against Photosensitization by Hematoporphyrin ...

Protective Effect of Cholesterol on Friend Leukemic Cells against Photosensitization by Hematoporphyrin Derivative S. Salzberg, F. Lejbkowicz, B. Ehrenberg, et al. Cancer Res 1985;45:3305-3310. Published online July 1, 1985.

Updated Version

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/45/7/3305

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].

Downloaded from cancerres.aacrjournals.org on July 10, 2011 Copyright © 1985 American Association for Cancer Research

[CANCER

RESEARCH

45, 3305-3310,

July 1985]

ProtectiveEffect of Cholesterolon Friend LeukemicCells against Photosensitizationby HematoporphyrinDerivative S. Salzberg,1 F. Lejbkowicz,

B. Ehrenberg,

and Z. Malik

Departments of Life Sciences [S. S., F. L, Z. M.Â¡and Physics [B. Â£.],Bar-Han University, Ramat-Gan 52 100, Israel

ABSTRACT The cytotoxic effects of hematoporphyrin derivative (HPD) on Friend erythroleukemic cells were studied. Upon binding of the porphyrin to the cells, the fluorescence spectra was shifted from 613 to 633 nm regarding the main band and from 676 to 667 nm concerning the secondary band. The kinetics of HPD binding was then determined. Maximum binding already occurred at 60 s after exposure of the cells to HPD. It could be demonstrated that the effect of the photoactivated HPD on cell viability was drug, dose, and light fluorescence dependent. Cellular protein synthesis and Friend virus complex release from the cells were equally inhibited by the photodynamic sensitization of the drug, indicating no specific effect on virus maturation. Since cholesterol affects the fluidity of cell membranes, it was important to study the effect of cholesterol enrichment on the photodynamic sensi tization by HPD. It was found that, while a 50% reduction in protein synthesis was monitored following treatment with 20 ng of HPD per ml and illumination by a 6-milliwatt white light for 60 s, no inhibition was observed following preenrichment of the cells with 0.5,1, or 2% of cholesterol hemisuccinate. The same trend of cholesterol protection was demonstrated with longer illumi nation periods up to 10 min. The protective effect of cholesterol hemisuccinate was also seen using scanning electron micros copy. It is thus concluded that the cholesterol hemisuccinate content of Friend erythroleukemic cell membranes is an important factor in regulating the cytotoxicity of photoactivated HPD.

INTRODUCTION The effectiveness of HPD2 PDT of malignant tumors in humans was shown by numerous reports (4, 7, 8, 11, 13, 24). Clinical studies clearly show the advantages of HPD PDT, which de stroys the cancer tissue without damaging the surrounding nor mal cells. In addition, the potential of HPD fluorescence to localize even small mÃ©tastases improves the success of the treatment (14, 24). The side effects of orthodox cancer treatment make the HPD PDT a preferential choice for medical cancer manage ment. Nevertheless, the empiric evidence of HPD localization and sensitization in transformed cells has not yet been explained. Some problems underlying the porphyrin effects still remain open: (a) the cytochemical basis for the preferential binding of HPD to cancer cells; (o) the retention potential for HPD of these cells, making possible the photodynamic killing; (c) whether the active fraction of HPD is processed intracellulariy; and (d)

whether macrophages or other nonmalignant cells are involved in the transfer of HPD to the malignant cells or in the killing effect. The low cholesterol content of leukemic cells determines a higher membrane fluidity, which is one main intrinsic difference of leukemic, as compared to normal, cells (3). In a previous study, chronic human lymphocytic leukemia lymphocytes were compared to normal cells. The findings have shown that the leukemic cells bound more 55Fe-protoporphyrin and were more sensitive to the cytotoxic effect of protoporphyrin (15). Further more, our preliminary results have indicated that FLC acquired resistance to HPD sensitization upon preincubation with choles terol-enriched media (18). In the present investigation, we ex panded our study on the cytotoxic effect of photoactivated HPD on these erythroleukemic cells. In addition, the possible resist ance of cholesterol-enriched cells, which mimics the structure of normal cells to HPD photosensitization, was also investigated.

MATERIALS AND METHODS Cells. FLC, line 745 (18), were grown in DMEM (Grand Island Biolog ical Co., Grand Island, NY) and supplemented with 10% NCS (Seralab, Crawly Down, United Kingdom). The cells were grown in 5-cm tissue culture dishes or in 10-ml flasks (NUNC, Roskilde, Denmark), in a humidified incubator enriched with 10% C02 at 37Â°C. The cells were subdivided twice a week by resuspension in fresh medium at a concen tration of approximately 1 x 106 cells/ml. Determination of Viable Cell Number. Cells were centrifugad at low speed (1000 x g for 5 min) and resuspended in DMEM (Grand Island Biological Co.) or PBS (140 mu NaCI-2.7 mm KCL-0.9 HIM CaCl2-0.06 rriM MgClz-8.0 rriM Na2HP04-1.5 mw KH2POÂ»3*~

760

X(nm)

Chart 1. Fluorescence spectrum (uncorrected) of HPD in water and bound to FLC. The fluorescence spectrum of HPD (25 ^g/ml) in water was measured first following excitation by an Ar* laser at 457 nm. FLC at 5 x 10s cells/ml were then added, and a second spectrum was taken. The excitation wavelength was 457 nm.

by centrifugation, and the amount of the FLV complex (2) present in the culture medium was determined by adding 1O-^l samples of the medium to 90 /il of a reaction mixture containing (final concentrations) 50 rriM Tris-hydrochloride (pH 8.2), 50 mw dithiothreitol (Sigma), 100 HIM NaCI, 0.5 rnw MnCI2, 0.1 mw each of dATP, dCTP, and dGTP (Sigma), 50 MUÃ• of [3H]dTTP per ml (50 Ci/mmol; Amersham Int., Amersham, United Kingdom), 0.03% (v/v) Nonidet P-40 (Fluka, A. G. Buchs, Switzerland), and 10 tÂ¡gof polyriboadenylic acid-oligodeoxythymidylic acid per ml (Pharmacia, PL-Biochemicals, Uppsala, Sweden). The mixture was in cubated at 37Â°Cfor 45 min, and 50-jil samples were spotted on DE-81 disk filters (Whatman, Maidstone, Kent, United Kingdom) and washed 5 times with 5% (w/v) Na2HPO4 and twice with double distilled water. The filters were dried, and radioactivity was determined in a toluene-based scintillation liquid. SEM. The cells were taken after the appropriate treatment with CHS, HPD, or both; washed with PBS, and fixed in 1% glutaraldehyde in phosphate buffer (pH 7.4). The cells were fixed with poly-L-lysine-coated glass coverslips and dehydrated with graded alcohol and Freon. After critical point drying, the cells were coated with gold and examined by a JSM-35 scanning electron microscope. 0.5

RESULTS

Time (mio)

Kinetics of HPD Binding to FLC. Since our study concerns the biological effects of HPD on FLC, it was important first to

Chart 2. Kinetics of HPD binding to FLC. The variation in the fluorescence intensity of HPD (25 ng/ml) was taken at 633 nm at room temperature, after the addition of 5 x 10Â«FLC per ml.

CANCER RESEARCH VOL. 45 JULY 1985 3306


CHS PROTECTION

OF FLC FROM HPD PHOTOSENSITIZATION

Effect of HPD on Cell Viability. Following the establishment of the kinetics of HPD binding to FLC, it was essential to determine next the effect of photoactivated HPD on the viability of FLC. Thus, cells were treated with various concentrations of HPD, followed by photoactivation for 3 different time periods. Immediately thereafter, the number of live cells was determined. Chart 3a demonstrates that both HPD concentration and the time of illumination are important factors affecting the number of viable cells. An immediate effect can be noticed which appears to be constant between 1 and 5 ^g of HPD per ml, followed by a gradual decrease up to 50 M9/ml. The results also indicate that a significant effect of illumination on cell viability occurs already when HPD-treated cells are exposed to light for 10 min. However, exposure to a longer period of time somewhat increases the susceptibility of cells to photoactivated HPD, with no significant difference observed between exposure times of 20 and 30 min. Another parameter for cell viability, which also represents the expression of cellular genes, is the rate of cell protein synthesis. In order to further strengthen the results obtained on the effect of photoactivated HPD on the number of viable cells, its effect on the incorporation of [3H]leucine into acid-insoluble material of HPD-treated cells was also determined. As illustrated in Chart 3b, similar results to those obtained on cell number (Chart 3a) are seen. A gradual decrease in the incorporation as HPD concentration increases is clearly demonstrated with the 3 dif ferent times of illumination used. Effect of HPD on the Release of FLV Complex. FLC harbors a virus complex which is constantly released from the cells to

the culture medium and can be monitored by the reverse transcriptase assay (13). It could be argued that photoactivated HPD inhibits specifically the release of the virus complex even under conditions where the cells remain practically unaffected, e.g., when cells are treated with a low concentration of HPD. We therefore established the effect of HPD on the release of the virus complex by determining reverse transcriptase activity in the culture medium of HPD-treated cells. The results presented in Chart 4 indicate that there is apparently no specific inhibitory effect of photoactivated HPD on virus release, since a similar pattern of HPD effect is observed on virus release as seen on cell viability. This is clearly demonstrated in Chart 5, where the ratio between reverse transcriptase activity and cell viability as

I

5

IO

50

HPD Concentration

IOO

(/ig/ml)

Chart 4. HPD effect on FLV complex release. Cells were treated with HPD and illuminated as described in Chart 3. The amount of virus released into the culture medium was measured by the reverse transcriptase (flT) assay. Reverse transcrip tase activity detected in the medium of control cultures was 8x10* cpm.

F 6

-t

b

Â¡

5

1-

IO

HPD Concentration

50

100

(Â¿ig/ml)

Chart 5. Ratio between the effects of HPD on virus release and cell viability. The percentage of inhibition of virus release as determined by reverse transcriptase (RT) activity was calculated from Chart 4 and was divided by the percentage of inhibition of cell viability calculated from Chart 3a. Symbols represent different illumination times, as described in Chart 3.

zz

â€” 3 1 S 2

I

9

10

HPD Concentration

SO

100

(Â¿tg/ml )

Charts. HPD effect on cell viability. FLC were treated with the indicated concentrations of HPD and immediately thereafter illuminated for either 10 min (Ã€), 20 min (â€¢). or 30 min (â€¢).Viable cell number (a) or incorporation of [3H]leucine into parallel cell cultures (b) was then determined. The number of viable cells and the amount of [3H|leucine incorporation in untreated (control) cultures were 8x10* cells/ml and 2.2 x 10* cpm, respectively.

01234 CHS Concentrotion(%) Charte. CHS effect on cell viability. Cells were treated with the indicated concentrations of CHS for either 1 (â€¢), 2 (â€¢), or 3 (A) h. The number of viable cells was then determined.

CANCER RESEARCH VOL. 45 JULY 1985 3307


CHS PROTECTION


is by itself toxic to FLC. SEM visualized the protective effect of CHS enrichment (Fig. 1). While untreated FLC revealed a typical membrane structure including microvilli (a), a damaged membrane with differently sized holes is apparent in photoactivated HPD-treated cells (b). No change is seen in FLC treated only with CHS (c), and a clear protective effect is observed with cells pretreated with CHS followed by sensitization with photoactivated HPD (d). DISCUSSION I1 IO' 20130' 05%CHS

1 IO120' 50'

I1 IO' 20' 30'

I%CHS 2% CHS CHS Concentration

10' 20' 30' 4%CHS

Chart 7. Protective effect of CHS against the photodynamic activity of HPD. FLC were treated with different concentrations of CHS for 90 min prior to addition of HPD (20 M9/ml). The cells were then illuminated for the indicated time periods, followed by the addition of [3H]leucine for 1 h. Incorporation of radioactivity into acid-insoluble material was monitored. D, percentage of inhibition of [3H]leucine incorporation obtained by HPD treatment following pretreatment with CHS.

alone; â€¢inhibition obtained by HPD

a function of HPD concentration is presented. The ratio remains constant at all concentrations used. We thus conclude that the effect of HPD on virus release is probably the result of the reduced cellular activity induced by photoactivated HPD. Effect of CHS on Cell Viability. CHS has been proven to alter the fluidity of cell membranes, thus resulting in a conformational change of a transformed cell membrane to a more normal state (21). It was therefore interesting to investigate the possible effect of pretreatment of FLC with CHS on the damage caused to these cells by photoactivated HPD. However, before such an experiment could be performed, it was essential to establish first the direct effect of CHS on cell viability. Cells were treated with different concentrations of CHS for different time periods at 4Â°C, and after washing, the number of viable cells was determined. The results illustrated in Chart 6 indicate that there is no effect on cell viability up to a concentration of 2% CHS, and only at 4% CHS is a reduction of 22% in cell number observed. Protective Effect of CHS from the Reduced Cellular Activity Induced by HPD. Based on the results presented in the previous paragraph, FLC were treated for 90 min with 4 different concen trations of CHS at 4Â°C,followed by treatment with 20 Â¿ig of HPD per ml. The cells were then illuminated for different time periods, and the number of viable cells was immediately thereafter deter mined. The results summarized in Chart 7 demonstrate that, under the appropriate conditions, CHS pretreatment can protect FLC from the reduced cellular activity induced by photoactivated HPD. This is clearly shown in the cases where CHS has been used at concentrations ranging from 0.5 to 2%. Maximum pro tections occurred when cells were treated with HPD, illuminated for 1 min, and immediately washed, and then protein synthesis was determined. Under these conditions, an anticellular effect is already evident (about 50% inhibition). However, no inhibition is seen following pretreatment with 1 or 2% CHS. Good protection is also demonstrated when treated cells are illuminated for 10 min, but only slight protection occurs at an illumination time of 30 min, probably as a result of the high toxicity induced by photoactivated HPD under these conditions. It is interesting to note that no protection is visible following pretreatment by 4% CHS. This may be due to the fact that this concentration of CHS CANCER

The present study demonstrates that cholesterol-enriched FLC are resistant to the initial anticellular effect of photoactivated HPD, as determined by using both biological and SEM methods. Ultrastructurally, the HPD-treated leukemic cells revealed a dam aged cytoplasmic membrane by SEM, whereas the cholesterolenriched cells remained unaltered in comparison to controls. The photodynamic killing effect of HPD was drug dose and light fluence dependent. Our results thus demonstrate that the cho lesterol cell content is an important factor in the photosensitization of leukemic cells exposed to prophyrin. It should be empha sized that, in the case of human lymphocytic leukemia diseases, the leukemic cells were found to be poor in cholesterol in comparison to nonleukemic cells (3).Cholesterol is known to regulate the membrane fluidity of cells (21) and thus affect the vertical displacement of the intramembranal proteins. Indeed, chronic leukemia lymphocytes were highly sensitive to the cytotoxic effect of protoporphyrin and hemin in comparison to normal lymphocytes (15). Thus, the present study mimics the situation shown to occur with human chronic lymphatic leukemia lympho cytes. The anticellular mechanism of the porphyrin-induced photodynamic effect is still not absolutely clear. There are some indications that this effect is due to the formation of singlet oxygen (25), which may damage membranes (17), cross-link membranal proteins (9), alter enzymatic activities (16), and affect the Na+-K+-ATPase pump (5). Although singlet oxygen is gen erally assumed to attack protein molecules, another oxygÃ©nation target could be unsaturated fatty acids and cholesterol (6, 23). Indeed, the photooxidation product of cholesterol in artificial membranes exposed to hematoporphyrin has been demon strated to be 3|8-hydroxy-5a-cholest-6-ene-hydroperoxide (23). Furthermore, it has been described that the oxygÃ©nation of cholesterol by singlet oxygen in these membranes was affected by both membrane fluidity and the amount of hematoporphyrin incorporated into liposomal membrane. Hematoporphyrin incor poration was decreased on increasing the ratio of cholesterol to phospholipid membrane. In addition, it was demonstrated that the optimal formation of singlet oxygen product depends largely on the ratio of cholesterol to phospholipid (23). From the above studies, it emerges that singlet oxygen attacks at least 2 classes of molecules present in the natural membrane: proteins and lipids. Cross-linking of proteins will most likely lead to a profound alteration in their biological activity and may produce an immediate lethal effect, as generally expressed by protein synthesis inhibition. On the other hand, it is conceivable that partial degradation of lipids and even cholesterol will not stop cellular metabolism as quickly as in the case of biologically active proteins. Thus, an enriched environment with cholesterol can be competitive to proteins as acceptor targets for singlet

RESEARCH VOL. 45 JULY 1985 3308


CHS PROTECTION


oxygen. This hypothesis is in accordance with our described findings of acquired resistance, for an illumination of up to 10 min, of HPD-exposed cells preenriched with cholesterol. In ad dition, reduced fluidity of membranes, induced by a higher cho lesterol content, will limit lateral mobility of proteins and diminish the probability of protein-protein interaction and cross-linking. It is interesting to note that HPD did not exert any specific inhibitory effect on the release of the FLV complex from FLC. The possibility that the release of this viral complex may be partly due to damage in membrane permeability rather than a direct effect on cell metabolism cannot be ruled out. However, as demonstrated in Chart 5, a constant ratio between virus release (reverse transcriptase activity) and cell viability is evident in all HPD concentrations used. We suggest, therefore, that the inhibition of virus release observed after HPD treatment is most likely the result of the anticellular effect expressed by photoactivated HPD. This is not the case with another natural substance, such as interferon, which has been shown to inhibit specifically the release of retroviruses from chronically infected cells (1). It is thus assumed that photoactivated HPD affects the general me tabolism of the treated cells, resulting in complete damage of the macromolecules synthesizing machinery. This study concentrated on the effect of photoactivated HPD on uninduced erythroleukemic cells. However, it has been well established that, upon induction with a variety of substances, the cells undergo a process of differentiation, including hemoglo bin synthesis (12). The question of whether HPD exerts a similar cytotoxic effect on differentiated FLC remains open.

ACKNOWLEDGMENTS We acknowledge hendler.

5.

6.

7. 8.

9.

10.

11. 12.

13. 14.

15. 16.

17.

18.

19.

the excellent editing and typing assistance of Bluma Leder20.

REFERENCES

21.

1. Aboud, M., Kimchi, R., Bakhanashvili, M., and Salzberg, S. Intracellular pro duction of virus particles and viral components in NIH/3T3 cells chronically infected with Moloney murine leukemia virus: effect of Â¡nterferon.J. Virol., 40: 830-838,1981. 2. Aboud, M., Weiss, D., and Salzberg, S. Rapid quantitation of interferon with chronically oncomavirus-producing cells. Infect. Immun., 13:1626-1632,1976. 3. Ben-Bassat, M., Polliak, A., Mitrani-Rosenbaum, S., Naparstek, E., Shouval, D., and Inbar, M. Fluidity of membrane lipids and lateral mobility of concanavalin A receptors in the cell surface of normal lymphocytes and lymphocytes from patients with malignant lymphomas and leukemias. Cancer Res., 37: 13071312,1977. 4. Benson, R. C., Kinsey, J. H., Cortese, D. A., Farrow, G. M., and Utz, D. C.

CANCER RESEARCH

22.

23.

24.

25.

Treatment of transitional cell carcinoma of the bladder with hematoporphyrin derivative phototherapy. J. Urol., 730: 1090-1095,1983. Breitbart, H., and Malik, Z. The effects of photoactivated protoporphyrin on reticulocyte membranes, intracellular activities, and hemoglobin precipitation. Photochem. Photobiol., 35: 365-369, 1982. Cannistraro, S., and Van de Vorst, A. Photosensitization by hematoporphyrin: ESR evidence for free radical induction in unsaturated fatty acids and for singlet oxygen production. Biochem. Biophys. Res. Commun., 74:1177-1185, 1977. Cortese, D. A., and Kinsey, J. H. Hematoporphyrin derivative phototherapy in the treatment of bronchogenic-carcinoma. Chest, 86: 8-13, 1984. Dougherty, T. J., Kaufman, J. E., Goldfarb, A., Weishaupt, K. R., Boyle, D., and Mittelman, A. Photoradiation therapy for the treatment of malignant tumors. Cancer Res., 38: 2628-2635, 1978. Dubbelman, T. M. A. R., De Goeig, A. F. P. M., and Van Steveninck, Y. Protoporphyrin-sensitized photodynamic modification of proteins in isolated human red blood cell membranes. Photochem. Photobiol., 28:197-204,1978. Ehrenberg, B., Malik, Z., and Nitzan, Y. Fluorescence spectral changes of hematoporphyrin derivative upon binding to lipid vesicles, Staphylococcus aureus, E. co/f cells. Photochem. Photobiol., in press, 1985. Forbes, I. M. Photochemotherapy of human cancers with hematoporphyrin derivative. Med. J. Aust., 740: 94-96, 1984. Friend, C., Scher, W., Holland, J. L, and Sato, T. Hemoglobin synthesis in murine virus-induced leukemic cells in vitro: stimulation of erythroid differentia tion by dimethyl sulfoxide. Proc. Nati. Acad. Sci. USA, 68: 378-383, 1971. Kessel, D. Hematoporphyrin and HPDâ€”photophysics, photochemistry, and phototherapy. Photochem. Photobiol., 39. 851-859,1984. Lin, C. W., Bellnier, D. A., Pront, G. R., Andrus, R. W., and Prescott, R. Cystoscopic fluorescence detector for photodetection of bladder. J. Urol., 737: 587-590,1984. Malik, Z., and Djaldetti, M. Cytotoxic effect of hemin and protoporphyrin on chronic, lymphocytic leukemia lymphocytes. Exp. Hematol., 8:867-879,1980. Malik, Z., and Breitbart, H. Cross-linking of hemoglobin and inhibition of globin synthesis in reticulocytes induced by photoactivated protoporphyrin. Acta Haematol., 64: 304-309, 1980. Malik, Z., and Djaldetti, M. Destruction of erythroleukemia, myelocytic, leu kemic, and Burkitt lymphoma cells by photoactivated protoporphyrin. Int. J. Cancer, 26: 495-500, 1980. Malik, Z., Lejbkowicz, F., and Salzberg, S. Cholesterol impregnation into erythroleukemic cell membrane induces resistance to hematoporphyrin pho todynamic effect. In: A. Andreoni and R. Cubeddu (eds.), Porphyrins in Tumor Phototherapy, pp. 185-191. New York: Plenum Publishing Corp., 1984. Miyoshi, N., Hisazumi, H., Veki, O., and Nakajma, K. Cellular binding of hematoporphyrin derivative in human bladder cancer cell lines KK-47. Photo chem. Photobiol., 39: 359-363, 1984. Moan, J., Smedshammer, L., and Christensen, T. Photodynamic effects on human cells exposed to light in the presence of hematoporphyrin. pH effects. Cancer Lett., 9. 327-332, 1980. Shinitzky, M., and Bemholz, Y. FLuidity parameters of lipid regions determined by fluorescence polarization. Biochim. Biophys. Acta, 575: 367-394,1978. Shinitzky, M., Skomick, Y., and Haran-Gera, N. Effective tumor immunization induced by cells of elevated membrane-lipid microviscosity. Proc. Nati. Acad. Sci. USA, 76: 5313-5316,1979. Suwa, K., Kimura, T., and Schaap, 0. P. Reaction of singlet oxygen with cholesterol in liposomal membranes. Effect of membrane fluidity on the photooxidation of cholesterol. Photochem. Photobiol., 28. 469-473, 1978. Tsuchiya, A., Obara, N., Miwa, M., Ohi, T., Kato, H., and Hayata, Y. Hemato porphyrin derivative and laser photoradiation in the diagnosis and treatment of bladder cancer. J. Urol., 730: 79-82,1982. Weishaupt, K. R., Gomer, C. Y., and Dougherty, T. Y. Identification of singlet oxygen as the cytotoxic agent in photoinactivation of a murine tumor. Cancer Res., 36. 2326-2329,1976.

VOL. 45 MAY 1985

3309


CHS PROTECTION


Fig. 1. Protective effect of CHS as depicted by SEM. a, untreated (control) FLC revealing an undamaged cell membrane with well-preserved microvilli (x 3,200); b, FLC sensitized by photoactivated HPD for 10 min, showing damaged membrane with holes of different diameters (x 11,000); c, FLC treated with 2% CHS alone. The cell membrane is undamaged. However, some changes in the shape and length of microvilli are noticed (x 4,000); d, FLC pretreated with 2% CHS, followed by sensitizaron with 20 ng of photoactivated HPD per ml for 10 min.

CANCER

RESEARCH

VOL. 45 JULY 1985

3310
