- Email: [email protected]
Electron spin resonance studies on photosensitized formation of hydroxyl radical by C-phycocyanin from Spirulina platensis
Biochimica et Biophysica Acta 1426 (1999) 205^211
Electron spin resonance studies on photosensitized formation of hydroxyl radical by C-phycocyanin from Spirulina platensis Su-ping Zhang, Jie Xie, Jian-ping Zhang, Jing-quan Zhao *, Li-jin Jiang Institute of Photographic Chemistry, Academia Sinica, Beijing 100101, People's Republic of China Received 23 July 1998; received in revised form 3 November 1998; accepted 3 November 1998
Abstract Visible light ( s 470 nm) irradiation of an oxygen-saturated solution of C-phycocyanin (C-PC) in the presence of the spin trap 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) gave an ESR spectrum characteristic of the DMPO-hydroxyl radical spin adduct DMPO-OH. The signal intensities of DMPO-OH adduct were enhanced by superoxide dismutase (SOD) and partly inhibited by catalase. It was partly responsible for the production of DMPO-OH that superoxide anion radical Oc3 2 dismutated to generate hydrogen peroxide (H2 O2 ) which decomposed ultimately to generate the highly reactive c OH. In addition, it can be concluded that singlet oxygen (1 O2 ) was an important intermediate according to the strong inhibitory action of 1,4-diazabicyclo[2.2.2]octane (DABCO) and histidine on DMPO-OH formation. The experimental results suggest that photodynamic action of C-PC proceed via both type I and type II mechanisms. Furthermore, the decay kinetics of DMPO-OH adduct, the effects of DMPO and C-PC concentrations as well as irradiation time on DMPO-OH adduct formation were also discussed. Concentration of C-PC should be an important factor to influence the ESR signal intensities of DMPO-OH. Therefore, it may be concluded that reasonably lower concentration of C-PC might prolong the duration of photosensitized formation of c OH and might strengthen the photodynamic action. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: C-Phycocyanin; Photodynamic action; Hydroxyl radical; ESR; Spin trapping
1. Introduction Photodynamic therapy (PDT) has been used with great success for the treatment of a variety of tumors, and attempts are constantly being made to extend this treatment modality to other clinical con-
Abbreviations: ESR, electron spin resonance; PDT, photodynamic therapy; PBP, phycobiliprotein; C-PC, C-phycocyanin; SOD, superoxide dismutase; HpD, hematoporphyrin derivative; DMPO, 5,5-dimethyl-1-pyrroline-N-oxide; DABCO, 1,4-diazabicyclo[2.2.2]octane * Corresponding author.
ditions [1,2]. Photofrin, the only photosensitizer approved for commercial use anywhere in the world for the therapy [3^5], while being clinically e¡ective, has major drawbacks of chemically impure, low absorption of light in `phototherapeutic window' (600^900 nm), skin phototoxicity and low selectivity with regard to uptake and retention by tumor vs. normal cells. These undesirable characteristics of photofrin have led to a search for more e¡ective photosensitizers with less side e¡ects. Recently, some studies [6^10] have demonstrated that phycobiliproteins (PBPs) exert much stronger photodynamic action on tumor cells than hematoporphyrin derivative (HpD) and might be used as
0304-4165 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 8 ) 0 0 1 5 3 - 6
BBAGEN 24743 18-12-98
206
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
one of promising candidates for new type of photodynamic reagent. Phycocyanin, the ¢rst employed one of PBPs in PDT, has several advantages over HpD, such as ready extraction and easy puri¢cation, high molar extinction coe¤cients, wide UV-visible absorption, no side e¡ects and signi¢cant reduction of normal tissue photosensitivity due to its fast metabolism in vivo [11]. However, study on the mechanism is still in the initial stage [12]. This has encouraged us to research the problem further. It is well known that reactive oxygen species are intimately associated with the photodynamic e¡ect of many sensitizers. A light-activated sensitizer can transfer energy from its triplet state to molecular oxygen with generation of singlet oxygen (1 O2 ) (type II mechanism) or interact with solvent or substrate by electron or proton transfer with generation of radicals (type I mechanism). In this paper, we used ESR associated with spintrapping techniques to detect c OH generated by irradiation of C-PC from Spirulina platenis. The experiments were designed to answer following questions: (1) where does c OH come from? (2) What is the detailed mechanism of the photodynamic action of CPC? (3) How did the related factors in£uence on the ESR signal intensity of DMPO-OH adduct. The experimental results might give signi¢cant enlightenment on the relationship of sensitizer dose and curative e¡ect in clinical application. 2. Materials and methods C-phycocyanin (C-PC) was isolated from laboratory cultures of the blue algae Spirulina platensis, according to Wang, Xia and colleagues [13,14].The purity of the protein was examined by absorption spectrum (Amax /A278 s 4.5) (Fig. 1). The powder of lyophilized protein was stored at 320³C and dissolved with phosphate bu¡er solution (pH 7.0) and water was freshly distilled before use. 5,5-dimethyl-1pyrroline-N-oxide (DMPO) was purchased from Aldrich and stored at 320³C under argon. 1,4-Diazabicyclo[2.2.2]octane (DABCO), histidine, superoxide dismutase (SOD) and catalase were purchased from Biotech. Technology Corporation, Chinese Academy of Sciences. Other reagents used, all of analytical grade, were obtained from Beijing Chemical Plant,
Fig. 1. Absorption spectrum of C-PC from Spirulina platensis dissolved in phosphate bu¡er solution (pH 7.0). The maximal absorption peak is located at 616 nm. A616 /A278 s 4.5.
China. The solutions injected quantitatively into specially made quartz capillaries were purged with argon or oxygen according to the experimental requirement. Irradiation was carried out by using a medium-pressure sodium lamp (450 W) on a `merrygo-round' apparatus and sample was immediately transferred to the microwave cavity after illumination. Light with wavelength less than 470 nm was cut o¡ by a long-pass ¢lter, and the apparatus was immersed in running water in a thermostat at 20³C. Measurements of the ESR spectra were carried out on a Bruker ESP 300E spectrometer operating at room temperature (X-band, microwave frequency, 9.5 GHz). 3. Results and discussion 3.1. Formation of hydroxyl radical (c OH) When an oxygen-saturated phosphate bu¡er (pH 7.0) solution of C-PC (2 mg/ml) and DMPO (60 mM) was irradiated, a four-line ESR spectrum in a 1:2:2:1 pattern (Fig. 2a) was immediately observed with hyper¢ne coupling constants aN = aH = 15.0, the characteristic of the hydroxyl radical spin adduct of DMPO (DMPO-OH). When ethanol (5%, v/v) was introduced into the system, the six-line ESR spectrum of DMPO adduct of CH3 WCHOH was observed
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
207
the literature [15]. Addition of mannitol, known scavenger of c OH [16], signi¢cantly reduced the ESR signal intensity of DMPO-OH adduct (spectrum not shown). In the absence of light, oxygen or C-PC, no detectable ESR signal was observed in the system, indicating that the formation of the DMPO-OH adduct is dependent on the presence of oxygen and light as well as C-PC. There are several potential sources for the ESR signal of DMPO-OH other than free c OH trapped by DMPO [17]. The absence and enhancement of DMPO-OH signal on argon and O2 , respectively indicated the importance of molecular oxygen in the mechanism, while a direct interaction between C-PC and DMPO does not seem possible. The lack of visualized DMPO-OOH spin adduct signal could not exclude the production of Oc3 2 . Indeed, the direct ESR detection of DMPO-OOH is still di¤cult because the short lifetime of the adduct and then may form DMPO-OH [18,19]. To access this, the e¡ects of SOD and catalase on DMPO-OH formation were investigated. The addition of SOD (40 Wg/ml), a speci¢c and e¤cient catalyzer of Oc3 2 disproportionation (Eq. 1), enhanced the quantity of DMPO-OH adduct (Fig. 2b), whereas catalase (50 Wg/ml), a speci¢c scavenger for H2 O2 (Fig. 2), inhibited the formation of DMPO-OH adduct by about 30%, which did not change with increasing the catalase concentration (Fig. 2c). It can be deduced from these results that the DMPO-OH generated partly from following procedure (Eqs. 1 and 3) rather than directly from decomposition of DMPO-OOH. SOD
Fig. 2. The irradiation was performed outside the microwave cavity. Instrumental settings: microwave power, 5.05 mW; modulation amplitude, 3.226 G; receiver gain, 2U105 . (a) ESR spectrum of hydroxyl radical generated by irradiation (6 min) of an oxygen-saturated phosphate bu¡er (pH 7.0) solution of C-PC (2 mg/ml) and DMPO (60 mM). (b) Same as a, but in presence of SOD (40 Wg/ml). (c) Same as a, but in presence of catalase (50 Wg/ml). (d) Same as a, but in presence of DABCO (50 mM). (e) Same as a, but in presence of histidine (10 mM).
with aN = 15.8 G and aH = 23.0 G which was generated from the hydrogen abstraction of ethanol by c OH (spectrum not shown). The values of the coupling constants are in good agreement with those in
2Oc3 2 2H ÿ!H2 O2 O2
1
catalase
2H2 O2 ÿ! 2H2 O O2
2
Fe2 H2 O2 ! Fe3 OH3 c
DMPO
OH ÿ! DMPO-OH
3
The singlet oxygen was considered to be formed by energy transfer from the triplet sensitizer to oxygen [12]: 0
hv
kisc
O2
PBP S0 ÿ! 1 PBP S1 ÿ!3 PBP T1 ÿ!1 O2
BBAGEN 24743 18-12-98
4
208
1
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
H
O2 DMPO ! DMPO-1 O2 ÿ! DMPO
DMPO-OH c OH ÿ! DMPO-OH
5
The latest studies indicated that the decomposition of the reaction product of DMPO with 1 O2 was a signi¢cant source of DMPO-OH [17,20]. The speci¢c 1 O2 quenchers (DABCO and histidine) have been used to examine this e¡ect. In the presence of DABCO (50 mM), the ESR signal intensity of the DMPO-OH was about 30% lower (Fig. 2d) than that of the sample without DABCO, while in the presence of histidine (10 mM), it was about 70% lower (Fig. 2e), indicating that 1 O2 could also contribute to the formation of DMPO-OH adduct (Eq. 5). The results agree quite well with that estimated by Zhang who believed DABCO (50 mM) could only quench about one half of produced 1 O2 [21]. Therefore, it can be estimated that type I and type II mechanisms for C-PC sensitization play a 30 and 70% part, respectively, in bu¡er aqueous solution. 3.2. The decay kinetics of the DMPO-OH adduct The decay of the adduct was measured by recording the amplitude of the ESR signal at di¡erent times
Fig. 4. Dependence of DMPO-OH signal amplitude on DMPO concentrations with the concentration of C-PC 10 mg/ml. Other experimental conditions are same as described in Fig. 2a.
after illumination (Fig. 3). The data could be well ¢tted with a single exponential function and the rate constant is 1.16U1033 s31 , which means the apparent decay of the adduct obeys the ¢rst-order kinetics. It has been reported [22] that the decay of the adduct had both ¢rst- and second-order components and that the half-life of the ¢rst-order component was 2.6 h, of which the rate constant was 7.4U1035 s31 . In fact, the two rate constants could not be directly compared because Finkelstein obtained two components to deal with the adduct decay while only one component was obtained from our results. 3.3. The e¡ects of the concentrations of DMPO on ESR signal intensities When other factors remained invariable, the relative ESR signal intensities of DMPO-OH adduct increased sharply with increasing concentrations of DMPO until a `saturated point' (Fig. 4), indicating the dependence of DMPO-OH radical adduct formation on the concentration of DMPO was reached.
Fig. 3. Decay of the DMPO-OH signal intensity with sample incubated in the dark for 210 s before measurement. Solid points stand for experimental data while dot line is from the exponential ¢t with M2 = 0.05. Other experimental conditions are the same as described in Fig. 2a.
3.4. The e¡ects of C-PC concentration on ESR signal intensities of DMPO-OH Concentration of C-PC was an important factor to in£uence the ESR signal intensities of DMPO-OH
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
Fig. 5. Dependence of the DMPO-OH signal amplitude on CPC concentrations. Other experimental conditions are same as those in Fig. 2a.
(Fig. 5). It was interesting to ¢nd that a maximum value appeared during the variation of C-PC concentration. The relative signal intensities of DMPO-OH adduct increase with the concentration of C-PC when concentrations of C-PC are lower than 2 mg/ ml. However, when concentrations of C-PC are higher than 2 mg/ml, the relative signal intensities decrease sharply and then slowly with increasing concentration of C-PC. The results could be explained based on the mechanism of photosensitization of a sensitizer (Sens) proposed by Foote [23]:
Sens* (a singlet or, more commonly, a triplet) can either react with the substrate or solvent (type I) or with molecular oxygen (type II). Spin elimination of DMPO-OH by sensitizer anion (Sensc3 ) may be used to explain why lower ESR signal intensity observed for the sample with higher concentration of C-PC. It was proved that the trapped DMPO spin adduct might undergo spin elimination by sensitizer anion or electron donor [24,25]. By the mechanism mentioned above, irradiation of oxygen-saturated sample would produce Sens* which transfer the energy to oxygen molecules and generate 1 O2 (type II). Sens*
209
consumption, in this way, would inhibit type I reaction. However, with the consumption of oxygen, the rate of 1 O2 formation decreases while that of radical ions increases, therefore the spin elimination becomes more signi¢cant. Obviously, the sample with low concentration of C-PC should consume oxygen less and slower than the high concentration sample. When the concentrations of C-PC are lower than 2 mg/ml, Sensc3 appears late due to the slow consumption of oxygen, thus the photosensitized formation of c OH continuously accumulates with the increasing concentration of C-PC and undergoes less despin action by Sensc3 , which makes the ESR signal intensities of DMPO-OH adduct increase with the increasing concentration of C-PC. While the concentrations of C-PC are higher than 2 mg/ml, fast oxygen consumption results in fast formation of Sensc3 , which leads to despin of DMPO-OH. In addition, the e¡ects of radical self-quenching caused by collision and Sens* (excited state) quenched by Sens (ground state) also increase at high concentration of C-PC, which could also result in the decrease of c OH formation. On the other hand, protein, composed of amino acids, can play a role of scavenger of free radicals and reactive oxygen, which may be another reason for the concentration dependence.
Fig. 6. Plot of the DMPO-OH signal amplitude as a function of irradiation time in a phosphate bu¡er (pH 7.0) solution using the concentration 6 mg/ml (line 1) and 10 mg/ml (line 2) of C-PC. Other experimental conditions are the same as those in Fig. 2a.
BBAGEN 24743 18-12-98
210
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
3.5. Dependence of ESR signal intensities of DMPO-OH on irradiation time
following mechanism when C-PC is irradiated with visible light ( s 470 nm):
The dependence of ESR signal intensities of DMPO-OH on irradiation time was observed on two samples with concentrations of 6 and 10 mg/ml of C-PC, respectively, as shown in Fig. 6. It can be seen that the maximal ESR signal intensity for the sample with 6 mg/ml C-PC appeared later and slightly stronger than that with 10 mg/ml C-PC. It can be explained as above that the maximal signal is stronger for the sample with 6 mg/ml C-PC than that with 10 mg/ml C-PC. In addition, fast oxygen exhaustion at higher concentration of C-PC may lead to an early end of the c OH formation, which in turn result in earlier appearance of the maximal signal of DMPO-OH. It is believed that 1 O2 is the major mediator of photochemical cell damage for many types of photosensitizers; however, oxygen species like the hydroxyl radical (c OH) can also induce deleterious e¡ects including lipid peroxidation and membrane damage [26,27]. Generally, tumor tissues are usually thought in oxygen-de¢cient environment. According to our results reported above, if the concentration of C-PC is high enough, the formation of c OH will end earlier due to fast oxygen exhaustion and might reduce photosensitized curative e¡ect. On the contrary, properly reducing concentration of C-PC can prolong the duration of photosensitized formation of c OH and might strengthen the lethal action of cancer cells. Judging from this, for the clinical application of photosensitizers, it is essential that the drug dose should be chosen carefully. As Dougherty reported [28], American hospitals reduced injected dose of Photofrin II from 1.5^2.0 to 1.0 mg/kg and apparently improved therapeutic ratio in treatment of tumors involving the skin. This may be relevant to the similar reasons mentioned above. 4. Conclusions
(2) It is estimated that C-PC is a 30%/70% type I/ type II sensitizer. (3) Decay of DMPO-OH adduct is according to the ¢rst-order kinetics and the rate constant is (1.16 þ 0.05)U1034 s31 . (4) When other factors kept invariable, the relative ESR signal intensities of DMPO-OH adduct increased sharply with increasing concentrations of DMPO until a `saturated point'. (5) Concentration of C-PC was an important factor to in£uence the ESR signal intensities of DMPOOH. When the concentration of C-PC is high enough, fast oxygen consumption may lead to photosensitized curative e¡ect lower due to the earlier end of c OH formation. (6) ESR signal intensities of DMPO-OH underwent maxima on irradiation time, which appeared later and slightly stronger for the sample with 6 mg/ml C-PC than that with 10 mg/ml C-PC. It can be further deduced that a reasonably lower concentration of C-PC might prolong the duration of photosensitized formation of c OH and might strengthen the photodynamic action. Acknowledgements This research was supported by the National Natural Science Foundation of China (NNSFC).
References
In this study, some points can be drawn based on the experimental results listed below. (1) Hydroxyl radical (c OH) can be generated by
[1] J.D. Spikes, G. Jori, Lasers Med. Sci. 2, (3) (1987) 3^15. [2] S.L. Marcus, Proc. IEEE 80 (1992) 869^889. [3] C.J. Gomer, T.J. Dougherty, Cancer Res. 39 (1979) 146^151.
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211 [4] D. Kessel, P. Thompson, B. Musselman, C.K. Chang, Photochem. Photobiol. 46 (1987) 563^568. [5] D.A. Bellnier, Y.-K. Ho, R.K. Pandey, J.R. Missert, T.J. Dougherty, Photochem. Photobiol. 50 (1989) 221^228. [6] N.C. Morcos, M. Berns, W.L. Henry, Lasers Surg. Med. 8, (1) (1988) 7^10. [7] N.C. Morcos, W.L. Henry, (1987) US Patent 4,886,831, Chem. Abstr. 113 (1990) 553j. [8] S. Zheng, X.H. Cai, L.M. He, Faming Zhuanli Shenqing Gongkai Shuomingshu, CN 1,091,976, Chem. Abstr. 122 (1994) 182179h. [9] X.H. Cai, L.M. He, J.L. Jiang, L.L. Yu, Z.M. Xu, S. Zheng, Zhongguo Haiyang Yaowu Zazhi 1 (1995) 15^18. [10] L.M. He, X.H. Cai, S. Zheng, L.L. Yu, Zhongguo Jiguang 22, (11) (1995) 847^849. [11] Q.S. Lin, J.P. Zhang, F.J. Zeng, L.J. Jiang, Acta Phytophysiol. Sinica 18, (3) (1992) 253^258. [12] J.A. He, Y.Z. Hu, L.J. Jiang, Biochim. Biophys. Acta 1320 (1997) 165^174. [13] G.C. Wang, B.C. Zhou, C.K. Zeng, Chin. Sci. Bull. 40 (1996) 741^743. in Chinese [14] A.D. Xia, J.C. Zhu, H.J. Wu, L.J. Jiang, Photograph. Sci. Photochem. 11, (1) (1993) 35^40. in Chinese [15] E. Finkelstein, G.M. Rosen, E.J. Rauckman, Arch. Biochem. Biophys. 200 (1980) 1^16.
211
[16] M. Sonoda, C.M. Krishra, P. Riesz, Photochem. Photobiol. 46 (1987) 625^631. [17] J. Feix, B. Kalyanaramen, Arch. Biochem. Biophys. 15 (1991) 43^51. [18] G. Buettner, Free Radic. Res. Commun. 19 (1993) 79^ 87. [19] G. Buettner, L. Oberley, Biochem. Biophys. Res. Commun. 83 (1978) 69^74. [20] P. Bilsky, K. Reszka, M. Bilska, C. Chignell, J. Am. Chem. Soc. 118 (1996) 1330^1338. [21] J.Z. Zhang, in: Introduction to Free Radical Biology (in Chinese), Zhongguo Kexue Jishu Daxue Yanjiushengyuan Huaxuebu, China, 1996, p. 51. [22] E. Finkelstein, G.M. Rosen, E.J. Rauckman, J. Am. Chem. Soc. 102 (1980) 4994^4999. [23] C.S. Foote, Photochem. Photobiol. 54, (5) (1991) 659. [24] N.H. Wang, Z.Y. Zhang, J. Photochem. Photobiol. B Biol. 14 (1992) 207. [25] K. Reszka, P. Kolodziejczyk, J.W. Lown, J. Free Radic. Biol. Med. 2 (1986) 267^274. [26] D. Dolphin, Can. J. Chem. 72 (1994) 1005^1013. [27] T. Kriska, L. Korecz, I. Nemes, D. Gal, Biochem. Biophys. Res. Commun. 215 (1995) 192^198. [28] T.J. Dougherty, Photochem. Photobiol. 45, (6) (1987) 879^ 889.
BBAGEN 24743 18-12-98
Electron spin resonance studies on photosensitized formation of hydroxyl radical by C-phycocyanin from Spirulina platensis Su-ping Zhang, Jie Xie, Jian-ping Zhang, Jing-quan Zhao *, Li-jin Jiang Institute of Photographic Chemistry, Academia Sinica, Beijing 100101, People's Republic of China Received 23 July 1998; received in revised form 3 November 1998; accepted 3 November 1998
Abstract Visible light ( s 470 nm) irradiation of an oxygen-saturated solution of C-phycocyanin (C-PC) in the presence of the spin trap 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) gave an ESR spectrum characteristic of the DMPO-hydroxyl radical spin adduct DMPO-OH. The signal intensities of DMPO-OH adduct were enhanced by superoxide dismutase (SOD) and partly inhibited by catalase. It was partly responsible for the production of DMPO-OH that superoxide anion radical Oc3 2 dismutated to generate hydrogen peroxide (H2 O2 ) which decomposed ultimately to generate the highly reactive c OH. In addition, it can be concluded that singlet oxygen (1 O2 ) was an important intermediate according to the strong inhibitory action of 1,4-diazabicyclo[2.2.2]octane (DABCO) and histidine on DMPO-OH formation. The experimental results suggest that photodynamic action of C-PC proceed via both type I and type II mechanisms. Furthermore, the decay kinetics of DMPO-OH adduct, the effects of DMPO and C-PC concentrations as well as irradiation time on DMPO-OH adduct formation were also discussed. Concentration of C-PC should be an important factor to influence the ESR signal intensities of DMPO-OH. Therefore, it may be concluded that reasonably lower concentration of C-PC might prolong the duration of photosensitized formation of c OH and might strengthen the photodynamic action. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: C-Phycocyanin; Photodynamic action; Hydroxyl radical; ESR; Spin trapping
1. Introduction Photodynamic therapy (PDT) has been used with great success for the treatment of a variety of tumors, and attempts are constantly being made to extend this treatment modality to other clinical con-
Abbreviations: ESR, electron spin resonance; PDT, photodynamic therapy; PBP, phycobiliprotein; C-PC, C-phycocyanin; SOD, superoxide dismutase; HpD, hematoporphyrin derivative; DMPO, 5,5-dimethyl-1-pyrroline-N-oxide; DABCO, 1,4-diazabicyclo[2.2.2]octane * Corresponding author.
ditions [1,2]. Photofrin, the only photosensitizer approved for commercial use anywhere in the world for the therapy [3^5], while being clinically e¡ective, has major drawbacks of chemically impure, low absorption of light in `phototherapeutic window' (600^900 nm), skin phototoxicity and low selectivity with regard to uptake and retention by tumor vs. normal cells. These undesirable characteristics of photofrin have led to a search for more e¡ective photosensitizers with less side e¡ects. Recently, some studies [6^10] have demonstrated that phycobiliproteins (PBPs) exert much stronger photodynamic action on tumor cells than hematoporphyrin derivative (HpD) and might be used as
0304-4165 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 1 6 5 ( 9 8 ) 0 0 1 5 3 - 6
BBAGEN 24743 18-12-98
206
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
one of promising candidates for new type of photodynamic reagent. Phycocyanin, the ¢rst employed one of PBPs in PDT, has several advantages over HpD, such as ready extraction and easy puri¢cation, high molar extinction coe¤cients, wide UV-visible absorption, no side e¡ects and signi¢cant reduction of normal tissue photosensitivity due to its fast metabolism in vivo [11]. However, study on the mechanism is still in the initial stage [12]. This has encouraged us to research the problem further. It is well known that reactive oxygen species are intimately associated with the photodynamic e¡ect of many sensitizers. A light-activated sensitizer can transfer energy from its triplet state to molecular oxygen with generation of singlet oxygen (1 O2 ) (type II mechanism) or interact with solvent or substrate by electron or proton transfer with generation of radicals (type I mechanism). In this paper, we used ESR associated with spintrapping techniques to detect c OH generated by irradiation of C-PC from Spirulina platenis. The experiments were designed to answer following questions: (1) where does c OH come from? (2) What is the detailed mechanism of the photodynamic action of CPC? (3) How did the related factors in£uence on the ESR signal intensity of DMPO-OH adduct. The experimental results might give signi¢cant enlightenment on the relationship of sensitizer dose and curative e¡ect in clinical application. 2. Materials and methods C-phycocyanin (C-PC) was isolated from laboratory cultures of the blue algae Spirulina platensis, according to Wang, Xia and colleagues [13,14].The purity of the protein was examined by absorption spectrum (Amax /A278 s 4.5) (Fig. 1). The powder of lyophilized protein was stored at 320³C and dissolved with phosphate bu¡er solution (pH 7.0) and water was freshly distilled before use. 5,5-dimethyl-1pyrroline-N-oxide (DMPO) was purchased from Aldrich and stored at 320³C under argon. 1,4-Diazabicyclo[2.2.2]octane (DABCO), histidine, superoxide dismutase (SOD) and catalase were purchased from Biotech. Technology Corporation, Chinese Academy of Sciences. Other reagents used, all of analytical grade, were obtained from Beijing Chemical Plant,
Fig. 1. Absorption spectrum of C-PC from Spirulina platensis dissolved in phosphate bu¡er solution (pH 7.0). The maximal absorption peak is located at 616 nm. A616 /A278 s 4.5.
China. The solutions injected quantitatively into specially made quartz capillaries were purged with argon or oxygen according to the experimental requirement. Irradiation was carried out by using a medium-pressure sodium lamp (450 W) on a `merrygo-round' apparatus and sample was immediately transferred to the microwave cavity after illumination. Light with wavelength less than 470 nm was cut o¡ by a long-pass ¢lter, and the apparatus was immersed in running water in a thermostat at 20³C. Measurements of the ESR spectra were carried out on a Bruker ESP 300E spectrometer operating at room temperature (X-band, microwave frequency, 9.5 GHz). 3. Results and discussion 3.1. Formation of hydroxyl radical (c OH) When an oxygen-saturated phosphate bu¡er (pH 7.0) solution of C-PC (2 mg/ml) and DMPO (60 mM) was irradiated, a four-line ESR spectrum in a 1:2:2:1 pattern (Fig. 2a) was immediately observed with hyper¢ne coupling constants aN = aH = 15.0, the characteristic of the hydroxyl radical spin adduct of DMPO (DMPO-OH). When ethanol (5%, v/v) was introduced into the system, the six-line ESR spectrum of DMPO adduct of CH3 WCHOH was observed
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
207
the literature [15]. Addition of mannitol, known scavenger of c OH [16], signi¢cantly reduced the ESR signal intensity of DMPO-OH adduct (spectrum not shown). In the absence of light, oxygen or C-PC, no detectable ESR signal was observed in the system, indicating that the formation of the DMPO-OH adduct is dependent on the presence of oxygen and light as well as C-PC. There are several potential sources for the ESR signal of DMPO-OH other than free c OH trapped by DMPO [17]. The absence and enhancement of DMPO-OH signal on argon and O2 , respectively indicated the importance of molecular oxygen in the mechanism, while a direct interaction between C-PC and DMPO does not seem possible. The lack of visualized DMPO-OOH spin adduct signal could not exclude the production of Oc3 2 . Indeed, the direct ESR detection of DMPO-OOH is still di¤cult because the short lifetime of the adduct and then may form DMPO-OH [18,19]. To access this, the e¡ects of SOD and catalase on DMPO-OH formation were investigated. The addition of SOD (40 Wg/ml), a speci¢c and e¤cient catalyzer of Oc3 2 disproportionation (Eq. 1), enhanced the quantity of DMPO-OH adduct (Fig. 2b), whereas catalase (50 Wg/ml), a speci¢c scavenger for H2 O2 (Fig. 2), inhibited the formation of DMPO-OH adduct by about 30%, which did not change with increasing the catalase concentration (Fig. 2c). It can be deduced from these results that the DMPO-OH generated partly from following procedure (Eqs. 1 and 3) rather than directly from decomposition of DMPO-OOH. SOD
Fig. 2. The irradiation was performed outside the microwave cavity. Instrumental settings: microwave power, 5.05 mW; modulation amplitude, 3.226 G; receiver gain, 2U105 . (a) ESR spectrum of hydroxyl radical generated by irradiation (6 min) of an oxygen-saturated phosphate bu¡er (pH 7.0) solution of C-PC (2 mg/ml) and DMPO (60 mM). (b) Same as a, but in presence of SOD (40 Wg/ml). (c) Same as a, but in presence of catalase (50 Wg/ml). (d) Same as a, but in presence of DABCO (50 mM). (e) Same as a, but in presence of histidine (10 mM).
with aN = 15.8 G and aH = 23.0 G which was generated from the hydrogen abstraction of ethanol by c OH (spectrum not shown). The values of the coupling constants are in good agreement with those in
2Oc3 2 2H ÿ!H2 O2 O2
1
catalase
2H2 O2 ÿ! 2H2 O O2
2
Fe2 H2 O2 ! Fe3 OH3 c
DMPO
OH ÿ! DMPO-OH
3
The singlet oxygen was considered to be formed by energy transfer from the triplet sensitizer to oxygen [12]: 0
hv
kisc
O2
PBP S0 ÿ! 1 PBP S1 ÿ!3 PBP T1 ÿ!1 O2
BBAGEN 24743 18-12-98
4
208
1
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
H
O2 DMPO ! DMPO-1 O2 ÿ! DMPO
DMPO-OH c OH ÿ! DMPO-OH
5
The latest studies indicated that the decomposition of the reaction product of DMPO with 1 O2 was a signi¢cant source of DMPO-OH [17,20]. The speci¢c 1 O2 quenchers (DABCO and histidine) have been used to examine this e¡ect. In the presence of DABCO (50 mM), the ESR signal intensity of the DMPO-OH was about 30% lower (Fig. 2d) than that of the sample without DABCO, while in the presence of histidine (10 mM), it was about 70% lower (Fig. 2e), indicating that 1 O2 could also contribute to the formation of DMPO-OH adduct (Eq. 5). The results agree quite well with that estimated by Zhang who believed DABCO (50 mM) could only quench about one half of produced 1 O2 [21]. Therefore, it can be estimated that type I and type II mechanisms for C-PC sensitization play a 30 and 70% part, respectively, in bu¡er aqueous solution. 3.2. The decay kinetics of the DMPO-OH adduct The decay of the adduct was measured by recording the amplitude of the ESR signal at di¡erent times
Fig. 4. Dependence of DMPO-OH signal amplitude on DMPO concentrations with the concentration of C-PC 10 mg/ml. Other experimental conditions are same as described in Fig. 2a.
after illumination (Fig. 3). The data could be well ¢tted with a single exponential function and the rate constant is 1.16U1033 s31 , which means the apparent decay of the adduct obeys the ¢rst-order kinetics. It has been reported [22] that the decay of the adduct had both ¢rst- and second-order components and that the half-life of the ¢rst-order component was 2.6 h, of which the rate constant was 7.4U1035 s31 . In fact, the two rate constants could not be directly compared because Finkelstein obtained two components to deal with the adduct decay while only one component was obtained from our results. 3.3. The e¡ects of the concentrations of DMPO on ESR signal intensities When other factors remained invariable, the relative ESR signal intensities of DMPO-OH adduct increased sharply with increasing concentrations of DMPO until a `saturated point' (Fig. 4), indicating the dependence of DMPO-OH radical adduct formation on the concentration of DMPO was reached.
Fig. 3. Decay of the DMPO-OH signal intensity with sample incubated in the dark for 210 s before measurement. Solid points stand for experimental data while dot line is from the exponential ¢t with M2 = 0.05. Other experimental conditions are the same as described in Fig. 2a.
3.4. The e¡ects of C-PC concentration on ESR signal intensities of DMPO-OH Concentration of C-PC was an important factor to in£uence the ESR signal intensities of DMPO-OH
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
Fig. 5. Dependence of the DMPO-OH signal amplitude on CPC concentrations. Other experimental conditions are same as those in Fig. 2a.
(Fig. 5). It was interesting to ¢nd that a maximum value appeared during the variation of C-PC concentration. The relative signal intensities of DMPO-OH adduct increase with the concentration of C-PC when concentrations of C-PC are lower than 2 mg/ ml. However, when concentrations of C-PC are higher than 2 mg/ml, the relative signal intensities decrease sharply and then slowly with increasing concentration of C-PC. The results could be explained based on the mechanism of photosensitization of a sensitizer (Sens) proposed by Foote [23]:
Sens* (a singlet or, more commonly, a triplet) can either react with the substrate or solvent (type I) or with molecular oxygen (type II). Spin elimination of DMPO-OH by sensitizer anion (Sensc3 ) may be used to explain why lower ESR signal intensity observed for the sample with higher concentration of C-PC. It was proved that the trapped DMPO spin adduct might undergo spin elimination by sensitizer anion or electron donor [24,25]. By the mechanism mentioned above, irradiation of oxygen-saturated sample would produce Sens* which transfer the energy to oxygen molecules and generate 1 O2 (type II). Sens*
209
consumption, in this way, would inhibit type I reaction. However, with the consumption of oxygen, the rate of 1 O2 formation decreases while that of radical ions increases, therefore the spin elimination becomes more signi¢cant. Obviously, the sample with low concentration of C-PC should consume oxygen less and slower than the high concentration sample. When the concentrations of C-PC are lower than 2 mg/ml, Sensc3 appears late due to the slow consumption of oxygen, thus the photosensitized formation of c OH continuously accumulates with the increasing concentration of C-PC and undergoes less despin action by Sensc3 , which makes the ESR signal intensities of DMPO-OH adduct increase with the increasing concentration of C-PC. While the concentrations of C-PC are higher than 2 mg/ml, fast oxygen consumption results in fast formation of Sensc3 , which leads to despin of DMPO-OH. In addition, the e¡ects of radical self-quenching caused by collision and Sens* (excited state) quenched by Sens (ground state) also increase at high concentration of C-PC, which could also result in the decrease of c OH formation. On the other hand, protein, composed of amino acids, can play a role of scavenger of free radicals and reactive oxygen, which may be another reason for the concentration dependence.
Fig. 6. Plot of the DMPO-OH signal amplitude as a function of irradiation time in a phosphate bu¡er (pH 7.0) solution using the concentration 6 mg/ml (line 1) and 10 mg/ml (line 2) of C-PC. Other experimental conditions are the same as those in Fig. 2a.
BBAGEN 24743 18-12-98
210
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211
3.5. Dependence of ESR signal intensities of DMPO-OH on irradiation time
following mechanism when C-PC is irradiated with visible light ( s 470 nm):
The dependence of ESR signal intensities of DMPO-OH on irradiation time was observed on two samples with concentrations of 6 and 10 mg/ml of C-PC, respectively, as shown in Fig. 6. It can be seen that the maximal ESR signal intensity for the sample with 6 mg/ml C-PC appeared later and slightly stronger than that with 10 mg/ml C-PC. It can be explained as above that the maximal signal is stronger for the sample with 6 mg/ml C-PC than that with 10 mg/ml C-PC. In addition, fast oxygen exhaustion at higher concentration of C-PC may lead to an early end of the c OH formation, which in turn result in earlier appearance of the maximal signal of DMPO-OH. It is believed that 1 O2 is the major mediator of photochemical cell damage for many types of photosensitizers; however, oxygen species like the hydroxyl radical (c OH) can also induce deleterious e¡ects including lipid peroxidation and membrane damage [26,27]. Generally, tumor tissues are usually thought in oxygen-de¢cient environment. According to our results reported above, if the concentration of C-PC is high enough, the formation of c OH will end earlier due to fast oxygen exhaustion and might reduce photosensitized curative e¡ect. On the contrary, properly reducing concentration of C-PC can prolong the duration of photosensitized formation of c OH and might strengthen the lethal action of cancer cells. Judging from this, for the clinical application of photosensitizers, it is essential that the drug dose should be chosen carefully. As Dougherty reported [28], American hospitals reduced injected dose of Photofrin II from 1.5^2.0 to 1.0 mg/kg and apparently improved therapeutic ratio in treatment of tumors involving the skin. This may be relevant to the similar reasons mentioned above. 4. Conclusions
(2) It is estimated that C-PC is a 30%/70% type I/ type II sensitizer. (3) Decay of DMPO-OH adduct is according to the ¢rst-order kinetics and the rate constant is (1.16 þ 0.05)U1034 s31 . (4) When other factors kept invariable, the relative ESR signal intensities of DMPO-OH adduct increased sharply with increasing concentrations of DMPO until a `saturated point'. (5) Concentration of C-PC was an important factor to in£uence the ESR signal intensities of DMPOOH. When the concentration of C-PC is high enough, fast oxygen consumption may lead to photosensitized curative e¡ect lower due to the earlier end of c OH formation. (6) ESR signal intensities of DMPO-OH underwent maxima on irradiation time, which appeared later and slightly stronger for the sample with 6 mg/ml C-PC than that with 10 mg/ml C-PC. It can be further deduced that a reasonably lower concentration of C-PC might prolong the duration of photosensitized formation of c OH and might strengthen the photodynamic action. Acknowledgements This research was supported by the National Natural Science Foundation of China (NNSFC).
References
In this study, some points can be drawn based on the experimental results listed below. (1) Hydroxyl radical (c OH) can be generated by
[1] J.D. Spikes, G. Jori, Lasers Med. Sci. 2, (3) (1987) 3^15. [2] S.L. Marcus, Proc. IEEE 80 (1992) 869^889. [3] C.J. Gomer, T.J. Dougherty, Cancer Res. 39 (1979) 146^151.
BBAGEN 24743 18-12-98
S. Zhang et al. / Biochimica et Biophysica Acta 1426 (1999) 205^211 [4] D. Kessel, P. Thompson, B. Musselman, C.K. Chang, Photochem. Photobiol. 46 (1987) 563^568. [5] D.A. Bellnier, Y.-K. Ho, R.K. Pandey, J.R. Missert, T.J. Dougherty, Photochem. Photobiol. 50 (1989) 221^228. [6] N.C. Morcos, M. Berns, W.L. Henry, Lasers Surg. Med. 8, (1) (1988) 7^10. [7] N.C. Morcos, W.L. Henry, (1987) US Patent 4,886,831, Chem. Abstr. 113 (1990) 553j. [8] S. Zheng, X.H. Cai, L.M. He, Faming Zhuanli Shenqing Gongkai Shuomingshu, CN 1,091,976, Chem. Abstr. 122 (1994) 182179h. [9] X.H. Cai, L.M. He, J.L. Jiang, L.L. Yu, Z.M. Xu, S. Zheng, Zhongguo Haiyang Yaowu Zazhi 1 (1995) 15^18. [10] L.M. He, X.H. Cai, S. Zheng, L.L. Yu, Zhongguo Jiguang 22, (11) (1995) 847^849. [11] Q.S. Lin, J.P. Zhang, F.J. Zeng, L.J. Jiang, Acta Phytophysiol. Sinica 18, (3) (1992) 253^258. [12] J.A. He, Y.Z. Hu, L.J. Jiang, Biochim. Biophys. Acta 1320 (1997) 165^174. [13] G.C. Wang, B.C. Zhou, C.K. Zeng, Chin. Sci. Bull. 40 (1996) 741^743. in Chinese [14] A.D. Xia, J.C. Zhu, H.J. Wu, L.J. Jiang, Photograph. Sci. Photochem. 11, (1) (1993) 35^40. in Chinese [15] E. Finkelstein, G.M. Rosen, E.J. Rauckman, Arch. Biochem. Biophys. 200 (1980) 1^16.
211
[16] M. Sonoda, C.M. Krishra, P. Riesz, Photochem. Photobiol. 46 (1987) 625^631. [17] J. Feix, B. Kalyanaramen, Arch. Biochem. Biophys. 15 (1991) 43^51. [18] G. Buettner, Free Radic. Res. Commun. 19 (1993) 79^ 87. [19] G. Buettner, L. Oberley, Biochem. Biophys. Res. Commun. 83 (1978) 69^74. [20] P. Bilsky, K. Reszka, M. Bilska, C. Chignell, J. Am. Chem. Soc. 118 (1996) 1330^1338. [21] J.Z. Zhang, in: Introduction to Free Radical Biology (in Chinese), Zhongguo Kexue Jishu Daxue Yanjiushengyuan Huaxuebu, China, 1996, p. 51. [22] E. Finkelstein, G.M. Rosen, E.J. Rauckman, J. Am. Chem. Soc. 102 (1980) 4994^4999. [23] C.S. Foote, Photochem. Photobiol. 54, (5) (1991) 659. [24] N.H. Wang, Z.Y. Zhang, J. Photochem. Photobiol. B Biol. 14 (1992) 207. [25] K. Reszka, P. Kolodziejczyk, J.W. Lown, J. Free Radic. Biol. Med. 2 (1986) 267^274. [26] D. Dolphin, Can. J. Chem. 72 (1994) 1005^1013. [27] T. Kriska, L. Korecz, I. Nemes, D. Gal, Biochem. Biophys. Res. Commun. 215 (1995) 192^198. [28] T.J. Dougherty, Photochem. Photobiol. 45, (6) (1987) 879^ 889.
BBAGEN 24743 18-12-98