 carcinosarcoma cells of the monocytoid origin induced by NDGA involves lipid peroxidation and depletion of GSH: Role of 12-lipoxygenase in regulating tumor cell survival. J. Cell. Physiol. 172, 155–170.[ISI][Medline]

Tang, D. G., Porter, A. T., and Honn, K. V. (1997). Critical role of arachidonate lipoxygenases in regulating apoptosis. Adv. Exp. Med. Biol. 407, 405–411.[ISI][Medline]

Thornberry, N. A., Bull, H. G., Calaycay, J. R., Chapman, K. T., Howard, A. D., Kostura, M. J., Miller, D. K., Molineaux, S. M., Weider, J. R., Aunins, J., Elliston, K. O. Ayala, J. M., Casano, F. J., Chin, J., Ding, G. J.-F., Egger, L. A., Gaffney, E. P., Limjuco, G., Palyha, O. C., Raju, S. M., Rolando, A. M., Salley, J. P., Yamin, T., Lee, T. D., Shively, J. E., MacCross, M., Mumford, R. A., Schmidt, J. A. and Tocci, M. J. (1992). A novel heterodimeric cysteine protease is required for interleukin-1 beta processing in monocytes. Nature 356, 768–774.[ISI][Medline]

Thornberry, N. A., and Lazebnik, Y. (1998). Caspases: Enemies within. Science 281, 1312–1316.[Abstract/Free Full Text]

Wagenknecht, B., Gulbins, E., Lang, F., Dichgans, J., and Weller, M. (1997). Lipoxygenase inhibitors block CD95 ligand-mediated apoptosis of human malignant glioma cells. FEBS Lett. 409, 17–23.[ISI][Medline]

Xiang, J., Chao, D. T., and Korsmeyer, S. J. (1996). Bax-induced cell death may not require interleukin 1ß-converting enzyme-like proteases. Proc. Natl. Acad. Sci. U S A 93, 14559–14563.[Abstract/Free Full Text]