Caspase-2 Acts Upstream of Mitochondria to Promote Cytochromec Release during Etoposide-induced Apoptosis*

DNA damage induced by the cancer chemotherapeutic drug etoposide triggers the onset of a series of intracellular events characteristic of apoptosis. Among the early changes observed is the release of cytochrome c from mitochondria, although the mechanism responsible for this effect is unclear. We demonstrate here a role for caspase-2 in etoposide-induced cytochrome crelease. In particular, Jurkat T-lymphocytes treated with an irreversible caspase-2 inhibitor, benzyloxycarbonyl-Val-Asp-Val-Ala-Asp-fluoromethyl ketone (z-VDVAD-fmk), or stably transfected with pro-caspase-2 antisense (Casp-2/AS) are refractory to cytochrome crelease stimulated by etoposide. Experiments performed using a reconstituted cell-free system indicate that etoposide-induced cytochrome c release by way of caspase-2 occurs independently of cytosolic factors, suggesting that the nuclear pool of pro-caspase-2 is critical to this process. Apart from inhibiting cytochrome c release, undermining caspase-2 activity results in an attenuation of downstream events, such as pro-caspase-9 and -3 activation, phosphatidylserine exposure on the plasma membrane, and DNA fragmentation. Taken together, our data indicate that caspase-2 provides an important link between etoposide-induced DNA damage and the engagement of the mitochondrial apoptotic pathway.

Caspases are cysteine-aspartate proteases that play critical roles during the initiation and execution of apoptosis. These enzymes share sequence similarity with the Caenorhabditis elegans cell death protease CED-3 and are synthesized as inactive precursor forms that must be proteolytically cleaved in order to be activated (1). Initiator pro-caspases 2, 8, 9, and 10 are activated with the help of adaptor molecules that bring these proteases into close proximity, permitting autoprocessing (2)(3)(4). With the exception of caspase-2, initiator caspases are responsible for cleaving and activating effector pro-caspases 3, 6, and 7. Effector caspases, in turn, cleave various proteins leading to morphological and biochemical features characteristic of apoptosis (5).
In recent years, it has become abundantly clear that caspase-9 activity is required for apoptosis induced by different stimuli (6). Specifically, damage to mitochondria results in the release of cytochrome c, which together with Apaf-1 and dATP lead to the recruitment and activation of pro-caspase-9 (7,8). Among the cytotoxic stimuli that can initiate the mitochondrial pathway is the chemotherapeutic drug and topoisomerase II poison etoposide (9). Normally, topoisomerase II prevents intertwining of DNA by generating transient double-stranded breaks through which an intact helix can pass (10). Etoposide was designed to exploit selectively the catalytic property of topoisomerase II by increasing the number and duration of DNA cleavage sites, resulting ultimately in permanent doublestranded breaks that are lethal to the cell (11). For several years, it was argued that cytochrome c release induced by etoposide was a caspase-independent event and that caspase-9, in general, was the most apical caspase in chemical-induced apoptosis (9,12,13). However, we recently demonstrated that etoposide-induced cytochrome c release involves distinct doseand caspase-dependent pathways (14). In particular, a low dose (10 M) of etoposide appears to exert its effect at the nuclear level, resulting in the release of a heat-labile factor(s) that, in turn, interacts with mitochondria to elicit cytochrome c release. Because this effect was inhibited by the general caspase inhibitor benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone (z-VAD-fmk), 1 and caspase-2 was the earliest caspase activated in Jurkat T-lymphocytes, we speculated that the release of cytochrome c triggered by a low dose of etoposide may require active caspase-2.
Two mRNA alternative splicing variants encode two caspase-2 proteins, caspase-2L and caspase-2S, with different effects on apoptotic cell death (15,16). The pro-apoptotic protein caspase-2L (referred to here as caspase-2) is the prevailing isoform expressed in most tissues (15). Subcellular fractionation studies have revealed that pro-caspase-2 is present in several intracellular compartments, including the mitochondrion, Golgi, cytosol, and nucleus (17,18). It is the only procaspase present constitutively in the nucleus. Interestingly, neurons from caspase-2 knock-out mice are actually more sensitive to death than neurons from wild-type mice, whereas caspase-2-deficient oocytes or lymphoblasts from the same animals are resistant to apoptosis induced by chemotherapeutic drugs or granzyme B and perforin, respectively (19).
The aim of the current study was to identify a role for caspase-2 in etoposide-induced apoptosis. The results indicate a decreased sensitivity to apoptosis induced by 10 M etoposide in cells with impaired caspase-2 activity. The level of inhibition precedes the engagement of mitochondria, because cytochrome c release was inhibited. Moreover, cells treated with benzyloxycarbonyl-Val-Asp-Val-Ala-Asp-fluoromethyl ketone (z-VDVAD-fmk) or stably transfected with pro-caspase-2 antisense (Casp-2⁄AS) exhibited significantly lower levels of phosphatidylserine (PS) exposure on the plasma membrane, as well as reduced caspase-9 and caspase-3 activities. In addition, reconstituted cell-free experiments revealed that etoposide-induced cytochrome c release from liver mitochondria was reduced when these organelles were incubated with nuclei isolated from Casp-2⁄AS-versus neotransfected cells. Combined, we propose that nuclear caspase-2 is an important upstream promoter of mitochondrial cytochrome c release in response to etoposide.

EXPERIMENTAL PROCEDURES
Cell Culture-Wild-type Jurkat T-lymphocytes or cells stably transfected with pro-caspase-2 antisense (Casp-2⁄AS) or control vector (neo) were cultured in RPMI 1640 complete medium supplemented with 10% (v⁄v) heat-inactivated fetal calf serum, 2% (w⁄v) glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin in a humidified air/CO 2 (19:1) atmosphere at 37°C. Cells were maintained in a logarithmic growth phase for all experiments. Apoptosis was induced with etoposide (10 M) (Bristol-Myers Squibb Co.), and ethanol (0.03% final concentration) was used as a vehicle control. In some cases, cells were first treated for 1 h with z-VAD-fmk, z-VDVAD-fmk, or z-LEHD-fmk (25 M) (Enzyme Systems Products, Dublin, CA) to inhibit caspase activity and Me 2 SO (0.2% final concentration) was used as a vehicle control.
Preparation of Cytosol for Cell-free System and Cytochrome c Measurement-Cells were collected and washed twice in ice-cold phosphatebuffered saline (PBS), resuspended in S-100 buffer (20 mM Hepes, pH 7.5, 10 mM KCl, 1.5 mM MgCl 2 , 1 mM EGTA, 1 mM EDTA, and a mixture of protease inhibitors) and incubated on ice for 15 min. Cells were centrifuged at 10,000 ϫ g for 15 min at 4°C. Supernatants were further centrifuged at 100,000 ϫ g for 1 h at 4°C and used for cell-free experiments or Western blot analysis.
Isolation of Jurkat Nuclei-Cells were washed once with ice-cold PBS and recovered at 500 ϫ g for 5 min. Pellets were resuspended in buffer A (10 mM Hepes, pH 7.6, 0.1 mM EDTA, 320 mM sucrose, 2 mM magnesium acetate, 1 mM DTT, and a mixture of protease inhibitors) at a concentration of 4 ϫ 10 7 ⁄ml. Cells were gently lysed by homogenization with a tight fitting pestle in a Dounce homogenizer, and plasma membrane rupture was monitored by phase-contrast microscopy. The resulting homogenate was diluted with 2 volumes of buffer B (10 mM Hepes, pH 7.6, 0.1 mM EDTA, 2.0 M sucrose, 5 mM magnesium acetate, 1 mM DTT) and transferred to centrifuge tubes. Prior to centrifugation, 3-5 ml of buffer B were added as a cushion to the bottom of the tube. Samples were centrifuged at 30,000 ϫ g for 45 min at 4°C, and the resulting nuclei were resuspended in storage buffer (50 mM Hepes, pH 7.6, 5 mM magnesium acetate, 1 mM DTT, 0.1 mM EDTA, 25% glycerol) at a final concentration of 200,000 nuclei/l. Aliquots were stored at Ϫ80°C until used.
Isolation of Rat Liver Mitochondria-The liver of a male Sprague-Dawley rat was minced on ice, resuspended in 50 ml of MSH buffer (210 mM mannitol, 70 mM sucrose, 5 mM Hepes, pH 7.5) supplemented with 1 mM EDTA, and homogenized with a glass Dounce homogenizer and Teflon pestle. Homogenates were centrifuged at 600 ϫ g for 8 min at 4°C. The supernatant was decanted and recentrifuged at 5500 ϫ g for 15 min to form a mitochondrial pellet that was resuspended in MSH buffer without EDTA and centrifuged again at 5500 ϫ g for 15 min. The final mitochondrial pellet was resuspended in MSH buffer at a protein concentration of 80 -100 mg⁄ml. Fresh mitochondria were prepared for each experiment and used within 4 h.
Apoptosis Measurements-Phosphatidylserine exposure on the outer leaflet of the plasma membrane was detected using the Annexin V: FITC Apoptosis Detection Kit II (BD PharMingen, San Jose, CA) according to the manufacturer's instructions. In brief, 5 ϫ 10 5 cells were pelleted following drug treatment and washed in PBS. Next, the cells were resuspended in 100 l of binding buffer containing annexin V-FITC and propidium iodide. Prior to fluorescence-activated cell sorter analysis, 400 l of binding buffer were added to the cells. Necrotic cells were excluded by gating before histogram analysis and never accounted for more than 4% of total cells.
Analysis of oligonucleosomal DNA fragmentation was performed as described previously (20). Briefly, 10 6 cells were collected by centrifugation at 1000 ϫ g, washed once in PBS, and resuspended in 250 l of TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). Subsequently, the suspension was mixed with an equal volume of lysis buffer (20 mM EDTA, 0.05% (v⁄v) Triton X-100, 10 mM Tris-HCl, pH 8.0) and incubated on ice for 30 min. Lysates were centrifuged at 13,000 ϫ g for 15 min at 4°C to separate intact chromatin (pellet) from DNA fragments (supernatant). The supernatant containing fragmented DNA was precipitated with ethanol overnight at Ϫ20°C. Precipitated and lyophilized DNA was resuspended in 20 l of TE buffer and incubated with 1 l of 50 mg⁄ml RNase A for 1 h at 37°C followed by 1 l of 25 mg⁄ml proteinase K. The samples were mixed at 1:5 ratio with loading buffer (40% sucrose, 0.25% bromphenol blue) and electrophoresed at 60 mA on 1.8% agarose gels. Separated DNA was stained with ethidium bromide and visualized by UV light.
Reconstituted Cell-free System-Standard reactions were carried out in a 35-l reaction volume with reaction buffer (20 mM Hepes, pH 7.6, 10 mM KCl, 2 mM magnesium acetate, 1 mM EGTA, 1 mM DTT, 250 mM sucrose, 10 mM succinate, 2 mM ATP, 10 mM creatine phosphate, 50 g/ml creatine kinase, and a mixture of protease inhibitors) in the absence or presence of Jurkat nuclei (10 6 ), isolated liver mitochondria (15 g of protein), and 25 g of wild-type Jurkat cytosol protein. Nuclei and mitochondria were suspended separately in reaction buffer prior to their addition to the reaction mix. Samples were incubated at 37°C for up to 2 h. Nuclei and mitochondria were removed by centrifugation at 12,500 ϫ g for 10 min at 4°C, and the supernatants were stored at Ϫ20°C until used for Western blot analysis.
Measurement of Caspase Activity-The measurement of DEVD-AMC, VDVAD-AMC, or LEHD-AMC (Peptide Institute, Osaka, Japan) cleavage was performed using a modified version of a fluorometric assay reported previously (21). 500,000 cells were pelleted and washed once with ice-cold PBS. For DEVD-AMC cleavage, cells were resuspended in 25 l of PBS, added to a microtiter plate, and combined with substrate dissolved in a standard reaction buffer (100 mM Hepes, pH 7.25, 10% sucrose, 10 mM dithiothreitol (DTT), 0.1% CHAPS). For VDVAD-AMC or LEHD-AMC cleavage, cells were resuspended in 25 l of PBS, added to a microtiter plate, and combined with substrate dissolved in a standard reaction buffer (100 mM MES, pH 6.5, 10% polyethylene glycol, 10 mM DTT, 0.1% CHAPS). Cleavage of the fluorogenic peptide substrates was monitored by AMC liberation in a Fluoroscan II plate reader (Labsystems, Stockholm, Sweden) using 355 nm excitation and 460 nm emission wavelengths. Fluorescence units were converted to pmol of AMC using a standard curve generated with free AMC. Data from duplicate samples were then analyzed by linear regression.

RESULTS AND DISCUSSION
The Caspase-2 Inhibitor z-VDVAD-fmk Prevents Apoptosis Induced by 10 M Etoposide-Uncertainty persists as to how DNA-damaging chemotherapeutic drugs, such as etoposide, initiate apoptosis by invoking the mitochondrial pathway and eliciting cytochrome c release. We demonstrated previously that the stimulation of cytochrome c release by etoposide occurs by distinct pathways that hinge upon the concentration of drug being used (14). Specifically, using wild-type Jurkat T-lymphocytes or a reconstituted cell-free system, cytochrome c release stimulated by a low (10 M) dose of etoposide appeared to require caspase activity and possibly that of caspase-2. To build on our previous study and to evaluate the contribution of caspase-2 to apoptosis induced by 10 M etoposide, we initially used the irreversible caspase-2 inhibitor z-VDVAD-fmk. Using FITC-labeled annexin V and flow cytometric analysis, we observed that pretreatment of wild-type Jurkat cells with 25 M z-VDVAD-fmk for 1 h completely blocked etoposide-induced PS exposure on the plasma membrane at 6 h (Fig. 1A). The level of PS exposure in cells treated with etoposide alone was ϳ37% (Fig. 1A). PS exposure was also evaluated at 3 h, and no difference between control and etoposide-treated cells was observed (data not shown). Consistent with the PS exposure data, 25 M z-VDVAD-fmk effectively blocked DNA fragmentation as assessed by oligonucleosomal ladder formation (Fig. 1B, lane 3 versus 2). Together, these data provide evidence that caspase-2 is important for etoposide-induced apoptosis.
Caspase-2 Is Critical for Etoposide-induced Cytochrome c Release in Intact Cells-A previous study reported the early activation of pro-caspase-2 in response to various stimuli. Activation occurred upstream of caspase-3 activity and thus suggested that pro-caspase-2 may act as an initiator caspase (22). In addition, the authors speculated that caspase-2 may be part of a signaling complex, similar to caspases 8 and 9, involved in transducing signals between death stimuli and downstream apoptotic events, although direct evidence to support this notion was not presented.
To determine whether caspase-2 was important for etoposide-induced cytochrome c release, wild-type Jurkat cells were treated with 10 M etoposide for 3 h in the absence or presence of 25 M z-VAD-fmk or z-VDVAD-fmk (Fig. 2). The results indicated that 10 M etoposide stimulated processing of procaspase-2, yielding bands of ϳ18 and 12 kDa in size (Fig. 2A,  lane 2). Pretreatment of cells with either z-VAD-fmk (Fig. 2A,  lane 3) or z-VDVAD-fmk (Fig. 2A, lane 4) prevented this event.
Processing of pro-caspase-2 was accompanied by the release of cytochrome c into the cytosol (Fig. 2B, lane 2), an effect that was undermined, but not completely prevented, when cells were first incubated with either z-VAD-fmk (lane 3) or z-VDVAD-fmk (lane 4) for 1 h. The absence of total protection by either inhibitor may be explained by recent observations demonstrating a transcription-independent role for p53, where in response to DNA damage this protein localizes to mitochondria to stimulate cytochrome c release and changes in membrane potential (23). Importantly, neither benzyloxycarbonyl-Asp-Glu-Val-Asp-fluoromethyl ketone (z-DEVD-fmk) nor benzyloxycarbonyl-Leu-Glu-His-Asp-fluoromethyl ketone (z-LEHDfmk), inhibitors of caspase-3-like (referred to here as caspase-3) and caspase-9 activities, respectively, was able to mimic the effect of z-VAD-fmk or z-VDVAD-fmk. Thus, it would appear that caspase-2 acts as an important upstream modulator of cytochrome c release in response to etoposide.
Inhibiting Caspase-2 Interferes With Etoposide-induced Procaspase-9 and Pro-caspase-3 Activation-Because caspase-2 inhibition attenuated cytochrome c release in response to etoposide, we conjectured that this would translate into reduced caspase-9 and -3 activities. To test this premise, caspase-2 (VDVADase), caspase-9 (LEHDase), and caspase-3 (DEVDase) activities were measured following treatment of wild-type Jurkat cells with 10 M etoposide for up to 6 h (Fig. 3). Consistent with our previous results (14), an increase in caspase-2 activity was detected prior to an increase in caspase-9 activity (Fig. 3,  A and C). Complementary experiments performed with z-VDVAD-fmk or z-LEHD-fmk revealed that both pro-caspase-9 and -3 activation, stimulated by 10 M etoposide, depend on active caspase-2 (Fig. 3, B and C). In other words, etoposideinduced caspase-9 and -3 activities were inhibited when cells were pretreated for 1 h with 25 M z-VDVAD-fmk. It should be noted that 25 M z-VDVAD-fmk prevented neither cytochrome c release nor pro-caspases 9 or 3 activation in response to 50 M etoposide, verifying the inability of z-VDVAD-fmk to inhibit these proteases directly. When cells were pretreated for 1 h with 25 M z-LEHD-fmk, caspase-9 and -3 activities were inhibited (Fig. 3, B and C). Surprisingly, z-LEHD-fmk also partially inhibited caspase-2 activity (Fig. 3A); however, this may reflect a pool of pro-caspase-2 that is normally activated by caspase-3-mediated cleavage (24). Taken together, these data suggest that the ordering of the proteolytic caspase cascade in response to a low dose of etoposide is as follows: caspase-2 3 caspase-9 3 caspase-3.
Decreased Expression of Pro-caspase-2 Mitigates Etoposideinduced Apoptosis-To assess whether Jurkat cells with a decreased level of pro-caspase-2 were similarly resistant to etoposide-induced apoptosis as wild-type cells treated with 25 M z-VDVAD-fmk, experiments were performed using neo-or Casp-2⁄AS-transfected cells described previously (Fig. 4A) (25). Although that study demonstrated an apparent role for caspase-2 in Fas-mediated cell death, the authors also reported that down-regulation of pro-caspase-2 had no effect on etoposide-induced apoptotic DNA fragmentation. It was speculated that this might be related to drug-induced transcriptional activation of Casp-2 (26). Although this possibility cannot be excluded, we were interested in whether the lack of protection could be attributed to the concentration (50 M) of etoposide employed, because we had demonstrated in our previous study that 50 M etoposide was directly toxic to mitochondria and stimulated a caspase-independent release of cytochrome c as well as earlier and more robust DEVDase activity (14).
Here, neo-and Casp-2⁄AS-transfected cells were treated with 10 M etoposide for up to 6 h and evaluated for any difference in apoptosis. Results indicated that both cell lines were sensitive to this concentration of etoposide as measured by PS exposure (Fig. 4, B and C). However, ϳ43% of neo-transfected cells (Fig. 4B) were positive for annexin V binding, whereas only ϳ28% of Casp-2⁄AS-transfected cells (Fig. 4C) exposed PS. The fact that Casp-2⁄AS cells were not entirely resistant to 10 M etoposide was not surprising because these cells maintain a certain level of pro-caspase-2, albeit considerably less than neo-transfected cells (Fig. 4A). It is worth mentioning that DNA fragmentation was also evaluated, and the results were consistent with those for PS exposure (data not shown). In either case, when neo or Casp-2⁄AS cells were pretreated with 25 M z-VDVAD-fmk for 1 h, PS exposure (Fig. 4, B and C) and DNA fragmentation were blocked.
Etoposide-induced Pro-caspase Activation Is Delayed and Less Robust in Casp-2⁄AS Cells-Because our inhibitor experiments (Fig. 3) revealed that pro-caspase-2 activation apparently occurs upstream of pro-caspases 9 and 3, the next step was to determine whether this ordering of pro-caspase activation was true in cells stably expressing pro-caspase-2 antisense. In other words, does an enforced lowering of procaspase-2 translate into reduced or delayed activation of procaspases 9 and 3 in response to 10 M etoposide? To test this possibility, neo-and Casp-2⁄AS-transfected cells were treated with etoposide for up to 6 h, and fluorometric analysis of different caspase activities was performed (Fig. 5). In agreement with the data presented in Fig. 3, Casp-2⁄AS cells exhibited lower levels of not only VDVADase but also DEVDase and LEHDase activities when compared with neo cells. In each case, increases in activity were observed earlier and were considerably more pronounced in neo-transfected cells (Fig. 5, A-C). In particular, by 6 h, neo cells exhibited an ϳ500 and ϳ1600% rise in VDVADase and DEVDase activities, respectively, whereas the rise of these activities in Casp-2⁄AS cells was more modest at ϳ250 and ϳ750%, respectively (Fig. 5, A  and B). Moreover, pretreatment of either cell line with 25 M z-VDVAD-fmk for 1 h prior to etoposide completely abolished any increase in VDVADase, DEVDase, or LEHDase activity (Fig. 5, A-C). Taken together, these data offer additional support for the notion that caspase-2 is the most apical caspase in etoposide-induced apoptosis.
Expression of Casp-2⁄AS Mitigates Cytochrome c Release in Response to Etoposide-To extend these results a step further, we were interested in determining whether stable expression of pro-caspase-2 antisense would attenuate etoposide-induced cytochrome c release to an extent similar to z-VDVAD-fmk or z-VAD-fmk (Fig. 2B). Two different, albeit complementary, approaches were used to address this question. First, intact neoor Casp-2⁄AS-transfected cells were treated with 10 M etoposide for 3 or 6 h in the absence or presence of 25 M z-VDVADfmk, and cytosolic extracts were prepared for cytochrome c measurement. Second, experiments were performed using a reconstituted cell-free system wherein isolated neo or Casp-2⁄AS Jurkat nuclei (10 6 ), isolated liver mitochondria (15 g of protein), and⁄or wild-type Jurkat cytosol (25 g of protein) were treated with 10 M etoposide for up to 2 h at 37°C.
Western blot analysis of cytosols isolated from neo-and Casp-2⁄AS-transfected cells at 3 h revealed that etoposide stim-ulated cytochrome c release in neo but not Casp-2⁄AS cells (Fig.  6A, lane 2 versus 1 and lane 5 versus 4). Moreover, pretreatment of neo-transfected cells with 25 M z-VDVAD-fmk attenuated cytochrome c release (Fig. 6A, lane 3 versus 2), an effect that was consistent with results obtained earlier using wildtype Jurkat cells (Fig. 2B, lane 4 versus 2). Whether apoptosisinducing factor (AIF) or second mitochondria-derived activator of caspase/direct inhibitor of apoptosis protein (IAP)-binding protein with low pI (Smac⁄DIABLO) was released at 3 h after etoposide treatment was also investigated. As seen in Fig. 6A, the release profile of Smac⁄DIABLO was similar to that observed for cytochrome c at 3 h (Fig. 6A), which is consistent with a previous report documenting Smac⁄DIABLO release as a general feature of apoptosis in Jurkat cells (27). In contrast, AIF was not released during etoposide-induced apoptosis. It should be noted that Western blot analysis of a subcellular  fraction containing mitochondria demonstrated the reactivity of the anti-AIF antibody with human AIF (data not shown). It is also worth mentioning that preliminary data recently generated in our laboratory suggest that AIF is only released during mitochondrial permeability transition pore opening, 2 which may account for the absence of its release in response to 10 M etoposide.
At 6 h after 10 M etoposide treatment, neither z-VDVADfmk nor pro-caspase-2 antisense alone was able to mitigate cytochrome c release (Fig. 6B, lanes 3 and 5). However, Casp-2⁄AS cells pretreated with 25 M z-VDVAD-fmk remained refractory to cytochrome c release (Fig. 6B, lane 6 versus 5). Together, this suggests that pro-caspase-2 antisense significantly delays but does not altogether prevent cytochrome c release induced by 10 M etoposide. Importantly, these data are in agreement with our previous report demonstrating the inability of 25 M z-VAD-fmk to prevent etoposide-induced cytochrome c release at 6 h, an effect attributed to the ability of a low dose (10 M) of etoposide to ultimately target mitochondria directly (14).
Experiments performed using a cell-free system consisting of isolated Jurkat nuclei and liver mitochondria yielded results for cytochrome c release that were consistent with the cytosolic extract data (Fig. 6, panel C versus A). Specifically, 10 M etoposide stimulated cytochrome c release following a 2-h incubation at 37°C when the reaction mixture contained nuclei isolated from neo-transfected cells (Fig. 6C, lane 2 versus 1) but not when Casp-2⁄AS nuclei were present (Fig. 6C, lane 5 versus  4). Interestingly, this effect was not altered by cytosolic factors, because etoposide-induced cytochrome c release occurred to a similar extent when a fraction containing wild-type Jurkat cytosol was present (data not shown). Treating the reaction mixture with 25 M z-VDVAD-fmk was sufficient to decrease cytochrome c release induced by etoposide (Fig. 6C, lane 3 versus 2). It should be mentioned that 10 M etoposide did not stimulate cytochrome c release when mitochondria were incubated in the absence of nuclei. Together, these data implicate a role for the nuclear pool of pro-caspase-2 during etoposideinduced cytochrome c release.
Concluding Remarks-That pro-caspase-2 is activated in response to a variety of pro-apoptotic stimuli, including tumor necrosis factor-␣, Fas ligand, growth factor withdrawal, and DNA-damaging agents, has been documented previously (22,25,28). However, assigning an emergent function to this protease during apoptosis has been difficult. This is due, in part, to the fact that despite its long pro-domain, which is characteristic of initiator caspases, caspase-2 has largely been assigned a downstream or feedback role because of the capacity for it to be cleaved by active caspase-3 (24,29). Additionally, caspase-2 was reported to act as both a positive and negative regulator of cell death in caspase-2-deficient mice (19). Also, unlike initiator caspases 8 and 9, caspase-2 cannot cleave effector caspases directly. Therefore, the aim of this study was to determine the role of caspase-2 in the apoptotic pathway triggered by the DNA-damaging agent etoposide.
We have now demonstrated that apoptosis induced by 10 M etoposide involves active caspase-2. Specifically, this protease is activated early within the pathway and acts upstream of mitochondria to regulate cytochrome c release both in intact Jurkat cells and in a reconstituted cell-free system. When etoposide-induced activation of pro-caspase-2 is subverted by z-VDVAD-fmk or stable transfection of pro-caspase-2 antisense, cytochrome c release and other manifestations of apoptosis are attenuated.
A previous report (30) indicated that simply overexpressing caspase-2 was sufficient to induce the translocation of Bid from cytosol to mitochondria. The same study revealed that caspase-2 is able to cleave Bid in vitro, albeit weakly, suggesting that caspase-2 may invoke the mitochondrial pathway by processing this protein. However, in agreement with our earlier study, we did not observe Bid cleavage following etoposide treatment (data not shown). Moreover, the fact that a cytosolic fraction was not necessary for etoposide-induced cytochrome c release in our cell-free system suggests that Bid is not required.
We propose a model wherein etoposide treatment stimulates the activation of nuclear pro-caspase-2. Subsequently, active caspase-2 or, more likely, a cleaved substrate translocates out of the nucleus to target mitochondria and stimulate cytochrome c release (Fig. 7). Coincidentally, during the preparation of this manuscript, two articles were published implicating caspase-2 as an effector of the mitochondrial pathway (31,32). However, although those papers and the current study each uniquely demonstrates that pro-caspase-2 is activated upstream of procaspase-9, the mechanism responsible for its activation is unknown. Because it was previously suggested that activation of pro-caspase-2 involves an adaptor molecule known as caspase and receptor-interacting protein (RIP) adaptor with death domain/RIP-associated Ich-1-homologous protein with death domain (CRADD/RAIDD) that is present in the cytosol and the nucleus (33)(34)(35), one could envisage a scenario in which activation of pro-caspase-2 occurs in a manner similar to the activation of initiator pro-caspases 8 and 9. Specifically, it is intriguing to think that etoposide-induced activation of procaspase-2 might involve the formation of a nuclear signaling complex, wherein upon drug treatment CRADD/RAIDD gathers pro-caspase-2 molecules into close proximity to allow autoprocessing. However, additional studies are needed to characterize the mechanism of pro-caspase-2 activation as well as subsequent events leading to the engagement of the mitochondrial apoptotic pathway.