Redox Regulation of Caspase-3(-like) Protease Activity: Regulatory Roles of Thioredoxin and Cytochrome c¹

Oxidative stress induces a variety of cellular responses, including active cell death as well as gene activation and proliferation (1, 2). We previously showed that hydrogen peroxide or a reversible thioloxidant, diamide, induces the tyrosine phosphorylation of several intracellular molecules in T cells, and a src family protein tyrosine kinase p56^lck is one of the proteins tyrosine phosphorylated by these oxidants, indicating that T cells have a redox-sensitive signaling mechanism (3, 4). Our previous report also showed that 150–200 μM of diamide induces apoptosis in Jurkat T cells, while at the concentration over 400 μM, diamide induces necrosis (5). The activation of caspases is known to be involved as a general mechanism in the induction of apoptosis (6, 7, 8, 9). Especially, caspase-3 (CPP32) plays a crucial role in the process of apoptosis. All caspases are cysteine protease and have critical cysteine residues in the catalytic domain sequence (9, 10). These results prompted us to assume that the activity of caspase may be regulated by the oxidative stress and the intracellular reduction/oxidation (redox) state.

Cellular redox state is regulated by cellular thiols, including glutathione (GSH)³ and thioredoxin (TRX). TRX is a small ubiquitous multifunctional protein having a redox-active disulfide/dithiol within its active site sequence, -Cys-Gly-Pro-Cys-, and operates together with NADPH and TRX reductase as a protein disulfide-reducing system (11). There is a collecting evidence that the TRX system is as important as the GSH system in the cellular redox regulation against oxidative stress. Human TRX was originally cloned as adult T cell leukemia-derived factor (ADF) produced by human T cell leukemia virus type I-transformed T cell (12). Mammalian TRX has been shown to be a stress-inducible protein (1, 13, 14, 15) and secreted from cells (16, 17, 18). Intracellular TRX plays an important role in the regulation of protein-nucleic acids interactions through the redox regulation of its cysteine residues (19, 20). TRX also shows a cytokine-like extracellular activity to promote cell growth (21, 22, 23) and protect cells against oxidative stress (13, 24, 25, 26).

In this work, we examined the regulatory mechanism of caspase-3(-like) protease in Jurkat cells by oxidative stress elicited by diamide. Our results indicate critical roles of intracellular reducing factors, including TRX, and of the release of cytochrome c from mitochondria into cytosol in the activation of caspase-3(-like) protease.

Agonistic anti-human Fas mAb (CH-11) (27) was purchased from Medical and Biological Laboratories (Nagoya, Japan). Anti-cytochrome c mAb and anti-caspase-3 polyclonal Ab were purchased from PharMingen (San Diego, CA). Anti-human TRX mAb (ADF-11 mAb, mouse IgG1) was provided by Fujirebio (Tokyo, Japan). A fluorogenic substrate for caspase-3 (CPP32) protease, acetyl-l-aspartyl-l-glutamyl-l-valyl-l-aspartic acid α-(4-methyl coumaryl-7-amide) (Ac-DEVD-MCA) (7), and dihydroethidium were purchased from Peptide Institute (Osaka, Japan) and Molecular Probes (Eugene, OR), respectively. Human rTRX (25) was provided by Ajinomoto, Basic Research Laboratory (Kawasaki, Japan). Diamide, reduced GSH, and dl-buthionine-[S,R]-sulfoximine (BSO) were purchased from Sigma (St. Louis, MO). 2-ME, DTT, and all other reagents were purchased from Nacalai Tesque (Kyoto, Japan), unless otherwise stated.

Jurkat, a human T cell line, was maintained in RPMI 1640 (Life Technologies, Grand Island, NY) with 10% heat-inactivated FCS and antibiotics (100 IU/ml penicillin and 100 mg/ml streptomycin) at 37°C in humidified atmosphere containing 5% CO₂.

Caspase-3(-like) protease activity was measured using the method described previously (28). Cells were collected and washed twice with ice-cold PBS(−), and resuspended in 50 mM Tris-HCl (pH 7.4), 1 mM EDTA, and 10 mM EGTA containing 50 μM digitonin. Then cells were incubated at 37°C for 10 min. Lysates were centrifuged at 15,000 rpm for 3 min, and supernatants were collected. A total of 10 μg protein was incubated with 50 μM Ac-DEVD-MCA at 37°C, and the release of 7-amino-4-methylcoumarin (AMC) was monitored by a spectrofluorometer (Hitachi F-2000) using an excitation wavelength of 380 nm and an emission wavelength of 460 nm. One unit was defined as the amount of enzyme required to release 5.2 pmol AMC/1 min/1 mg lysate protein at 37°C.

Jurkat cells cultured with diamide or anti-Fas mAb were collected and washed with ice-cold RPMI 1640 with 10% FCS. After centrifuge, the pellet of cells was fixed by immersion in 2% glutaraldehyde, 0.1 M phosphate buffer, pH 7.2, for 2 h at 4°C. The specimens were washed in 0.1 M phosphate buffer, pH 7.2, and then postfixed by 1% OsO₄, 0.1 M sucrose, 0.1 M phosphate buffer, pH 7.2, for 1.5 h at 4°C. After fixation, the specimens were dehydrated by ethanol and embedded in Epon. Thin sections were cut with diamond knife, stained with uranyl acetate and lead citrate, and examined with a transmission-electron microscope (Hitachi H-7000).

Intracellular redox state was estimated by levels of intracellular ROI, which were monitored by flow-cytometric analysis with dihydroethidium (also called hydroethidine), as described previously (29, 30). Dihydroethidium is a sodium brohydride-reduced derivative of ethidium bromide. Dihydroethidium permeates the cell membrane easily, and in the presence of superoxide anion (O₂⁻) it can be directly oxidized to red fluorescent ethidium bromide, which is intercalated into DNA to enhance red fluorescence of cell nucleus. In this way, cellular fluorescence intensity indicates the levels of intracellular ROI and can be monitored by flow cytometry. For assays, Jurkat cells were cultured with the indicated concentration of diamide or 100 ng/ml anti-Fas mAb for 3 h, and 2 μM dihydroethidium was added for the last 20 min at 37°C. Some samples of cells were precultured with BSO for 24 h, and after wash with medium without FCS, cells were cultured with the indicated concentration of diamide, followed by the treatment of dihydroethidium. After wash with PBS(−) containing 1% BSA and 0.1% NaN₃, the cells were analyzed on a flow cytometer (FACSCalilbur; Becton Dickinson, Mountain View, CA) measuring the fluorescence at 585 ± 21 nm with an excitation wavelength of 488 nm.

For TRX immunoblotting, the total cell lysate was prepared. Cells were collected and washed twice with ice-cold PBS(−), and then lysed with a solubilizing solution (10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Nonidet P-40, 1 mM EDTA, 1 mM EGTA, 0.1 mM PMSF, 8 μg/ml aprotinin, and 2 μg/ml leupeptin) on ice for 30 min. The extracts were cleared by centrifugation.

For cytochrome c assay, soluble cytosolic fraction was prepared as described (31). Cells were collected and washed twice with ice-cold PBS(−), then the cell pellet was suspended in 5 vol of ice-cold buffer (20 mM HEPES-KOH (pH 7.4), 10 mM KCl, 1.5 mM MgCl₂, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 0.1 mM PMSF, 8 μg/ml aprotinin, and 2 μg/ml leupeptin). After sitting on ice for 15 min, the cells were disrupted by douncing 15 times with a loose-fitting pestle (Dounce type homogenizer, Wheaton, NJ). The suspension was centrifuged twice at 1000 × g for 5 min at 4°C to remove the nuclei. The supernatant was further centrifuged at 10⁵ × g for 1 h (Beckman TL-100 Ultracentrifuge). The resulting supernatant was used as soluble cytosolic fraction.

For caspase-3 immunoblotting, cells were collected and washed twice with ice-cold PBS(−), and resuspended in 50 mM Tris-HCl (pH 7.4), 1 mM EDTA, and 10 mM EGTA containing 50 μM digitonin. Then cells were incubated at 37°C for 10 min. Lysates were centrifuged at 15,000 rpm for 3 min, and supernatants were collected.

Equal amounts of protein were subjected to electrophoresis on a 15% SDS-polyacrylamide gel, and then electrophoretically transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA). After blocking with 5% BSA and 5% skim milk in Tris-buffered saline containing 0.05% Tween-20 at 4°C overnight, the membrane was incubated with anti-TRX or anti-cytochrome c mAb, followed by horseradish peroxidase-conjugated anti-mouse Ig (Amersham, Little Chalfont, U.K.). For caspase-3 immunoblotting, the membrane was incubated with anti-caspase-3 polyclonal Ab, followed by horseradish peroxidase-conjugated anti-rabbit Ig (Amersham). Detection of chemoluminescence was performed with an ECL Western blot detection kit (Amersham), according to the supplier’s recommendation.

We previously demonstrated that diamide induces cell death and DNA fragmentation in Jurkat T cells (5). To investigate the regulatory mechanism of diamide-induced cell death, we first examined the caspase-3(-like) protease activity in Jurkat cells cultured with various concentrations of diamide. As shown in Fig. 1,A, the activation of caspase-3(-like) protease was observed. The activity was highest in cells cultured with 200 μM diamide, and gradually decreased in cells with higher concentration of diamide. The activity could not be detected in cells cultured with 500 μM diamide. As shown in Fig. 1 B, caspase-3(-like) protease began to be activated 4 h after the culture with 200 μM diamide, and the peak of the activity was observed after about 10 h of culture.

Transmission-electron microscopy demonstrated that nuclei of Jurkat cells showed the characteristic morphologic change of apoptosis: condensation and fragmentation of chromatin in nuclear periphery, intact nuclear membranes, and almost intact mitochondria (Fig. 2,B), which is similar to those cultured with 100 ng/ml anti-Fas mAb (Fig. 2,D). When the cells were examined 3 h after exposure to 200 μM diamide, the fragmentation of nuclei was not detected yet, although the condensation of chromatin was observed (data not shown). After 7 h of the culture with 200 μM diamide or anti-Fas mAb, the fragmentation of nuclei was remarkable, and some population of apoptotic cells began to lose the integrity of plasma membrane (Fig. 2, B and D).

In contrast to the culture with 200 μM diamide, Jurkat cells cultured with 500 μM diamide (Fig. 2,C) revealed typical features of necrosis, as expected from our previous observation (5). In cells cultured with 500 μM diamide for 3 h, the vacuole formation and the swelling of mitochondria were seen, whereas the integrity of nuclear membrane was still maintained (data not shown). After 7 h of the culture with 500 μM diamide, the vacuole formation became remarkable, and the integrity of nuclear and plasma membrane as well as the structure of mitochondria such as cristae were destroyed (Fig. 2 C).

The morphologic changes induced by diamide correlate with the caspase-3(-like) protease activity (Figs. 1 A and 2, B and C) as well as the DNA fragmentation, as previously reported (5).

Diamide oxidizes the intracellular thiols, including GSH and TRX. We previously showed the transient decrease of cellular GSH and the oxidation of TRX during diamide-induced apoptosis (5). It is supposed that the decrease of these reducing factors causes the generation of intracellular ROI. The generation of intracellular ROI in Jurkat cells was estimated by flow-cytometric analysis with dihydroethidium. As shown in Fig. 3, Jurkat cells cultured with 200 μM diamide for 3 h displayed only a slight increase, indicating that generated ROI might be buffered by intracellular reducing factors. In contrast, Jurkat cells cultured with 500 μM diamide for 3 h displayed a considerable increase in the intracellular ROI.

As shown in Fig. 4,A, immunoblot analysis demonstrated that the exposure of cells to 200 μM diamide in culture medium resulted in the release of cytochrome c from mitochondria into cytosol as early as 1 h. When cells were cultured with 500 μM diamide, the release of cytochrome c was also detected as early as 1 h (Fig. 4 B).

As shown in Fig. 5, the processed large subunits of caspase-3 (32, 33) began to be detected after 3 h of culture with 200 μM diamide as well as anti-Fas mAb. This processing of caspase-3 correlates with the time course of the caspase-3(-like) protease activity (Fig. 1 B). However, when cells were cultured with 500 μM diamide, the processing was not detected.

To test the possibility that the activity of caspase is regulated by the redox state of thiol in its active site cysteine, we examined the effect of various redox-active reagents on the activity of caspase-3(-like) protease in the lysate of Jurkat cells cultured with anti-Fas mAb. Reducing agents such as 2-ME, human rTRX, or GSH in cell lysate per se only marginally increased the activity (Fig. 6, A–C).

On the contrary, caspase-3(-like) protease activity was dose dependently decreased by the addition of diamide to the lysate of Fas-activated cells (Fig. 6,D). The decreased enzymatic activity was recovered by the addition of thiol-reducing agents such as DTT or TRX (Fig. 6, E and F). Less than 500 μM of diamide itself did not decrease the fluorescence of AMC nor affected the assay (data not shown).

We previously showed that apoptosis is induced by much lower concentration of diamide in BSO-precultured Jurkat cells as compared with BSO-untreated cells (5). The preculture with BSO itself caused neither the induction of apoptosis nor the activation of caspase-3(-like) protease in Jurkat cells (5, data not shown). Intracellular total GSH decreased to about 1.4% after the culture with 5 mM BSO for 24 h (data not shown). As shown in Fig. 7,A, in GSH-depleted cells precultured with 5 mM BSO for 24 h, caspase-3(-like) protease was activated in cells cultured with 5–50 μM diamide, whereas the enzymatic activity could not be detected in cells cultured with more than 100 μM of diamide. As compared with Fig. 1 A, lower concentration of diamide activated caspase-3(-like) protease in GSH-depleted cells than BSO-untreated cells. Morphologic observation also showed that GSH-depleted cells with 20 μM diamide were apoptotic, and those with 200 μM diamide were necrotic (data not shown).

As shown in Fig. 7 B, GSH-depleted Jurkat cells cultured with 20 μM diamide for 3 h displayed a slight increase in the intracellular ROI, estimated by flow-cytometric analysis with dihydroethidium. In contrast, GSH-depleted cells cultured with 200 μM diamide for 3 h displayed a considerable increase in the intracellular ROI.

Immunoblot analysis demonstrated that a marginal release of cytochrome c into cytosol occurred when cells were cultured with 5 mM BSO for 24 h, and that the release was markedly facilitated by the following culture with 20 μM diamide (Fig. 7 C).

Immunoblot analysis demonstrated that intracellular TRX protein levels were conserved in Jurkat cells cultured with apoptosis-inducible concentration of diamide (Fig. 8). When cells were precultured with BSO, intracellular protein levels of TRX were also preserved in cells cultured with 20 μM diamide.

We have shown that 150–200 μM of diamide induces apoptosis in Jurkat T cells, while at the concentration over 400 μM, diamide induces necrosis (5). In this study, we show that caspase-3(-like) protease is activated in Jurkat T cells cultured with 200 μM diamide, while the protease activity is not detected in the cells cultured with 500 μM diamide (Fig. 1,A). Recently, Hampton et al. (34) have shown that 50 μM hydrogen peroxide induces apoptosis in Jurkat cells with caspase activation, and higher concentration of hydrogen peroxide causes necrosis without caspase activity. Flow-cytometric analysis indicates the possibility that oxidative stress induces apoptosis as long as intracellular reducing factors can buffer the generated ROI, while it induces necrosis when a physiologic intracellular reducing environment collapses under the excessively generated ROI (Figs. 3 and 7 B). The disruption of enzymatic caspase activity induced by excessive oxidation may cause the shift from apoptosis to necrosis (35). Our results suggest the importance of an intracellular reducing environment for the activity of caspases.

Caspases have cysteine residues in the catalytic domains (9, 10). It has been reported that the activity of caspase-1 is lost when the active site cysteine is replaced by other amino acid (10). In this study, we show that caspase-3(-like) protease activity is regulated by redox condition of cysteine by use of a thiol-oxidizing agent, diamide, and thiol-reducing agents, including DTT as well as endogenous reducing factors, GSH and TRX (Fig. 6). Our results indicate that the enzymatic activity of caspase-3(-like) protease is highly redox sensitive and the protease is active in a reducing environment.

Diamide has exclusively inhibitory, but not enhancing effect on the caspase activity when added to the cell lysate (Fig. 6,D). Caspase-3(-like) protease, however, was markedly activated when Jurkat cells were cultured with a limited range of the concentration of diamide (Fig. 1). Our results indicate that diamide activates caspase-3(-like) protease not directly, but via some cellular event induced by diamide. One of the possibilities involved in the activation of caspase-3(-like) protease by diamide is the release of cytochrome c from mitochondria into cytosol (31, 36, 37, 38, 39). It has been reported that diamide induces apoptosis in mouse thymocyte with mitochondrial membrane potential disruption (40, 41). There might be a redox-sensitive molecule in the outer membrane of mitochondria, which is responsible for the release of mitochondrial cytochrome c, because cytochrome c resides in the space between the outer and inner membranes of mitochondria. TRX2 (42), which was recently cloned and localized in mitochondria, may modulate apoptosis-inducing signal via mitochondria in cooperation with other thiol molecules. Although the release of cytochrome c was detected as early as 1 h (Fig. 4,A), it took 4 h for caspase-3(-like) protease to be activated in cells cultured with 200 μM diamide (Fig. 1,B), suggesting that caspase-3(-like) protease activity is regulated by several factors, including the intracellular redox status in addition to the release of cytochrome c from mitochondria into cytosol. When cells were cultured with 500 μM diamide, the processing of caspase-3 was not detected in spite of the release of mitochondrial cytochrome c (Figs. 4 B and 5). The processing of caspase-3 seems to be suppressed under a thiol-oxidized state (43). These results indicate that diamide also decreases the activity of caspase-9, which is also cysteine protease, or interferes with the cytochrome c/Apaf-1/caspase-9 complex, the likely activator of caspase-3.

Other possibilities involved in the activation of caspase-3(-like) protease by diamide is c-Jun N-terminal kinase (JNK) or p38 MAP kinase (44, 45). Quite recently, it has been reported that reduced form of TRX is associated with apoptosis signal-regulating kinase (ASK) 1 (46), which was identified as MAP kinase kinase kinase to activate the JNK and p38 MAP kinase pathways (47). When TRX is oxidized by oxidative stress, ASK 1 is dissociated from oxidized TRX and activated to induce apoptosis signal, suggesting that TRX is a direct inhibitor of ASK 1. These results are consistent with our previous statement that oxidized TRX may trigger apoptosis in response to sulfhydryl oxidation (5).

GSH system is one of the intracellular thiol-reducing factors to antagonize the excessive oxidation. When Jurkat cells were cultured with 150 or 200 μM diamide, intracellular total GSH level transiently decreased slightly, but maintained during the process of apoptosis (5, S.U., unpublished observations). We comparatively studied the effect of GSH depletion in Jurkat cells cultured with BSO, which inhibits GSH synthesis and depletes intracellular reduced form of GSH. In GSH-depleted cells, lower concentration of diamide can induce the activation of caspase-3(-like) protease and the release of cytochrome c as well as apoptosis than BSO-untreated cells (Fig. 7). Our results indicate that GSH-depleted cells are more sensitive to oxidative stress. Interestingly, the intracellular levels of TRX were maintained even when cells were treated with BSO, suggesting that the maintenance of an intracellular reducing environment by TRX is quite important for the activation of caspase-3(-like) protease. Cisplatin, which is one of the most widely used anti-cancer drugs and which cytotoxicity is attributed to generation of ROI (15), has been reported to induce apoptosis as well as necrosis in cancer cells, but to induce mainly necrosis in BSO-treated cells (48). When Jurkat cells were precultured with N-acetyl-l-cysteine (NAC), apoptosis induced by 200 μM diamide was only slightly inhibited (S.U., unpublished observations). It has been reported that Fas-mediated apoptosis in T cell as well as NK cell is inhibited by the preculture with NAC and enhanced by the preculture with BSO (49, 50, 51), and that Fas expression is down-regulated by the culture with NAC (50). Taken together, it is likely that cells containing much thiols, including GSH and TRX, tend to survive from a variety of stress or death signal.

In conclusion, the activation of caspase-3(-like) protease is mediated, at least partly, by the release of cytochrome c from mitochondria into cytosol, which is sensitive to oxidative stress, and endogenous reducing factors, including TRX, are required for caspase activity to induce apoptosis.

We thank Drs. Keizo Furuke, Norihito Sato, Takashi Inamoto, and Akira Yamauchi for helpful discussions; Drs. Shigeru Tsuyuki and Hirotaka Kazama for technical advice; Dr. Makio Fujioka for help with electron microscopy; and Ms. Yoko Kanekiyo for secretarial help.