Role of reactive oxygen species in cis-dichlorodiammineplatinum-induced cytotoxicity on bladder cancer cells.

This study was undertaken to investigate the intracellular induction of reactive oxygen species (ROS) by cis-dichlorodiammineplatinum (CDDP) and the augmentation of their cytotoxicity in bladder cancer cells (KU7) by enhancement of ROS generation by the glutathione (GSH) depletors buthionine sulphoximine (BSO) and diethylmaleate (DEM). CDDP-induced cytotoxicity in KU7 cells and its modulation by GSH depletors were determined using spectrophotometric measurement with crystal violet staining. The effects of GSH depletors on intracellular GSH levels were confirmed using the GSH reductase-DTNB recycling method. Intracellular ROS generation induced by CDDP with or without GSH depletors was estimated from the amount of intracellular dichlorofluorescein (DCF), an oxidized product of dichlorofluorescein (DCFH), which was measured with an anchored cell analysis and sorting system. The cytotoxic effects of CDDP (IC50 15.0 +/- 2.5 microM) were significantly enhanced by BSO (IC50 9.3 +/- 2.6 microM, P < 0.01) and DEM (IC50 10.3 +/- 0.3 microM, P <0.01). BSO and DEM produced a significant depletion in intracellular GSH levels (9.6 +/- 0.4 nmol 10(-6) cells, 17.9 +/- 1.0 nmol 10(-6) cells) compared with the controls (30.5 +/- 0.6 nmol 10(-6) cells). Intracellular DCF production in KU7 cells treated with CDDP (1.35 +/- 0.33 microM) was significantly enhanced by the addition of BSO (4.43 +/- 0.33 microM) or DEM (3.12 +/- 0.22 microM) at 150 min. These results suggest that ROS may play a substantial role in CDDP-induced cytotoxicity and that GSH depletors augment its cytotoxicity through an enhancement of ROS generation in bladder cancer cells.

Reactive oxygen species (ROS) have been reported by Fridovich (1978) to cause cell damage, and many other reports have diversely demonstrated a clear relationship between ROS and tissue damage. ROS-mediated cell death may be important in the pathogenesis of several neurodegenerative disorders (Coyle and Puttfarrcken, 1993;Busciglio and Yankner, 1995). Moreover, oxidative tissue injury has been observed to be central in reperfusion injury of ischaemic tissues in myocardium, kidney and brain (Bulkley, 1983). Indeed, reperfused tissues have been protected in ischaemic diseases including renal transplantation and myocardial ischaemia by inhibitors of ROS (McCord, 1985). In the cytotoxicity of effective anti-tumour agents including adriamycin (ADR) (Meijer et al, 1987), bleomycin (Sausville et al, 1978) and tumour necrosis factor (TNF) (Yamauchi et al, 1989), ROS have been noted as playing a contributing role.
CDDP is an anti-tumour agent effective for treating various human cancers of the brain, head and neck, stomach, ovary, testis and bladder. Its anti-tumour activity is attributed primarily to its ability to form DNA-platinum cross-linked adducts (Zwelling et al, 1979). In addition, CDDP is reported to cause apoptosis (Evans and Dive, 1993). On the other hand, CDDP-induced nephrotoxicity (Sugihara et al, 1987) and emesis ) -the resulting major side effectsare also believed to be related to ROS generation. Uslu and Bonavida (1996) have suggested that the synergistic cytotoxic effects of TNF-a and CDDP on ovarian cancer cells are associated with ROS induction.
The mechanism of CDDP-induced cytotoxicity, however, has not been fully elucidated. The present study was therefore undertaken to investigate a role of ROS in the CDDP-induced cytotoxicity on bladder cancer cells.
Cell line KU7 cells, which are an established bladder cancer cell line from a patient with transitional cell carcinoma of the bladder (Shibayama et al, 1991), were incubated in a monolayer cell culture and maintained in 75-cm2 Corning culture flasks (Tokyo, Japan) containing RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum and carbenicillin sodium in a humidified atmosphere of 5% carbondioxide-95% air at 37°C. The culture medium was renewed every 3 days.

Cytotoxicity assay
Cytotoxicity was determined as reported previously (Nakashima et al, 1991) with a modification of the photometric procedure described by Nedwin et al (1985). Aliquots of KU7 cells (1 x 104 per well) were seeded in a total volume of 100 gl of medium, in a 96-multiwell plate with round bottoms (Corning), and the experiments were performed under the same conditions as described previously. Following a 24-h preincubation, the cells were exposed to CDDP in the concentration range of 3.3-53.3 giM for 24 h. To assess the effect of BSO or DEM on CDDP-induced cytotoxicity, the cells were preincubated with 9.3 mm BSO for 4 h or with 0.2 mm DEM for 4 h. After the preincubation period, the supematant was discarded and medium containing CDDP, in the concentration range described above, was then added to the wells. After incubation, the plates were washed and cytotoxicity was determined by staining the plates with 0.2% crystal violet (in 2% ethanol). The absorbance value of each well was determined at 550 nm with a 405-nm reference beam by a microplate reader (Bio-Rad, Tokyo, Japan). A linear relationship (r = 0.953) was seen between the absorbance value and the cell number excluding trypan blue. Cell growth was expressed as the percentage of absorbance value compared with controls.

Measurement of intracellular glutathione (GSH)
KU7 cells were seeded in culture dishes and exposed to GSH depletors (BSO, 9.3 mm; DEM, 0.2 mM) for different periods (0, 4, 8, 12 h). After exposure to GSH depletors, the supernatant was removed and the cells were detached with trypsin. The cells were collected by centrifugation and washed with PBS. This cell pellet was suspended in PBS (1 x 106 cells ml-1). The cells were homogenized using a Polytron homogenizer and centrifuged at 4°C (5000 g, 30 min). The supematant was then harvested and the GSH concentration was determined with the glutathione reductase-DTNB recycling method, as reported previously (Tietze, 1969). The supernatant (1.5 ml) was incubated with 40 ,l of DTNB at a concentration of 3.8 mm for 2 min. An aliquot (200 gl) of glutathione reductase (6 units ml-') and 100 pl of NADPH (5.4 mM) were then added to the mixture, and the concentration of GSH was estimated by using spectrophotometric measurement.

Measurement of intracellular reactive oxygen species
To determine the net intracellular levels of ROS generated by CDDP, we used DCFH-DA, which is permeable in cells and interacts with intracellular ROS, to generate fluorescent DCF as reported previously (Bass et al, 1983;Cathcart et al, 1983). DCFH-DA is a stable non-fluorescent product that is activated to non-fluorescent DCFH by alkaline hydrolysis penetrating the cell membrane. DCFH is oxidized rapidly to highly fluorescent DCF by ROS.
The relationship between fluorescence intensity and the various concentrations of DCF was determined in each experiment. The amount of DCF produced was then calculated from the fluorescence intensity, and the amount of ROS was subsequently estimated from DCF production. Fluorescence intensity in the cells was determined by using an anchored cell analysis and sorting system (ACAS 570, Meridian Instruments, Okemos, MI, USA) with an excitation wavelength of 488 nm and an emission wavelength of 550 nm. KU7 cells were seeded in glass-bottomed DCF production in KU7 cells treated with CDDP (16.7 gM) were significantly higher than those of controls (P < 0.05

Cytotoxicity assay
CDDP induced cytotoxic effects in KU7 cells in a dose-dependent manner. The effects of BSO or DEM in CDDP-induced cytotoxicity were assessed (Figure 1). BSO at a concentration of 18.6 mM or higher and DEM at a concentration of 0.4 mm or higher demonstrated cytotoxic effects on KU7 cells after 4 h. However, BSO (9.3 mM) or DEM (0.2 mM) showed no cytotoxic effects on KU7 cells after 4 h. We then used BSO and DEM at these concentrations. The IC50 of CDDP in the presence of BSO (9.3 mM) was 9.3 ± 2.6 JIM (n = 5), which was significantly lower than that of CDDP alone (15.0 ± 2.5 JIM, n = 12, P < 0.01). The IC50 of CDDP  Figure 4 Effects of GSH depletors on CDDP-induced DCF production. BSO (9.3 mM) and DEM (0.2 mM) both produced a significant enhancement in DCF production (P < 0.05). Each value represents the mean ± s.d.
*Significantly different from CDDP alone (DEM). tSignificantly different from CDDP alone (BSO) in the presence of DEM (0.2 mM) was 10.3 ± 0.3 ,UM (n = 7), which was significantly lower than that of controls (P < 0.01). These data indicate that these GSH depletors significantly enhanced CDDPinduced cytotoxicity.

Measurement of intracellular ROS generation
In the preliminary examination, hydrogen peroxide (30 ,UM) caused a significant increase in DCF production ( Figure 2). CDDP produced a significant increase in DCF production in a timedependent manner. Intracellular DCF production induced by CDDP (16.7 gIM) for 150 min was 1.35 ± 0.05 gM (P < 0.01), which was significantly higher than that of the controls (0.13 ± 0.04 ,UM) (Figure 3). BSO and DEM also produced a significant enhancement in the CDDP-induced DCF production in a time-dependent manner (Figure 4). Intracellular DCF production induced by CDDP in the presence of 9.3 mM BSO and 0.2 mM DEM at 150 min was 4.43 ± 0.33 ,UM and 3.12 ± 0.22 ,UM respectively, significantly higher than that induced by CDDP alone (1.35 ± 0.05 gIM). These data suggest that CDDP induces ROS generation and that GSH depletors significantly enhance CDDPinduced ROS generation.
British Journal of Cancer (1997) 76(2), 206-210 * DISCUSSION CDDP is a platinum complex that consists of two carrier ligands of ammonia and two leaving groups of chloride. The mechanisms of CDDP-induced cytotoxicity have long been recognized to be due to the conversion of CDDP to a di-ucl-acquo complex of CDDP, which forms an interstrand cross-link with double-strand DNA to prevent DNA synthesis (Zwelling et al, 1979;Micetich et al, 1983). Moreover, this reaction is likely to cause apoptosis (Evans and Dive, 1993). However, until now it has not been possible to describe fully the mechanism of CDDP-induced cytotoxicity, although a considerable amount of research has thus far been performed. The cytotoxicity of ADR has been reported to depend on its enzymatic defence against ROS, and levels of scavenging enzymes, including glutathione peroxidase or glutathione-S-transferase, modulate the cytotoxicity of ADR (Mimnaugh et al, 1989). Tumour necrosis factor (TNF) was also noted to induce ROS generation, and it was presumed that ROS play a salient role in the tumour cell killing induced by TNF (Yamauchi et al, 1989). Therefore ROS, reportedly, significantly serves in the cytotoxicity of several chemotherapeutic agents (Sausville et al, 1978;Meijer et al, 1987) and TNF (Yamauchi et al, 1989). It could thus be anticipated to enhance the cytotoxic effects through the increase in ROS generation induced by anti-tumour agents.
CDDP is also known to induce nephrotoxicity. Recent reports have shown that CDDP causes nephrotoxicity through ROS generation (Sugihara et al, 1987). It has been demonstrated that CDDP induced nephrotoxicity can be presented by GSH in renal tissues (Anderson et al, 1990). Furthermore, it has been reported that GSH reacts with ROS and plays a role in the reduction of hydrogen peroxide and organic peroxides (Meister, 1983;Martensson, 1991) and that CDDP is inactivated by direct binding to GSH (Meijer et al, 1990). Ishikawa et al (1993) showed that CDDP reacts with intracellular GSH and that the resulting glutathione-platinum complex is actively exported from leukaemia cells via a GS-X pump. In addition, GSH has been considered to contribute to multidrug resistance in human renal cell carcinoma (Micksch et al, 1990) and lung cancer (Meijer et al, 1990).
Recently, GSH depletors have been found to decrease mitochondrial GSH and damage mitochondria, which are a resource of ROS (Meister, 1995). The lethal effects of excessive GSH deficiency in animals such as new born rats and guinea pigs appear to be related to multiorgan failure, involving mitochondrial and other damage in liver, kidney and lung (Martensson et al, 1991). Our results also provide significant evidence that GSH depletors induce intracellular GSH deficiency and enhance CDDP-induced cytotoxicity in bladder cancer cells, whereas GSH depletor alone had no significant cytotoxicity effect at the concentration and exposure time used in this study.
CDDP-induced emesis has been reported to be enhanced by ferric chloride, which is known to catalyse the production of cytotoxic ROS, and to be ameliorated by desferrioxamine, an iron chelator (Matsuki et al, 1994). These results suggest that ROS are involved in CDDP-induced emesis. In CDDP-induced acute renal failure, mitochondrial respiration has been reported to be reduced (Gordon et al, 1986). Moreover, TNF-ax-induced inhibition of mitochondrial electron transport, together with the effects observed for different mitochondrial inhibitors, favours the proposition that TNF-ac also damages the mitochondrial respiratory chain which, consequently, results in increased production of ROS inside the mitochondrion (Schulze-Osthoff et al, 1992). Uslu and Bonavida (1996) reported that the cytotoxicity of CDDP and TNFax against ovarian cancer cells was inhibited in the presence of mitochondrial respiratory chain inhibitors, suggesting that the synergistic cytotoxic effects of CDDP and TNF-a are associated with ROS.
Determining the amount of ROS has been difficult because the ROS-mediated reaction is quite rapid in living cells (Fridovich, 1978). The amount of ROS induced by TNF in tumorigenic fibroblast cells and sarcoma cells was estimated by the formation of methane from dimethylsulphoxide in a previous investigation (Yamauchi et al, 1989). In the present study, the net intracellular generation of ROS was estimated by DCF production from DCFH. Our study demonstrates that CDDP significantly increases ROS generation in bladder cancer cells when compared with controls, and that combination with CDDP and GSH depletors significantly augments CDDP-induced ROS generation.
In an effort to shed light on this deficiency, CDDP has been found to stimulate ROS generation in bladder cancer cells. The present study suggests that ROS may play an important role in CDDP-induced cytotoxicity and that GSH depletors augment this cytotoxicity by enhancing ROS generation in bladder cancer cells.