Omi/HtrA2 is a mitochondrial proapoptotic serine protease that is able to induce both caspase-dependent and caspase-independent cell death. After apoptotic stimuli, Omi is released to the cytoplasm where it binds and cleaves inhibitor of apoptosis proteins. In this report, we investigated the role of Omi in renal cell death following cisplatin treatment. Using primary mouse proximal tubule cells, as well as established renal cell lines, we show that the level of Omi protein is upregulated after treatment with cisplatin. This upregulation is followed by the release of Omi from mitochondria to the cytoplasm and degradation of XIAP. Reducing the endogenous level of Omi protein using RNA interference renders renal cells resistant to cisplatin-induced cell death. Furthermore, we show that the proteolytic activity of Omi is necessary and essential for cisplatin-induced cell death in this system. When renal cells are treated with Omi's specific inhibitor, ucf-101, they become significantly resistant to cisplatin-induced cell death. Ucf-101 was also able to minimize cisplatin-induced nephrotoxic injury in animals. Our results demonstrate that Omi is a major mediator of cisplatin-induced cell death in renal cells and suggest a way to limit renal injury by specifically inhibiting its proteolytic activity.
- serine protease
- mitochondrial protein
- proximal tubular cells
omi/htra2 is a mitochondrial serine protease originally isolated through its interaction with Mxi2 (11), an alternatively spliced form of the p38 stress-activated kinase (10). Omi is expressed ubiquitously and the amount of protein increases when cells are exposed to heat shock or treated with tunicamycin (14). Omi is the mammalian homolog of the prokaryotic HtrA proteins. HtrAs are chaperones that are essential for bacterial survival after heat shock or oxidative stress (6, 23, 34, 43). In addition to Omi, the family of eukaryotic HtrAs includes two other members: HtrA1/L56 (18, 59) and HtrA3/PRSP (31). Both HtrA1 and HtrA3 are secreted proteins whose normal function is not known. Downregulation or absence of HtrA1 has been reported in some human metastatic melanoma tumors (1).
Recent studies identified Omi as a nuclear encoded mitochondrial serine protease that is released into the cytoplasm on induction of apoptosis (16, 30, 46, 52, 53). In the cytoplasm, Omi activates caspase-9 by interacting with and degrading inhibitor of apoptosis proteins (IAPs) (44, 58). The interaction of Omi with IAPs is mediated via an AVPS NH2-terminal sequence of mature Omi protein that is similar to the corresponding motif in other known IAP-binding proteins including Reaper (54), Hid (55), and Grim (4) in Drosophila, as well as the mammalian proteins Smac/Diablo (9) and caspase-9 (45). Furthermore, Omi can promote apoptosis in a caspase-independent pathway through its ability to function as a protease (53). Omi protein can be defined by three different and functionally distinct domains. An NH2-terminal domain (aa 1–133) carries the mitochondrial targeting sequence and is immediately cleaved after Omi enters the mitochondria (6, 23, 34, 43). The second domain of Omi is a catalytic domain and shows the highest homology with the bacterial HtrAs as well as with the other two eukaryotic homologs (11, 31). A PDZ domain at the COOH terminus of Omi represents the third domain of the protein and has a unique binding specificity (20). Structural and biochemical studies defined the role of this PDZ domain as a regulator of the proteolytic activity of the enzyme (15, 26). The proteolytic activity of Omi is essential for its normal function to promote both caspase-dependent and caspase-independent cell death (16, 53). Recently, we described the isolation and characterization of a specific nonpeptidyl inhibitor of Omi's proteolytic activity (4a). This inhibitor, we call ucf-101, can enter mammalian cells where it colocalizes with endogenous Omi and inhibits its activity (4a). The ability of ucf-101 to specifically inhibit the proteolytic activity of Omi in vitro, as well as in vivo, makes this compound a very useful reagent to delineate the function of Omi in cell injury and apoptosis.
In the present study, we explored the potential role of Omi in renal tubular cells following cisplatin-induced cell death. Cisplatin [cis-diammine dichloroplatinum(II)] is a chemotherapeutic drug used to treat several solid tumors including testicular, lung, head, neck, and cervical cancers (32, 37). Because cisplatin has an inherent dose-dependent toxicity, its use is limited by its nephrotoxicity. The mechanism by which cisplatin causes renal injury is not clear; it can induce apoptosis as well as necrosis, and proximal tubules are particularly sensitive (7, 29). Understanding the mechanism of cisplatin-induced renal cell death could lead to the development of drugs that can be used to protect patients undergoing chemotherapy from developing acute renal failure (ARF).
Our results demonstrate that Omi is expressed throughout the mouse kidney and its subcellular localization is regulated by cisplatin treatment of renal cells. Furthermore, using RNA interference, to reduce the amount of endogenous Omi protein, or the specific inhibitor ucf-101 afforded significant protection on renal cells, both in vitro and in vivo, from the cytotoxic effects of cisplatin.
MATERIALS AND METHODS
Human embryonic kidney (HEK)-293 cells were grown using DMEM (GIBCO, Carlsbad, CA) supplemented with 10% FBS (Hyclone, Logan, UT), 2 mM l-glutamine, 1.5 g/l sodium bicarbonate, 1 mM sodium pyruvate, 50 U/ml penicillin, and 50 μg/ml streptomycin (GIBCO). HK-2 cells were grown in DMEM media supplemented with 10% FBS (Hyclone), 15 mM HEPES, 2 mM l-glutamine, 50 U/ml penicillin, 50 μg/ml streptomycin (GIBCO), 0.4 μg/ml hydrocortisone, 5 μg/ml insulin, and 5 μg/ml apotransferrin (Sigma, St. Louis, MO).
Isolation of mouse kidney proximal tubular cells.
Mouse proximal tubular (MPT) cells were isolated from collagenase-digested fragments derived from the cortices of kidneys of C57BL6 mice as previously described (28, 42). Briefly, kidneys were dissected to obtain cortical tissue, which was digested with a solution of collagenase (Worthington Biochemical, Lakewood, NJ) and soybean trypsin inhibitor (GIBCO) at 37°C for 45 min. Cells were grown in serum-free mixture (1:1) of DMEM and Ham's F-12 containing 15 mM HEPES, 2 mM l-glutamine, 5 μg/ml insulin, 50 nM hydrocortisone, 5 μg/ml apotransferrin, 50 U/ml penicillin, and 50 μg/ml streptomycin. The medium was replaced every 2 days and MPT cells were used after they reached confluence, between day 7 and day 10 after culture.
Confocal microscopy of MPT cells.
For immunofluorescence, MPT cells were grown for 7 days on microscope glass cover slides. Adherent cells were washed in PBS, fixed in 4% paraformaldehyde, and made permeable using ice-cold acetone. Nonspecific binding was blocked with 2% BSA in PBS; cells were then stained using rabbit anti-Omi antibody at room temperature for 2 h. After three washes with PBS, cy3-conjugated anti-rabbit antibody (Jackson ImmunoResearch, West Grove, PA) was added for 1 h. Following three washes with PBS, 100 nM of MitoTracker (Molecular Probes, Eugene, OR) was added for 20 min and the coverslips were placed on microscope slides using Fluoromount-G mounting solution. Slides were observed in an LSM510 confocal laser-scanning microscope (Zeiss).
Mouse tissue immunohistochemistry.
Mouse kidneys were briefly perfused in situ with PBS (0.9% NaCl in 10 mM sodium phosphate buffer, pH 7.4), followed by PLP (2% paraformaldehyde, 70 mM l-lysine, and 10 mM sodium periodate) as fixative (10). Kidney slices were immersed overnight in PBS containing 30% sucrose and then frozen in liquid nitrogen. Semi-thin (1 μm) sections were cut from tissue embedded in LX-112, placed on microscope slides, dried in air, and stored at −20°C. Sections were treated with PBS containing 0.1% SDS for 5 min and incubated for 20 min in PBS containing 2% BSA to reduce nonspecific staining. This was followed by 2-h incubation at room temperature with the various antibodies. Omi antibody was raised against His-Omi134–458 (5) and used at 1:100 dilution; mouse anti-gp330 antibody (22) was used at 1:200 dilution. After three washes with PBS, cy3-conjugated anti-rabbit (Jackson ImmunoResearch) and Oregon-green-conjugated anti-mouse (Molecular Probes) were used at the concentration recommended by the manufacturer. Finally, the sections were washed three times with PBS and a coverslip was placed on the microscope slides using Fluoromount-G. MitoTracker green FM (Molecular Probes) staining was performed following the antibodies staining for 20 min using 100 nM of dye diluted in PBS. Slides were observed using a confocal laser-scanning microscope (Zeiss).
Western blot analysis.
MPT or HK-2 cells were lysed using a Triton X-100-based lysis buffer (1% Triton X-100, 10% glycerol, 150 mM NaCl, 20 mM Tris·HCl, pH 7.5, 2 mM EDTA) in the presence of a protease inhibitor mix (Roche Diagnostics, Indianapolis, IN). Approximately 15 μg of whole cell extract were resuspended in SDS sample buffer and boiled for 3 min. Samples were resolved by SDS-PAGE (10) and electrotransferred onto polyvinylidene difluoride membranes (Pall Life Sciences) using a Semi-Dry cell Transfer Blot (Bio-Rad); 2% nonfat dry milk in TBST buffer was used to block any nonspecific binding. The membrane was incubated with Omi antibody (1:5,000) or XIAP antibody (Chemicon International) (44), followed by a secondary peroxidase-conjugated goat anti-rabbit (Jackson ImmunoResearch) (1:15,000) and visualized by enhanced chemiluminescence (Pierce, Rockford, IL).
To suppress endogenous Omi expression, siRNA oligos were used that specifically target the Omi mRNA. Oligonuleotides were made by Ambion and had the following sequence: Omi-siRNA sense 5′-AAcggcucaggauucgugg-3′ and Omi-siRNA antisense 5′-CCACGAAUCCUGAGCCGUU-3′ corresponding to residues 281–287 of the coding region of human Omi. HEK-293 cells were plated in six-well plates the day before and then were transfected with siRNA (1,000 ng/well) using Oligofectamine reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. After 36 h, transfected cells were exposed to 50 μM cisplatin for 14 h and then processed for either SDS-PAGE and Western blot analysis or stained with annexin V and 7-amino-actinomycin D (7-AAD) (BD Biosciences, San Diego, CA) followed by Flow Cytometry to quantify the percentage of apoptotic and necrotic cells.
Cell death assay and flow cytometry.
Cell death was estimated using annexin V (apoptotic cells) and 7-AAD (necrotic cells) staining (8, 12, 17). Samples were analyzed on a FACSCalibur Flow Cytometer (BD Biosciences). MPT or HK-2 cells were plated in appropriate medium using 12-well plates and when the cells reached 90% confluence they were treated with ucf-101 (25 and 50 μM for MPT, 50 and 70 μM for HK-2) for 20 min followed by cisplatin for 14 h. Cisplatin was used on MPT at concentrations of 25, and 50 μM, whereas on HK-2 cells the concentrations were 25, 50, 70 μM; vehicle alone was used as the control. After 14 h, cells were detached, washed twice with ice-cold PBS, resuspended in 1× binding buffer (BD Biosciences), then stained with PE-conjugated annexin V and 7-AAD according to BD Bioscience protocol.
Cisplatin-induced nephrotoxicity and terminal transferase-mediated dUTP nick end-labeling assay.
All experiments were performed using male BALB/c mice (Charles River Laboratory). Animals were anesthetized with pentobarbital sodium (50 mg/kg ip) on the day of surgery. A miniosmotic pump (1 μl/h; Durect Alzet Osmotic Pumps, Cupertino, CA) filled with 1 mM ucf-101 was implanted subcutaneously into each animal. Twenty-four hours after implantation of the miniosmotic pumps, the mice were treated with 20 mg/kg body weight cisplatin (Sigma) intraperitoneally. Control animals were treated with 0.9% NaCl. To evaluate nephrotoxicity, blood was collected from the retrobulbar vein plexus on day 2 and day 4 after cisplatin treatment and plasma creatinine was measured using a Beckman Coulter Creatinine Analyzer (3, 35).
For terminal transferase-mediated dUTP nick end-labeling (TUNEL) assays, kidneys were perfused via the left ventricle with 30 ml of PBS for 2 min at 37°C and then with PLP (2% paraformaldehyde-75 mM l-lysine-10 mM sodium periodate) fixative. Kidneys were removed from the animals and placed in PLP overnight at 4°C. The kidneys were then washed and stored in PBS containing 0.02% sodium azide at 4°C. For TUNEL staining, hemisected fixed tissue was washed with PBS three times for 5 min each, placed overnight in PBS, embedded in paraffin, and then cut into 4-μm sections using a microtome. Sections were mounted on Fisher Superfrost Plus (Fisher). Sections were stained with TUNEL reagent according to the manufacturer's protocol (Roche Diagnostics). Briefly, the kidney sections were deparaffinized and incubated with TUNEL reagent for 1 h at 37°C. TUNEL-positive cells were counted under a fluorescence microscope (Zeiss). Cell death was also monitored using periodic acid-Schiff (PAS) reagent. Kidney tissues were fixed in methyl Carnoy's solution, embedded in paraffin, sectioned, and stained with PAS reagent following standard procedures (40, 47).
All quantitative data are expressed as means ± SD. Differences among groups were analyzed by one-way ANOVA followed by Tukey's post hoc test. A value of P < 0.05 was considered significant.
Localization of Omi in mouse kidney mitochondria.
Immunohistochemical staining of frozen mouse kidney sections using Omi-specific antibodies was performed to characterize the expression of the protein as well as its subcellular localization. To identify proximal tubules, kidney sections were stained with anti-gp330 antibodies (13, 22, 36). Figure 1 shows Omi is expressed throughout the kidney including the proximal tubules (Fig. 1, top right). The subcellular localization of Omi was investigated using MitoTracker, a mitochondrial-specific dye (Fig. 2, top left). Omi staining of the mouse kidney sections showed extensive colocalization with MitoTracker staining, suggesting that the protein is predominantly present in the mitochondria (Fig. 1, bottom left).
Subcellular localization of Omi in MPT cells is regulated by chemical stress.
The subcellular localization of Omi in MPT cells was investigated using Omi antibody and MitoTracker (Molecular Probes). Figure 3A shows the distinct punctate staining of Omi characteristic of mitochondrial staining that colocalizes with MitoTracker. The predominant presence of Omi in the mitochondria of MPT cells is consistent with previous reports (16, 46). Figure 3B shows MPT cells that were treated with 20 μM cisplatin (apoptotic stress). The intensity of Omi-specific staining increases and becomes more diffuse as the protein is released to the cytoplasm.
Induction of Omi protein in renal cells after cisplatin treatment.
To monitor the level of Omi protein during induction of apoptosis, MPT cells were treated with cisplatin and cell extracts were prepared at various time intervals after treatment. The amount of Omi protein was analyzed by SDS-PAGE and Western blot analysis using specific antibodies. Figure 4A, lane 1, shows the level of Omi protein is low in resting MPT cells. After cisplatin treatment, there is a significant increase in the amount of Omi protein (lanes 2-5). Omi's protein induction reaches a peak at 50 μM cisplatin treatment (lane 4). We also used HK-2 cells, a human proximal cell line (38), in a similar experiment. HK-2 cells have a higher basal level of Omi protein than MPT cells (Fig. 4B, lane 1). After treatment with cisplatin, there is also induction of Omi in HK-2 cells (Fig. 4B, lanes 2-4).
Use of RNA interference to reduce the endogenous protein level of Omi.
Because cisplatin-induced apoptosis in renal cells coincides with induction of Omi protein, we investigated whether upregulation of Omi protein is necessary for apoptosis induced by this agent. For these experiments, we used RNA interference and HEK-293 cells. These cells were used because both MPT and HK-2 cells are highly resistant to transfection. After transfecting HEK-293 cells with Omi-specific siRNA, cell extracts were prepared and the endogenous level of Omi protein was monitored by SDS-PAGE and Western blot analysis. Figure 5A shows that transfection of Omi-specific siRNA substantially reduces the level of Omi protein. The same cells were also treated with two different concentrations of cisplatin followed by annexin V+7-AAD staining and analyzed by FACS. The percentage of apoptotic and necrotic cells after cisplatin treatment was significantly lower in the population of the HEK-293 cells, where Omi levels had been reduced with the use of siRNA (Fig. 5B).
Ucf-101 inhibitor protects renal cells from cisplatin-induced cell death.
We used the specific inhibitor ucf-101 to block the protease activity of Omi in MPT as well as HK-2 cells. The cells were treated with two different concentrations of ucf-101, followed by cisplatin treatment. Cells were stained with annexin V (apoptotic cells) and 7-AAD (necrotic cells) and then analyzed by FACS. Figure 6A showed a gradual increase in the number of MPT cells undergoing cell death as the concentration of cisplatin increased from 25 to 50 μM. In the presence of ucf-101, the number of apoptotic and necrotic cells was significantly reduced, suggesting that ucf-101 affords protection on MPT cells from cisplatin-induced cell death. We also used HK-2 cells in a similar experiment to investigate whether ucf-101 can also protect them from cisplatin-induced cell death. These cells were treated with higher concentrations of cisplatin (50 and 70 μM) than the MPT cells, resulting in substantial apoptosis as well as necrosis. In the presence of ucf-101, the percentage of apoptotic and necrotic HK-2 cells was significantly reduced after cisplatin treatment (Fig. 6B).
Ucf-101 inhibits cisplatin-induced XIAP degradation.
HK-2 cells were treated with 25 and 50 μM of cisplatin in the presence or absence of ucf-101 as described above. Cell extracts were prepared (see materials and methods), and ∼20 μg of total cell lysates were analyzed by SDS-PAGE and Western blot using XIAP antibodies (BD Biosciences). Figure 7 shows the level of XIAP is substantially reduced in HK-2 cells treated with 25 or 50 μM cisplatin (lanes 2 and 3). When used together with 50 μM cisplatin, ucf-101 (70 μM) was able to block XIAP degradation (lane 4).
Effect of ucf-101 on renal function of mice treated with cisplatin.
We treated mice with ucf-101 to assess any effect it might have in minimizing renal damage induced by cisplatin. Because the pharmacokinetic properties of ucf-101 are unknown, we decided to use miniosmotic pumps implanted subcutaneously in the animals to provide a steady and continuous release of the drug. Plasma creatinine levels were used as an indicator of nephrotoxic injury (35, 48). The results of these experiments show that mice treated with cisplatin and ucf-101 were more resistant to nephrotoxicy than animals treated with cisplatin alone (Fig. 8). The protection of renal function by ucf-101 was more pronounced 4 days after cisplatin treatment. Creatinine levels were 50% lower in mice treated with ucf-101 than in the control animals (Fig. 8). More detailed experiments to investigate the optimum dose of ucf-101, as well as the best mode to introduce this drug into the animals, are currently in progress in our laboratory.
We also used the TUNEL assay to quantify apoptosis and necrosis in fixed kidney tissues from the animals used in the experiments described above. Figure 9 shows TUNEL staining of kidney sections from animals treated with vehicle (a) or ucf-101 alone (c). Vehicle or ucf-101 alone, at the concentration used, had no effect in the kidneys of animals. When cisplatin is used, many TUNEL-positive cells are clearly seen in the kidney sections consistent with its cytotoxic effect (Fig. 9b). When cisplatin and ucf-101 were used together, the number of TUNEL-positive cells is significantly reduced (Fig. 9d). The same tissue sections were also stained with PAS and counterstained with hematoxylin (40, 47). Death cells were identified by their dense nuclei and loss of cytoplasm. Figure 9e shows PAS staining after ucf-101 and cisplatin treatments; Fig. 9f shows kidney sections treated with cisplatin alone. Figure 9B presents a quantitative analysis of these results.
Renal tubular cells die by both apoptosis and necrosis. The relative contribution of each of these two forms of cell death to the cell damage that follows experimentally induced ischemia or toxic injury is not very clear (41, 49–51). In this report, we investigated the potential role of Omi/HtrA2 in renal cell death. Omi is a recently described mitochondrial serine protease that is able to induce caspase-dependent as well as caspase-independent cell death (46, 53). On induction of apoptosis, Omi translocates to the cytoplasm where it binds and cleaves IAP proteins, relieving their inhibitory effect on caspases (16, 30). The protease activity of Omi is central to its function; it is necessary for its processing to a mature protein as well as the degradation of IAPs (44, 58). A specific inhibitor of Omi has been isolated and characterized in our laboratory. This nonpeptidyl molecule, ucf-101, is able to inhibit the protease activity of Omi in vitro and in vivo (4a). Furthermore, ucf-101 can easily enter mammalian cells, which makes it very useful for physiological studies of apoptosis.
Cisplatin is a chemotherapeutic agent used to treat various solid tumors. A side effect of this drug is its nephrotoxicity. Cisplatin can cause renal proximal tubular cell apoptosis at low concentration and necrosis at higher concentration (21, 25, 29). The mechanism by which cisplatin causes apoptosis is not yet clear, although it probably includes DNA damage leading to activation of p53, oxidative stress, or changes in signal transduction (7, 21, 24, 57). Although Omi has been identified as a downstream target of p53 (19), its potential involvement in cisplatin-induced apoptosis in renal cells has never been investigated. Our results show that Omi is expressed in the proximal tubule cells of kidneys, an area that sustains most damage during ischemia or toxic insults (27, 33, 39, 56).
In our experiments, we found cisplatin induced expression of Omi protein in a dose-dependent manner in both primary MPT cells and an established human proximal tubular cell line. The increased expression of Omi also coincided with the translocation of the protein from mitochondria to the cytoplasm. We also show that Omi protein is necessary for cisplatin-induced renal cell apoptosis, because reduction in the protein level using RNA interference minimized the cell death of HEK-293 cells. When MPT cells or HK-2 cells were treated with the ucf-101 inhibitor before cisplatin treatment, the percentage of apoptotic cells was dramatically reduced. This protective effect of ucf-101 was specific because it correlated with the ability of this drug to inhibit cisplatin-induced XIAP degradation. XIAP is a bona fide substrate of Omi and its degradation leads to activation of caspase-3 and -9 resulting in cell death (16, 17). Our results are in accord with a recent study that showed ucf-101 protected human neutrophils from TNF-α-induced apoptosis by inhibiting Omi's proteolytic activity (2). We also investigated whether ucf-101 can protect renal cells when administered into mice together with cisplatin. These experiments performed in animals clearly showed that ucf-101 provided significant protection of renal cells from the toxic effects of cisplatin. The protective effect of ucf-101, when used in animals, might not be limited to renal function. A recent study (29a) shows ucf-101 was able to protect cardiac myocytes from ischemia-reperfusion-induced apoptosis. The antiapoptotic ability of ucf-101 relies entirely on its ability to function as a specific Omi inhibitor. The possibility that ucf-101 might also have a nonspecific activity against some other uncharacterized protease or kinase cannot be excluded. Results from experiments using several ucf-101 analogs (4a) suggest the in vivo antiapoptotic property of the ucf compounds are proportional to their effectiveness as specific inhibitors of Omi's proteolytic activity in vitro (results not shown).
Our studies suggest that Omi serine protease plays a significant role in the cisplatin-induced cell death of renal cells. Furthermore, the proteolytic activity of Omi is necessary and essential for its proapoptotic function in this system. By inhibiting Omi's proteolytic activity using the ucf-101 specific inhibitor, renal cells become more resistant to the toxic effects of cisplatin.
This work was supported by the National Institutes of Health Grant R01-DK-55734–01 (to A. S. Zervos).
We thank members of the Zervos lab for comments and suggestions.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society