Lysosomal enzymes promote mitochondrial oxidant production, cytochrome c release and apoptosis Ming Zhao 1 , Fernando Antunes 1,2 , John W. Eaton 1,3 and Ulf T. Brunk 1 1 Division of Pathology II, Faculty of Health Sciences, Linko ¨ ping University, Sweden; 2 Grupo de Bioquı ´ mica e Biologia Teo ´ ricas – Instituto Bento da Rocha Cabral and Department of Chemistry and Biochemistry, Faculty of Sciences, University of Lisbon, Portugal; 3 James Graham Brown Cancer Center, University of Louisville, Louisville, KY, USA Exposure of mammalian cells to oxidant stress causes early (iron catalysed) lysosomal rupture followed by apoptosis or necrosis. Enhanced intracellular production of reactive oxygen species (ROS), presumably of mitochondrial origin, is also observed when cells are exposed to nonoxidant pro- apoptotic agonists of cell death. We hypothesized that ROS generation in this latter case might promote the apoptotic cascade and could arise from effects of released lysosomal materials on mitochondria. Indeed, in intact cells (J774 macrophages, HeLa cells and AG1518 fibroblasts) the lysosomotropic detergent O-methyl-serine dodecylamide hydrochloride (MSDH) causes lysosomal rupture, enhanced intracellular ROS production, and apoptosis. Furthermore, in mixtures of rat liver lysosomes and mitochondria, selective rupture of lysosomes by MSDH promotes mitochondrial ROS production and cytochrome c release, whereas MSDH has no direct effect on ROS generation by purifed mito- chondria. Intracellular lysosomal rupture is associated with the release of (among other constituents) cathepsins and activation of phospholipase A2 (PLA2). We find that addi- tion of purified cathepsins B or D, or of PLA2, causes substantial increases in ROS generation by purified mito- chondria. Furthermore, PLA2 ) but not cathepsins B or D ) causes rupture of semipurified lysosomes, suggesting an amplification mechanism. Thus, initiation of the apoptotic cascade by nonoxidant agonists may involve early release of lysosomal constituents (such as cathepsins B and D) and activation of PLA2, leading to enhanced mitochondrial oxidant production, further lysosomal rupture and, finally, mitochondrial cytochrome c release. Nonoxidant agonists of apoptosis may, thus, act through oxidant mechanisms. Keywords: apoptosis; cathepsins; lysosomes; lysosomotropic detergents; oxidative stress. In the last two decades, the phenomenon of apoptosis has attracted great interest and many intricate molecular events underlying the process have been elucidated [1–8]. Several crucial steps are thought to involve mitochondrial release of pro-apoptotic factors, although the exact mechanisms involved in this release are less well understood. In this regard, there is substantial evidence that, at least in some circumstances, the discharge into the cytosol of lysosomal constituents may be an early and, perhaps, initiating event in apoptosis, and that mitochondrial release of pro-apoptotic factors might be a consequence of earlier lysosomal destabilization [9–18]. In further, albeit indirect, support of this, it was recently found that activation of the pro-apoptotic tumour supressor protein, p53, also results in early lysosomal rupture, although through still unknown mechanisms [14]. In the case of simple oxidant-induced apoptosis, lyso- somal rupture occurs in two sequential phases [19,20], where the second one includes activation of phospholipase A2 (PLA2) with production of free arachidonic acid (AA) [21,22]. Theoretically, released lysosomal enzymes, PLA2, and AA all might be capable of destabilizing mitochondrial membranes. Interestingly, over-expression of the anti- apoptotic protein, Bcl-2, abrogates the secondary phase of lysosomal rupture, the activation of PLA2, and the mitochondrial release of cytochrome c [19,21,22]. However, the precise mechanisms through which Bcl-2 mediates these effects are presently unknown. Remarkably, in apoptosis caused by a number of nonoxidative agents, there appears to be increased intracel- lular generation of reactive oxygen species (ROS), probably of mitochondrial origin [23–30]. Although the mechanisms responsible for enhanced mitochondrial ROS production during the process of apoptosis remain unknown, this phenomenon raises the possibility that internally generated ROS, like exogenously added oxidants, may act through a common pathway–lysosomal destabilization. The present investigations were aimed at identifying intracellular events that might connect exposure of cells to nonoxidative agonists of apoptosis and intracellular ROS production. As mentioned above, there is abundant evi- dence that ) at least in some circumstances ) lysosomal rupture might be an early, perhaps even initiating, event in Correspondence to M. Zhao, Division of Pathology II, Faculty of Health Sciences, Linko ¨ ping University, SE-581 85 Linko ¨ ping, Sweden. Fax: +46 13 22 15 29, Tel.: +46 13 22 15 15, E-mail: ming.zhao@inr.liu.se Abbreviations: AA, arachidonic acid; DHE, dihydroethidium; HRP, horseradish peroxidase; LE, lysosomal enzymes; LEF, lysosome- mitochondria enriched fraction; MSDH, O-methyl-serine dodecylamide hydrochloride; PLA2, phospholipase A2; ROS, reactive oxygen species. (Received 28 April 2003, revised 11 July 2003, accepted 24 July 2003) Eur. J. Biochem. 270, 3778–3786 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03765.x the apoptotic cascade. Therefore, in the present investiga- tions we have used a synthetic lysosomotropic detergent, O-methyl-serine dodecylamide hydrochloride (MSDH) to specifically induce lysosomal rupture and ensuing apoptosis [12,31,32]. This was done in order to determine whether internal oxidative stress of mitochondrial origin might arise as a consequence of lysosomal rupture and act as an amplifying loop causing further lysosomal breach. Here, we present evidence that released lysosomal enzymes ) both directly and through activation of PLA2 ) may trigger enhanced mitochondrial production of superoxide and hydrogen peroxide, and cause the release of cytochrome c. Materials and methods Materials Chemicals were from Sigma unless stated otherwise. RPMI 1640 medium, Hepes, foetal bovine serum, glutamine, penicillin, and streptomycin were from Gibco. BODIPY FL phallacidin and dihydroethidium (DHE) were from Molecular Probes. Monoclonal anti-cytochrome c Igs were from Pharmingen, and horseradish peroxidase (HRP)- conjugated goat anti-mouse Igs were from DAKO. Percoll was from Amersham Pharmacia Biotech. Cell cultures Human foreskin fibroblasts (AG-1518, passages 14–20; Coriell Institute, Camden, NJ, USA), J774 cells (a murine histiocytic lymphoma cell line), and human epithelial cells (HeLa) were cultured at 37 °C in humidified air with 5% CO 2 in RPMI 1640 medium supplemented with 2 m M glutamine, 50 IUÆmL )1 penicillin-G, 50 lgÆmL )1 strepto- mycin, and 10% foetal bovine serum. Cells were subcul- tured once a week. Twenty-four hours before experiments, cells were trypsinized and seeded into 35-mm Petri dishes or 96-well plates (Costar, Cambridge, MA, USA) at a density of 10 000 cells per cm 2 . Apoptosis assays DNA fragmentation was assessed using propidium iodide staining of nuclear DNA, essentially as described by Nicoletti et al. [33]. Briefly, cell pellets from individual wells were gently resuspended in 1.5 mL of a hypotonic and membrane-disrupting solution of propidium iodide (50 lgÆmL )1 in 0.1% sodium citrate with 0.1% Triton X-100) in 12 · 75 mm polypropylene tubes. The tubes were kept overnight in the dark at 4 °C before flow-cytometric analyses. The propidium iodide-induced red fluorescence of suspended individual nuclei was measured by flow cyto- fluorometry, using the FL3 channel. Nuclei with partly degraded DNA were counted, and their frequency was expressed as a percentage of the total number of nuclei analysed in at least 10 000 cells. Actin staining AG1518 fibroblasts were seeded in 35-mm Petri dishes and cultured for 24 h before being exposed to 30 l M MSDH in ordinary medium for 3 h. Cellular actin was stained with BODIPY FL phallacidin. Cells were fixed for 10 min in 4% formaldehyde in NaCl/P i , permeabilized for 10 min in 0.3% Triton X-100 in phosphate-buffered saline (NaCl/P i ), and stained for 30 min with BODIPY FL phallacidin (final concentration 0.6 lgÆmL )1 )at37°C. After staining, cells were washed twice in NaCl/P i , and visualized and documented (k EX 495 nm; k EM 520 nm) using a Nikon microphot-SA fluorescence microscope with a Hamamatsu ORCA-100 color digital camera and Adobe PHOTOSHOP software. Evaluation of oxidative stress AG1518 fibroblasts, J774 and HeLa cells were seeded in 96-well plates and cultured for 24 h under standard conditions before being exposed to 30 l M MSDH and 10 l M DHE (in complete medium). Fluorescence intensity, indicating oxidation of DHE was assayed at various periods of time after addition of MSDH and DHE on a VICTOR 1420 (Wallac Sverige AB, Upplands Va ¨ sby, Sweden) fluorescent plate-reader (k EX 485 nm; k EM 620 nm). In some experiments, cells were observed and documented under green light excitation (k EX 546 nm; k EM 590 nm) using fluorescence microscopy as described above. Preparation of rat liver lysosome-mitochondria enriched fraction Livers were removed from 160–200-g female Sprague– Dawley rats (starved overnight), homogenized in 0.3 M sucrose (1 : 9, w/v) and centrifuged at 500 g for 10 min. The supernatants were again centrifuged at 3500 g for 10 min, the pellets discarded, and the lysosome/mitochon- dria-containing supernatants centrifuged at 10 000 g for 10 min. The pellets were washed, suspended and re-centri- fuged at 10 000 g for 10 min and finally resuspended in the sucrose solution to a protein concentration of  1.5 mgÆmL )1 . The resultant lysosome/mitochondria enriched fraction (LEF) was found to be stable (no release of lysosomal enzymes) for up to 4 h in the homogenization solution at 4 °C, while some release of lysosomal enzymes occurredwithin2hat37°C. Preparation of a purified mitochondria fraction Mitochondria were purified from rat liver using a combi- nation of differential and Percoll gradient centrifugation [34,35]. All procedures were carried out at 4 °C. Briefly, fresh liver was minced and homogenized in M-SHE buffer (0.21 M mannitol, 0.07 M sucrose, 10 m M Hepes pH 7.4, 1m M EDTA, 1 m M EGTA, 0.15 m M spermine, 0.75 m M spermidine). Unbroken cells and nuclei were pelleted at 500 g for 10 min. The supernatant was centrifuged at 10 000 g to pellet mitochondria and lysosomes which were resuspended and washed twice with M-SHE buffer. A 2-mL suspension was then layered onto 37.5 mL of Percoll solution (50% Percoll, 50% 2 · M-SHE) and centrifuged for 1 h at 50 000 g in a Ti-60 rotor. The brown mitochondrial band was collected, either by fractionating the gradient or by direct syringe aspiration. The purified mitochondria were pooled, diluted 10-fold with M-SHE buffer, again pelleted by centrifugation and, finally, Ó FEBS 2003 Lysosomes, apoptosis and mitochondria-mediated oxidative stress (Eur. J. Biochem. 270) 3779 resuspended in M-SHE buffer to a protein concentration of  1.5 mgÆmL )1 . The degree of lysosomal contamination of the purified mitochondria fraction was estimated by assaying b-galactosidase/protein and compared to that of LEF. Enzymatic detection of lysosomal integrity and estimation of fraction purity The integrity of lysosomes in the LEF preparation was assessed by assaying released b-galactosidase. LEF (200 lL) was incubated for 3 h at 37 °C with either PLA2 (0.2 UÆmL )1 ), 30 l M MSDH, 2.5 lgÆmL )1 cathep- sin B, or 2.5 lgÆmL )1 cathepsin D and then centrifuged at 14 000 g for 10 min. Stock solutions of the cathepsins were made up in NaCl/P i pH 6.0, whereas MSDH and PLA2 were in NaCl/P i pH 7.4. The supernatants were removed, and 1 mL distilled water with Triton X-100 (final concen- tration 0.1%) was added to the pellets to cause complete lysis of remaining intact lysosomes. Activities of b-galac- tosidase were measured as described previously [22] on the ruptured lysosomal pellet and on the supernatant. The results were expressed as percentage released over total b-galactosidase. Mitochondrial generation of H 2 O 2 Mitochondrial production of H 2 O 2 was assayed essentially as described elsewhere [36]. Briefly, 1.33 UÆmL )1 HRP, 0.066 mgÆmL )1 q-hydroxyphenylacetate, 0.013 mgÆmL )1 superoxide dismutase, and 1 mg mitochondrial protein were added to 2.4 mL respiratory buffer (0.07 M sucrose, 0.23 M mannitol, 30 m M Tris/HCl, 4 m M MgCl 2 ,5m M KH 2 PO 4 ,1m M EDTA, 0.5% BSA, pH 7.4) in a spectro- fluorophotometer cuvette at 37 °C. Succinate (final concen- tration 6.67 m M ) and antimycin A (final concentration 0.83 lgÆmL )1 ) were added, and H 2 O 2 -induced fluorescence recorded (k EX 320 nm; k EM 400 nm) during the first 10 min after mixing. Western blotting for cytochrome c Two-hundred microlitres LEF, or purified mitochondria, were incubated for 3 h at 37 °C with either 30 l M MSDH, PLA2 (0.2 UÆmL )1 ), 2.5 lgÆmL )1 cathepsin B, or 2.5 lgÆmL )1 cathepsin D. Stock solutions of the cathepsins were made up in NaCl/P i pH 6.0, while MSDH and PLA2 were in NaCl/P i pH 7.4. Following centrifugation at 14 000 g for 10 min, the supernatants were separated by Fig. 1. MSDH induces apoptosis and stress fibre formation in fibroblasts. (A) Cells were seeded into 35-mm Petri dishes at a density of 10 000 cellsÆcm )2 .After24h,30l M MSDH was added to complete culture medium (2 mL), and cells were incubated for another 10 h under standard culture conditions. The Nicoletti technique for apoptotic nuclei was applied. One representative experiment out of three is shown. (B) Cells were seeded in 35-mm Petri dishes and kept for 24 h before being exposed to 30 l M MSDH for 3 h. Actin staining was then performed as described in Materials and methods. 3780 M. Zhao et al. (Eur. J. Biochem. 270) Ó FEBS 2003 SDS/PAGE (12% acrylamide) and transferred onto Immo- bilon membranes (2 h; 200 mA). Membranes then were incubated at room temperature for 1 h in blocking buffer [5% low-fat milk powder in Tris-buffered saline (TBS)] and for another 2 h in dilution buffer (0.5% low-fat milk powder in TBS) containing a 1 : 400 dilution of a mono- clonal anti-cytochrome c Ig. After washing in TBS with 0.06% Tween 20, Immobilon membranes were incubated for 1 h at room temperature in a buffer containing a 1 : 1500 dilution of peroxidase-conjugated secondary antibodies. After washing, peroxidase-dependent chemilu- minescence was detected by using enhanced chemilumines- cence Western blotting reagents and hyperfilm according to the manufacturer’s instructions (Amersham Pharmacia Biotech). Statistical analysis All experiments were repeated at least three times. Values are given as arithmetic mean ± SD. Significance Fig. 2. MSDH induces intracellular ROS production. Cells were seeded into 96-well plates at a density of 10 000 cellsÆcm )2 . After 24 h, cells were exposed simultaneously to 30 l M MSDH and 10 l M DHE under otherwise standard culture conditions while control cells were exposed to DHE only. (A) Fluorescence intensity arising from oxidized dihydroethidium in J774, HeLa and AG1518 cells was measured at indicated time points. (B) J774 cells were visualized and photographed after 3 h exposure to MSDH (n ¼ 3). Very similar results were obtained with HeLa and AG1518 cells under the same conditions although detectable oxidant generation occurred earlier. Ó FEBS 2003 Lysosomes, apoptosis and mitochondria-mediated oxidative stress (Eur. J. Biochem. 270) 3781 of differences between groups was determined using Student’s two-tailed t-test. *P £ 0.05; **P £ 0.01; ***P £ 0.001. Results Cultured cells exposed to the synthetic lysosomotropic detergent, MSDH, undergo lysosomal rupture and ensuing apoptosis or necrosis depending upon the extent of lysosomal destabilization [12]. In the present experiments, we induced apoptosis in fibroblasts, J774 cells, and HeLa cells by exposing them to 30 l M MSDH. After 8 h of MSDH exposure, nuclear propidium iodide staining and flow cytometry (to detect DNA fragmentation) revealed apoptotic nuclei appearing as a broad, hypodiploid DNA smear in front of a narrow peak of diploid DNA (Fig. 1A shows results in fibroblasts). At an early stage in this process, well before the appearance of frank apoptosis, fibroblasts showed significantly increased numbers of stress fibres (Fig. 1B). Fig. 3. MSDH induces mitochondrial ROS production by rupturing lysosomes. Purified mitochondria (1.0 mg proteinÆmL )1 ) or a lysosome/mito- chondria-enriched fraction (1.0 mg proteinÆmL )1 ) were incubated with either of MSDH (30 l M ), PLA2 (0.2 UÆmL )1 ), or cathepsin B or D (12.5 lgÆmL )1 ;pH6.0)for3h.(A)H 2 O 2 production, (B) cytochrome c release, and (C) lysosomal stability were assayed as described in Materials and methods (n ¼ 3). 3782 M. Zhao et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Because oxidative stress has been reported to induce stress fibre formation [37], we suspected that the MSDH exposure might be causing increased intracellular generation of ROS. This latter was monitored by following changes in DHE-induced fluorescence. When oxidized, this compound intercalates into DNA and RNA, resulting in an increase in quantum yield. Fluorescence intensity was measured kineti- cally at indicated time points. Increased ROS production occurred after 1 h of MSDH-exposure in fibroblasts (AG1518) and epithelial cells (HeLa), but was significant only after 3 h in macrophages (J774) (Fig. 2A). Note that in fibroblasts and HeLa cells, and also in J774 cells (results not shown), the oxidation of DHE eventually reached a steady state consistent with only a transient production of ROS. Figure 2B shows DHE-exposed J774 cells after 3 h expo- sure to MSDH, when there were still no morphological signs of apoptosis. Theoretically, the increased oxidant generation might arise from effects of released lysosomal enzymes (directly or by activation of PLA2) on mitochondrial ROS production or, alternatively, from direct effects of MSDH on the mitochondria. To discriminate between these possibilities, we added MSDH to purified rat liver mitochondria (4.5- fold purified from lysosomal contamination as compared to the LEF preparation, results not shown). Under these conditions, no changes in mitochondrial production of H 2 O 2 (Fig. 3A) or release of cytochrome c (Fig. 3B) took place. Because we previously observed that lysosomal contents cause activation of PLA2 in J774 cells [22], we also exposed mitochondria to that enzyme and found it to enhance mitochondrial production of ROS (Fig. 3A) and to release cytochrome c as well (Fig. 3B). These findings strongly suggest that MSDH affects mitochondria by first destabilizing lysosomes and causing the release of hydrolytic enzymes which, in turn, attack mitochondria or activate PLA2. Activated PLA2 may further promote this cascade of events, attacking both mitochondrial and lysosomal membranes and causing further lysosomal rupture. This supposed sequence of events was confirmed by adding MSDH to a lysosome/mitochondria-enriched rat liver fraction, where it was found to induce enhanced mito- chondrial production of H 2 O 2 (Fig. 3A), release of cyto- chrome c (Fig. 3B), and lysosomal rupture (Fig. 3C). To test further the idea that released lysosomal hydrolases might enhance mitochondrial ROS production, release of cytochrome c, and activation of PLA2 (all of which may promote the apoptotic cascade), we tested the effects of two lysosomal cathepsins (cathepsin B, a cysteine protease, and cathepsin D, an aspartic protease) on purified mitochondria. Both proteases caused substan- tial increases in mitochondrial production of H 2 O 2 (Fig. 3A) and release of cytochrome c (Fig. 3B). However, neither cathepsin B nor D caused detectable lysosomal rupture in LEF preparations (Fig. 3C), although, as expected, both MSDH and PLA2 did induce lysosomal rupture (Fig. 3C). Thus, cathepsins B and D do not directly cause rupture of lysosomes in an LEF preparation. However, the possibility remains that the intracellular release of other lysosomal hydrolases may do so, or that lysosomal proteases might secondarily destabilize lysosomes through, for example, enhanced oxidative stress or activation of PLA2 following Fig. 4. The lysosomal/mitochondrial axis theory of apoptosis. Both the internal and external pathways may involve lysosomal rupture. Released lysosomal enzymes (LE) may: (a) attack mitochondria directly, inducing oxidative stress and release of cytochrome c (this study and [12,20–22,49– 52]); (b) activate lytic pro-enzymes, such as PLA2, which may attack both mitochondria or lysosomes (this study and [22]); (c) activate Bid [53]; (d) directly activate caspases [15,16,54,55]. It is also possible that released lysosomal enzymes backfire on still intact lysosomes, causing further rupture. Caspase 8 may somehow induce lysosomal rupture [56,57] or the activation of death receptors may cause production of sphingosine [58], which is a lysosomotropic detergent [59]; while p53 causes lysosomal labilization by unknown mechanisms [14]. Other mechanisms may also be involved in lysosomal labilization in relation to apoptosis. Ó FEBS 2003 Lysosomes, apoptosis and mitochondria-mediated oxidative stress (Eur. J. Biochem. 270) 3783 mitochondrial attack by cathepsins and PLA2. Indeed, low, steady-state oxidative stress has been shown to destabilize lysosomes [20] and relocation of lysosomal enzymes to the cytosol was earlier shown to activate PLA2 [22]. Discussion We previously suggested that oxidative stress-induced apoptosis might be initiated by iron-catalysed lysosomal rupture [9,10]. It has since been found that early release to the cytosol of lytic lysosomal enzymes may be characteristic of apoptosis caused by a variety of stimuli [10,12–14, 19,21,22,38–40]. In these latter circumstances, it appears that relocation of lysosomal enzymes to the cytosol may, as in the case of oxidant-induced apoptosis, precede changes of mitochondrial membrane potential, release of cyto- chrome c, and all the morphological signs of apoptosis. These considerations raised the question of whether there might be some ROS-dependent mechanisms common to apoptosis caused by oxidants and that caused by nonoxi- dant agents. In most cells, the predominant source of intracellular ROS generation is the mitochondrial electron transport chain which, even under normal conditions, may ÔleakÕ 1–2% of all electrons as ROS [41–43] (although there is controversy regarding this estimate and the absolute percentage may well be lower [44]). Not only will exogenous oxidants, such as H 2 O 2 , directly induce apoptosis, but enhanced intracellular production of ROS occurs when cells are exposed to a number of pro-apoptotic agents including tumour necrosis factor-a [23], ceramide [24], growth factor withdrawal, HIV infection, and lipopolysaccharide [25–30]. In these cases it is unclear whether such oxidative stress is the cause or an effect of apoptosis. We hypothesized that released lysosomal enzymes or PLA2 directly or indirectly activated by such enzymes [22] might attack mitochondria and induce not only release of cytochrome c, but also enhanced formation of ROS. Released arachidonic acid may further exaggerate this process [45]. These ROS of mitochondrial origin could promote further lysosomal rupture but could also have the secondary effect of maintaining any cytochrome c released by the mitochondria in the oxidized form (although we should note that the cellular cytoplasm contains an abun- dance of reducing agents which could counteract this). Cytochrome c is involved in the activation of caspase-9 [7,46] and is considered a key component of the apoptotic cascade. Ordinarily, any cytochrome c released from mitochondria in oxidized form would rapidly be reduced by the reductive cytosolic milieu. However, it has been proposed that cytochrome c needs to remain oxidized in order to promote apoptosis [46], and the oxidizing equivalents generated by mitochondria may have precisely this effect. MSDH is a lysosomotropic detergent that rapidly induces specific lysosomal rupture and therefore is a very useful tool for detailed kinetic studies of the consequences of lysosomal rupture. The pKa of MSDH is 5.8–5.9 [31,32], allowing it to accumulate in charged form intralysosomally (pH  4.5) due to proton trapping [47], while its accumulation in the cytosol (pH  7.2) is negligible. In protonated, charged form MSDH acts as a much stronger detergent than when uncharged, further targeting the action of this agent to the lysosomal compartment [31]. We previously reported that released lysosomal enzymes activate PLA2 causing further lysosomal fragmentation [22]. The new data presented here confirm and extend those findings and show that relocated lysosomal enzymes work in concert with activated PLA2, causing the release of cytochrome c, enhanced mitochondrial formation of ROS, and promoting further lysosomal degradation. With regard to the mechanisms involved in enhanced mitochondrial ROS production, one particularly likely possibility is that of generation of free fatty acids. At least in pancreatic beta cell mitochondria, free fatty acids have been shown to increase ROS generation, perhaps through electron leak involving complex I of the respiratory chain [48]. Whether the progressive lysosomal destabilization is dependent exclu- sively on upstream actions of cathepsins B and D, or whether other lysosomal constituents might similarly desta- bilize mitochondria and lysosomes is not yet clear. Our present understanding concerning the involvement of lysosomes in apoptosis is summarized in Fig. 4. As shown, the initiation of apoptosis by exogenous oxidants, and by at least some other agonists, may involve early lysosomal rupture. The release of lysosomal enzymes (LE) into the cell cytoplasm may set off a cascade of intracellular degradative events. These LE may: (a) attack mitochondria directly, inducing release of cytochrome c; (b) directly and/ or indirectly cause enhanced formation of mitochondrial ROS (and further oxidant-induced lysosomal destabiliza- tion); (c) activate lytic pro-enzymes, such as PLA2, which in turn would attack both mitochondria and lysosomes; (d) activate Bid and/or other pro-apoptotic proteins; and (e) directly activate pro-caspases. Notably, this sequence of early events (except for cytochrome c release) may be relatively independent of the classical apoptotic cascade involving caspase activation. In many circumstances, this Ôlysosomal-mitochondrial axisÕ apoptotic pathway, invol- ving combined effects of caspases, lysosomal hydrolases and mitochondrial ROS generation, may be of central import- ance in the final execution of the apoptotic cascade wherein a lysosomal/mitochondrial cross-talk may constitute an amplifying loop. Acknowledgements We thank G. Dubowchik (Bristol-Myers Squibb; Pharmaceutical Research Institute) for the kind gift of MSDH. This study was supported by a grant from the Swedish Cancer Foundation (grant no. 4296). JWE was the recipient of a visiting professorship from the Linko ¨ ping University Hospital and is supported by The Common- wealth of Kentucky Research Challenge Trust Fund. References 1. Nicholson, D.W. (2000) From bench to clinic with apoptosis- based therapeutic agents. Nature 407, 810–816. 2. Nicholson, D.W. (2001) Apoptosis. Baiting death inhibitors. Nature 410, 33–34. 3. Savill, J. & Fadok, V. (2000) Corpse clearance defines the meaning of cell death. Nature 407, 784–788. 4. Baumann, S., Krueger, A., Kirchhoff, S. & Krammer, P.H. (2002) Regulation of T cell apoptosis during the immune response. Curr. Mol. Med. 2, 257–272. 3784 M. Zhao et al. (Eur. J. Biochem. 270) Ó FEBS 2003 5. Yuan, J. & Yankner, B.A. (2000) Apoptosis in the nervous system. Nature 407, 802–809. 6. Kaufmann, S.H. & Hengartner, M.O. (2001) Programmed cell death: alive and well in the new millennium. Trends Cell Biol. 11, 526–534. 7. Hengartner, M.O. (2000) The biochemistry of apoptosis. Nature 407, 770–776. 8. Zakeri,Z.&Lockshin,R.A.(2002)Celldeathduringdevelop- ment. J. Immunol. Methods 265, 3–20. 9. Zdolsek, J., Zhang, H., Roberg, K. & Brunk, U. (1993) H 2 O 2 - mediated damage to lysosomal membranes of J-774 cells. Free Radic. Res. Commun. 18, 71–85. 10. O ¨ llinger, K. & Brunk, U.T. (1995) Cellular injury induced by oxidative stress is mediated through lysosomal damage. Free Radic. Biol. Med. 19, 565–574. 11.Brunk,U.T.,Dalen,H.,Roberg,K.&Hellquist,H.B.(1997) Photo-oxidative disruption of lysosomal membranes causes apoptosis of cultured human fibroblasts. Free Radic. Biol. Med. 23, 616–626. 12. Li, W., Yuan, X., Nordgren, G., Dalen, H., Dubowchik, G.M., Firestone, R.A. & Brunk, U.T. (2000) Induction of cell death by the lysosomotropic detergent MSDH. FEBS Lett. 470, 35–39. 13. Brunk, U.T., Neuzil, J. & Eaton, J.W. (2001) Lysosomal involvement in apoptosis. Redox Report 6, 91–97. 14. Yuan, X.M., Li, W., Dalen, H., Lotem, J., Kama, R., Sachs, L. & Brunk, U.T. (2002) Lysosomal destabilization in p53-induced apoptosis. Proc. Natl Acad. Sci. USA 99, 6286–6291. 15. Ishisaka,R.,Utsumi,T.,Kanno,T.,Arita,K.,Katunuma,N., Akiyama, J. & Utsumi, K. (1999) Participation of a cathepsin 1-type protease in the activation of caspase-3. Cell Struct. Funct. 24, 465–470. 16. Katunuma, N., Matsui, A., Le, Q.T., Utsumi, K., Salvesen, G. & Ohashi, A. (2001) Novel procaspase-3 activating cascade mediated by lysoapoptases and its biological significances in apoptosis. Adv. Enz. Reg. 41, 237–250. 17. Bursch, W. (2001) The autophagosomal-lysosomal compartment in programmed cell death. Cell Death Differ. 8, 569–581. 18. Salvesen, G.S. (2001) A lysosomal protease enters the death scene. J. Clin. Invest. 107, 21–22. 19. Zhao, M., Eaton, J.W. & Brunk, U.T. (2000) Protection against oxidant-mediated lysosomal rupture: a new anti-apoptotic activity of Bcl-2? FEBS Lett. 485, 104–108. 20. Antunes, F., Cadenas, E. & Brunk, U.T. (2001) Apoptosis induced by exposure to a low steady-state concentration of H 2 O 2 is a consequence of lysosomal rupture. Biochem. J. 365, 549–555. 21. Zhao, M., Eaton, J.W. & Brunk, U.T. (2001) Bcl-2 phosphory- lation is required for inhibition of oxidative stress-induced lyso- somal leak and ensuing apoptosis. FEBS Lett. 509, 405–412. 22. Zhao, M., Brunk, U.T. & Eaton, J.W. (2001) Delayed oxidant- induced cell death involves activation of phospholipase A2. FEBS Lett. 509, 399–404. 23. Obrador, E., Navarro, J., Mompo, J., Asensi, M., Pellicer, J.A. & Estrela, J.M. (1998) Regulation of tumour cell sensitivity to TNF- induced oxidative stress and cytotoxicity: role of glutathione. Biofactors 8, 23–26. 24. Andrieu-Abadie, N., Gouaze, V., Salvayre, R. & Levade, T. (2001) Ceramide in apoptosis signaling: relationship with oxidative stress. Free Radic. Biol. Med. 31, 717–728. 25. Muller, J.M., Ziegler-Heitbrock, H.W.L. & Baeuerle, P.A. (1993) Nuclear factor kappa B, a mediator of lipopolysaccharide effects. Immunobiology 187, 233–256. 26. Albrecht, H., Tschopp, J. & Jongeneel, C.V. (1994) Bcl-2 protects from oxidative damage and apoptotic cell death without inter- fering with activation of NF-kappa B by TNF. FEBS Lett. 351, 45–48. 27. Atabay, C., Cagnoli, C.M., Kharlamov, E., Ikonomovic, M.D. & Manev, H. (1996) Removal of serum from primary cultures of cerebellar granule neurons induces oxidative stress and DNA fragmentation: protection with antioxidants and glutamate receptor antagonists. J. Neurosci. Res. 43, 465–475. 28. Degli Esposti, M. & McLennan, H. (1998) Mitochondria and cells produce reactive oxygen species in virtual anaerobiosis: relevance to ceramide-induced apoptosis. FEBS Lett. 430, 338–342. 29. Garcia-Ruiz, C., Colell, A., Mari, M., Morales, A. & Fernandez- Checa, J.C. (1997) Direct effect of ceramide on the mitochondrial electron transport chain leads to generation of reactive oxygen species. Role of mitochondrial glutathione. J. Biol. Chem. 272, 11369–11377. 30. Dobmeyer, T.S., Findhammer, S., Dobmeyer, J.M., Klein, S.A., Raffel, B., Hoelzer, D., Helm, E.B., Kabelitz, D. & Rossol, R. (1997) Ex vivo induction of apoptosis in lymphocytes is mediated by oxidative stress: role for lymphocyte loss in HIV infection. Free Radic. Biol. Med. 22, 775–785. 31. Firestone, R.A., Pisano, J.M. & Bonney, R.J. (1979) Lysosomo- tropic agents. 1. Synthesis and cytotoxic action of lysosomotropic detergents. J. Med. Chem. 22, 1130–1133. 32. Wilson, P.D., Firestone, R.A. & Lenard, J. (1987) The role of lysosomal enzymes in killing of mammalian cells by the lyso- somotropic detergent N-dodecylimidazole. J. Cell Biol. 104, 1223–1229. 33. Nicoletti, I., Migliorati, G., Pagliacci, M.C., Grignani, F. & Riccardi, C. (1991) A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 139, 271–279. 34. Hempel, S.L., Buettner, G.R., O’Malley, Y.Q., Wessels, D.A. & Flaherty, D.M. (1999) Dihydrofluorescein diacetate is superior for detecting intracellular oxidants: comparison with 2¢,7¢-dichloro- dihydrofluorescein diacetate, 5 (and 6)-carboxy-2¢,7¢-dichlorodi- hydrofluorescein diacetate, and dihydrorhodamine 123. Free Radic. Biol. Med. 27, 146–159. 35. Gasnier, F., Rousson, R., Lerme, F., Vaganay, E., Louisot, P. & Gateau-Roesch, O. (1993) Use of Percoll gradients for isolation of human placenta mitochondria suitable for investigating outer membrane proteins. Anal. Biochem. 212, 173–178. 36. Hyslop, P.A. & Sklar, L.A. (1984) A quantitative fluorimetric assay for the determination of oxidant production by poly- morphonuclear leukocytes: its use in the simultaneous fluorimetric assay of cellular activation processes. Anal. Biochem. 141, 280– 286. 37. Huot, J., Houle, F., Marceau, F. & Landry, J. (1997) Oxidative stress-induced actin reorganization mediated by the p38 mitogen- activated protein kinase/heat shock protein 27 pathway in vas- cular endothelial cells. Circ. Res. 80, 383–392. 38. Brunk, U.T. & Svensson, I. (1999) Oxidative stress, growth factor starvation and Fas activation may all cause apoptosis through lysosomal leak. Redox Report 4, 3–11. 39. Yuan, X.M., Li, W., Brunk, U.T., Dalen, H., Chang, Y.H. & Sevanian, A. (2000) Lysosomal destabilization during macrophage damage induced by cholesterol oxidation products. Free Radic. Biol. Med. 28, 208–218. 40. Neuzil, J., Zhao, M., Ostermann, G., Sticha, M., Gellert, N., Weber, C., Eaton, J.W. & Brunk, U.T. (2002) Alpha-tocopheryl succinate, an agent with in vivo anti-tumour activity, induces apoptosis by causing lysosomal instability. Biochem. J. 362, 709–715. 41. Boveris, A. & Chance, B. (1973) The mitochondrial generation of hydrogen peroxide. General properties and effect of hyperbaric oxygen. Biochem. J. 134, 707–716. 42. Chance, B., Sies, H. & Boveris, A. (1979) Hydroperoxide meta- bolism in mammalian organs. Physiol. Rev. 59, 527–605. Ó FEBS 2003 Lysosomes, apoptosis and mitochondria-mediated oxidative stress (Eur. J. Biochem. 270) 3785 43. Liu, Y., Fiskum, G. & Schubert, D. (2002) Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80, 780–787. 44. Hansford, R.G., Hogue, B.A. & Mildaziene, V. (1997) Depen- dence of H 2 O 2 formation by rat heart mitochondria on substrate availability and donor age. J. Bioenerg. Biomembr. 29, 89–95. 45. Scorrano, L., Penzo, D., Petronilli, V., Pagano, F. & Bernardi, P. (2001) Arachidonic acid causes cell death through the mitochon- drial permeability transition. Implications for tumor necrosis factor-alpha aopototic signaling. J. Biol. Chem. 276, 12035–12040. 46. Hancock, J.T., Desikan, R. & Neill, S.J. (2001) Does the redox status of cytochrome C act as a fail-safe mechanism in the regu- lation of programmed cell death? Free Radic. Biol. Med. 31, 697– 703. 47. de Duve, C., de Barsy, T., Poole, B., Trouet, A., Tulkens, P. & Van Hoof, F. (1974) Commentary. Lysosomotropic agents. Biochem. Pharmacol. 23, 2495–2531. 48. Koshkin,V.,Wang,X.,Scherer,P.E.,Chan,C.B.&Wheeler, M.B. (2003) Mitochondrial functional state in clonal pancreatic beta-cells exposed to free fatty acids. J. Biol. Chem. 278, 19709– 19715. 49. Roberg,K.&O ¨ llinger, K. (1998) Oxidative stress causes reloca- tion of the lysosomal enzyme cathepsin D with ensuing apoptosis in neonatal rat cardiomyocytes. Am. J. Pathol. 152, 1151–1156. 50. Roberg, K., Johansson, U. & O ¨ llinger, K. (1999) Lysosomal release of cathepsin D precedes relocation of cytochrome c and loss of mitochondrial transmembrane potential during apoptosis induced by oxidative stress. Free Radic. Biol. Med. 27, 1228–1237. 51. Roberg, K. (2001) Relocalization of cathepsin D and cytochrome c early in apoptosis revealed by immunoelectron microscopy. Lab. Invest. 81, 149–158. 52. Roberg, K., Kagedal, K. & O ¨ llinger, K. (2002) Microinjection of cathepsin D induces caspase-dependent apoptosis in fibroblasts. Am. J. Pathol. 161, 89–96. 53. Stoka, V., Turk, B., Schendel, S.L., Kim, T.H., Cirman, T., Snipas, S.J., Ellerby, L.M., Bredesen, D., Freeze, H., Abraham- son,M.,Bromme,D.,Krajewski,S.,Reed,J.C.,Yin,X.M.,Turk, V. & Salvesen, G.S. (2001) Lysosomal protease pathways to apoptosis. Cleavage of bid, not pro-caspases, is the most likely route. J. Biol. Chem. 276, 3149–3157. 54. Zhou, Q. & Salvesen, G.S. (1997) Activation of pro-caspase-7 by serine proteases includes a non-canonical specificity. Biochem. J. 324, 361–364. 55. Katunuma, N., Matsui, A., Kakegawa, T., Murata, E., Asao, T. & Ohba, Y. (1999) Study of the functional share of lysosomal cathepsins by the development of specific inhibitors. Adv. Enzyme Regul. 39, 247–260. 56. Guicciardi, M.E., Deussing, J., Miyoshi, H., Bronk, S.F., Svingen, P.A., Peters, C., Kaufmann, S.H. & Gores, G.J. (2000) Cathepsin B contributes to TNF-alpha-mediated hepatocyte apoptosis by promoting mitochondrial release of cytochrome c. J. Clin. Invest. 106, 1127–1137. 57. Cuvillier, O., Edsall, L. & Spiegel, S. (2000) Involvement of sphingosine in mitochondria-dependent Fas-induced apoptosis of type II Jurkat T cells. J. Biol. Chem. 275, 15691–15700. 58. Werneburg, N.W., Guicciardi, M.E., Bronk, S.F. & Gores, G.J. (2002) Tumor necrosis factor-a-associated lysosomal permeabili- zation is cathepsin B dependent. Am. J. Physiol. Gastrointest. Liver Physiol. 283, 947–956. 59. Ka ˚ gedal, K., Zhao, M., Svensson, I. & Brunk, U.T. (2001) Sphingosine-induced apoptosis is dependent on lysosomal pro- teases. Biochem. J. 359, 335–343. 3786 M. Zhao et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . inducing release of cytochrome c; (b) directly and/ or indirectly cause enhanced formation of mitochondrial ROS (and further oxidant- induced lysosomal destabiliza- tion); (c) activate lytic pro -enzymes, . mitochondrial ROS production, release of cytochrome c, and activation of PLA2 (all of which may promote the apoptotic cascade), we tested the effects of two lysosomal cathepsins (cathepsin B, a cysteine. of lysosomal constituents (such as cathepsins B and D) and activation of PLA2, leading to enhanced mitochondrial oxidant production, further lysosomal rupture and, finally, mitochondrial cytochrome