0031-8655/92 $05.00+0.00
Photochemistry and Photobiology Vol. 55, No. 2. pp. 231-237, 1992
Printed in Great Britain.
All
rights reserved
Copyright
0 1992 Pergamon Press plc
PROTECTION BY THE FLUORIDE ION AGAINST PHTHALOCYANINE-INDUCED PHOTODYNAMIC KILLING OF CHINESE HAMSTER CELLS EHUDBEN-HuRI*,MARIANE. CLAY^, EDUARDO F. VICIOSO~, ANTONIO R. A N T U N E Z ~ , BORISD. RlHTER4, MALCOLM E. KENNEY~* and NANCY L. OLEINICK~* 'Nuclear Research Center-Negev, P.O. Box 9001, Beer-Sheva 84190, Israel, ZDivisions of Biochemical Oncology and Radiation Oncology, Department of Radiology, Case Western Reserve University School of Medicine, Cleveland, OH 44106, USA, 3Department of Radiology and Medical Physics, Faculty of Medicine, University of Malaga, 29080 Malaga, Spain and "Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA (Received 15 April 1991; accepted 19 July 1991)
Abstract-When a dilute F- solution was added to a culture of Chinese hamster cells that had been preincubated with an aluminium phthalocyanine sensitizer derived from AIPcCI, the photosensitivity of the cells was markedly reduced compared to control cells not treated with F-. Under the same treatment conditions, the reduction in [3H]thymidine incorporation into cellular DNA caused by light and this sensitizer and the production of DNA-protein crosslinks caused by light and this sensitizer were also inhibited by F-. In contrast, the killing of Chinese hamster cells, the reduction of thymidine incorporation by the cells, and the production of DNA-protein crosslinks in the cells caused by the combination of light and either Photofrin I1 or the silicon phthalocyanine HOSiPcOSi(CH3)z(CHz)3N(CH3)2 were not inibited by F-. We conclude that the aluminium phthalocyanine sensitizer used is largely or completely AIPc(OH)(H,O), that it is converted to a fluoro complex by F-, and that this compound probably is a less efficient generator of photochemical damage at a critical cellular target(s) than is AIPc(OH)(H,O). The inhibition of thymidine incorporation and DNA-protein crosslink formation indicates that the effects of F- can be expressed at intracellular sites. It is further concluded that the silicon phthalocyanine sensitizer and Photofrin I1 do not interact significantly with F-.
INTRODUCTION
Phthalocyanines are efficient photosensitizers of cultured mammalian cells (Ben-Hur and Rosenthal, 1985; Brasseur et al., 1985). Because of this ability, as well as their intense absorption in the far red, the lack of cutaneous phototoxicity, and the ease with which they can be chemically modified, phthalocyanines are increasingly being studied as secondgeneration sensitizers for photodynamic therapy (PDT)? of cancer (Spikes, 1986; Ben-Hur, 1991). Phthalocyanine photosensitization of mammalian cells causes damage to many cellular structures, including the plasma membrane (Ben-Hur and Rosenthal, 1986), mitochondria (Ben-Hur et al., 1987a), and DNA. Damage to DNA can be expressed as alkali-labile sites, single-strand breaks, *To whom correspondence should be addressed. TAbbreviafions: AIPc, the aluminium phthalocyanine sensitizer formed when AlPcCl is dissolved in DMF; AIPcCl, chloroaluminium phthalocyanine; AIPcS,, chloroaluminium phthalocyanine tetrasulfonate isomer mixture; DMF, dimethylformamide; DPC, DNA-protein crosslinks; H,EDTA, ethylenediamine tetraacetic acid; HEPES, N-2-hydroxyethyl piperazineN-2-ethane sulfonic acid; HpD, hematoporphyrin derivative; PBS, phosphate-buffered saline; PDT, photodynamic therapy; PFII, Photofrin 11; SiPc, HOSiPcOSi(CH3),(CH2),N(CH3)2;SSC, standard saline citrate; TCA, tnchloroacetic acid; Tris, Tris-
(hydroxymethy1)aminornethane.
and DNA-protein crosslinks (DPC) (Ben-Hur et al., 1987b; Hunting et al., 1987; Ramakrishnan et al., 1988,1989). Although mutagenesis was not detected at either the hemizygous hprt (hypoxanthine guanine phosphoribosyl transferase) locus or the Na'/K'-ATPase locus after photosensitization of Chinese hamster V79 cells with an aluminium phthalocyanine (Ben-Hur et al., 1987b), mutations were found to be induced at the heterozygous tk (thymidine kinase) locus in two strains of L5178Y mouse lymphoma cells (Evans et al., 1989; Rerko et al., 1991; Oleinick er al., 1991). Recently, it was found that photohemolysis induced in human erythrocytes with a sulfonated aluminium phthalocyanine is markedly inhibited by F(Ben-Hur et al., 1991a). Addition of F- immediately prior to irradiation of cells pretreated with the sensitizer was sufficient for complete inhibition of the photohemolysis. Maximal inhibition of photohemolysis required that F- be present during light exposure. It was concluded that F- interferes with an early step in the photochemical mechanism inducing hemolysis. Because photohemolysis results from membrane damage, it was hypothesized that F- exerts its action at the membrane. Indeed, crosslinking of spectrin monomers in erythrocyte ghosts photosensitized with AlPcS4 is inhibited by F- (Ben-Hur and Orenstein, 1991). However, in spite of what is known, a complete account of the cellular events that are associated
23 1
EHUDBEN-HURer al.
232
with the photocytotoxicity of phthalocyanines cannot be given yet. I n an effort to further understand the effect of F- o n the photocytotoxicity of phthalocyanines and thus the larger problem of the mechanisms of the phototoxicity, we have treated V79 fibroblasts with light and AIPc, SiPc, or PFII as sensitizers. W e have asked whether or not the effect of F- is limited t o erythrocytes, whether or not the photodynamic action of photosensitizers other than aluminium phthalocyanines can also be inhibited by the presence of F-, and whether or not the effects of F- can be expressed a t intracellular sites removed from the plasma membrane. W e have found that light and any o n e of the three sensitizers killed t h e cells, reduced the incorporation of [3H]thymidine into DNA, and caused the production of DNAprotein crosslinks in t h e cells. With AIPc, Finhibited all three effects. MATERIALS AND METHODS
Cell culture. Chinese hamster V79 lung fibroblasts were grown in McCoy's 5A medium (Gibco Laboratories, Grand Island, NY) augmented with 10% calf serum and HEPES to a final concentration of 20 mM, pH 7.4. Cultures were routinely incubated in a humidified atmosphere of 5% co2/95% air, at 37°C. For experiments, cultures were seeded with an appropriate number of cells, as described below, and incubated for 4 h to allow for cell attachment before the addition of photosensitizers. Photorensitizers. Photofrin 11 (Photofrin Polyporphyrin, PFII) was obtained as a lyophilized powder (QLT Phototherapeutics, Inc., Vancouver, BC, Canada). A solution of 2.5 mg/mL was prepared in 5 % dextrose, sterilized by filtration, and used immediately. AlPcCl was obtained as a crystalline solid (Eastman Kodak, Rochester. NY) and was recrystallized from 1-chloronaphthalene. A stock sensitizer solution was prepared from it by dissolving it in DMF at a rate of 1 mmoUL. This solution retained its full photosensitizing capacity for at least 1 month. The sensitizer in this solution is referred to as AIPc. The silicon phthalocyanine [HOSiPcOSi(CH,),(CH,),N(CH,),I ( S i c ) was synthesized as described elsewhere (Oleinick el a / . , in preparation). A 0.5 m M stock solution of it in DMF was prepared and stored at 4°C. It retained its full photosensitizing capacity for at least several months when stored thus. but lost capacity over a period of days when brought to room temperature. Cultures were treated with AIPc. SiPc, or PFII by adding an appropriate aliquot (1 kL, 1 kL, or 10 kL, respectively) of the stock solution to each mL of medium overlying the cells and returning the culture dishes to the incubator for an additional 1120 h. Control cultures not treated with a photosensitizer were included in each experiment. Cell irradiation procedure. The cells which had been treated with PFII were irradiated by placing the tissue culture dish containing them on a glass exposure tray located 72.5 cm above a 500 W tungsten-halogen lamp light source for periods of 2-20 min. The fluence rate at the level of the cell monolayers was -88 W/m2. The cells which had been treated with the other two sensitizers were irradiated in the same way except that irradiation times of 1.5-10 min were used, and the light was passed through a filter (Lee primary red filter No. 106, Vincent Lighting, Cleveland, OH) that removed light of wavelengths SiPc > AlPc. (Hunting et al., 1987), AIPcS, is not clastogenic, neither sister chromatid exchanges (Berg el al., DISCUSSION 1988) nor micronuclei (Ben-Hur et al., 1988) being produced by its photodynamic action. However, the Effect of Fobservation that the photodynamic effects of AlPc The results presented show that under the exper- are mutagenic in certain strains of L5178Y cells imental conditions employed here, light and AlPc (Evans et al., 1989; Oleinick et al., 1991) suggests were cytotoxic at fluences >6 kJ/m2, that they that DNA damage can have a primary influence on inhibited ["Hlthymidine incorporation at fluences of the viability of cells under some conditions. It is 3 3 kJ/m2, and that they produced DPC at fluences noteworthy that the sensitizers differed considerably of 2 3 kJ/rn2. Thus the inhibition of [3H]thymidine in their efficiency in producing DNA damage at incorporation and the production of DPC caused by equitoxic fluences, i. e. the photodynamic effects low light levels and AlPc appear to reflect sub-lethal were greatest with PFII, less with SiPc, and least with AIPc. A similar trend has been noted with cell damage. The results demonstrate that F- provided protec- regard to mutagenic potential. In L5178Y cells, PFII tion from the killing of V79 cells by light and AlPc was found to be more mutagenic at the thymidine at all fluences at which light and AlPc reduced kinase locus than AlPc at equitoxic doses (Rerko cell survival. Thus, F- provided marked protection et al., 1992). Therefore, a one-to-one relationship against cell killing from a fluence of 9 k J h 2 up to between DNA damage and cytotoxicity does not the highest fluence tested, 24 kJ/m2. The results appear likely. also show that F- provided protection from the Whether or not the DNA damage induced by inhibition of [3H]thymidine incorporation caused by AlPc and light is directly responsible for cell killing, light and AlPc as well as from the production of the results show that nuclear events induced by DPC caused by light and AIPc. These results are AlPc and light can be inhibited by F-. Although not surprising in view of the previously observed protection provided by F- against the photo- the crosslinking of spectrin monomers in the human hemolysis of human erythrocytes produced by light erythrocyte membrane by the photodynamic action of AIPcS, is inhibited in the presence of F- (Benand AIPcS4 (Ben-Hur et al., 1991). For DPC production, protection was provided by Hur, 1991), study of human erythrocytes did not F- at fluences at which light and AlPc reduced cell permit analysis of photodynamic damage to organsurvival, but no significant protection was provided elles other than the plasma membrane. The present at those which were sub-lethal. For [3H]thymidine results show that the action of F- is not limited to incorporation, the protection of F- only occurred inhibition of photodynamic damage in the plasma at fluences which were supra-lethal (318 kJ/m2). membrane, presumably because F- can enter the There was little or no effect of F- on any of the cell during the course of irradiation (see above). PAP 55:2-F
EHUDBEN-HURet al.
236 Cause of effect of F-
The effect of F- on the action of light and AlPc could arise as a result of various mechanisms. Fcould scavenge singlet oxygen or oxygen radicals. However, there is no evidence to suggest that this is the case. The observation that photodynamic damage induced by PFII was not affected by Fstrongly indicates that it does not scavenge lo2,as PFII mediates its effect via '02 production (Weishaupt et a f . . 1976). It also could interact with AlPc and prevent it from causing damage to a critical cellular site. This appears to be a reasonable possibility, as it has been demonstrated that Freduces the binding of AIPcS., to proteins, inhibits the photobleaching of AlPc when the sensitizer is bound to bovine serum albumin, and interferes with the photodynamic crosslinking of spectnn monomers when sensitized by A1PcS4 (Ben-Hur et a l . , 1991b). These effects of F- are not due to inhibition of primary photochemical processes. Thus, Faffects neither AIPcS4-induced photodynamic oxidation of tryptophan (Ben-Hur ef al., 1991b) nor the life-time of its excited triplet (Ferraudi ef al., 1988). Therefore. it is important to consider the probable structure and the chemical properties of the actual photosensitizer that is active in the cells. This in turn requires consideration of the chemical properties of AlPcCl and the way in which it was used in treating cells. AlPcCl has long been known to be subject to hydrolysis to hydrated hydroxides (Barrett et al., 1936; Owen and Kenney, 1962). As a result, because the DMF used as a stock solution solvent can be assumed to have been wet (it was handled with ordinary techniques), and because the growth medium in which the cells were held while being exposed to the photosensitizer was mostly water, it appears probable that the active sensitizer is largely or completely A1Pc(OH)(H20) and that this compound is formed from AlPcCl according to the equation AlPcCl + 2Hz0 + AIPc(OH)(H20)
+ HCI
(1)
(Alternatively. it could be largely or completely a higher hydrate formed in a similar fashion.) AIPc(OH)( H 2 0 ) is presumed to contain octahedral Al with OH and H 2 0 ligands that are tram to each other. Although the kinetics of the exchange of the trans ligands on A1Pc(OH)(H20) have not been studied in any case. it is known that the ligands are labile (Owen and Kenney, 1962; Linsky et al., 1980). Furthmore. fluoroaluminium complexes, such as [AIF(H20)5]2+and [AIF,aq]-, exist in aqueous solution (Matwiyoff and Wageman, 1970). With this in mind. it seems probable that AIPc(OH)(H20) is converted into a fluoro complex in the presence of F - . Consistent with complex formation is the observation that AlPc derivatives display a 5 nm
blue shift in their excitation and emission spectra upon addition of F- (Ben-Hur et al., 1991b). Once the possibility is accepted that the sensitizer in the F--containing solution is largely a fluoro complex, reasonable explanations of the F- effect can be offered. For example, the effect could arise because the fluoro complex is insoluble and is aggregated inside the cell. Dye aggregates are photochemically inactive. Another possibility is that the binding of the complex to a critical cell site could be weakened and thus its ability to induce type I (electron transfer) reactions would be reduced (Ben-Hur et al., 1991b). A third possibility is that the complex has less favorable photophysical characteristics than AIPc(OH)(H20) and as a consequence is simply a poorer sensitizer. As discussed above, this does not appear to be the case. The earlier observation that F- must be added prior to the exposure of human erythrocytes to light in order to inhibit maximally the photohemolysis induced by light and AIPcS4 (Ben-Hur et al., 1991a) is consistent with the formation of such a fluoro complex. This complex appears to be very strong, as removal of F- prior to exposure of erythrocytes to light does not abolish its effect (E. Ben-Hur, unpublished). In this connection it should be pointed out that since AI-CI and Ga-CI bonds are generally readily hydrolyzed, the sensitizer effects commonly ascribed to AlPcCl and GaPcCl and their substituted derivatives probably arise in many cases from the corresponding hydrated hydroxides (e.g. in the cases of AlPcCl and GaPcCl from AIPc(OH)(H20) and GaPc(OH)(H20)). This distinction may be important because the strength and mode of bonding of the two types of sensitizers at the target sites and their photophysical characteristics probably are quite different. Our results show that there is little or no effect of F- on the photodynamic actions of PFII and SiPc. This is probably due to the lack of high affinity sites for F- in the two compounds. Finally, it should be noted that the F- effect could have basic as well as practical implications for PDT. On the basic level, F- protects selectively against AlPc photodynamic inactivation of some plasma membrane functions and could, therefore, serve as a probe for critical targets. On the practical side, F- can totally protect against A1PcS4-induced photohemolysis, but only a small protection by Fis seen against A1PcS4-inducedphotodynamic inactivation of enveloped viruses (Ben-Hur et al., 1992). Fluoride could therefore be used to improve the therapeutic ratio in treating blood for transfusion by AIPcS4-PDT for elimination of viruses, such as the human immunodeficiency virus (Horowitz ef al., 1991). CONCLUSIONS
The results of this study lead to the following four conclusions: (1) the active sensitizer in aqueous
Phthalocyanine and fluoride
237
phyrin. In Light in Biology and Medicine (Edited by R. H. Douglas, J. Moan and F. Dall’Acqua) Vol. 1, pp. 95-103. Plenum Press, New York. Brasseur, N., H. Ali, D. Autenrieth, R. Langlois and J. E. van Lier (1985) Biological activities of phthalocyanines. 111. Photoinactivation of V-79 Chinese hamster cells by tetrasulfophthalocyanines.Photochem. Photobiol. 42, 515-521. Chiu, S. M., L. R. Friedman, N. M. Sokany, L. Y. Xue and N. L. Oleinick (1986) Nuclear matrix proteins are crosslinked to transcriptionally active gene sequences by ionizing radiation. Radiat. Res. 107, 24-38. Evans, H. H., R. M. Rerko, J. Mend, M. E. Clay, A. R. Antunez and N. L. Oleinick (1989) Cytotoxic and Acknowledgements-This research was supported by mutagenic effects of the photodynamic action of chloroGrants PO1 CA-48735 and P30 CA-43703 from the aluminium phthalocyanine and visible light in L5178Y National Cancer Institute, DHHS, and by a generous gift cells. Photochem. Photobiol. 49, 43-48. from the Marguerite M. Wilson Foundation. Ferraudi, G., G. A. Arguello, H. Ali and J. E. van Lier (1988) Types I and I1 sensitized photooxidation of REFERENCES aminoacid by phthalocyanines: a flash photochemical study. Photochem. Photobiol. 47, 657-660. Barrett, P. A., C. E. Dent and R. P. Linstead (1936) Horowitz, B., B. Williams, S. Ryukin, A. M. Prince, D. Pascual, J. Geacintov and J. Valinsky (1991) InactiPhthalocyanine as a co-ordinating group. A general vation of viruses in blood with aluminium phthalocyinvestigation of the metallic derivatives. Phthalocyanine derivatives. Transfusion 31, 102-108. anines, Part VII. J . Chem. SOC.1719-1736. Ben-Hur, E. (1991) Basic photobiology and mechanisms Hunting, D. J., B. J. Gowans, N. Brasseur and J. E. van Lier (1987) DNA damage and repair following treatof action of phthalocyanines. In Photodynamic Therapy: ment of V79 cells with sulfonated phthalocyanines. PhoBasic Principles and Clinical Applications (Edited by T. rochem. Photobiol. 45, 769-773. J. Dougherty and B. W. Henderson). Marcel Dekker, Linsky, J. P., T.R. Paul, R. S. Nohr and M. E. Kenney New York. In press. (1980) Studies of a series of halo aluminium, gallium, Ben-Hur, E., T.M. A. R. Dubbelman and J. Van Stevenand indium phthalocyanines. Inorg. Chem. 19, inck (1991b) The effect of fluoride on binding and 3131-3135. photodynamic action of phthalocyanines with proteins. Matwiyoff, N. A. and W. E. Wageman (1970) Nuclear Photochem. Photobiol. 54, 703-707. magnetic resonance studies of aluminium (111) fluoride Ben-Hur, E., A. Freud, A. Canfi and A. Livne (1991a) ion complexes in aqueous solutions. Inorg. Chem. 9. The role of glycolysis and univalent ions in phthalocy1031-1 036. anine-sensitized photohemolysis of human erythrocytes. Oleinick, N. L., M. L. Aganval, A. R. Antunez, M. E . In[. J. Radial. Biol. 59, 797-806. Clay, H. H. Evans, E. J. Harvey, R. M. Rerko and Ben-Hur, E., T. Fujihara, F. Suzuki and M. M. Elkind L.-Y. Xue (1991) Effects of photodynamic treatment on (1987b) Genetic toxicology of the photosensitization of DNA. In Laser-Tissue Interactions, SPIE Vol. 1427. Chinese hamster cells by phthalocyanines. Photochem. 90-100. Photobiol. 45, 227-230. Ben-Hur, E., M. Green, A. Prager, R. Kol and I. Rosen- Owen, J. E. and M. E. Kenney (1962) Aluminium and silicon hydroxy and oxy systems of the phthalocyanino thal (1987a) Phthalocyanine photosensitization of mamtype. Inorg. Chem. 1, 334-336. malian cells: biochemical and ultrastructural effects. Ramakrishnan, N., M. E. Clay, L. Y. Xue, H. H. Evans, Photochem. Photobiol. 46, 651-656. A. R. Antunez and N. L. Oleinick (1988) Induction of Ben-Hur, E., R. C. Hoeben, H. van Ormondt, T.M. A. DNA-protein cross-links in Chinese hamster cells by R. Dubbelman and J. Van Steveninck (1992) Photodythe photodynamic action of chloroaluminium phthalocynamic inactivation of retroviruses by phthalocyanines: anine and visible light. Phofochem. Photobiol. 48, the effects of sulfonation, metal ligand and fluoride. J. 297-303. Photochem. Photobiol. B: Biol. In press. Ben-Hur, E., R. Kol. R. Marko, E. Riklis and I. Rosen- Ramakrishnan, N., N. L. Oleinick, M. E. Clay, M. F. Horng, A. R. Antunez and H. H. Evans (1989) DNA thal (1988) Combined action of phthalocyanine photolesions and DNA degradation in mouse lymphoma sensitization and gamma radiation on mammalian cells. L5178Y cells after photodynamic treatment sensitized Int. J . Radiat. Biol. 54, 21-30. by chloroaluminium phthalocyanine. Photochem. PhoBen-Hur, E. and A. Orenstein (1991) The endothelium tobiol. 50, 373-378. and red blood cells as potential targets in PDT-induced Rerko, R. M., M. E. Clay, A. R. Antunez, N. L. Oleinick vascular stasis. Int. J. Radiat. Biol. 60,293-301. and H. H. Evans (1992) Photofrin I1 photosensitization Ben-Hur, E. and I. Rosenthal(1985) The phthalocyanines: is mutagenic at the tk locus in mouse L5178Y cells. a new class of mammalian cells photosensitizers with a Photochem. Photobiol. 55, 75-80. potential for cancer phototherapy. Int. J . Radial. Biol. Spikes, J. E. (1986) Yearly review: phthalocyanines as 47, 145-147. photosensitizers in biological systems and for the photoBen-Hur, E. and I. Rosenthal (1986) Photohemolysis of dynamic therapy of tumors. Photochem. Photobiol. 43, human erythrocytes induced by aluminium phthalocy691-699. anine tetrasulfonate. Cancer Lett. 30, 321-327. Berg, K., E. Hovig and J. Moan (1988) Sister chromatid Weishaupt, K. R., C. J. Gomer and T. J. Dougherty (1976) Identification of singlet oxygen as the cytotoxic exchanges induced by photodynamic treatment of cells agent in photo-inactivation of a murine tumor. Cancer in the presence of Photofrin 11, aluminium phthalocyRes 36, 23262329. anine tetrasulfonate and tetra (3-hydroxyphenyl) por-
media treated with solutions prepared from AlPcCl is AIPc(OH)(H20) or a higher hydrate; (2) the cell killing and two biochemical responses caused by this sensitizer and light are strongly inhibited by F-; (3) the cell killing and the same two biochemical responses caused by light and either SiPc or PFII are not inhibited by F-; and (4) the inhibition of the photoresponses of AlPc by F- is not limited to the plasma membrane but is effective at intracellular sites.
Photochemistry and Photobiology Vol. 55, No. 2. pp. 231-237, 1992
Printed in Great Britain.
All
rights reserved
Copyright
0 1992 Pergamon Press plc
PROTECTION BY THE FLUORIDE ION AGAINST PHTHALOCYANINE-INDUCED PHOTODYNAMIC KILLING OF CHINESE HAMSTER CELLS EHUDBEN-HuRI*,MARIANE. CLAY^, EDUARDO F. VICIOSO~, ANTONIO R. A N T U N E Z ~ , BORISD. RlHTER4, MALCOLM E. KENNEY~* and NANCY L. OLEINICK~* 'Nuclear Research Center-Negev, P.O. Box 9001, Beer-Sheva 84190, Israel, ZDivisions of Biochemical Oncology and Radiation Oncology, Department of Radiology, Case Western Reserve University School of Medicine, Cleveland, OH 44106, USA, 3Department of Radiology and Medical Physics, Faculty of Medicine, University of Malaga, 29080 Malaga, Spain and "Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA (Received 15 April 1991; accepted 19 July 1991)
Abstract-When a dilute F- solution was added to a culture of Chinese hamster cells that had been preincubated with an aluminium phthalocyanine sensitizer derived from AIPcCI, the photosensitivity of the cells was markedly reduced compared to control cells not treated with F-. Under the same treatment conditions, the reduction in [3H]thymidine incorporation into cellular DNA caused by light and this sensitizer and the production of DNA-protein crosslinks caused by light and this sensitizer were also inhibited by F-. In contrast, the killing of Chinese hamster cells, the reduction of thymidine incorporation by the cells, and the production of DNA-protein crosslinks in the cells caused by the combination of light and either Photofrin I1 or the silicon phthalocyanine HOSiPcOSi(CH3)z(CHz)3N(CH3)2 were not inibited by F-. We conclude that the aluminium phthalocyanine sensitizer used is largely or completely AIPc(OH)(H,O), that it is converted to a fluoro complex by F-, and that this compound probably is a less efficient generator of photochemical damage at a critical cellular target(s) than is AIPc(OH)(H,O). The inhibition of thymidine incorporation and DNA-protein crosslink formation indicates that the effects of F- can be expressed at intracellular sites. It is further concluded that the silicon phthalocyanine sensitizer and Photofrin I1 do not interact significantly with F-.
INTRODUCTION
Phthalocyanines are efficient photosensitizers of cultured mammalian cells (Ben-Hur and Rosenthal, 1985; Brasseur et al., 1985). Because of this ability, as well as their intense absorption in the far red, the lack of cutaneous phototoxicity, and the ease with which they can be chemically modified, phthalocyanines are increasingly being studied as secondgeneration sensitizers for photodynamic therapy (PDT)? of cancer (Spikes, 1986; Ben-Hur, 1991). Phthalocyanine photosensitization of mammalian cells causes damage to many cellular structures, including the plasma membrane (Ben-Hur and Rosenthal, 1986), mitochondria (Ben-Hur et al., 1987a), and DNA. Damage to DNA can be expressed as alkali-labile sites, single-strand breaks, *To whom correspondence should be addressed. TAbbreviafions: AIPc, the aluminium phthalocyanine sensitizer formed when AlPcCl is dissolved in DMF; AIPcCl, chloroaluminium phthalocyanine; AIPcS,, chloroaluminium phthalocyanine tetrasulfonate isomer mixture; DMF, dimethylformamide; DPC, DNA-protein crosslinks; H,EDTA, ethylenediamine tetraacetic acid; HEPES, N-2-hydroxyethyl piperazineN-2-ethane sulfonic acid; HpD, hematoporphyrin derivative; PBS, phosphate-buffered saline; PDT, photodynamic therapy; PFII, Photofrin 11; SiPc, HOSiPcOSi(CH3),(CH2),N(CH3)2;SSC, standard saline citrate; TCA, tnchloroacetic acid; Tris, Tris-
(hydroxymethy1)aminornethane.
and DNA-protein crosslinks (DPC) (Ben-Hur et al., 1987b; Hunting et al., 1987; Ramakrishnan et al., 1988,1989). Although mutagenesis was not detected at either the hemizygous hprt (hypoxanthine guanine phosphoribosyl transferase) locus or the Na'/K'-ATPase locus after photosensitization of Chinese hamster V79 cells with an aluminium phthalocyanine (Ben-Hur et al., 1987b), mutations were found to be induced at the heterozygous tk (thymidine kinase) locus in two strains of L5178Y mouse lymphoma cells (Evans et al., 1989; Rerko et al., 1991; Oleinick er al., 1991). Recently, it was found that photohemolysis induced in human erythrocytes with a sulfonated aluminium phthalocyanine is markedly inhibited by F(Ben-Hur et al., 1991a). Addition of F- immediately prior to irradiation of cells pretreated with the sensitizer was sufficient for complete inhibition of the photohemolysis. Maximal inhibition of photohemolysis required that F- be present during light exposure. It was concluded that F- interferes with an early step in the photochemical mechanism inducing hemolysis. Because photohemolysis results from membrane damage, it was hypothesized that F- exerts its action at the membrane. Indeed, crosslinking of spectrin monomers in erythrocyte ghosts photosensitized with AlPcS4 is inhibited by F- (Ben-Hur and Orenstein, 1991). However, in spite of what is known, a complete account of the cellular events that are associated
23 1
EHUDBEN-HURer al.
232
with the photocytotoxicity of phthalocyanines cannot be given yet. I n an effort to further understand the effect of F- o n the photocytotoxicity of phthalocyanines and thus the larger problem of the mechanisms of the phototoxicity, we have treated V79 fibroblasts with light and AIPc, SiPc, or PFII as sensitizers. W e have asked whether or not the effect of F- is limited t o erythrocytes, whether or not the photodynamic action of photosensitizers other than aluminium phthalocyanines can also be inhibited by the presence of F-, and whether or not the effects of F- can be expressed a t intracellular sites removed from the plasma membrane. W e have found that light and any o n e of the three sensitizers killed t h e cells, reduced the incorporation of [3H]thymidine into DNA, and caused the production of DNAprotein crosslinks in t h e cells. With AIPc, Finhibited all three effects. MATERIALS AND METHODS
Cell culture. Chinese hamster V79 lung fibroblasts were grown in McCoy's 5A medium (Gibco Laboratories, Grand Island, NY) augmented with 10% calf serum and HEPES to a final concentration of 20 mM, pH 7.4. Cultures were routinely incubated in a humidified atmosphere of 5% co2/95% air, at 37°C. For experiments, cultures were seeded with an appropriate number of cells, as described below, and incubated for 4 h to allow for cell attachment before the addition of photosensitizers. Photorensitizers. Photofrin 11 (Photofrin Polyporphyrin, PFII) was obtained as a lyophilized powder (QLT Phototherapeutics, Inc., Vancouver, BC, Canada). A solution of 2.5 mg/mL was prepared in 5 % dextrose, sterilized by filtration, and used immediately. AlPcCl was obtained as a crystalline solid (Eastman Kodak, Rochester. NY) and was recrystallized from 1-chloronaphthalene. A stock sensitizer solution was prepared from it by dissolving it in DMF at a rate of 1 mmoUL. This solution retained its full photosensitizing capacity for at least 1 month. The sensitizer in this solution is referred to as AIPc. The silicon phthalocyanine [HOSiPcOSi(CH,),(CH,),N(CH,),I ( S i c ) was synthesized as described elsewhere (Oleinick el a / . , in preparation). A 0.5 m M stock solution of it in DMF was prepared and stored at 4°C. It retained its full photosensitizing capacity for at least several months when stored thus. but lost capacity over a period of days when brought to room temperature. Cultures were treated with AIPc. SiPc, or PFII by adding an appropriate aliquot (1 kL, 1 kL, or 10 kL, respectively) of the stock solution to each mL of medium overlying the cells and returning the culture dishes to the incubator for an additional 1120 h. Control cultures not treated with a photosensitizer were included in each experiment. Cell irradiation procedure. The cells which had been treated with PFII were irradiated by placing the tissue culture dish containing them on a glass exposure tray located 72.5 cm above a 500 W tungsten-halogen lamp light source for periods of 2-20 min. The fluence rate at the level of the cell monolayers was -88 W/m2. The cells which had been treated with the other two sensitizers were irradiated in the same way except that irradiation times of 1.5-10 min were used, and the light was passed through a filter (Lee primary red filter No. 106, Vincent Lighting, Cleveland, OH) that removed light of wavelengths SiPc > AlPc. (Hunting et al., 1987), AIPcS, is not clastogenic, neither sister chromatid exchanges (Berg el al., DISCUSSION 1988) nor micronuclei (Ben-Hur et al., 1988) being produced by its photodynamic action. However, the Effect of Fobservation that the photodynamic effects of AlPc The results presented show that under the exper- are mutagenic in certain strains of L5178Y cells imental conditions employed here, light and AlPc (Evans et al., 1989; Oleinick et al., 1991) suggests were cytotoxic at fluences >6 kJ/m2, that they that DNA damage can have a primary influence on inhibited ["Hlthymidine incorporation at fluences of the viability of cells under some conditions. It is 3 3 kJ/m2, and that they produced DPC at fluences noteworthy that the sensitizers differed considerably of 2 3 kJ/rn2. Thus the inhibition of [3H]thymidine in their efficiency in producing DNA damage at incorporation and the production of DPC caused by equitoxic fluences, i. e. the photodynamic effects low light levels and AlPc appear to reflect sub-lethal were greatest with PFII, less with SiPc, and least with AIPc. A similar trend has been noted with cell damage. The results demonstrate that F- provided protec- regard to mutagenic potential. In L5178Y cells, PFII tion from the killing of V79 cells by light and AlPc was found to be more mutagenic at the thymidine at all fluences at which light and AlPc reduced kinase locus than AlPc at equitoxic doses (Rerko cell survival. Thus, F- provided marked protection et al., 1992). Therefore, a one-to-one relationship against cell killing from a fluence of 9 k J h 2 up to between DNA damage and cytotoxicity does not the highest fluence tested, 24 kJ/m2. The results appear likely. also show that F- provided protection from the Whether or not the DNA damage induced by inhibition of [3H]thymidine incorporation caused by AlPc and light is directly responsible for cell killing, light and AlPc as well as from the production of the results show that nuclear events induced by DPC caused by light and AIPc. These results are AlPc and light can be inhibited by F-. Although not surprising in view of the previously observed protection provided by F- against the photo- the crosslinking of spectrin monomers in the human hemolysis of human erythrocytes produced by light erythrocyte membrane by the photodynamic action of AIPcS, is inhibited in the presence of F- (Benand AIPcS4 (Ben-Hur et al., 1991). For DPC production, protection was provided by Hur, 1991), study of human erythrocytes did not F- at fluences at which light and AlPc reduced cell permit analysis of photodynamic damage to organsurvival, but no significant protection was provided elles other than the plasma membrane. The present at those which were sub-lethal. For [3H]thymidine results show that the action of F- is not limited to incorporation, the protection of F- only occurred inhibition of photodynamic damage in the plasma at fluences which were supra-lethal (318 kJ/m2). membrane, presumably because F- can enter the There was little or no effect of F- on any of the cell during the course of irradiation (see above). PAP 55:2-F
EHUDBEN-HURet al.
236 Cause of effect of F-
The effect of F- on the action of light and AlPc could arise as a result of various mechanisms. Fcould scavenge singlet oxygen or oxygen radicals. However, there is no evidence to suggest that this is the case. The observation that photodynamic damage induced by PFII was not affected by Fstrongly indicates that it does not scavenge lo2,as PFII mediates its effect via '02 production (Weishaupt et a f . . 1976). It also could interact with AlPc and prevent it from causing damage to a critical cellular site. This appears to be a reasonable possibility, as it has been demonstrated that Freduces the binding of AIPcS., to proteins, inhibits the photobleaching of AlPc when the sensitizer is bound to bovine serum albumin, and interferes with the photodynamic crosslinking of spectnn monomers when sensitized by A1PcS4 (Ben-Hur et a l . , 1991b). These effects of F- are not due to inhibition of primary photochemical processes. Thus, Faffects neither AIPcS4-induced photodynamic oxidation of tryptophan (Ben-Hur ef al., 1991b) nor the life-time of its excited triplet (Ferraudi ef al., 1988). Therefore. it is important to consider the probable structure and the chemical properties of the actual photosensitizer that is active in the cells. This in turn requires consideration of the chemical properties of AlPcCl and the way in which it was used in treating cells. AlPcCl has long been known to be subject to hydrolysis to hydrated hydroxides (Barrett et al., 1936; Owen and Kenney, 1962). As a result, because the DMF used as a stock solution solvent can be assumed to have been wet (it was handled with ordinary techniques), and because the growth medium in which the cells were held while being exposed to the photosensitizer was mostly water, it appears probable that the active sensitizer is largely or completely A1Pc(OH)(H20) and that this compound is formed from AlPcCl according to the equation AlPcCl + 2Hz0 + AIPc(OH)(H20)
+ HCI
(1)
(Alternatively. it could be largely or completely a higher hydrate formed in a similar fashion.) AIPc(OH)( H 2 0 ) is presumed to contain octahedral Al with OH and H 2 0 ligands that are tram to each other. Although the kinetics of the exchange of the trans ligands on A1Pc(OH)(H20) have not been studied in any case. it is known that the ligands are labile (Owen and Kenney, 1962; Linsky et al., 1980). Furthmore. fluoroaluminium complexes, such as [AIF(H20)5]2+and [AIF,aq]-, exist in aqueous solution (Matwiyoff and Wageman, 1970). With this in mind. it seems probable that AIPc(OH)(H20) is converted into a fluoro complex in the presence of F - . Consistent with complex formation is the observation that AlPc derivatives display a 5 nm
blue shift in their excitation and emission spectra upon addition of F- (Ben-Hur et al., 1991b). Once the possibility is accepted that the sensitizer in the F--containing solution is largely a fluoro complex, reasonable explanations of the F- effect can be offered. For example, the effect could arise because the fluoro complex is insoluble and is aggregated inside the cell. Dye aggregates are photochemically inactive. Another possibility is that the binding of the complex to a critical cell site could be weakened and thus its ability to induce type I (electron transfer) reactions would be reduced (Ben-Hur et al., 1991b). A third possibility is that the complex has less favorable photophysical characteristics than AIPc(OH)(H20) and as a consequence is simply a poorer sensitizer. As discussed above, this does not appear to be the case. The earlier observation that F- must be added prior to the exposure of human erythrocytes to light in order to inhibit maximally the photohemolysis induced by light and AIPcS4 (Ben-Hur et al., 1991a) is consistent with the formation of such a fluoro complex. This complex appears to be very strong, as removal of F- prior to exposure of erythrocytes to light does not abolish its effect (E. Ben-Hur, unpublished). In this connection it should be pointed out that since AI-CI and Ga-CI bonds are generally readily hydrolyzed, the sensitizer effects commonly ascribed to AlPcCl and GaPcCl and their substituted derivatives probably arise in many cases from the corresponding hydrated hydroxides (e.g. in the cases of AlPcCl and GaPcCl from AIPc(OH)(H20) and GaPc(OH)(H20)). This distinction may be important because the strength and mode of bonding of the two types of sensitizers at the target sites and their photophysical characteristics probably are quite different. Our results show that there is little or no effect of F- on the photodynamic actions of PFII and SiPc. This is probably due to the lack of high affinity sites for F- in the two compounds. Finally, it should be noted that the F- effect could have basic as well as practical implications for PDT. On the basic level, F- protects selectively against AlPc photodynamic inactivation of some plasma membrane functions and could, therefore, serve as a probe for critical targets. On the practical side, F- can totally protect against A1PcS4-induced photohemolysis, but only a small protection by Fis seen against A1PcS4-inducedphotodynamic inactivation of enveloped viruses (Ben-Hur et al., 1992). Fluoride could therefore be used to improve the therapeutic ratio in treating blood for transfusion by AIPcS4-PDT for elimination of viruses, such as the human immunodeficiency virus (Horowitz ef al., 1991). CONCLUSIONS
The results of this study lead to the following four conclusions: (1) the active sensitizer in aqueous
Phthalocyanine and fluoride
237
phyrin. In Light in Biology and Medicine (Edited by R. H. Douglas, J. Moan and F. Dall’Acqua) Vol. 1, pp. 95-103. Plenum Press, New York. Brasseur, N., H. Ali, D. Autenrieth, R. Langlois and J. E. van Lier (1985) Biological activities of phthalocyanines. 111. Photoinactivation of V-79 Chinese hamster cells by tetrasulfophthalocyanines.Photochem. Photobiol. 42, 515-521. Chiu, S. M., L. R. Friedman, N. M. Sokany, L. Y. Xue and N. L. Oleinick (1986) Nuclear matrix proteins are crosslinked to transcriptionally active gene sequences by ionizing radiation. Radiat. Res. 107, 24-38. Evans, H. H., R. M. Rerko, J. Mend, M. E. Clay, A. R. Antunez and N. L. Oleinick (1989) Cytotoxic and Acknowledgements-This research was supported by mutagenic effects of the photodynamic action of chloroGrants PO1 CA-48735 and P30 CA-43703 from the aluminium phthalocyanine and visible light in L5178Y National Cancer Institute, DHHS, and by a generous gift cells. Photochem. Photobiol. 49, 43-48. from the Marguerite M. Wilson Foundation. Ferraudi, G., G. A. Arguello, H. Ali and J. E. van Lier (1988) Types I and I1 sensitized photooxidation of REFERENCES aminoacid by phthalocyanines: a flash photochemical study. Photochem. Photobiol. 47, 657-660. Barrett, P. A., C. E. Dent and R. P. Linstead (1936) Horowitz, B., B. Williams, S. Ryukin, A. M. Prince, D. Pascual, J. Geacintov and J. Valinsky (1991) InactiPhthalocyanine as a co-ordinating group. A general vation of viruses in blood with aluminium phthalocyinvestigation of the metallic derivatives. Phthalocyanine derivatives. Transfusion 31, 102-108. anines, Part VII. J . Chem. SOC.1719-1736. Ben-Hur, E. (1991) Basic photobiology and mechanisms Hunting, D. J., B. J. Gowans, N. Brasseur and J. E. van Lier (1987) DNA damage and repair following treatof action of phthalocyanines. In Photodynamic Therapy: ment of V79 cells with sulfonated phthalocyanines. PhoBasic Principles and Clinical Applications (Edited by T. rochem. Photobiol. 45, 769-773. J. Dougherty and B. W. Henderson). Marcel Dekker, Linsky, J. P., T.R. Paul, R. S. Nohr and M. E. Kenney New York. In press. (1980) Studies of a series of halo aluminium, gallium, Ben-Hur, E., T.M. A. R. Dubbelman and J. Van Stevenand indium phthalocyanines. Inorg. Chem. 19, inck (1991b) The effect of fluoride on binding and 3131-3135. photodynamic action of phthalocyanines with proteins. Matwiyoff, N. A. and W. E. Wageman (1970) Nuclear Photochem. Photobiol. 54, 703-707. magnetic resonance studies of aluminium (111) fluoride Ben-Hur, E., A. Freud, A. Canfi and A. Livne (1991a) ion complexes in aqueous solutions. Inorg. Chem. 9. The role of glycolysis and univalent ions in phthalocy1031-1 036. anine-sensitized photohemolysis of human erythrocytes. Oleinick, N. L., M. L. Aganval, A. R. Antunez, M. E . In[. J. Radial. Biol. 59, 797-806. Clay, H. H. Evans, E. J. Harvey, R. M. Rerko and Ben-Hur, E., T. Fujihara, F. Suzuki and M. M. Elkind L.-Y. Xue (1991) Effects of photodynamic treatment on (1987b) Genetic toxicology of the photosensitization of DNA. In Laser-Tissue Interactions, SPIE Vol. 1427. Chinese hamster cells by phthalocyanines. Photochem. 90-100. Photobiol. 45, 227-230. Ben-Hur, E., M. Green, A. Prager, R. Kol and I. Rosen- Owen, J. E. and M. E. Kenney (1962) Aluminium and silicon hydroxy and oxy systems of the phthalocyanino thal (1987a) Phthalocyanine photosensitization of mamtype. Inorg. Chem. 1, 334-336. malian cells: biochemical and ultrastructural effects. Ramakrishnan, N., M. E. Clay, L. Y. Xue, H. H. Evans, Photochem. Photobiol. 46, 651-656. A. R. Antunez and N. L. Oleinick (1988) Induction of Ben-Hur, E., R. C. Hoeben, H. van Ormondt, T.M. A. DNA-protein cross-links in Chinese hamster cells by R. Dubbelman and J. Van Steveninck (1992) Photodythe photodynamic action of chloroaluminium phthalocynamic inactivation of retroviruses by phthalocyanines: anine and visible light. Phofochem. Photobiol. 48, the effects of sulfonation, metal ligand and fluoride. J. 297-303. Photochem. Photobiol. B: Biol. In press. Ben-Hur, E., R. Kol. R. Marko, E. Riklis and I. Rosen- Ramakrishnan, N., N. L. Oleinick, M. E. Clay, M. F. Horng, A. R. Antunez and H. H. Evans (1989) DNA thal (1988) Combined action of phthalocyanine photolesions and DNA degradation in mouse lymphoma sensitization and gamma radiation on mammalian cells. L5178Y cells after photodynamic treatment sensitized Int. J . Radiat. Biol. 54, 21-30. by chloroaluminium phthalocyanine. Photochem. PhoBen-Hur, E. and A. Orenstein (1991) The endothelium tobiol. 50, 373-378. and red blood cells as potential targets in PDT-induced Rerko, R. M., M. E. Clay, A. R. Antunez, N. L. Oleinick vascular stasis. Int. J. Radiat. Biol. 60,293-301. and H. H. Evans (1992) Photofrin I1 photosensitization Ben-Hur, E. and I. Rosenthal(1985) The phthalocyanines: is mutagenic at the tk locus in mouse L5178Y cells. a new class of mammalian cells photosensitizers with a Photochem. Photobiol. 55, 75-80. potential for cancer phototherapy. Int. J . Radial. Biol. Spikes, J. E. (1986) Yearly review: phthalocyanines as 47, 145-147. photosensitizers in biological systems and for the photoBen-Hur, E. and I. Rosenthal (1986) Photohemolysis of dynamic therapy of tumors. Photochem. Photobiol. 43, human erythrocytes induced by aluminium phthalocy691-699. anine tetrasulfonate. Cancer Lett. 30, 321-327. Berg, K., E. Hovig and J. Moan (1988) Sister chromatid Weishaupt, K. R., C. J. Gomer and T. J. Dougherty (1976) Identification of singlet oxygen as the cytotoxic exchanges induced by photodynamic treatment of cells agent in photo-inactivation of a murine tumor. Cancer in the presence of Photofrin 11, aluminium phthalocyRes 36, 23262329. anine tetrasulfonate and tetra (3-hydroxyphenyl) por-
media treated with solutions prepared from AlPcCl is AIPc(OH)(H20) or a higher hydrate; (2) the cell killing and two biochemical responses caused by this sensitizer and light are strongly inhibited by F-; (3) the cell killing and the same two biochemical responses caused by light and either SiPc or PFII are not inhibited by F-; and (4) the inhibition of the photoresponses of AlPc by F- is not limited to the plasma membrane but is effective at intracellular sites.
