Photochemistry and Photobiology, 2002, 76(4):

423–430

Nitric Oxide Modulates Tumor Cell Death Induced by Photodynamic Therapy Through a cGMP-dependent Mechanism{ Edgar R. Gomes, Ramiro D. Almeida, Arse´lio P. Carvalho and Carlos B. Duarte* Center for Neuroscience of Coimbra, Department of Zoology, University of Coimbra, Coimbra, Portugal Received 9 January 2002; accepted 29 May 2002

ABSTRACT

tributed to an increase in the expression of heme oxygenase1, heat shock protein 70 or Bcl-2, this was not the case under our experimental conditions. These results show that NO decreases the extent of apoptotic cell death after PDT treatment through a PKG-dependent mechanism, upstream or at the level of caspase activation.

Photodynamic therapy (PDT) of cancer is a very promising technique based on the formation of singlet oxygen induced by a sensitizer after irradiation with visible light. The stimulation of tumor growth by nitric oxide (NO) was reported recently, and NO was shown to have a protective effect against PDT-induced tumor death. We investigated a putative direct effect of NO on tumor cell death induced by PDT, using the human lymphoblastoid CCRF-CEM cells and bisulfonated aluminum phthalocyanine (AlPcS2) as a sensitizer. Cells were incubated with AlPcS2 in the presence or absence of NO donors ((Z)-1-[(2-aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate, hydroxylamine and S-nitroso-Nacetylpenicillamine) or L-arginine. Under these conditions, in the absence of NO donors or L-arginine the cells died rapidly by apoptosis upon photosensitization. In the presence of NO donors or L-arginine, apoptotic cell death after photosensitization was significantly decreased. Modulation of cell death by NO was not due to S-nitrosylation of caspases and occurred at the level or upstream of caspase-9 processing. The protective effect of NO was reversed by incubating the cells with 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one, an inhibitor of guanylyl cyclase, or with KT5823, an inhibitor of protein kinase G (PKG). Incubation with 8-bromo-cyclic guanosine monophosphate, a membrane permeable cyclic guanosine monophosphate analog, also decreased cell death induced by PDT. Although the protective effect of NO against apoptotic cell death in several models has been at-

INTRODUCTION In photodynamic therapy (PDT) of cancer, a sensitizer, light and oxygen are used to induce death of tumor cells. The sensitizer is intravenously administered to the patient and accumulates in the malignant tissue within a few hours. Then, the tumor area is specifically irradiated with visible light with the appropriate wavelength from a laser or a high-capacity lamp coupled to a fiber optic delivery system. When the sensitizer absorbs light, the energy is transferred to oxygen, generating singlet oxygen (1O2) that causes cell damage and tumor destruction (1). Tumor destruction by PDT can occur by different mechanisms. It can be due to a direct effect on the malignant tumor cells or an indirect mechanism, whereby the collapse of tumor vasculature leads to tumor destruction (1). Singlet oxygen is the main reactive element produced by the photodynamic reaction, and because it can only diffuse about 20 nm during its lifetime (2,3), the mechanism of tumor destruction depends on the localization of sensitizers in tumors. In general, the direct mechanism occurs when more hydrophobic sensitizers are used because these sensitizers are accumulated in tumor cells. The indirect mechanism prevails in PDT treatments with more hydrophilic sensitizers, which are mainly localized in the vascular stroma (4,5). Among the aluminum phthalocyanines, the hydrophilic tetrasulfonated derivative is localized in the vascular stroma and induces tumor regression by the indirect mechanism of vascular destruction (6). In contrast, the bisulfonated aluminum phthalocyanine (AlPcS2) is localized in tumor cells; therefore, it destroys the tumor by a direct mechanism (e.g. 3). Nitric oxide (NO) plays an important role in tumor cell biology because it increases blood flow in the tumor, thereby facilitating metastasis (7) and promoting tumor growth (8). Also, nitric oxide synthase (NOS) is expressed at higher levels in tumors than in normal tissue (7). Tumors generating low levels of NO are much more sensitive to PDT than those containing high levels of NO, and the administration of the NOS inhibitor NG-nitro-L-arginine, together with PDT treatment, enhances tumor regression (9,10). Administration of the NO inhibitor

{Posted on the website on 19 July 2002. *To whom correspondence should be addressed at: Center for Neuroscience of Coimbra, Department of Zoology, University of Coimbra, 3004-517 Coimbra, Portugal. Fax: 351-239-822776; e-mail: [email protected] Abbreviations: AlPcS2, bisulfonated aluminum phthalocyanine; 8-BrcGMP, 8-bromo-cyclic guanosine monophosphate; cGMP, cyclic guanosine monophosphate; DETA-NONOate, (Z)-1-[(2-aminoethyl)-N(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate; DEVD-AFC, CBZAsp-Glu-Val-Asp-7-amino-4-trifluoromethyl coumarin; DNA-PK, DNAdependent protein kinase; DNA-PKcs, DNA-dependent protein kinase catalytic subunit; DTT, dithiothreitol; HEPES, N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid; HO-1, heme oxygenase-1; HSC70, constitutive heat shock protein 70; HSP70, heat shock protein 70; NF-jB, nuclear factor–jB; NO, nitric oxide; NOS, nitric oxide synthase; ODQ, 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one; PDT, photodynamic therapy; PKG, protein kinase G; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; SNAP, S-nitroso-N-acetylpenicillamine. Ó 2002 American Society for Photobiology 0031-8655/02

$5.0010.00

423

424 Edgar R. Gomes et al. reduces the blood flow in the tumor, and this could explain the increase in PDT efficiency (9). Another possible effect of NO on the PDT treatment is a direct effect of NO produced by tumor cells and by endothelial cells of tumor microvasculature on tumor cell death because NO is able to directly inhibit cell death (11). To address this hypothesis, we studied the effect of NO on tumor cell death using an in vitro photosensitization system (12), with the human lymphoblastoid CCRF-CEM cells and AlPcS2 as a sensitizer. This sensitizer is known to directly induce tumor destruction (13). We observed that preincubation of cells with NO donors or L-arginine, the NOS substrate, decreased cell death induced by photosensitization. Using a combination of pharmacological and biochemical approaches, we determined that the protective mechanism activated by NO was mediated by the activation of protein kinase G (PKG). Our findings support previous evidence indicating that NO is able to interfere with PDT and suggest that this effect is at least in part due to a direct action of NO on tumor cell death.

MATERIALS AND METHODS Materials. The human lymphoblastoid CCRF-CEM cells were obtained from American Type Culture Collection (Manassas, VA), and fetal calf serum was from Biochrom (Berlin, Germany). AlPcS2 was a gift from Johan E. Van Lier (University of Sherbrooke, Quebec, Canada). 1H[1,2,4]Oxadiazolo[4,3-a]quinoxalin-1-one (ODQ), 8-bromo-cyclic guanosine monophosphate (8-Br-cGMP) and KT5823 were purchased from Biomol (Plymouth Meeting, PA). S-Nitroso-N-acetylpenicillamine (SNAP), hydroxylamine and (Z)-1-[(2-aminoethyl)-N-(2-ammonioethyl)amino]diazen-1-ium-1,2-diolate (DETA-NONOate) were obtained from Alexis (San Diego, CA). SYTO-13 and propidium iodide were obtained from Molecular Probes (Leiden, The Netherlands). CBZ-Asp-Glu-Val-Asp-7-amino4-trifluoromethyl coumarin (DEVD-AFC) was purchased from Enzyme System Products (Livermore, CA). All other chemicals were from Sigma (Sintra, Portugal) or Merck (Darmstadt, Germany). Cell culture. CCRF-CEM cells were cultured in Roswell Park Memorial Institute–1640 medium supplemented with 10% fetal calf serum, 23.8 mM NaHCO3 and penicillin–streptomycin (50 U/mL and 50 lg/mL) in a 95% air and 5% CO2 atmosphere. PDT treatment. Cells (5 3 105 cells/mL) were incubated with 5 lM AlPcS2 in culture medium for 18 h, with or without NO donors, L-arginine, ODQ, 8-Br-cGMP or KT5823. After centrifugation the pellet was resuspended in fresh culture medium, with or without supplementation of NO donors or L-arginine, at the initial concentrations. Cells were irradiated with visible light using a cutoff filter (,600 nm), with a fluence rate of 30 mW/cm2 and a total light dose of 10 J/cm2. Apoptotic nuclei quantification. Cells were washed and resuspended in Na1 medium (NaCl 140 mM, KCl 5 mM, MgCl2 1 mM, CaCl2 1 mM, glucose 5.5 mM, N-2-hydroxyethylpiperazine-N9-2-ethanesulfonic acid [HEPES] 15 mM, pH 7.4) supplemented with 200 nM SYTO-13 and 1.6 lg/mL propidium iodide. Apoptotic and necrotic cells were scored under a fluorescence microscope using a triple-wavelength filter set (XF 63; Omega Optical, Brattleboro, VT). Analysis of DNA fragmentation. DNA fragmentation was detected using a cell death detection enzyme-linked immunosorbent assay kit (Roche, Carnaxide, Portugal), according to the manufacturer’s instructions. Total cell extracts preparation. Cells were washed twice with incomplete KPM (50 mM KCl; 50 mM 1,4-piperazinediethane-sulfonic acid, pH 7.0; 10 mM ethyleneglycol-bis(b-aminoethyl ether)-N,N,N9,N9-tetraacetic acid; 2 mM MgCl2), followed by another wash with complete KPM (incomplete KPM supplemented with 1 lg/mL chymostatin, 1 lg/mL leupeptin, 1 lg/mL antipain, 1 lg/mL pepstatin, 100 lM phenylmethylsulfonyl fluoride [PMSF]). Pellets were resuspended in complete KPM containing 0.5% Triton X-100 and frozen (2808C) and thawed (158C) three times. The cell lysates were centrifuged at 15 800 g for 10 min. The supernatants were then collected, and the protein concentration was quantified using the Bradford assay (Bio Rad, Amadora, Portugal).

Microsomal extract preparation. Cells were washed as described for the preparation of total cell extracts. Resuspended pellets were sonicated four times for 5 s followed by a centrifugation at 10 000 g for 15 min. Supernatants were collected and centrifuged at 100 000 g for 1 h. Pellets were resuspended in complete KPM containing 0.5% Triton X-100, and the protein concentration was quantified using the Bradford assay. Caspase-3–like activity assay. Twenty micrograms of the total cell extracts in 20 lL of complete KPM containing 0.5% Triton X-100 was incubated with or without 20 mM dithiothreitol (DTT) for 30 min at 48C. Forty microliters of the reaction buffer (25 mM HEPES, 0.1% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonic acid) supplemented with 100 lM PMSF and 100 lM DEVD-AFC was added and incubated for 30 min at 378C. Four hundred microliters of the reaction buffer was then added, and AFC fluorescence was measured (kex 5 390 nm; kem 5 475 nm). Immunoblot. Total cell extracts or microsomal extracts were diluted 1:5 (vol/vol) in 63 concentrated sample buffer (0.35 M Tris, 30% [wt/ vol] glycerol, 10% [wt/vol] sodium dodecyl sulfate [SDS], 0.6 M DTT, 0.01% [wt/vol] bromophenol blue) and heated at 958C for 5 min. Samples (10 lg for caspase-3, caspase-9 and Bcl-2; 30 lg for heme oxygenase-1 [HO-1]; 40 lg for heat shock protein 70 [HSP70] and PKG) were separated on a 7% (HSP70 and PKG), 10% (HO-1), 12% (Bcl-2) or 15% (caspase-3 and caspase-9) SDS–polyacrylamide gel and transferred to a polyvinylidene difluoride membrane (Amersham Biosciences, Carnaxide, Portugal). Immunodetection was performed, according to manufacturer’s instructions (ECF Western Blotting kit, Amersham Biosciences), using primary antibodies for caspase-3 (Pharmingen, San Diego, CA), caspase9 (gift from Yuri Lazebnick, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), HO-1 (Stressgen, Victoria, Canada), HSP70 (Zymed, South San Francisco, CA) and a-tubulin (Zymed) at 1:2000 dilution, Bcl-2 (Zymed) at 1:6000 dilution, PKG (Calbiochem, La Jolla, CA) at 1:300 dilution and actin (Roche) at 1:15 000 dilution. Membranes were scanned in a Storm 860 Gel and Blot Imaging System (Amersham Biosciences) and quantified with ImageQuant software. When stripping was performed, membranes were washed in stripping buffer (100 mM glycine, pH 2.9) for 30 min. Statistical analysis. Results are presented as means 6 SEM for at least three independent experiments. Statistical significance was determined using two-tailed paired Student’s t-test, using GraphPad Prism 2.01. (*P , 0.05, **P , 0.01, ***P , 0.001.)

RESULTS Modulation of PDT-induced apoptotic cell death by NO It is known that PDT can induce either apoptotic or necrotic cell death depending on the sensitizer, cell line and irradiation conditions (14,15). To study cell death induced by PDT, we used CCRFCEM cells and AlPcS2 as a sensitizer. Cells were incubated with AlPcS2 for 18 h before irradiation, and cell death was measured 5 h after irradiation with 30 mW/cm2 and 10 J/cm2 by double staining with propidium iodide and SYTO-13. Under these conditions, we observed that more than 50% of the cells exhibited the characteristic apoptotic morphology (Fig. 1, control). On the other hand, the number of necrotic cells was less than 5%, the same as that in nonirradiated cells (data not shown). To evaluate the effect of NO on cell death induced by PDT, we incubated the cells with AlPcS2 and different NO donors (DETA-NONOate, SNAP and hydroxylamine) or L-arginine, the NOS substrate (Fig. 1). In the absence of light neither NO donors nor L-arginine inhibited cell growth or AlPcS2 uptake or induced cell death (data not shown). We observed that the number of apoptotic nuclei in photosensitized cells decreased when the cells were preincubated with the NO donors or L-arginine. To further support the evidence that NO was interfering with the apoptotic machinery in our system, we measured DNA fragmentation, a marker of apoptotic cell death (Fig. 2). Photosensitization of CCRF-CEM cells with AlPcS2 alone induced DNA fragmentation, as measured 5 h after irradiation. DNA fragmen-

Photochemistry and Photobiology, 2002, 76(4) 425

Figure 1. NO modulates cell death induced by PDT, as determined by nuclear morphology. Cells were preincubated with 5 lM AlPcS2, with or without different NO donors or L-arginine, for 18 h. (a) DETA-NONOate, 50 lM; (b) SNAP, 25 lM; (c) hydroxylamine, 100 or 200 lM; (d) L-arginine, 5 or 10 mM. The percentage of cells with apoptotic morphology (as determined by double staining with propidium iodide and SYTO-13) was measured 5 h after irradiation. Data are presented as means 6 SEM from independent experiments (n 5 4–8). *P , 0.5, **P , 0.01 when compared with the control.

tation was decreased in cells preincubated with AlPcS2 and hydroxylamine, supporting the idea that NO interferes with cell death induced by PDT. These results demonstrate that cells treated with NO donors are consistently more resistant to apoptosis induced by PDT. In the mitochondria-mediated apoptotic pathway, cytochrome c is released into the cytosol and together with Apaf-1, dATP and procaspase-9 forms a large oligomeric complex, the apoptosome (16). Procaspase-9, an initiator caspase, is processed and activates downstream effector caspases (caspase-3, caspase-6 and caspase-7) (17,18). These effector caspases are responsible for the cleavage of several proteins in the cell, leading to the morphological changes characteristic of apoptosis. To determine at which step NO interferes with the proteolytic pathways, we studied the processing of caspase-3, an effector caspase, and caspase-9, an initiator caspase (Fig. 3). When cells were preincubated with hydroxylamine before photosensitization, we observed a time-dependent decrease in caspase-3 processing as compared with control cells maintained in the absence of NO donors. This decrease was confirmed by the quantification of processed caspase subunits p20 and p17 in the blot, relative to total caspase-3 (procaspase plus processed caspase) (Fig. 3b,c). The same effect was also observed on caspase-9 processing (Fig. 3e,f). These results demonstrate that the modulation of cell death by NO occurs upstream or at the level of caspase-9 processing. The reduction in caspase-3 processing and its activity is probably caused by the decrease in caspase-9 processing. Mechanisms of protection induced by NO Several hypotheses have been raised to explain the mechanisms of protection induced by NO (11,19). NO can inhibit the activa-

Figure 2. NO modulates DNA fragmentation induced by PDT. Cells were preincubated with 5 lM AlPcS2 and 200 lM hydroxylamine (Hx) for 18 h. Percentage of DNA fragmentation was measured 5 h after irradiation using a DNA fragmentation kit and plotted as the percentage of PDT-induced DNA fragmentation. Data are presented as means 6 SEM from independent experiments (n 5 3).

tion of caspases directly by S-nitrosylation (20,21) or through a cyclic guanosine monophosphate (cGMP)-dependent mechanism, probably by activating PKG (22–24). NO can also modulate gene expression by a cGMP-dependent mechanism (25,26) or by S-nitrosylation (27). Because the results shown in Fig. 3 indicated that caspase-3 processing was decreased in cells pretreated with NO donors, probably due to a decrease in caspase-9 processing, we hypothesized that NO might interfere with the apoptotic pathway at the level or upstream of caspase-9 processing. But these experiments did not rule out the possibility that NO may interfere with the apoptotic pathway because of direct inhibition of effector caspases through S-nitrosylation. To test the role of caspase S-nitrosylation in the decrease of cell death by NO, we measured caspase-3–like activity in cell extracts from photosensitized cells, preincubated with or without 200 lM hydroxylamine, 3.5 h after irradiation (Fig. 4). As expected, the caspase-3–like activity decreased in extracts prepared from hydroxylamine-treated cells. Preincubation of cell lysates with 20 mM DTT, which removes the thiol-bound NO groups from proteins (24), did not restore the activity of caspase-3–like proteases in hydroxylamine-treated cells. Therefore, S-nitrosylation of caspase-3–like proteases is not involved in the modulation of cell death by NO. NO reacts with the heme group of the soluble guanylyl cyclase, thereby activating the enzyme that converts guanosine triphosphate into cGMP. This cyclic nucleotide activates cGMPgated cation channels, cGMP-dependent phosphodiesterases and PKG (28). To study whether a cGMP-dependent mechanism activated by NO could be responsible for the decrease in cell death observed after PDT, we used a pharmacological approach that allowed us to dissect the different steps of the cGMP-dependent pathway. The role of soluble guanylyl cyclase was studied using ODQ, an inhibitor of this enzyme (29,30), and 8-Br-cGMP, a plasma membrane permeable analog of cGMP (31). The cells were incubated with ODQ (10 lM) for 18 h, together with AlPcS2, in the

426 Edgar R. Gomes et al.

Figure 4. NO decreases caspase-3–like activity through a S-nitrosylation–independent mechanism. Cells were preincubated with 5 lM AlPcS2 (open bars) or 5 lM AlPcS2 and 200 lM hydroxylamine (filled bars) for 18 h. Caspase-3–like activity was measured in cell extracts prepared 3.5 h after irradiation, with or without incubation with 20 mM DTT, using the fluorogenic substrate DEVD-AFC. Data are presented as means 6 SEM from independent experiments (n 5 4).

Figure 3. NO decreases caspase-3 and caspase-9 processing. Cells were preincubated with 5 lM AlPcS2 (AlPcS2) or with 5 lM AlPcS2 and 200 lM hydroxylamine (AlPcS2 1 Hx) for 18 h. Total cell extracts were prepared at the indicated time periods, after irradiation, and immunoblotted with (a) anti–caspase-3 or (d) anti–caspase-9 antibodies. Membranes were stripped and reprobed with anti–a-tubulin antibody. Panels (b) and (c) show the quantification of the results presented in (a). The bands from cells preincubated (filled bars) or not (open bars) with hydroxylamine, corresponding to the (b) p20 and (c) p17 fragments, were quantified as a percentage of total caspase-3 (procaspase band plus processed fragments p20 and p17). Panels (e) and (f) show the quantification of the results presented in (d). The bands from cells preincubated (filled bars) or not (open bars) with hydroxylamine, corresponding to the (e) p37 and (f) p35 fragments, were quantified as a percentage of total caspase-9 (procaspase band plus processed fragments p37 and p35). Data shown are representative of three independent experiments.

presence or absence of 200 lM hydroxylamine, and cell death was measured 5 h after irradiation (Fig. 5a). The protective effect of the NO donor was abolished when ODQ was present during the incubation with hydroxylamine, and cell death was not significantly different from the control situation. Incubation of the cells with ODQ and AlPcS2, in the absence of hydroxylamine, did not alter photosensitization-evoked cell death. In another experiment (Fig. 5b) cells were incubated with different concentrations of 8-Br-cGMP (600 lM and 1 mM) together with AlPcS2 for 18 h, and cell death was measured 5 h after irradiation. This cGMP analog decreased cell death to levels similar to those observed when the NO donors were used (Fig. 1). Taken together, these results strongly suggest the involvement of a cGMPdependent mechanism in the NO-induced modulation of CCRFCEM cell death by PDT. It is known that PKG activation is probably involved in the

cGMP-dependent mechanism of inhibition of cell death (23,24). Two different forms of PKG have been identified in humans, PKG I and PKG II. Alternative splicing of PKG I creates two different isoforms, PKG Ia and PKG Ib (31). Because it has been described that PKG can be down-regulated or lost during the growth of primary cultures and after prolonged cell culturing (32), we analyzed the presence of PKG I in CCRF-CEM cells by immunoblot, using an antibody that reacts with both isoforms of PKG Type I (33) (Fig. 5c). A single band was observed with the expected molecular weight of 75 kDa, confirming the expression of PKG I in CCRF-CEM cells. To further investigate the involvement of PKG in the protection of CCRF-CEM cells by NO, we used KT5823, an inhibitor of PKG (Fig. 5d). When the cells were incubated with the PKG inhibitor in the presence of hydroxylamine, the number of photosensitized cells with apoptotic morphology was the same as in the control (photosensitized cells without hydroxylamine and KT5823) or in cells incubated only with the PKG inhibitor. Therefore, this inhibitor reverted the protective effect observed when the cells were only incubated with hydroxylamine. Because it is known that PKG can be involved in the modulation of gene expression (25), we investigated whether NO increases the expression levels of HO-1, Bcl-2 or HSP70 (Fig. 6). These proteins have been described to be up-regulated and involved in the mechanisms of cellular protection by NO against cell death (34–36). Cells were incubated with hydroxylamine or DETA-NONOate for 18 h, and the expression of proteins was analyzed by immunoblot. We could not detect measurable amounts of HO-1 in microsomal fractions prepared from control or NO donor–treated cells (Fig. 6a). But as expected, incubation of the cells with hemin, a substrate of HO-1, induced its expression. The levels of Bcl-2 also were similar in control and DETA-NONOate–stimulated cells (Fig. 6b). This observation was confirmed by the quantification of Bcl-2 blots relative to the amount of actin (data not shown). When total cell extracts were immunoblotted with an anti-HSP70 antibody, we could not observe any detectable levels of the protein, although we observed a band corresponding to constitutive heat shock protein 70

Photochemistry and Photobiology, 2002, 76(4) 427

Figure 5. Involvement of a cGMP-dependent mechanism and PKG activation in the NO-induced decrease in cell death by PDT. Cells were preincubated with 5 lM AlPcS2, in the presence or absence of 200 lM hydroxylamine (Hx) and 10 lM ODQ (a), or with different concentrations of 8-Br-cGMP (600 lM and 1 mM) (b) for 18 h. In a different experiment, cells were treated as in (a), using 600 nM KT5823 (KT) instead of ODQ (d). Cell death was measured by apoptotic nuclei quantification 5 h after irradiation. Data are presented as means 6 SEM from independent experiments (n 5 4–6). To confirm the expression of PKG, total cell extracts were prepared from CEM cells, separated in an 8% SDS-polyacrylamide gel and immunoblotted with an anti-PKG antibody (c). *P , 0.5, **P , 0.01 when compared with the control.

(HSC70) (Fig. 6c). When total cell extracts were prepared from cells heat shocked during 10 min at 448C, followed by 6 h of incubation at 378C to allow cell recovery, we observed two bands corresponding to HSC70 and HSP70. These results demonstrate that in our system the mechanism of modulation of cell death activated by NO does not involve the induction of the expression of HO-1, Bcl-2 and HSP70.

DISCUSSION In this study we showed that the photosensitization of CCRFCEM cells with AlPcS2 induces apoptosis, and this effect was diminished upon preincubation of the cells with NO donors or Larginine, the NOS substrate. In previous studies it was shown that PDT could induce either necrosis or apoptosis. But the main mechanism of cell death that occurs in vitro is apoptosis (37). Necrosis occurs when the photodamage is induced in plasma membrane or in lysosomes. Therefore, sensitizers that are localized in these areas within the cell induce necrosis. When the sensitizer is localized in the mitochondria, PDT activates the apoptotic machinery (38,39). AlPcS2 used in this work was shown to have a diffuse distribution within the cytoplasm, probably localized in the mitochondria, and did not colocalize with lysosomal markers (40). It has also been shown using electron microscopy that AlPcS2 photosensitization induces the swelling of the mitochondria and of the endoplasmic reticulum, as well as

Figure 6. NO does not induce the expression of HO-1, Bcl-2 or HSP70. Cells were incubated with 200 lM hydroxylamine (Hx) or 50 lM DETA-NONOate (DETA), or left without treatment (Ct), for 18 h. Cells were also incubated with 10 lM hemin for 12 h (a) or heat shocked 10 min at 448C, followed by 6 h of incubation at 378C (c). (a) Microsomal or (b, c) total cell extracts were prepared and immunoblotted with (a) anti–HO-1, (b) anti–Bcl-2 or (c) anti-HSP70 antibodies. (b) Membranes were also reprobed with anti-actin antibody. Data shown are representative of three independent experiments.

cytoplasmic membrane damage (41). Therefore, the localization of AlPcS2 is well correlated with the apoptotic cell death observed in our work. Other reports have also observed apoptotic cell death induced with AlPcS2 (42) and with other phthalocyanines (37). In this work we observed a decrease in cell death induced by PDT when cells were preincubated with different NO donors, as observed by nuclear morphology (Fig. 1) and DNA fragmentation (Fig. 2). We also observed a decrease in the processing of caspase-3 and caspase-9 (Fig. 3). Therefore, we can conclude that NO interferes with the apoptotic pathway upstream or at the level of caspase-9 activation. In other situations where different apoptotic stimuli were used, NO was also shown to interfere with apoptosis upstream or at the level of caspase activation (23,24,43–46). Low levels of NO donors or endogenous NO synthesis were first shown to inhibit apoptosis in human B lymphocytes (45), and since then several reports using different cells and apoptotic insults have shown the same inhibitory action of NO (11). In particular, NO inhibits apoptosis induced by oxidative stress (47–49). Accordingly, cell death induced by hydrogen peroxide can be inhibited by low doses of NO released by NONOate derivatives and NO donors containing an S-nitroso functional group such as SNAP. Hydrogen peroxide reacts with metalloproteins, in particular iron-containing proteins, forming hypervalent complexes or the hydroxyl radical (or both), which can lead to lipid peroxidation, protein oxidation and DNA damage. NO can react and neutralize these complexes because of the formation of iron–

428 Edgar R. Gomes et al. nitrosyl complexes, therefore, preventing metal-catalyzed reactive oxygen species formation (50,51). NO can also react with the superoxide radical, therefore, acting as a superoxide scavenger, and may react with lipid alkoxyl and peroxyl radicals, thereby blocking the lipid radical chain propagation reaction (52). Because PDT induces the production of reactive oxygen species, leading to lipid peroxidation and protein oxidation, the mechanism of protection by NO observed in our work could be explained by the interaction of this molecule with the radical species formed during PDT. But this is not the case because preincubation of cells with NO donors for a few minutes before irradiation did not interfere with cell death (data not shown). Only prolonged incubations (18 h) with NO donors protected the cells against photosensitization-induced apoptotic death. Therefore, the mechanism of protection does not involve the direct interaction of NO with reactive oxygen species. In this work we observed that NO modulates tumor cell death induced by PDT through a cGMP-dependent mechanism (Fig. 5) and not by S-nitrosylation of caspase-3–like proteases (Fig. 3) or by up-regulation of HO-1, Bcl-2 and HSP70 (Fig. 6). Three criteria have been proposed by Smolenski et al. (32) for determining the involvement of PKG in the NO-cGMP–dependent mechanisms: (1) PKG expression and activation by a cGMPelevating agent in the cell or tissue investigated; (2) specific activation or inhibition of PKG that mimics or blocks the effect of the cGMP-elevating agent; and (3) reduction or absence of an effect of cGMP in PKG-deficient systems. In the present work, we were able to fulfill two of these criteria. We showed that PKG is present in CCRF-CEM cells (Fig. 5c), 8-Br-cGMP, an activator of PKG, mimics the protective effect induced by NO (Fig. 5b) and KT5823, an inhibitor of PKG, blocks this effect (Fig. 5d). We have also shown that ODQ, an inhibitor of soluble guanylyl cyclase, blocked the protective effect of NO. In contrast with our results, a recent work showed that PKG I is not expressed in Jurkat-E6 and A3.01 T cell lines (53). Taken together, these results strongly suggest an involvement of PKG in the mechanisms activated by NO that are responsible for the modulation of cell death induced by PDT. The mechanisms activated by PKG are presently unknown, but probably PKG phosphorylates unidentified substrates, leading to the activation of a cascade of events that will inhibit the apoptotic pathway. Although this is likely to be responsible for the interference of NO with apoptotic cell death induced by PDT, very few substrates for this kinase are presently known (31,54). Until now, very few studies, where NO inhibited apoptosis through a cGMP-dependent mechanism, have shown the involvement of PKG according to the criteria proposed by Smolenski et al. (32). These criteria have only been fulfilled in studies performed in hepatocytes (24) and cortical neurons (47). In this work we also measured the expression of HO-1, HSP70 and Bcl-2, and we did not observe any alteration in the expression pattern of these proteins upon exposure of the cells to NO (Fig. 6). But other proteins like Cox-2 and the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) were not studied. Cox-2 is a good candidate because its up-regulation is mediated by AP-1 and nuclear factor–jB (NF-jB) (55,56). DNA-dependent protein kinase (DNA-PK) is probably not involved in the mechanism of protection because it is involved in the mechanism of DNA repair (57), and cell death induced by AlPcS2 probably occurs as a result of the damage to mitochondria, leading to the activation of the apoptotic machinery (38).

DNA damage can occur during PDT treatment, but it is probably a consequence of the treatment and is not responsible for the activation of cell death (1,37). It is also known that NO can regulate gene expression through S-nitrosylation (58). Low doses of NO activated NF-jB in lymphocytes (59) because of the S-nitrosylation of p21ras (60). On the other hand, higher doses of NO inhibit the activation of NFjB in hepatocytes (61), T cells (62) and endothelial (63) and vascular smooth muscle cells (64) through different mechanisms involving S-nitrosylation (58). Other transcription factors like p53 are also regulated by S-nitrosylation (58). In our work these pathways are probably not involved because we were able to completely block the protective mechanism induced by NO with KT5823, an inhibitor of PKG (Fig. 5d), and 8-Br-cGMP mimicked the protective effect of NO (Fig. 5b), showing that the activation of a cGMP-dependent mechanism plays a major role. In conclusion, we demonstrated that NO can directly decrease cell death induced by PDT. Therefore, together with the putative effect of NO on tumor microvasculature, this direct effect may be important in developing new strategies to increase the efficiency of PDT treatment of tumors. Acknowledgement—This work was supported by PRAXIS XXI.

REFERENCES 1. Moore, J. V., C. M. L. West and C. Whitehurst (1997) The biology of photodynamic therapy. Phys. Med. Biol. 42, 913–935. 2. Moan, J. and K. Berg (1991) The photodegradation of porphyrins in cells can be used to estimate the lifetime of singlet oxygen. Photochem. Photobiol. 53, 549–553. 3. Peng, Q. and J. Moan (1995) Correlation of distribution of sulphonated aluminium phthalocyanines with their photodynamic effect in tumour and skin of mice bearing CaD2 mammary carcinoma. Br. J. Cancer 72, 565–574. 4. Ahmad, N. and H. Mukhtar (2000) Mechanism of photodynamic therapy-induced cell death. Methods Enzymol. 319, 342–358. 5. Peng, Q., J. Moan, G. Farrants, H. E. Danielsen and C. Rimington (1990) Localization of potent photosensitizers in human tumor LOX by means of laser scanning microscopy. Cancer Lett. 53, 129–139. 6. Peng, Q., J. Moan, J. M. Nesland and C. Rimington (1990) Aluminum phthalocyanines with asymmetrical lower sulfonation and with symmetrical higher sulfonation: a comparison of localizing and photosensitizing mechanism in human tumor LOX xenografts. Int. J. Cancer 46, 719–726. 7. Tozer, G. M., V. E. Prise and D. J. Chaplin (1997) Inhibition of nitric oxide synthase induces a selective reduction in tumor blood flow that is reversible with L-arginine. Cancer Res. 57, 948–955. 8. Jadeski, L. C., K. O. Hum, C. Chakraborty and P. K. Lala (2000) Nitric oxide promotes murine mammary tumour growth and metastasis by stimulating tumour cell migration, invasiveness and angiogenesis. Int. J. Cancer 86, 30–39. 9. Korbelik, M., C. S. Parkins, H. Shibuya, I. Cecic, M. R. Stratford and D. J. Chaplin (2000) Nitric oxide production by tumour tissue: impact on the response to photodynamic therapy. Br. J. Cancer 82, 1835–1843. 10. Henderson, B., T. M. Sitnik-Busch and L. A. Vaughan (1999) Potentiation of photodynamic therapy antitumor activity in mice by nitric oxide synthase inhibition is fluence rate dependent. Photochem. Photobiol. 70, 64–71. 11. Liu, L. and J. S. Stamler (1999) NO: an inhibitor of cell death. Cell Death Differ. 6, 937–942. 12. Gomes, E. R., T. Cruz, C. F. Lopes, A. P. Carvalho and C. B. Duarte (1999) Photosensitization of lymphoblastoid cells with phthalocyanines at different saturating incubation times. Cell Biol. Toxicol. 15, 249–260. 13. Chan, W. S., N. Brasseur, C. Lamadeleine, R. Ouellet and J. E. Vanlier (1997) Efficacy and mechanism of aluminium phthalocyanine

Photochemistry and Photobiology, 2002, 76(4) 429

14.

15.

16.

17. 18.

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

29. 30.

31.

32.

33.

34.

35.

and its sulphonated derivatives mediated photodynamic therapy on murine tumours. Eur. J. Cancer 33, 1855–1859. Luo, Y. and D. Kessel (1997) Initiation of apoptosis versus necrosis by photodynamic therapy with chloroaluminum phthalocyanine. Photochem. Photobiol. 66, 479–483. Kessel, D., Y. Luo, Y. Q. Deng and C. K. Chang (1997) The role of subcellular localization in initiation of apoptosis by photodynamic therapy. Photochem. Photobiol. 65, 422–426. Zou, H., Y. C. Li, H. S. Liu and X. D. Wang (1999) An APAF-1– cytochrome c multimeric complex is a functional apoptosome that activates procaspase-9. J. Biol. Chem. 274, 11549–11556. Thornberry, N. A. and Y. Lazebnik (1998) Caspases: enemies within. Science 281, 1312–1316. Earnshaw, W. C., L. M. Martins and S. H. Kaufmann (1999) Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu. Rev. Biochem. 68, 383–424. Melino, G., M. V. Catani, M. Corazzari, P. Guerrieri and F. Bernassola (2000) Nitric oxide can inhibit apoptosis or switch it into necrosis. Cell. Mol. Life Sci. 57, 612–622. Rossig, L., B. Fichtlscherer, K. Breitschopf, J. Haendeler, A. M. Zeiher, A. Mulsch and S. Dimmeler (1999) Nitric oxide inhibits caspase-3 by S-nitrosation in vivo. J. Biol. Chem. 274, 6823–6826. Dimmeler, S., J. Haendeler, M. Nehls and A. M. Zeiher (1997) Suppression of apoptosis by nitric oxide via inhibition of interleukin1beta–converting enzyme (ICE)-like and cysteine protease protein (CPP)-32–like proteases. J. Exp. Med. 185, 601–607. Genaro, A. M., S. Hortelano, A. Alvarez, C. Martinez and L. Bosca (1995) Splenic B lymphocyte programmed cell death is prevented by nitric oxide release through mechanisms involving sustained Bcl-2 levels. J. Clin. Investig. 95, 1884–1890. Kim, Y. M., H. T. Chung, S. S. Kim, J. A. Han, Y. M. Yoo, K. M. Kim, G. H. Lee, H. Y. Yun, A. Green, J. Li, R. L. Simmons and T. R. Billiar (1999) Nitric oxide protects PC12 cells from serum deprivation-induced apoptosis by cGMP-dependent inhibition of caspase signaling. J. Neurosci. 19, 6740–6747. Kim, Y. M., R. V. Talanian and T. R. Billiar (1997) Nitric oxide inhibits apoptosis by preventing increases in caspase-3–like activity via two distinct mechanisms. J. Biol. Chem. 272, 31138–31148. Komalavilas, P., P. K. Shah, H. Jo and T. M. Lincoln (1999) Activation of mitogen-activated protein kinase pathways by cyclic GMP and cyclic GMP–dependent protein kinase in contractile vascular smooth muscle cells. J. Biol. Chem. 274, 34301–34309. Gudi, T., G. K. Hong, A. B. Vaandrager, S. M. Lohmann and R. B. Pilz (1999) Nitric oxide and cGMP regulate gene expression in neuronal and glial cells by activating type II cGMP-dependent protein kinase. FASEB J. 13, 2143–2152. Marshall, H. E. and J. S. Stamler (2001) Inhibition of NF-kappa B by S-nitrosylation. Biochemistry 40, 1688–1693. Lohmann, S. M., A. B. Vaandrager, A. Smolenski, U. Walter and H. R. De Jonge (1997) Distinct and specific functions of cGMPdependent protein kinases. Trends Biochem. Sci. 22, 307–312. Koesling, D. (1998) Modulators of soluble guanylyl cyclase. Arch. Pharmacol. 358, 123–126. Lucas, K. A., G. M. Pitari, S. Kazerounian, I. Ruiz-Stewart, J. Park, S. Schulz, K. P. Chepenik and S. A. Waldman (2000) Guanylyl cyclases and signaling by cyclic GMP. Pharmacol. Rev. 52, 375–414. Pfeifer, A., P. Ruth, W. Dostmann, M. Sausbier, P. Klatt and F. Hofmann (1999) Structure and function of cGMP-dependent protein kinases. Rev. Physiol. Biochem. Pharmacol. 135, 105–149. Smolenski, A., A. M. Burkhardt, M. Eigenthaler, E. Butt, S. Gambaryan, S. M. Lohmann and U. Walter (1998) Functional analysis of cGMP-dependent protein kinases I and II as mediators of NO/cGMP effects. Arch. Pharmacol. 358, 134–139. Keilbach, A., P. Ruth and F. Hofmann (1992) Detection of cGMP dependent protein kinase isozymes by specific antibodies. Eur. J. Biochem. 208, 467–473. Foresti, R. and R. Motterlini (1999) The heme oxygenase pathway and its interaction with nitric oxide in the control of cellular homeostasis. Free Radic. Res. 31, 459–475. Suschek, C. V., V. Krischel, D. Bruch-Gerharz, D. Berendji, J. Krutmann, K. D. Kroncke and V. Kolb-Bachofen (1999) Nitric oxide fully protects against UVA-induced apoptosis in tight correlation with Bcl-2 up-regulation. J. Biol. Chem. 274, 6130–6137.

36. Kim, Y. M., M. E. Devera, S. C. Watkins and T. R. Billiar (1997) Nitric oxide protects cultured rat hepatocytes from tumor necrosis factor-alpha–induced apoptosis by inducing heat shock protein 70 expression. J. Biol. Chem. 272, 1402–1411. 37. Oleinick, N. L. and H. H. Evans (1998) The photobiology of photodynamic therapy: cellular targets and mechanisms. Radiat. Res. 150, S146–S156. 38. Kessel, D. and Y. Luo (1998) Mitochondrial photodamage and PDTinduced apoptosis. J. Photochem. Photobiol. B: Biol. 42, 89–95. 39. Kessel, D. and Y. Luo (2001) Intracellular sites of photodamage as a factor in apoptotic cell death. J. Porphyrins Phthalocyanines 5, 181–184. 40. Peng, Q., G. W. Farrants, K. Madslien, J. C. Bommer, J. Moan, H. E. Danielsen and J. M. Nesland (1991) Subcellular localization, redistribution and photobleaching of sulfonated aluminum phthalocyanines in a human melanoma cell line. Int. J. Cancer 49, 290–295. 41. Peng, Q., J. Moan and J. M. Nesland (1996) Correlation of subcellular and intratumoral photosensitizer localization with ultraestructural features after photodynamic therapy. Ultrastruct. Pathol. 20, 109– 129. 42. Ketabchi, A., A. MacRobert, P. M. Speight and J. H. Bennett (1998) Induction of apoptotic cell death by photodynamic therapy in human keratinocytes. Arch. Oral Biol. 43, 143–149. 43. Li, J. R., C. A. Bombeck, S. F. Yang, Y. M. Kim and T. R. Billiar (1999) Nitric oxide suppresses apoptosis via interrupting caspase activation and mitochondrial dysfunction in cultured hepatocytes. J. Biol. Chem. 274, 17325–17333. 44. Kim, Y. M., T. H. Kim, D. W. Seol, R. V. Talanian and T. R. Billiar (1998) Nitric oxide suppression of apoptosis occurs in association with an inhibition of Bcl-2 cleavage and cytochrome c release. J. Biol. Chem. 273, 31437–31441. 45. Mannick, J. B., K. Asano, K. Izumi, E. Kieff and J. S. Stamler (1994) Nitric oxide produced by human B lymphocytes inhibits apoptosis and Epstein–Barr virus reactivation. Cell 79, 1137–1146. 46. Hebestreit, H., B. Dibbert, I. Balatti, D. Braun, A. Schapowal, K. Blaser and H. U. Simon (1998) Disruption of Fas receptor signaling by nitric oxide in eosinophils. J. Exp. Med. 187, 415–425. 47. Fernandez-Tome, P., I. Lizasoain, J. C. Leza, P. Lorenzo and M. A. Moro (1999) Neuroprotective effects of DETA-NONOate, a nitric oxide donor, on hydrogen peroxide–induced neurotoxicity in cortical neurones. Neuropharmacology 38, 1307–1315. 48. Wink, D. A., I. Hanbauer, M. C. Krishna, W. G. DeGraff, J. Gamson and J. B. Mitchell (1993) Nitric oxide protects against cellular damage and cytotoxicity from reactive oxygen species. Proc. Natl. Acad. Sci. USA 90, 9813–9817. 49. Joshi, M. S., J. L. Ponthier and J. R. Lancaster (1999) Cellular antioxidant and pro-oxidant actions of nitric oxide. Free Radic. Biol. Med. 27, 1357–1366. 50. Kanner, J., S. Harel and R. Granit (1991) Nitric oxide as an antioxidant. Arch. Biochem. Biophys. 289, 130–136. 51. Wink, D. A., Y. Vodovotz, M. B. Grisham, W. DeGraff, J. C. Cook, R. Pacelli, M. Krishna and J. B. Mitchell (1999) Antioxidant effects of nitric oxide. Methods Enzymol. 301, 413–424. 52. Rubbo, H., R. Radi, M. Trujillo, R. Telleri, B. Kalyanaraman, S. Barnes, M. Kirk and B. A. Freeman (1994) Nitric oxide regulation of superoxide and peroxynitrite-dependent lipid peroxidation. Formation of novel nitrogen-containing oxidized lipid derivatives. J. Biol. Chem. 269, 26066–26075. 53. Fischer, T. A., A. Palmetshofer, S. Gambaryan, E. Butt, C. Jassoy, U. Walter, S. Sopper and S. M. Lohmann (2001) Activation of cGMP-dependent protein kinase Ib inhibits interleukin 2 release and proliferation of T cell receptor–stimulated human peripheral T cells. J. Biol. Chem. 276, 5967–5974. 54. Wang, X. and P. J. Robinson (1997) Cyclic GMP-dependent protein kinase and cellular signaling in the nervous system. J. Neurochem. 68, 443–456. 55. von Knethen, A., D. Callsen and B. Brune (1999) NF-kappaB and AP-1 activation by nitric oxide attenuated apoptotic cell death in RAW 264.7 macrophages. Mol. Biol. Cell 10, 361–372. 56. Pilz, R. B., M. Suhasini, S. Idriss, J. L. Meinkoth and G. R. Boss (1995) Nitric oxide and cGMP analogs activate transcription from AP-1–responsive promoters in mammalian cells. FASEB J. 9, 552– 558.

430 Edgar R. Gomes et al. 57. Xu, W., L. Liu, G. C. M. Smith and I. G. Charles (2000) Nitric oxide upregulates expression of DNA-PKcs to protect cells from DNA-damaging anti-tumour agents. Nat. Cell Biol. 2, 339–345. 58. Marshall, H. E., K. Merchant and J. S. Stamler (2000) Nitrosation and oxidation in the regulation of gene expression. FASEB J. 14, 1889–1900. 59. Lander, H. M., P. Sehajpal, D. M. Levine and A. Novogrodsky (1993) Activation of human peripheral blood mononuclear cells by nitric oxide–generating compounds. J. Immunol. 150, 1509–1516. 60. Lander, H. M., J. S. Ogiste, S. F. Pearce, R. Levi and A. Novogrodsky (1995) Nitric oxide–stimulated guanine nucleotide exchange on p21ras. J. Biol. Chem. 270, 7017–7020.

61. Taylor, B. S., Y. M. Kim, Q. Wang, R. A. Shapiro, T. R. Billiar and D. A. Geller (1997) Nitric oxide down-regulates hepatocyte-inducible nitric oxide synthase gene expression. Arch. Surg. 132, 1177–1183. 62. Sekkai, D., F. Aillet, N. Israel and M. Lepoivre (1998) Inhibition of NF-kappaB and HIV-1 long terminal repeat transcriptional activation by inducible nitric oxide synthase 2 activity. J. Biol. Chem. 273, 3895–3900. 63. Peng, H. B., P. Libby and J. K. Liao (1995) Induction and stabilization of I kappa B alpha by nitric oxide mediates inhibition of NFkappa B. J. Biol. Chem. 270, 14214–14219. 64. Shin, W. S., Y. H. Hong, H. B. Peng, R. De Caterina, P. Libby and J. K. Liao (1996) Nitric oxide attenuates vascular smooth muscle cell activation by interferon-gamma. The role of constitutive NF-kappa B activity. J. Biol. Chem. 271, 11317–11324.