Endoplasmic Reticulum Stress–Associated Caspase 12 Mediates Cisplatin-Induced LLC-PK1 Cell Apoptosis

Hua Liu and Radhakrishna Baliga Department of Pediatrics, University of Mississippi Medical Center, Jackson, Mississippi

Reactive oxygen metabolites are important mediators in cisplatin-induced apoptosis in renal tubular epithelial cells (LLC-PK1). Mitochondria have been implicated to play a principal role in cisplatin-induced apoptosis. Caspase 12, an endoplasmic reticulum (ER)-specific caspase, participates in apoptosis under ER stress. Cytochrome P450 system is crucial to the generation of reactive oxygen metabolites and is present at high concentration in the ER. The direct role of caspase 12 in any model of renal injury has not previously been described. In this study, cleavage of procaspase 12 preceded that of caspases 3 and 9 after cisplatin treatment of LLC-PK1 cells. The active form of caspase 8 was not detected throughout the course of study. Preincubation of the LLC-PK1 cells with the caspase 9 inhibitor did not attenuate caspase 3 activation and provided no significant protection. Caspase 3 inhibitor provided only modest protection against cisplatin-induced apoptosis. LLC-PK1 cells that were transfected with anti–caspase 12 antibody significantly attenuated cisplatin-induced apoptosis. Taken together, these data indicate that caspase 12 plays a pivotal role in cisplatin-induced apoptosis. It is proposed that the oxidative stress that results from the interaction of cisplatin with the ER cytochrome P450 leads to activation of procaspase 12, resulting in apoptosis. J Am Soc Nephrol 16: 1985–1992, 2005. doi: 10.1681/ASN.2004090768

isplatin is a widely used chemotherapeutic agent in activated. The extrinsic, or receptor-mediated, pathway, involves the treatment of a variety of solid human tumors (1). cellular ligands such as TNF-␣ or FasL, resulting in caspase 8 C The therapeutic effectiveness of cisplatin is blunted by activation (19). The intrinsic, or mitochondrial, pathway involves the nephrotoxicity that develops primarily at the S3 segment of release of the mitochondrial protein cytochrome c, resulting in the proximal tubule (2,3). Reactive oxygen metabolites (ROM) activation of caspase 9 (20). Both caspases 8 and 9 then activate are important mediators of cisplatin-induced renal tubular cell caspase 3, the major effector caspase that is responsible for the apoptosis (4,5). The mechanism that triggers apoptosis in re- destruction of various substrates. sponse to cisplatin is not well defined. Mitochondria have been Recent studies have indicated a close link between ER stress and implicated as principal sensors and thus critical in the genera- caspase 12 activation resulting in apoptotic cell death (21–24). tion of the apoptotic cascade (6–8). The prevailing assumption Procaspase 12 is located predominantly on the cytoplasmic side of has been that all electron-transfer processes that lead to ROM the ER and is expressed at high levels in the kidney, especially the production are localized to the mitochondria (9,10). Recent renal tubular epithelial cells (25). The ER also regulates apoptosis studies, including that of ours, suggest that the cytochrome by sensitizing the mitochondria to a variety of extrinsic and in- P450 (CYP) system is crucial to the ROM-mediated cell signal- trinsic stimuli and also initiating cell death signals of its own (26). ing and that CYP is present in high concentration in the endo- We demonstrated previously that the microsomal CYP2E1 is a site plasmic reticulum (ER) of most animal cells (11–14). and a source for ROM generation in cisplatin-induced apoptosis Caspases are a family of cysteine proteases that play an impor- (5). The direct role of caspase 12 in cisplatin-induced apoptosis in tant role in the programmed cell death (15–17). Their activation renal tubular epithelial cells was not explored previously. Hence, signals a point of no return in the apoptotic pathway. The caspase our study was designed to explore the role of caspase 12 in family is broadly divided into two groups, namely the initiator cisplatin-induced apoptosis. Treatment of the LLC-PK1 cells with caspases (caspases 8, 9, and 12) and the effector caspases (caspases cisplatin resulted in cleavage of procaspase 12 and apoptosis. 3, 6, and 7). The initiator caspases undergo activation in response Preincubation of the cells with caspase 9 inhibitor did not inhibit to an apoptotic stimuli and in turn activate the effector caspases caspase 3 activation and provided no significant protection. that are responsible for the cleavage of a wide variety of physio- Caspase 3 inhibitor provided only modest protection. LLC-PK1 logic substrates (18). There are two relatively well-recognized cell cells that were transfected with anti–caspase 12 antibody (Ab) death pathways that depend on the type of procaspase that is significantly attenuated cisplatin-induced apoptosis. We also looked at the interrelation between ER stress and procaspase 12

Received September 14, 2004. Accepted April 7, 2005. cleavage. Procaspase 12 was cleaved in cisplatin-treated CYP2e1 wild-type (WT) mice, and this cleavage was significantly pro- Published online ahead of print. Publication date available at www.jasn.org. tected in the CYP2e1 knockout (KO) animals. We therefore postu- Address correspondence to: Dr. Radhakrishna Baliga, Department of Pediatrics, late that the oxidative stress that results from the interaction of Division of Nephrology, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216-4505. Phone: 601-984-5970; Fax: 601-815-5902; cisplatin with ER CYP leads to the activation of procaspase 12, E-mail: [email protected] resulting in apoptosis.

Materials and Methods Transferase-Mediated dUTP Nick-End Labeling Assay Cell Culture and Treatment Transferase-mediated dUTP nick-end labeling (TUNEL) was per- LLC-PK1 cells, purchased from American Type Culture Collection formed by using ApoAlert DNA Fragmentation Assay Kit (Clontech, (CRL 1392; Manassas, VA), were maintained in Medium 199 supple- Palo Alto, CA), by which fluorescein-dUTP incorporation at the free ends of fragmented DNA is visualized by fluorescence microscopy. mented with 3% FBS in a humidified atmosphere of 95% air/5% CO2 at 37°C and fed at intervals of 3 d (27). The cells were maintained in TUNEL stain was performed following the manufacturer’s protocol. 75-cm2 tissue culture flasks, and the monolayer was subcultured using 0.05% trypsin, 0.53 mM EDTA, in HBSS. All experiments were carried Detection of DNA Fragmentation out on confluent cell monolayer between passages 203 and 215. Apo- LLC-PK1 cells were homogenized and lysed with a buffer that con- ptosis was induced in LLC-PK1 cells by incubation with 50 ␮M cispla- tained 10 mM Tris-HCl (pH 7.5), 1 mM EDTA, 0.1 mM NaCl, and 0.5% tin at 37°C for 12 h (28). Caspase inhibitions were performed by SDS. The lysates were incubated with proteinase K (0.2 mg/ml) at 50°C preincubation of caspase 3 inhibitor DEVD-CHO (60 ␮M; Bachem, overnight. DNA was isolated from the lysates following the procedures as Torrance, CA) and caspase 9 inhibitor LEHD-CHO (60 ␮M; Peptide described by Ramachandra and Studzinski (31). The DNA obtained was Inc., Osaka, Japan) for1hat37°C. resuspended in a buffer (10 mM Tris-HCl and 1 mM EDTA) at 1 ␮g/ml and incubated with 0.1 U/10 ␮g DNA of DNAase-free RNAase cocktail (cat. no. 2286; Ambion, Austin, TX). Electrophoresis was performed in Determination of Caspase Activity 1.6% agarose gels, and DNA was visualized with ethidium bromide. Caspases 3, 8, and 9 activities were tested by use of the Caspase Colorimetric Activity Assay Kits (Chemicon International Inc., Te- Immunocytochemistry mecula, CA) following the manufacturer’s protocols. Equal numbers of Cultured LLC-PK1 cells were fixed in B5 solution, and cell block was ϫ 6 cells (0.5 to 2 10 ) from each group were treated with cell lysis buffer made by imbedding in tissue medium. Sections were cut on a glass ϫ on ice for 10 min and centrifuged at 10,000 g for 5 min. The super- slide. After deparaffinization and antigen retrieval, the sections were natants were incubated with respective caspase substrates (DEVD-pNA immunolabeled and visualized according to an avidin-biotin complex for caspase 3, IETD-pNA for caspase 8, and LEHD-pNA for caspase 9) (ABC) method (32). at 37°C for 1 h. The activities were assayed by use of a spectrophotom- eter microtiter plate reader at 405 nm. The specificity of changes in the Transfection of Anti–Caspase 12 Ab into LLC-PK1 Cells caspase activities was confirmed by addition of a specific inhibitor to Anti–caspase 12 Ab (AB3612, Lot 22101182; Chemicon) was delivered the caspase in parallel experiments. into LLC-PK1 cells by using transfection agent Chariot (30025; Active Motif, Carlsbad, CA) following the manufacturer’s protocol. Briefly, the initial transfection mixture was prepared by adding 100 ␮l of anti–caspase Preparation of Cellular Fractions 12 Ab dilution (1.2 ␮l of stock Ab in 98.8 ␮l of PBS) to 100 ␮l of Chariot For the detection of caspases, cell fractions were prepared as de- ␮ ␮ dilution (6 l of Chariot stock solution in 96 l of distilled H2O). The scribed previously (25). In brief, LLC-PK1 cells or kidney cortex was mixture was incubated at room temperature for 30 min to allow the lysed and homogenized in buffer A (50 mM Tris-HCl [pH 8.0], 1 mM formation of Chariot-Ab complex. LLC-PK1 cells in six-well culture plate mercaptoethanol, 1 mM EDTA, 0.32 M sucrose, and 0.1 mM PMSF). (40 to 50% confluence) were overlaid with the mixture (200 ␮l/well), Nuclear fraction was collected from the pellet after first centrifugation added with 400 ␮l of serum-free culture medium (final Ab dilution 1:500), at 900 ϫ g for 10 min. The supernatants were centrifuged again at incubated at 37°C in a humidified atmosphere that contained 5% CO2 for ϫ g 5000 for 10 min, and the pellet was collected as mitochondrial 1 h, then added with 1 ml of completed growth medium and continued to fraction. The resulting supernatants were centrifuged at 105,000 ϫ g for 60 min, and the microsomal and soluble fractions were the pellet and the supernatant, respectively. The purity of these fractions was controlled by the presence of known compartment-specific proteins using Western blot analysis: Cytochrome c for mitochondria and grp78 (Bip) for ER (25). For measurement of cytochrome c release, cytosolic mitochondrial fractions were prepared as described elsewhere (29). In brief, cells were washed with PBS and collected, incubated in 200 ␮lof mitochondrial buffer (68 mM sucrose, 200 mM mannitol, 50 mM KCl, 1 mM EGTA, 1 mM dithiothreitol, and protease inhibitor cocktail [P-8340; Sigma, St. Louis, MO]), homogenized, and centrifuged at 4°C at 800 ϫ g. The supernatant was centrifuged at 14,000 ϫ g for 10 min at 4°C, and the resulting pellet and supernatant were collected as mitochondrial and cytosolic fractions, respectively.

Western Blot Western blot was performed by the chemiluminescence method as described elsewhere (30). The antibodies used were as follows: Rabbit anti–caspase 12 (AB3612, Chemicon) and mouse anti-actin (MAB1501; Figure 1. Cleavage of procaspases 12 and 3 in CYP2e1 knockout Chemicon); rabbit anti–caspase 3 (sc-7148), caspase 8 (sc-7890), caspase (KO) mice after cisplatin (Cis) treatment. Procaspase 12 cleav- 9 (sc-8355), and rabbit anti–cytochrome C (sc-7159; Santa Cruz Biotech- age was markedly attenuated (A) and there was no activation nology, Santa Cruz, CA); and rabbit anti-grp78 (Stressgen, Victoria, BC, of caspase 3 (B) in the cisplatin-treated KO mice as compared Canada). with that of wild-type (WT) mice. J Am Soc Nephrol 16: 1985–1992, 2005 Caspase 12 in Cisplatin-Induced Apoptosis 1987

Figure 2. Localization of procaspase 12. Western blot analysis of cell fraction from normal rat kidney cortex using anti–caspase 12 antibodies (Ab) showed that procaspase 12 was present exclusively in the microsomal fraction but not in the nuclear, mitochondrial, and soluble fractions. The purity of these fractions was controlled by the presence of known compartment- specific proteins: Cytochrome c for mitochondria and grp78 Figure 3. Representative sections of transferase-mediated dUTP (Bip) for endoplasmic reticulum (ER). nick-end labeling (TUNEL) staining for detection of apoptotic LLC-PK1 cells 12 h after cisplatin treatment. Apoptotic cells were identified by the nuclear fluorescence staining. incubate at 37°C for 2 h. The cells were washed twice with PBS before the cisplatin treatment. The transfection efficacy of Chariot in LLC-PK1 cells was tested by using ␤-galactosidase in a pilot study, and the transfection Detection and Localization of Procaspase 12 rate for ␤-galactosidase was Ͼ90%. Procaspase 12 was detected predominantly in the microsomal fraction of the normal kidney cortex but not in the nuclear, CYP2e1 KO Mice mitochondrial, and soluble fractions by Western blot analysis Mice (strain 129/SV) carrying a deletion of CYP2e1 gene were pro- (Figure 2). vided by Dr. Gonzalez (National Institutes of Health, Bethesda, MD). The CYP2e1 gene was isolated from a 129/SV mouse (CYP2e1 WT, Cisplatin Induced Apoptosis CYP2e1ϩ/ϩ) genomic library that contained the complete coding re- Cisplatin induced apoptosis in a dose-dependent manner as gion. The gene was disrupted by the replacement of exon 2 with the detected by TUNEL staining (17 Ϯ 4.5% at 25 ␮M, 31 Ϯ 6.7% at PGK-NEO cassette (33). Mice homozygous for the disrupted CYP2e1 50 ␮M, and 85 Ϯ 10% at 100 ␮M). Significant apoptosis with allele were born and developed normally with no obvious phenotypic intense TUNEL-positive nuclear staining was noted in the cells divergence from WT mice. The clone of KO mice was maintained by that were treated with 50 ␮M cisplatin for 12 h (Figure 3). breeding CYP2e1Ϫ/Ϫ male mice with CYP2e1Ϫ/Ϫ female mice. Immu- noblotting and Northern blot confirmed the complete absence of CYP2E1 protein and mRNA in the mice. The animal study was ap- proved by the Institutional Review Board, and the experimental procedures were conducted in accordance with our institutional guide- lines.

Results Cisplatin-Induced ER Stress Results in the Cleavage of Procaspase 12 We recently demonstrated an important role of microsomal CYP2E1 as a source of ROM generation in cisplatin-induced renal tubular apoptosis (5). Treatment of CYP2e1 KO mice with cisplatin resulted in marked reduction of ROM generation and significant attenuation of apoptosis when compared with CYP2e1 WT mice. Hence, we looked at the role of caspase 12 in these animals that were treated with cisplatin. Indeed, pro- Figure 4. Time course of caspase activity after cisplatin treatment. caspase 12 was markedly cleaved in the WT mice that were Cells were treated with 50 ␮M cisplatin and assayed for caspase treated with cisplatin, and this cleavage was significantly di- activity at different time points. Caspase 3 and caspase 9 were minished in the KO animals (Figure 1A). These results indicate significantly activated at 9 h and continued to increase. There was that the ER-associated caspase 12 plays an important role in only a modest increase in the caspase 8 activity at 24 h as com- cisplatin-induced apoptosis. The activation of caspase 3 was pared with caspases 3 and 9. Values are means Ϯ SEM from four also attenuated in these KO mice (Figure 1B). separate measurements. *P Ͻ 0.05 versus control (0 h). 1988 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 1985–1992, 2005

Figure 5. Time course of caspase activation by Western blot analysis. (A) The representative blots demonstrate that the cleaved form of caspase 12 was noted at 9 h after cisplatin (50 ␮M) treatment and increased progressively up to 24 h. The active form of caspase 3 was observed at 12 h of the cisplatin treatment, with progressive increase up to 24 h. There was no activation of caspase 8 as indicated by neither presence of active form nor decrease in the band of its proform. Caspase 9 was activated at 12 h as indicated by a decrease in density of the proform of caspase 9 with absence at 24 h. This is because the Ab to caspase 9 was probably specific to the proform only and did not recognize the active form of caspase 9. (B) Densitometric analyses of procaspase 12 cleavage was based on three independent experiments for the Western blots. The values of density for caspase 12 were corrected by loading control actin. (C) Western blot showed that the anti–caspase 9 Ab did not recognize the active form of caspase 9 by using a known caspase 9 activator.

Time Course of Caspase Activation and Cytochrome C ptosis. Western blot analysis indicated that cytochrome c was Release released from the mitochondria into the cytosol in a time- We initially examined the activities of caspases 3, 9, and 8 in dependent manner (Figure 6), thus implicating the caspase response to cisplatin treatment in the LLC-PK1 cells. Caspases 9–dependent mitochondrial pathway. Our data thus suggest 3 and 9 were significantly activated at 9 h and continued to that the cleavage of the ER-associated procaspase 12 precedes increase up to 24 h (Figure 4). There was only a modest increase that of caspases 3 and 9. in the caspase 8 activity at 24 h (Figure 4). Western blot analysis was performed to examine the cleavage Cisplatin-Induced Apoptosis Is Caspase 12–Dependent of procaspase 12 and its relationship to caspases 3, 9, and 8 Effect of Caspase 9 Inhibitor. Pretreatment of the cells during cisplatin-induced injury to the LLC-PK1 cells. As shown with caspase 9 inhibitor prevented cisplatin-induced caspase 9 in Figure 5A, the cleaved form of procaspase 12 was noted at activation as measured by its activity and Western blot analysis 9 h after cisplatin treatment and increased progressively there- (Figure 7, A and C). Caspase 9 inhibitor, however, did not after up to 24 h. The densitometric analysis of the blot demon- completely attenuate the caspase 3 activity (Figure 7B) and the strated a graduate decrease in procaspase 12 density and an formation of the active form of caspase 3 as determined by increase in the cleaved form of caspase 12 starting at 9 h (Figure 5B). The active form of caspase 3 was observed at 12 h after the cisplatin treatment with progressive increase up to 24 h (Figure 5A). The proform of caspase 9 showed a decrease in density of the band at 12 h with absence at 24 h (Figure 5A). This is because the Ab to caspase 9 was probably specific to the proform only and did not recognize the active form of caspase 9. The anti–caspase 9 Ab specificity was confirmed by a study using a known caspase 9 activator staurosporine. As shown in Figure 5C, the procaspase 9 was significantly reduced in staurosporine-treated (5 ␮M) cells as compared with the control, whereas no cleaved caspase 9 was observed. There was no activation of caspase 8 as indicated by neither presence of active Figure 6. Cytochrome c release by Western blot analysis. The form nor decrease in the band of its proform (Figure 5A). The decrease in the density of cytochrome c band from mitochon- weak activity of caspase 8 coincided with the lack of its activa- drial fraction corresponded with the increase in the density of tion by way of Western blot, indicating that the extrinsic path- the band in cytosolic fraction, indicating cytochrome c release way may not play a significant role in cisplatin-mediated apo- beginning at 12 h after cisplatin treatment. J Am Soc Nephrol 16: 1985–1992, 2005 Caspase 12 in Cisplatin-Induced Apoptosis 1989

Western blot (Figure 7D), thus indicating that the activation of fragmentation and TUNEL staining (Figure 8, C and F). The caspase 3 may also be from an unknown pathway or caspase 12. procaspase 12 was significantly decreased in the cisplatin- The cleavage of procaspase 12 after pretreatment with caspase treated LLC-PK1 cells associated with the presence of cleaved 9 inhibitor was no different from that induced by cisplatin caspase 12, and this cleavage was prevented in the cells that alone (Figure 7E). Caspase 9 inhibitor had no protective effect were transfected by caspase 12 Ab (Figure 10A). The induction on cisplatin-induced apoptosis as indicated by DNA laddering of the active form of caspase 3 was also markedly reduced (Figure 8A) and TUNEL staining (Figure 8D). (Figure 10B) in anti–caspase 12 Ab–transfected cells that were Effect of Caspase 3 Inhibitor. Cisplatin-induced caspase 3 treated with cisplatin. There were no significant differences in activation as measured by Western blot analysis and its activity the cytochrome c release in the cytosol and the activation of was markedly attenuated by pretreatment of the cells with caspase 9 in the caspase 12 Ab–transfected cells and the control caspase 3 inhibitor (Figure 9, A and B). Pretreatment of the cells cells that were treated with cisplatin (Figure 10, C and D). with caspase 3 inhibitor did not alter the cleaved form of procaspase 12 when compared with cisplatin alone (Figure 9C). Discussion Caspase 3 inhibitor provided modest protection against cispla- ER is one of the largest cell organelles that is responsible for tin-induced apoptosis as demonstrated by TUNEL staining the production of cellular components, proteins, lipids, and (Figure 8E). However, there was no effect on cisplatin-induced sterols (34). Its proper function is essential for the cell. Various DNA damage as measured by DNA laddering (Figure 8A). This agents, including oxidants, can interfere with the ER function, suggests that there exists another pathway or caspase 12 that leading to ER stress and cell death (34). Caspase 12 is an may act directly on the cells and result in apoptosis. ER-specific caspase that is localized to the cytosolic face of the Effect of Caspase 12 Ab. Transfection of the anti–caspase ER, making it vulnerable to the ER stress and the activation of 12 Ab using a Chariot kit significantly attenuated the cisplatin- the caspase cascade (25). In our study, procaspase 12 was induced apoptosis as demonstrated by TUNEL staining (Figure localized predominantly to the microsomal fraction of the kid- 8F) and DNA laddering (Figure 8B). Chariot itself had no effect ney cortex but not in the nuclear, mitochondrial, and soluble on cisplatin-induced apoptosis as confirmed by both DNA fractions by Western blot analysis.

Figure 7. Effect of caspase 9 inhibitor LEHD-CHO on caspase activation. Pretreatment of the cells with LEHD-CHO completely prevented cisplatin-induced caspase 9 activation as measured by its activity (A) and Western blot (C). Administration of cisplatin to the cells that were pretreated with LEHD-CHO did not completely abolish the activation of caspase 3 (B and D). The cleavage of procaspase 12 was no different in the cisplatin-treated cells with or without pretreatment by the caspase 9 inhibitor (E). *P Ͻ ϩ 0.05 versus control; P Ͻ 0.05 versus cisplatin alone. 1990 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 1985–1992, 2005

cisplatin had significant reduction of ROM generation and provided marked protection against apoptosis (5). Procaspase 12 was markedly cleaved in the WT mice, with significant reduction of the cleavage in the KO animals that were treated with cisplatin, indicating that the ER stress activates caspase 12 that plays a prominent role in cisplatin-induced apoptosis. Having demonstrated the link between ER stress and procaspase 12 cleavage resulting in apoptosis, we next examined the downstream targets that were yet to be identified in cisplatin-induced apoptosis after procaspase 12 cleavage. The cleavage of procaspase 12 preceded that of caspases 3 and 9, whereas the activation of caspase 8 was not detected, indicating the lack of association with the death receptor pathway. Cytochrome c was released into the cytosol in a time-dependent manner, thus implicating the involvement of the mitochondrial apoptotic pathway. Caspase 9 inhibitor did not completely attenuate caspase 3 activity, and the formation of the active form of the enzyme thus suggests that the activation of caspase 3 may result from an unknown pathway, possibly through caspase 12. This result is supported by the observation of Kaushal et al. (28), in which caspase 3 activation preceded that of caspase 9. Caspase 9 inhibitor had no protective effect as indicated by DNA laddering and TUNEL staining. However, caspase 3 inhibitor provided only modest protection as demonstrated by TUNEL staining with no effect on cisplatin-induced DNA fragmentation as tested by DNA laddering, thus suggesting that there exists another pathway or caspase 12 that may act directly on these cells. Our results are similar to those of Schnellman et al. (35), who demonstrated that cisplatin-induced apoptosis is partly caspase 3 dependent and that the activation of caspase 3 could be independent of the mitochondrial dysfunction as well Figure 8. Effect of caspase inhibitors on DNA fragmentation. Cisplatin (50 ␮M, 12 h) induced apoptosis as indicated by DNA as caspases 8 and 9 activation. fragmentation (A and B). Both caspases 3 and 9 inhibitors did LLC-PK1 cells that were stably transfected with caspase 12 not significantly reduce the DNA damage from cisplatin treat- Ab prevented the cleavage of procaspase 12 and significantly ment (A). However, transfection of anti–caspase 12 Ab to LLC- attenuated the cisplatin-induced apoptosis as demonstrated by PK1 cells prevented cisplatin-induced DNA fragmentation (B). both TUNEL staining and DNA laddering, thus indicating the Chariot by itself did not have any effect on cisplatin-induced key role of caspase 12 in cisplatin-induced apoptosis. Several DNA fragmentation (C). Effect of caspase inhibitors on cisplatin mechanisms have been considered to be responsible for the (Cis)-induced apoptosis by TUNEL staining. Pretreatment of cytotoxic effects of cisplatin, including DNA damage by intra- caspase 9 inhibitor did not significantly decrease the cisplatin- and interstrand cross-link (36). However, Shoshan et al. (37) induced apoptosis as indicated by the percentage of TUNEL demonstrated the ability of cisplatin to induce ER stress–medi- positive staining (D). There was only modest protection by ated apoptosis in enucleated tumor cells independent of the caspase 3 inhibitor (E). LLC-PK1 cells that were transfected with anti–caspase 12 Ab demonstrated marked resistance to direct DNA damaging activity. This report thus further cisplatin-induced DNA damage (F). Chariot by itself did not strengthens our observation on the role of ER-specific caspase have any effect on cisplatin-induced apoptosis as measured by 12 in this model of injury. As mentioned earlier, inhibition of TUNEL staining (F). The percentage of apoptotic cells was caspase 3 provided only modest protection; in contrast, caspase calculated by the cells with positive nuclear staining in five to 12 Ab–transfected cells significantly attenuated cisplatin-in- six randomly chosen microscopy fields. Values are means Ϯ duced apoptosis. This suggests that caspase 12 could execute Ͻ ϩ Ͻ SEM. *P 0.05 versus controls; P 0.05 versus cisplatin alone. apoptosis directly or through unknown intermediates. Apoptotic agents that perturb ER functions can also induce release of cytochrome c from the mitochondria (22,38). It is possi- Previous studies using renal tubular epithelial cells have ble that the modulation of calcium homeostasis and ROM gener- shown that cisplatin-induced apoptosis is caspase dependent ation can mount an oxidative stress response that damages mem- and that mitochondria play a prominent role (28,30). We re- branes, including that of the mitochondria (39). This is substantiated cently demonstrated an important role of the microsomal CYP by the fact that caspase 3 activation was attenuated in the CYP2e1 2E1 as the source of ROM in cisplatin-induced nephrotoxicity KO mice, in which ROM generation was markedly reduced. In and apoptosis (5). CYP 2e1 KO mice that were treated with our study, no significant differences were observed in the cyto- J Am Soc Nephrol 16: 1985–1992, 2005 Caspase 12 in Cisplatin-Induced Apoptosis 1991

Figure 9. Effect of caspase 3 inhibitor DEVD-CHO on caspase activation. Cisplatin-induced caspase 3 activation was reduced by pretreatment of the cells with DEVD-CHO (A and B). The cleavage of procaspase 12 was similar to that induced by cisplatin alone (C).

permeability and the activation of caspase 9 are not the downstream effects of procaspase 12 cleavage. Our data thus indicate the existence of a novel apoptotic pathway in which caspase 12 functions as the initiator caspase in response to a toxic insult to the LLC-PK1 cells. The ER-specific caspase 12 not only plays a critical role but also helps to define the precise mechanism of apoptosis in this model of renal injury. We propose that the oxidative stress that results from the interaction of cisplatin with CYP leads to the activation of caspase 12 that acts directly on the cell and also via the effector caspase 3 resulting in apoptosis. The generation of ROM also leads to the mitochondrial membrane damage with cytochrome c release that activates caspase 9. These two arms of the ER stress response operate independent of each other, leading to apoptosis. Recent observation indicates that caspase 4 in humans is homologous to murine caspase 12, and it is activated in an ER stress– specific manner, suggesting that it might be the human form of caspase 12 (40). This is certainly encouraging as it will provide new strategies or insights to prevent or ameliorate cisplatin-induced injury to the renal tubular epithelial cells.

Acknowledgments This study is partially supported by a project grant from the Kidney Care Foundation (KCF-98029). An abstract of this study was presented at the 2004 annual meeting of The American Society of Nephrology, St. Louis, MO, November 2004. We are grateful to Dr. F.J. Gonzalez for the KO mice. We thank Dr. S.V. Shah for critique and valuable suggestions. We greatly appreciate Dr. Steven Figure 10. Effect of anti–caspase 12 Ab on caspase cleavage. The A. Bigler and Dr. M. Baliga for assistance in the immunohistochemistry. procaspase 12 was significantly reduced in the cisplatin (Cis)-treated LLC-PK1 cells associated with the presence of cleaved caspase 12, and this cleavage was prevented by pretreatment of the cells with References caspase 12 Ab (A). The induction of the active form of caspase 3 was 1. Lebwohl D, Canetta R: Clinical development of platinum markedly reduced in anti–caspase 12 Ab–transfected cells that were complexes in cancer therapy: An historical perspective and treated with cisplatin (B). There were no significant differences in the an update. Eur J Cancer 34: 1522–1534, 1998 cytochrome c release in the cytosol and the decrease in the procaspase 2. Ries F, Klastersky J: Nephrotoxicity induced by cancer 9 between caspase 12 Ab–transfected cells and the control cells that chemotherapy with special emphasis on cisplatin toxicity. were treated with cisplatin (C and D). Am J Kidney Dis 8: 368–379, 1986 3. Safirstein R, Winston J, Goldstein M, Moel D, Dikman S, Guttenplan J: Cisplatin nephrotoxicity. Am J Kidney Dis 8: chrome c release and the caspase 9 activation in the cells that were 356–367, 1986 transfected with caspase 12 Ab as compared with control cells, 4. Ueda N, Kaushal GP, Shah SV: Apoptotic mechanisms in thus indicating that the changes in the mitochondrial membrane acute renal failure. Am J Med 108: 403–415, 2000 1992 Journal of the American Society of Nephrology J Am Soc Nephrol 16: 1985–1992, 2005

5. Liu H, Baliga R: Cytochrome P450 2E1 null mice provide 24. Rao RV, Peel A, Logvinova A, del Rio G, Hermel E, Yokota T, novel protection against cisplatin-induced nephrotoxicity Goldsmith PC, Ellerby LM, Ellerby HM, Bredesen DE: Cou- and apoptosis. Kidney Int 63: 1687–1696, 2003 pling endoplasmic reticulum stress to the cell death program: 6. Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD: The Role of the ER chaperone GRP78. FEBS Lett 514: 122–128, 2002 release of cytochrome c from mitochondria: A primary site 25. Nakagawa T, Zhu H, Morishima N, Li E, Xu J, Yankner BA, for Bcl-2 regulation of apoptosis. Science 275: 1132–1136, 1997 Yuan J: Caspase-12 mediates endoplasmic-reticulum- 7. Reed JC: Cytochrome c: Can’t live with it—Can’t live with- specific apoptosis and cytotoxicity by amyloid-beta. Nature out it. Cell 91: 559–562, 1997 403: 98–103, 2000 8. Takeda M, Kobayashi M, Shirato I, Osaki T, Endou H: 26. Breckenridge DG, Stojanovic M, Marcellus RC, Shore GC: Cisplatin-induced apoptosis of immortalized mouse prox- Caspase cleavage product of BAP31 induces mitochondrial imal tubule cells is mediated by interleukin-1 beta convert- fission through endoplasmic reticulum calcium signals, ing enzyme (ICE) family of proteases but inhibited by enhancing cytochrome c release to the cytosol. J Cell Biol overexpression of Bcl-2. Arch Toxicol 71: 612–621, 1997 160: 1115–1127, 2003 9. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, 27. Liu H, Baliga M, Bigler SA, Baliga R: Effect of cytochrome Alnemri ES, Wang X: Cytochrome c and dATP-dependent P450 (CYP) 2E1 inhibitors on cisplatin (CP) induced cyto- formation of Apaf-1/caspase-9 complex initiates an apo- toxicity in proximal tubular (LLC-PK1) cells. Anticancer Res ptotic protease cascade. Cell 91: 479–489, 1997 22: 863–868, 2002 10. Finkel T: Oxygen radicals and signaling. Curr Opin Cell Biol 28. Kaushal GP, Kaushal V, Hong X, Shah SV: Role and regula- 10: 248–253, 1998 tion of activation of caspases in cisplatin-induced injury to 11. Rashba-Step J, Cederbaum AI: Generation of reactive oxygen renal tubular epithelial cells. Kidney Int 60: 1726–3617, 2001 intermediates by human liver microsomes in the presence of 29. Park MS, De Leon M, Devarajan P: Cisplatin induces apo- NADPH or NADH. Mol Pharmacol 45: 150–157, 1994 ptosis in LLC-PK1 cells via activation of mitochondrial 12. Bondy SC, Naderi S: Contribution of hepatic cytochrome pathways. J Am Soc Nephrol 13: 858–865, 2002 P450 systems to the generation of reactive oxygen species. 30. Bradd SJ, Dunn MJ: Analysis of membrane proteins by Biochem Pharmacol 48: 155–159, 1994 western blotting and enhanced chemiluminescence. Meth- 13. Baliga R, Zhang Z, Baliga M, Shah SV: Evidence for cyto- ods Mol Biol 19: 211–218, 1993 chrome P-450 as a source of catalytic iron in myoglobinuric 31. Ramachandra S, Studzinski GP: Morphological and bio- acute renal failure. Kidney Int 49: 362–369, 1996 chemical criteria of apoptosis. In: Cell Growth and Apoptosis, 14. Baliga R, Zhang Z, Baliga M, Ueda N, Shah SV: Role of 1st Ed., edited by Studzinski GP, Oxford University Press, cytochrome P-450 as a source of catalytic iron in cisplatin- New York, 1995, pp 119–141 induced nephrotoxicity. Kidney Int 54: 1562–1569, 1998 32. Hsu SM, Raine L, Fanger H: A comparative study of the 15. Thornberry NA, Lazebnik Y: Caspases: Enemies within. peroxidase-antiperoxidase method and an avidin-biotin com- Science 281: 1312–1316, 1998 plex method for studying polypeptide hormones with radio- 16. Fraser A, Evan G: A license to kill. Cell 85: 781–784, 1996 immunoassay antibodies. Am J Clin Pathol 75: 734–738, 1981 17. Chinnaiyan AM, Dixit VM: The cell-death machine. Curr 33. Lee SS, Buters JT, Pineau T, Fernandez-Salguero P, Gonza- Biol 6: 555–562, 1996 18. Morishima N, Nakanishi K, Takenouchi H, Shibata T, Yasu- lez FJ: Role of CYP2E1 in the hepatotoxicity of acetamino- hiko Y: An endoplasmic reticulum stress-specific caspase cas- phen. J Biol Chem 271: 12063–12067, 1996 cade in apoptosis. Cytochrome c-independent activation of 34. Pahl HL: Signal transduction from the endoplasmic retic- caspase-9 by caspase-12. J Biol Chem 277: 34287–34294, 2002 ulum to the cell nucleus. Physiol Rev 79: 683–701, 1999 19. Earnshaw WC, Martins LM, Kaufmann SH: Mammalian 35. Cummings BS, Schnellmann RG: Cisplatin-induced renal caspases: Structure, activation, substrates, and functions cell apoptosis: Caspase 3-dependent and -independent during apoptosis. Annu Rev Biochem 68: 383–424, 1999 pathways. J Pharmacol Exp Ther 302: 8–17, 2002 20. Zou H, Li Y, Liu X, Wang X: An APAF-1.cytochrome c 36. Boulikas T, Vougiouka M: Cisplatin and platinum drugs at multimeric complex is a functional apoptosome that acti- the molecular level. Oncol Rep 10: 1663–1682, 2003 vates procaspase-9. J Biol Chem 274: 11549–11556, 1999 37. Mandic A, Hansson J, Linder S, Shoshan MC: Cisplatin in- 21. Rao RV, Castro-Obregon S, Frankowski H, Schuler M, duces endoplasmic reticulum stress and nucleus-indepen- Stoka V, del Rio G, Bredesen DE, Ellerby HM: Coupling dent apoptotic signaling. J Biol Chem 278: 9100–9106, 2003 endoplasmic reticulum stress to the cell death program. An 38. Ito Y, Pandey P, Mishra N, Kumar S, Narula N, Kharbanda Apaf-1-independent intrinsic pathway. J Biol Chem 277: S, Saxena S, Kufe D: Targeting of the c-Abl tyrosine kinase 21836–21842, 2002 to mitochondria in endoplasmic reticulum stress-induced 22. Hacki J, Egger L, Monney L, Conus S, Rosse T, Fellay I, apoptosis. Mol Cell Biol 21: 6233–6242, 2001 Borner C: Apoptotic crosstalk between the endoplasmic 39. Kowaltowski AJ, Vercesi AE: Mitochondrial damage in- reticulum and mitochondria controlled by Bcl-2. Oncogene duced by conditions of oxidative stress. Free Radic Biol Med 19: 2286–2295, 2000 26: 463–471, 1999 23. Yoneda T, Imaizumi K, Oono K, Yui D, Gomi F, Katayama T, 40. Hitomi J, Katayama T, Eguchi Y, Kudo T, Taniguchi M, Tohyama M: Activation of caspase-12, an endoplastic reticu- Koyama Y, Manabe T, Yamagishi S, Bando Y, Imaizumi K, lum (ER) resident caspase, through tumor necrosis factor Tsujimoto Y, Tohyama M: Involvement of caspase-4 in receptor-associated factor 2-dependent mechanism in re- endoplasmic reticulum stress-induced apoptosis and sponse to the ER stress. J Biol Chem 276: 13935–13940, 2001 Abeta-induced cell death. J Cell Biol 165: 347–356, 2004