Vibrio vulnificus cytolysin induces superoxide anion-initiated apoptotic signaling pathway in human ECV304 cells.

Previous studies showed that exposure to Vibrio vulnificus cytolysin (VVC) caused characteristic morphologic changes and dysfunction of vascular structures in lung. VVC showed cytotoxicity for mammalian cells in culture and acted as a vascular permeability factor. In this study, the underlying mechanisms of VVC-induced cytotoxicity was investigated on ECV304 cell, a human vascular endothelial cell line. When cells were exposed to 0.4 hemolytic units (HU) of VVC, consecutive apoptotic events were observed; the elevation of superoxide anion (O (-.)(2)), the release of cytochrome c, the activation of caspase-3, the cleavage of poly(ADP-ribose) polymerase, and the DNA fragmentation. The pretreatment with 4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO), O(-.) 2) scavenger, completely abolished O(-.)(2) levels and downstream apoptotic events. Moreover, pretreatment with cyclosporin A (CsA), a mitochondrial permeability transition inhibitor, was capable of attenuating O(-.)(2)-mediated cytochrome c release and caspase-3 activation, and consequent apoptosis. Apoptosis, as demonstrated by oligonucleosomal DNA fragmentation and fluorescence microscopy, was induced 24 h after VVC treatment, which was also prevented by caspase-3 inhibitor, Ac-DEVD-CHO. Caspase-1 inhibitor, Ac-YVAD-CHO, did not protect ECV 304 cells from apoptosis. These results suggest a scenario where VVC-induced apoptosis is triggered by the generation of O(-.)(2), release of cytochrome c from mitochondria, activation of caspase-3, degradation of poly(ADP-ribose) polymerase, and DNA fragmentation. The induction of apoptosis in endothelial cells by VVC may provide a pivotal mechanism for understanding the pathophysiology of septicemia.

Vibrio vulnificus, previously designated as a lactose-positive vibrio, is a halophilic bacterium which inhabits marine or estuarine areas. V. vulnificus causes wound infections and septicemia when he comes into contact with seawater or when contaminated seafood is consumed (1,2). A number of factors have been implicated as contributing to disease caused by V.
V. vulnificus cytolysin (VVC) 1 is a water-soluble polypeptide with a M r of 51,000 (9,10). VVC is extremely toxic, and even a submicrogram amounts cytolysin is fatal to mice when injected intravenously. We already reported that hemolysis caused by VVC is colloid-osmotic in nature and those cytolysins, after binding to membranes, were oligomerized to form small pores on cell membranes (11,12).
Apoptosis is a physiological cell death and characterized by chromatin condensation, membrane blebbing, cell shrinkage, and DNA fragmentation (13). Although the morphologic appearance of apoptotic cells is well described, the signaling pathway or pathways leading to it are not yet fully understood. Apoptosis can be initiated by oxidative stress mediated by the generation of reactive oxygen intermediates (14). This hypothesis arose from several observations that linked oxidative stress to apoptosis. These include the induction of oxidative stress by apoptotic stimuli such as tumor necrosis factor-␣ (15); apoptosis inhibition by antioxidants such as N-acetylcysteine (16), catalase (17), and 3,3,5,5-tetramethylpyrroline N-oxide (TEMPO) (18); and the direct induction of apoptosis by hydrogen peroxide (19).
Mitochondria are major sources of superoxide anion (O 2 . ), which is generated during mitochondrial respiration (20). O 2 . that is generated in excess acts as mediators of the apoptotic signal pathway. It seems that various proapoptotic stimuli provoke alterations of the permeability of the mitochondrial outer membrane that permits release of apoptotic proteins such as cytochrome c and apoptosis inducing factor (21,22). The released cytochrome c activates caspase, a family of cysteine proteases and inhibition of apoptosis by Bcl-2 or cyclosporin A (CsA) may be mediated by blocking the release of cytochrome c (23)(24)(25). Caspase is synthesized as catalytically inactive proenzymes comprising a large and small subunit with a variable length amino-terminal prodomain. On activation, the prodomain is lost by catalytic cleavage of carboxyl-terminal at an aspartate residue, with heterodimerization of the large and small subunits to form the active enzyme (26). Once activated, caspase-3 cleaves many substrate proteins including poly(ADP-ribose) polymerase (PARP) (27), lamin (28), DNA fragmentation factor (DFF/ICAD) (29), and gelsolin (30).
Although some studies on the action of the VVC against mammalian cells such as red blood cells (11,31,32) and mast cells (33,34) have been carried out, little is known about the relationship between its lethal activity and cytotoxic mechanism involving pore formation. It has been reported that VVC has lethal activity by increasing vascular permeability and neutrophil sequestration in the lungs of mice (35). Kim (36) already reported that VVC had a cytotoxic effect in pulmonary endothelial cells, which was associated with the formation of small transmembrane pores.
The purpose of this study was to examine whether VVC at lower concentrations induces apoptosis in endothelial cells. Our data suggests that elevated O 2 . levels in ECV304 cells following VVC exposure act as a signal to trigger the release of cytochrome c from mitochondria. We also provide evidence that cytochrome c acts at the upstream of caspases in inducing apoptosis.

EXPERIMENTAL PROCEDURES
Culture Conditions-The human endothelial cell line ECV304 was purchased from American Type Culture Collection. Cells were placed into 75-cm 2 tissue culture flasks and grown at 37°C under a humidified, 5% CO 2 atmosphere in Dulbecco's modified Eagle's medium (Life Technologies, Inc., Grand Island, NY) supplemented with 10% fetal bovine serum and 2 mM glutamine, 10,000 units/ml penicillin, 10 mg/ml streptomycin, and 2.5 g/ml amphotericin B.
Bacterial Strain and Culture-A virulent strain of V. vulnificus MO6 -24/O was kindly donated by J. G. Morris, Jr. (Veterans Affairs Medical Center and Division of Infectious Disease, University of Maryland, School of Medicine, Baltimore, MD). V. vulnificus was cultured in heart infusion diffusate broth (Life Technologies, Inc., Gaithersburg, MD) at 37°C for 4 h as described by Kreger et al. (37).
Purification of VVC-The cytolysin was purified to homogeneity from the culture supernatant by ammonium sulfate fractionation, DEAEcellulose chromatography, and phenyl-Sepharose CL-4B chromatography as described by Kim et al. (34). The purified cytolysin had a specific activity of 81,000 HU/mg of protein with 30% recovery. The hemolytic activity of cytolysin against mouse erythrocytes was determined by the method of Bernheimer and Schwartz (38).
Cell Viability Assay-General viability of cultured cells was determined by reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to formazan (39). After 24 h incubation with VVC, cells (10 5 cells/well) in 96-well plates were washed twice with phosphate-buffered saline. MTT (100 g/0.1 ml phosphate-buffered saline) was added to each well. Cells were incubated at 37°C for 1 h, and 100 l of dimethyl sulfoxide was added to dissolve the formazan crystals. Detection of Cytochrome c Release-The release of mitochondrial cytochrome c was determined by Western blot (42). Briefly, at the end of various designated treatments, cells (1.5 ϫ 10 7 cells) were collected with trypsin, washed with phosphate-buffered saline, and resuspended in ice-cold homogenizing buffer (250 mM sucrose, 20 mM Hepes-KOH (pH 7.5), 10 mM KCl, 1.5 mM MgCl 2 , 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, and 1 g/ml leupeptin). After 30 min incubation on ice, cells were homogenized with a glass Dounce homogenizer (30 strokes). The homogenate was subjected to a series of centrifugations (1,000 ϫ g for 10 min, then 10,000 ϫ g for 20 min, and finally 100,000 ϫ g for another 60 min, 4°C) for the collection of the cytosolic fraction. Fifty g of protein was loaded on 15% SDS gel. After electrophoretic separation, the proteins were transferred to nitrocellulose membrane (Millipore, Bedford, MA) using a semi-dry blotting apparatus (Bio-Rad, Munich, Germany), and the blot was incubated with mouse anticytochrome c antibody (Pharmingen, San Diego, CA), followed by reaction with alkaline phosphataseconjugated secondary antibody.
Caspase Activity Assay-After treatment with VVC, cells (5 ϫ 10 6 ) were washed with ice-cold phosphate-buffered saline and lysed in Triton X-100 buffer (0.5% Triton X-100, 10 mM EDTA, and 10 mM Tris-HCl, pH 7.5) for 30 min on ice. Cell lysates were mixed with caspase assay buffer (10% glycerol, 2 mM dithiothreitol, and 20 mM HEPES, pH 7.5) containing caspase substrate 20 M Ac-DEVD-AFC (Pharmingen Inc., San Diego, CA), and incubated for 1 h at 37°C. Enzyme catalyzed release of AFC was monitored using a spectrofluorometer with an excitation wavelength of 400 nm and an emission wavelength of 505 nm.
Western Blot Analysis of PARP Cleavage-Cytosolic extracts were separated by gel electrophoresis on a 10% reducing SDS-polyacrylamide gel. Subsequently the proteins were transferred onto a nitrocellulose membrane. Prior to incubation with PARP antibodies (Transduction Lab, Lexington, KY), the membrane was blocked with 2% bovine serum albumin for 30 min. After washing, the proteins were detected with an alkaline phosphatase-coupled secondary antibody.
Fluorescent Staining of Cells with H-33342-Apoptotic cells are characterized by unique morphological changes within their nuclei, consisting of chromatin condensation and nuclear fragmentation. These changes can also be identified using the fluorescent DNA-binding dye H-33342 (bisbenzamide, Calbiochem, San Diego, CA), which stains the nuclei of normal cells as well as those that have died by apoptosis (43,44). Briefly, adherent cells were stained with H-33342 (1 g/ml) for 10 min at 37°C. Cells were then washed with phosphate-buffered saline twice and fixed with 3% paraformaldehyde. Cells were observed and photographed using fluorescence microsopy.
Protein Determination-Protein concentration in the ECV304 cell cytosol was determined by the method of Bradford (45) with bovine serum albumin as the standard. All of the samples were assayed in triplicate.
Statistical Analysis-Statistical analysis of the data was performed with Student's t test and ANOVA. Differences with p Ͻ 0.05 were considered statistically significant.

Reduction of Cell
Survival by VVC-MTT conversion assay was used to determine the viability of ECV304 cells exposed to VVC. As shown in Fig. 1 Fig. 3, Fig. 5, lane 3 in Fig. 6, and Fig. 7C). VVC-induced Cytochrome c Release-Mitochondrial apoptosis is accompanied by an efflux of cytochrome c from the intermembrane space of mitochondria into the cytosolic compartment (24,25). As shown in Fig. 3 . is responsible for VVC-mediated cytochrome c release. Activation of Caspase-3 during VVC-induced Apoptosis-Several studies have shown that the release of mitochondrial cytochrome c triggers activation of caspase-3 (23-25). We therefore determined caspase activity using fluorogenic caspase-3 substrate, Ac-DEVD-AFC. As shown in Fig. 4A, VVC caused slow activation of caspase-3 during the first 3 h but increased 3-4-fold during the next 3 h and remained at that level by 12 h. This protease activity increased dose dependently and reached maximal at 0.4 HU (Fig. 4B). However, caspase-1 activity was not increased by VVC (data now shown). Consistent with O 2 . production and cytochrome c release, TEMPO, CsA, and Ac-DEVD-CHO almost completely abolished the activation of caspase-3 (Fig. 5A). These results were confirmed by Western blotting of PARP cleavage (Fig. 5B). PARP is a well known endogenous substrate of caspase-3 and its inactivation by caspase-3 mediated cleavage prevents futile DNA repair cycle (27). Evident cleavage of PARP was found in those cells treated with 0.4 HU for 6 h. The pattern of PARP cleavage was found to be very consistent with that of caspase-3 activity. Again, no cleavage of PARP was observed when cells were treated with TEMPO, CsA, and Ac-DEVD-CHO. VVC-induced PARP cleavage was not affected by Ac-YVAD-CHO, caspase-1 inhibitor, implicating a specific role of caspase-3 in VVC-induced apoptosis.
DNA Fragmentation and Morphological Change in ECV 304 Cells Exposed to VVC-To investigate whether ECV304 cells undergo apotosis after treatment with VVC, internucleosomal DNA fragmentation was determined by agarose gel electrophoresis. DNA extract from ECV304 cells treated with 0.4 HU of VVC for 24 h generated a characteristic ladder pattern of discontinuous DNA fragments on agarose gel electrophoresis (Fig. 6). VVC-induced apoptosis was further confirmed using fluorescence microscopy, which is based on the fact that the nucleus of apoptotic cells stained with a high-intensity than that of normal cells (46 -48). Normal nuclei stained with H-33342 fluorescence faintly and demonstrated a delicate chro- matin structure (Fig. 7A). The nuclei of VVC-exposed cell fluorescence more intensely and also demonstrated two classic features of apoptosis: chromatin condensation and fragmentation (Fig. 7B). The increased intensity under these conditions is believed to be both the result of chromatin condensation as well as of an increased rate of influx of H-33342 into apoptotic cells. Preincubation with TEMPO and CsA completely inhibited VVC-induced apoptosis (lanes 3 and 4 in Fig. 6, and C and D in Fig. 7) as Ac-DEVD-CHO did (lane 5 in Fig. 6, and E in Fig. 7). In contrast, Ac-YVAD-CHO did not prevent VVC-induced apoptosis, suggesting that VVC-induced apoptosis was caspase-3 specific. At higher concentrations (more than 2 HU), VVC induced necrosis rather than apoptosis (data now shown precedes the release of cytochrome c and subsequently the activation of caspase-3, and plays a crucial role in the apoptotic process. This was supported by the kinetics of VVC-induced apoptosis that the maximal production of O 2 . , the release of cytochrome c, and the peak activity of caspase-3 were observed after 5 min, 3 h, and 6 h after VVC treatment, respectively. Several biochemical changes were present in VVC-treated ECV304 cells with the appearance of the apoptotic phenotype. Even though DNA and membrane changes strongly suggest an apoptotic death, many researchers consider caspase activation as an even more reliable hallmark of apoptosis (49). During VVC-induced cell death, caspase-3 activation seems to be critical to mediate apoptosis. Using both the DNA fragmentation assay and morphological study, it was observed that ECV304 cell apoptosis could be prevented by inhibition of caspase activity with caspase-3 inhibitor, but not with caspase-1 inhibitor. Fluorogenic substrates also showed increased caspase-3 activity in time and dose dependent manners, but no change in caspase-1 activity. Activation of caspase leads to the cleavage of several cellular substrates, one of which is PARP. Treatment of ECV 304 cells with VVC induced proteolytic cleavage of PARP, which was reversed by pretreatment with caspase-3 inhibitor. Although our findings point to a critical role for caspase-3 in ECV 304 cell apoptosis by VVC, they do not rule out the possibility that another caspase family protease also plays an important role in cell death. Actually, caspase-6, caspase-7, and caspase-8 may be inhibited by Ac-DEVD-CHO.
Certainly, our data does not exclude the possibility that other regulators of the apoptotic signaling cascade may exert a role similar to that of caspase-3 for the execution of the apoptotic program by VVC. It is known that pore-forming proteins, perforins, and ionophores contribute to apoptosis induction through the marked alterations in calcium levels (50 -52) or by affecting directly or indirectly cellular metabolic processes such as ATP levels and mitochondrial function (53,54). At higher concentrations (more than 2 HU), VVC induced necrosis. 2 VVC caused calcium influx from outside and reduced cellular ATP levels in those necrotic conditions. Therefore, VVC-mediated calcium influx may trigger necrotic pathway rather than apoptotic signaling events.
Cytotoxicity toward nucleated cells has been demonstrated for other pore-forming cytolysins such as staphylococcal ␣-toxin (55), streptolysin-O (56), and Vibrio cholerae E1 Tor cytolysin (57). Among them, staphylococcal ␣-toxin was reported to induce apoptosis in human T lymphocytes with low doses (58). On the basis of permeability studies, those authors associated the induction of apoptosis with the formation of very small pores on the plasma membrane, but they did not explain the molecular pathway leading to apoptosis.
Until now, little is known about the cytotoxic mechanism of VVC on mammalian cells. In the present study, we showed for the first time that endothelial cell damage mediated by VVC occurred through apoptosis. Our previous study showed that increased pulmonary vascular permeability and neutrophil sequestration in lung tissue are important in the lethal activity of cytolysin injected intravenously (35). Thus, VVC secreted during V. vulnificus infection may participate in increasing vascular permeability and subsequent lung damage. We have no data as to whether or not VVC-induced apoptosis also occurs in vivo. However, considering the fact that tissue necrosis is a clinical feature characterized primarily in severe V. vulnificus infection, the presence of apoptotic process in endothelial cells may give us important clues in understanding the pathophysiology of V. vulnificus infection.
Taken together, these data suggest that O 2 . generation coupled with the release of cytochrome c could be a strong amplification signal for the efficient activation of the caspase cascade during VVC-induced apoptosis. Additional studies are needed to investigate the implication of VVC in the triggering of endothelial cell apoptosis during infection with V. vulnificusin in vivo.