- Email: [email protected]
Interaction of Nitric Oxide with Photoexcited Rose Bengal: Evidence for One-Electron Reduction of Nitric Oxide to Nitroxyl Anion
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS
Vol. 324, No. 2, December 20, pp. 367–373, 1995
Interaction of Nitric Oxide with Photoexcited Rose Bengal: Evidence for One-Electron Reduction of Nitric Oxide to Nitroxyl Anion Ravinder Jit Singh,1 Neil Hogg, and B. Kalyanaraman2 Biophysics Research Institute, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, Wisconsin 53226-0509
Received July 24, 1995, and in revised form September 15, 1995
The interaction of nitric oxide (•NO) with Rose Bengal (RB) in the presence of electron donors was investigated. Upon illumination of a mixture of RB and •NO with visible light, an enhancement in the rate of •NO consumption was observed that increased with increasing RB concentration. In the presence of electron donors (NADH, glutathione, or ascorbate), the rates of • NO depletion increased further. NADH enhanced •NO depletion to a greater extent than either glutathione or ascorbate. Photoactivated RB under anaerobic conditions reacts with NADH to form the RB anion radical (RB•0), which has a characteristic visible absorption band centered at 418 nm. Rose Bengal anion radical disproportionates to give RB and a colorless reduced form of RB, RBH0. The net result of this process is the photobleaching of RB. The presence of •NO during irradiation of RB and NADH introduced a lag time into the kinetics of RB photobleaching. The length of this lag time was proportional to the concentration of •NO. A similar lag time, which was also dependent on the • NO concentration, was observed in the kinetics of formation of RB•0. The three-line electron spin resonance (ESR) spectrum of RB•0, with an intensity ratio 1:2:1, was obtained during irradiation of RB and NADH under anaerobic conditions. •NO introduced a concentration-dependent lag time into the kinetics of the appearance of this ESR signal. We propose that •NO oxidizes RB•0 to regenerate RB and thus inhibit photobleaching until •NO is consumed. This reaction predicts the formation of NO0, the one-electron reduced form of •NO. Nitrous oxide, a characteristic dimerization product of NO0, was detected by gas chromatography. This evidence indicates the occurrence 1 Visiting Fellow, Department of Chemistry, Guru Nanak Dev University, Amristar, India. 2 To whom correspondence should be addressed. Fax: 414-2668515. E-mail: [email protected].
of a Type I mechanism between photoactivated RB and • NO. q 1995 Academic Press, Inc. Key Words: nitric oxide; Rose Bengal; photosensitization; nitrous oxide; nitroxyl anion.
Nitric oxide (•NO),3 a physiologically relevant free radical that is involved in inflammation (1), neurotransmission (2), and the maintenance of vascular tone (3), has also been implicated in the mechanisms of many pathologies including atherosclerosis (4) and ischemia/reperfusion injury (5). The contradictory behavior of •NO has been rationalized in terms of reactions of one-electron-reduced and the one-electron-oxidized forms of •NO, i.e., nitroxyl anion (NO0) and nitrosonium cation (NO/), respectively (6). The impaired release of endothelial-dependent relaxing factor (EDRF), known to be •NO, has been implicated in the vasoconstrictive effects of photodynamic therapy (PDT) (7). The interaction of •NO with PDT sensitizers may be responsible for the effects of PDT on the vasculature. RB has been widely used as a photodynamic sensitizer due to a large molar extinction coefficient and high triplet quantum yield (8). Both the Type II (1O2 mechanism) and Type I (electron-transfer mechanism) chemistry of RB have been thoroughly investigated in aerobic media (9). RB in the triplet state interacts with triplet oxygen to form singlet oxygen with a quantum yield of 0.76. In the presence of a reduc3 Abbreviations used: EDRF, endothelium-derived relaxing factor; ESR, electron spin resonance; GSH, glutathione (reduced form); GSNO, nitrosoglutathione; HNO, nitrosyl hydride; NADH, b-nicotinamide adenine dinucleotide (reduced form); •NO, nitric oxide; N2O, nitrous oxide; NO0, nitroxyl anion; 0OONO, peroxynitrite; PBS, phosphate-buffered saline; PDT, photodynamic therapy; RB, Rose Bengal; RB•0, Rose Bengal anion radical; RBH0, reduced form of RB.
367
0003-9861/95 $12.00 Copyright q 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ m4606$9221
11-21-95 13:39:34
arca
AP: Archives
368
SINGH, HOGG, AND KALYANARAMAN
ing agent, electron transfer to triplet RB occurs leading to the formation of the RB anion radical (RB•0), which reacts with oxygen to give superoxide (9). •NO has been widely used as a physical quencher for the excited singlet and triplet states of photosensitizers in photochemical reactions (10). We report here that •NO can also react with RB•0 to regenerate RB and prevent photobleaching. Concomitantly, •NO is reduced to NO0 (via a Type I mechanism), which subsequently dimerizes to form nitrous oxide (N2O) (11). MATERIALS AND METHODS Materials. RB was obtained from Aldrich Chemical Co. (Milwaukee, WI) and was used without further purification. b-Nicotinamide adenine dinucleotide (reduced form) (NADH), glutathione (reduced form) (GSH), and sodium ascorbate were purchased from Sigma Chemical Co. (St. Louis, MO). •NO gas was supplied by Matheson Gas (Joliet, IL). Argon and N2O were obtained from AIRCO (Murray Hill, NJ). Solutions of •NO were prepared after purging through a solution of NaOH to remove higher oxides of •NO and quantified as reported earlier (12). Irradiation procedures. Before incidence on the reaction mixture, light was passed through a copper sulfate (100 g/liter) solution and a long pass filter of l ú 408 nm. For •NO electrode measurements, samples were purged with argon in a thermostated YSI (Yellow Springs, OH) oxygen-electrode chamber modified by the addition of a glass window and irradiated with a 300-W Hg arc lamp source. The kinetics of the reaction of •NO with RB in the presence of light were studied by immersing the •NO electrode (World Precision, Instruments, Inc., Sarasota, FL) through an air-tight seal into the thermostated chamber. Reagents were added through the seal. Spectrophotometric studies were performed using a Hewlett–Packard diode array spectrophotometer. Solution containing RB and NADH in phosphate-buffered saline (PBS) (20 mM, 145 mM NaCl, pH 7.4) in a sealed cuvette was degassed and purged with argon in the dark. Appropriate concentrations of •NO solution were added to the sealed cuvette through a septum, and the changes in the visible spectra during continuous irradiation were monitored. N2O measurements. Nitrous oxide was detected using a Varian 3700 gas chromatograph equipped with a Porapak Q column (6* 1 1 89) and a thermal conductivity detector operating at 307C with a flow rate of 5 ml/min (13). Peaks were identified by comparison with the retention time of authentic N2O. The solution containing RB and NADH in an 8-ml sealed glass vial was degassed and purged with argon gas in the dark. •NO solution was added to this mixture and the glass vial was irradiated for 15 min. A 100-ml portion of the head space gas was injected using an air-tight Hamilton syringe. Quantification of N2O production was performed by determining the total amount of N2O (i.e., N in the following equation) in the gas and aqueous phases using the equation N Å Cg(Vg / aV1), where Cg equals the concentration of N2O in the gas phase, Vg is the volume of the gas phase, V1 is the volume of the liquid phase, and a (0.544 at 257C) is the Bunsen absorption coefficient of N2O (14). Kinetic simulations. Kinetic simulations were performed using software written by Frank Neese, Faculta¨t fu¨r Biologie, Universita¨t Konstanz, Konstanz, Germany. ESR measurements. ESR spectra were recorded on a Varian E109 spectrometer operating at 9.5 GHz and employing 100 kHz field modulation. Samples were prepared in the nitrogen glove box and taken up in a 100-ml capillary, which was sealed at both ends with
/ m4606$9221
11-21-95 13:39:34
arca
FIG. 1. Effect of RB concentration on the rate of decay of •NO. The solution containing RB (0–10 mM) and •NO (20 mM) in deoxygenated PBS was irradiated with light (l ú 408 nm, I0 Å 100 W/m2) and the rate of consumption of •NO was monitored with an •NO electrode. The initial rate was plotted as a function of RB concentration. Inset: (a) The kinetics of decay in the presence of light alone, and (b) the effect of addition of RB (10 mM) to the same solution.
Miniseal (Baxter, IL). A 4-mm quartz tube containing the capillary was irradiated inside the ESR cavity using a monochromator.
RESULTS
Figure 1 shows the effect of RB on the stability of NO as monitored by a •NO electrode. The background rate of •NO decay under anaerobic conditions at 207C was not affected by light (l ú 408 nm, I0 Å 100 W/m2) or by RB in the dark (data not shown). However, RB (10 mM) in the presence of light greatly increased the rate of •NO decay (Fig. 1, inset). The increased rate of • NO decay was a linear function of RB concentration in the range of 0.2 to 3.0 mM and reached a plateau at ú4 mM (Fig. 1). NADH (70 mM) in the presence of RB (1.5 mM) and light increased the rate of •NO depletion from 0.02 to 0.7 mM s01. The rate of decay of •NO was also increased in the presence of GSH (70 mM) and ascorbate (200 mM) to 0.06 and 0.38 mM s01, respectively. This is analogous to the effects observed with RB and oxygen in the presence of reducing agents (15). The characteristic time-dependent optical absorbance changes that occur upon photoactivation of RB in the presence of •NO and NADH are shown in Fig. 2. RB has an absorbance of 550 nm (e Å 76,000 01 M cm01) (9). During irradiation, this absorbance peak decays with time and can be used to monitor photo•
AP: Archives
PHOTOACTIVATION OF NITRIC OXIDE
369
spectrum with a 1:2:1 intensity ratio (Fig. 5, inset). Based on data in the literature, this spectrum was analyzed in terms of an electron interacting with two equivalent protons (a1H Å a8H Å 3.1 G) (9). The formation of the radical was monitored by fixing the magnetic field at the peak of the center line (Fig. 5, inset). As shown, •NO introduced a lag time to the kinetics of the appearance of this radical (Fig. 5) in a manner identical to that observed by optical spectroscopy (Fig. 4). The N2O, formed from dimerization of NO0, was detected by gas chromatography (Fig. 6). Irradiation of RB (100 mM) and NADH (1 mM) in the presence of •NO (350 mM) gave a yield of 180 { 10 nmol of N2O after 15 min. This indicates that approximately 36% of the added •NO has been converted to NO0. Irradiation of either RB or NADH in the presence of •NO gave a much smaller yield of N2O (21.0 { 1.9 nmol) (Fig. 6). DISCUSSION FIG. 2. Changes in the characteristic absorption peaks for RB, RB•0, and NADH on irradiation. The solution containing RB (8 mM), NADH (80 mM), and •NO (16 mM) in degassed PBS was continuously irradiated with light (l ú 408 nm, I0 Å 30 W/m2) and the changes in the uv/visible absorption spectrum were monitored every 5 s. The figure shows a smoothed-surface representation of the data, and lines do not represent individual scans.
bleaching. The peak at 340 nm is largely due to the presence of NADH, and the changes in absorbance reflect NADH consumption. However, the many interfering absorbances in this region make such changes difficult to quantify. During the course of irradiation, a third peak, centered at 418 nm, forms and decays with time. This peak has been attributed previously to RB•0 (9). The extinction coefficient for this species has been determined to be 37,600 M01 cm01 by pulse radiolysis (16). Ultraviolet-visible absorption spectrophotometry can be used to monitor continuously the changes in RB and RB•0 during irradiation of RB and NADH. We have used this method to investigate the effects of •NO on the kinetics of this reaction. Figure 3A shows the effect of 0–80 mM •NO on the kinetics of RB photobleaching in the presence of NADH. The addition of •NO introduced a lag time into the kinetics of RB photobleaching. The length of this lag time was dependent on the amount of •NO added. The kinetic traces of RB•0 formation and decay from the experiments shown in Fig. 3A are shown in Fig. 4A. •NO introduced a lag time into the kinetics of formation of RB•0, the length of which was proportional to the concentration of •NO. The corresponding lag times apparent in Figs. 3A and 4A were similar in magnitude. The reaction between RB•0 and •NO was also monitored by ESR spectroscopy. Under anaerobic conditions, RB•0 exhibits a characteristic three-line ESR
/ m4606$9221
11-21-95 13:39:34
arca
The data presented here are consistent with the following mechanism (Eqs. [1] to [7]). Visible radiation
FIG. 3. Effect of •NO concentration on the kinetics of RB bleaching. (A) RB (8 mM), •NO (0–80 mM), and NADH (80 mM) were incubated in PBS. Upon continuous irradiation (l ú 408 nm, I0 Å 30 W/m2), the changes in absorbance at 550 nm were recorded as a function of time. (B) Kinetic simulation of the data in A using the mechanism defined in the text.
AP: Archives
370
SINGH, HOGG, AND KALYANARAMAN
k8
NO0 / NO0 / 2H/ r N2O / H2O
[8]
In the presence of NADH, triplet RB is reduced to RB•0 generating NAD• (Eq. [2], k2 Å 4000 M01 s01). NAD• may also react with triplet RB generating RB•0 and NAD/ (Eq. [3], k3 Å 9.6 1 108 M01 s01). In the absence of •NO, RB•0 disproportionates to give RB and RBH0, the colorless end product (Eq. [4], k4 Å 3.1 1 104 M01 s01). However, in the presence of •NO, RB•0 reacts with • NO forming RB and NO0 (Eq. [5]). NAD• formed in Eq. [2] may either disproportionate (k6 Å 1 1 108 M01 s01) or react with •NO. N2O is subsequently formed from the rapid disproportionation of NO0 as shown in Eq. [8] (k8 Å 2 1 109 M01 s01) (17). Kinetic simulations of the above mechanism (see Appendix), using published rate constants, are shown in Figs. 3B and 4B. The rate constants k1 , k5 , and k7 were allowed to vary to give the best fit to the experimental data. The simulations shown in Figs. 3B and 4B correspond to the experimental data in Figs. 3A and 4A. There is good agreement between the simulated and
FIG. 4. Effect of •NO concentration on the kinetics of the formation of RB•0. (A) RB (8 mM), •NO (0–80 mM), and NADH (80 mM) were incubated in PBS. Upon continuous irradiation (l ú 408 nm, I0 Å 30 W/m2), the changes in absorbance at 418 nm were recorded as a function of time. (B) Kinetic simulation of the data in A using the mechanism defined in the text.
excites ground-state RB to the triplet state (Eq. [1]). The rate for this reaction will depend on the intensity of light incident on the solution and on the optical characteristics of RB. k1
RB r 3RB k2
RB / NADH r RB•0 / NAD• / H/
3
k3
NAD• / RB r NAD/ / RB•0 k4
RB•0 / RB•0 / H/ r RB / RBH0 k5
RB•0 / •NO r NO0 / RB k6
NAD• / NAD• / H/ r NADH / NAD/ k7
NAD• / •NO r NAD/ / NO0
/ m4606$9221
11-21-95 13:39:34
[1] [2] [3] [4] [5] [6] [7]
arca
FIG. 5. Effect of •NO on the ESR signal of RB•0 formation. RB (50 mM) was incubated with NADH (250 mM) with various concentrations of •NO (0–1000 mM) under anaerobic conditions in PBS. The change in the intensity of the central field line of RB•0 was monitored. The sample was irradiated with monochromatic light (l Å 550 nm) (I0 Å 40 W/m2) inside the ESR cavity. Spectrometer conditions: modulation amplitude, 4 G; time constant, 0.128 s; microwave power, 10 mW. (Inset) Structure and ESR spectrum of RB•0. ESR spectrum was obtained upon irradiation of RB (50 mM) and NADH (250 mM) under nitrogen. Spectrometer conditions: modulation amplitude, 1 G; time constant, 0.128 s; microwave power, 5 mW.
AP: Archives
371
PHOTOACTIVATION OF NITRIC OXIDE
The photosensitized energy transfer between triplet RB and nitrosoglutathione (GSNO) has been proposed to account for the enhanced release of •NO from GSNO (Eqs. [11] and [12]). RB* / GSNO r RB / GSNO* GSNO* r GS• / •NO
FIG. 6. N2O formation from the reaction between •NO, NADH, and RB. The anaerobic solution containing RB (100 mM) and NADH (1 mM) in the presence of •NO (350 mM) was irradiated with light (l ú 408 nm, I Å 50 W/m2) for 15 min after which N2O formed in the head space was measured by gas chromatography.
experimental data at all concentrations of •NO. Differences in the exact shapes of the curves are apparent; however, it is conceivable that the above mechanism is a simplification of the system. However, we believe that the above mechanism models all of the salient features of the observed reaction. From these kinetic simulations, the rate constant for the reaction between •NO and RB•0 (k5) was determined to be 3.8 1 104 M01 s01. This is almost equal to the rate constant for the disproportionation of RB•0 and smaller than the rate constant for the corresponding reaction with oxygen. The rate constant between NAD• and NO• was determined to be 1.3 1 108 M01 s01. Nitric oxide was also consumed by the irradiation of RB in the absence of reducing agents (Fig. 1), albeit at a slower rate. The most likely explanation for this is that triplet RB is known to react with ground-state RB to give RB•0 and RB•/ (Eq. [9]). Reaction of •NO with either or both of these species would lead to •NO consumption. Direct energy transfer between RB and •NO (analogous to the Type II mechanism) may also lead to consumption of •NO, although the fate of the excitedstate NO• remains uncertain. RB / 1RB r RB•0 / RB•/
3
[9]
H/
/ m4606$9221
11-21-95 13:39:34
[10]
arca
[12]
Both electron-transfer and energy-transfer reactions seem plausible during RB-sensitized activation of •NO (Scheme 1). The one-electron reduction potential of RB/ RB•0 is negative (00.51 V) (16). NO0, being isoelectronic with molecular oxygen, can exist in the singlet (1NO0) or in the triplet (3NO0) form. The one-electron reduction potential (E7 relative to NHE) for the •NO/ 3 NO0 couple is /0.39 V and E7 for the •NO/1NO0 couple is 00.35 V (19). Therefore, the reduction of •NO to NO0 by RB•0 is thermodynamically feasible (20). The consequences of NO0 production in vivo remain unknown. It can bind to ferric iron complexes (especially heme iron) forming nitrosyl product or react with oxygen (21) to form peroxynitrite, a potent biological oxidant. Scheme 1 summarizes the possible Type I and Type II reaction mechanisms for the photoactivation of • NO and nitrosothiols in the presence of RB and indicates the potential formation of NO0 and peroxynitrite. Hearse and co-workers have developed a model for inducing sudden and controlled photodynamic oxidative stress in isolated hearts and cardiac myocytes (22). They have shown that photoactivation of RB leads to rapid electrophysiological changes, arrhythmias, and progressive decline in coronary flow in aerobic isolated rat and rabbit hearts. The reactive oxygen species (singlet oxygen or superoxide) were suggested to be responsible for the damage to the endothelium and/or vascular smooth muscle. Photoactivation of RB inhibits Na– Ca exchange current in myocytes possibly through alterations in sulfhydryl groups (23–25).
Photosensitized reactions between RB and nitrosothiol have been reported previously (18). RB•0 has been shown to reduce nitroxide to hydroxylamine via the Type I mechanism (Eq. [10]). RB•0 / úN{O• r RB / úN{OH
[11]
SCHEME I
AP: Archives
372
SINGH, HOGG, AND KALYANARAMAN
These pathophysiological effects could also be explained by the intermediacy of 0OONO. Peroxynitrite has been shown to oxidize the protein sulfhydryl group to disulfides (26). •NO is constantly generated in the vasculature by the constitutive endothelial •NO synthase, which maintains the vascular tone (3). In the present study, we have shown that •NO could be photoreduced to NO0, which can also form 0OONO on reaction with oxygen. Alternatively, 0OONO could be • formed by the reaction between O•0 2 and NO (27). The pathophysiological changes observed during photoactivation of RB in isolated hearts are consistent with the rapid depletion of EDRF vis a vis •NO present in the endothelium. As mentioned previously, PDT has been shown to impair the release of EDRF in endothelium (7). We postulate that reactions shown in Scheme I could also account for the pathophysiological changes in the myocardium caused by photoactivation of RB. The present findings may also be extended to other photochemical and biochemical systems. Radical anions are formed during photoreduction of sensitizers such as hematoporphyrin, uroporphyrin, etc. (28, 29). It is conceivable that •NO can be photoactivated to peroxynitrite via a similar mechanism (Scheme 1) in the presence of these photosensitizers. Radical anions are also formed as obligatory intermediates during reductive activation of xenobiotics (e.g., adriamycin, paraquat) (30). The toxicity of these compounds has been linked to their ability to form superoxide by redox activation. It is plausible that •NO may undergo a similar type of reduction to NO0 and 0OONO during redox activation of such xenobiotics. In conclusion, we have shown that RB, in the presence of light and reducing agents, can reduce •NO to NO0 via a Type I mechanism. Under these conditions, N2O, a diagnostic product of NO0 formation, has been detected. APPENDIX
The following differential equations were constructed from Eqs. [1]–[7]. This set of differential equations is not analytically soluble. Therefore, we have used numerical methods to calculate the kinetics of each species. d[RB] Å 0k1[RB] / k4[RB•0]2 dt / k5[RB•0][•NO] 0 k3[RB][NAD•] [1] d[3RB] Å k1[RB] 0 k2[3RB][NADH] dt
[2]
d[NADH] Å 0k2[3RB][NADH] / k6[NAD•][NAD•] [3] dt
/ m4606$9221
11-21-95 13:39:34
arca
d[RB•0] Å /k2[3RB][NADH] / k3[RB][NAD•] dt 0 2k4[RB•0]2 0 k5[RB•0][•NO]
[4]
•
d[NAD ] Å /k2[3RB][NADH] 0 k3[RB][NAD•] dt 0 2k6[NAD•]2 0 k7[NAD•][•NO]
[5]
/
d[NAD ] Å /k3[3RB][NAD•] dt / k6[NAD•]2 / k7[NAD•][•NO]
[6]
d[RBH0] Å /k4[RB•0]2 dt
[7]
d[•NO] Å 0k5[RB•0][•NO] 0 k7[NAD•][•NO] dt
[8]
d[NO0] Å /k5[RB•0][•NO] / k7[NAD•][•NO] dt
[9]
ACKNOWLEDGMENTS The research has been supported by NIH Grants CA49089, HL45048, and RR01008.
REFERENCES 1. Hibbs, J. B., Jr., Taintor, R. R., and Vavrin, Z. (1987) Science 235, 550–565. 2. Knowles, R. G., Placios, M., Palmer, R. M. J., and Moncada, S. (1989) Proc. Natl. Acad. Sci. USA 86, 5129–5162. 3. Palmer, R. M. J., Ferrige, A. G., and Moncada, S. (1987) Nature 327, 524–526. 4. Cook, J. P., and Tsao, P. S. (1994) Atheroscler. Thromb. 5, 653– 654. 5. Konorev, E. A., Tarpey, M. M., Joseph, J., Baker, J. E., and Kalyanaraman, B. (1995) J. Pharmacol. Exp. Ther. 274, 200– 206. 6. Stamler, J. S., Singel, D. J., and Loscalzo, J. (1992) Science 258, 1898–1902. 7. Gilissen, M. J., Merbel-de Wit, L. E. A., Star, W. M., Koster, J. F., and Sluiter, W. (1993) Cancer Res. 53, 2548–2552. 8. Neckers, D. C. (1989) J. Photochem. Photobiol. 47, 1–29. 9. Sarna, T., Zajac, J., Bowman, M. K., and Truscott, T. G. (1991) J. Photochem. Photobiol. 60A, 295–310. 10. Kusuhara, S., and Hardwick, R. (1964) J. Chem. Phys. 41, 2386– 2389. 11. Fukuto, J. M., Chaing, K., Hszieh, R., Wong, P., and Chandhuri, G. (1992) J. Pharmacol. Exp. Ther. 263, 546–551. 12. Hogg, N., Singh, R. J., Joseph, J., Neese, F., and Kalyanaraman, B. (1995) Free Rad. Res. Commun. 22, 47–56. 13. Fukuto, J. M., Wallace, G. C., Hsziaeh, R., and Chaudhuri, G. (1992) Biochem. Biopharmacol. 43, 607–613. 14. Pufhal, R. A., Wishnok, J. S., and Marletta, M. A. (1995) Biochemistry 34, 1930–1941. 15. Lee, P. C., and Rodgers, M. A. (1987) Photochem. Photobiol. 45, 79–87.
AP: Archives
PHOTOACTIVATION OF NITRIC OXIDE 16. Lambert, C., Sarna, T., and Truscott, T. G. (1990) J. Chem. Soc. Faraday Trans. 86, 3879–3882. 17. Bazylinski, D. A., and Hollocher, T. C. (1985) Inorg. Chem. 24, 4285–4288. 18. Singh, R. J., Hogg, N., Joseph, J., and Kalyanaraman, B. (1995) FEBS Lett. 360, 47–51. 19. Stanburg, D. M. (1989) Adv. Inorg. Chem. 33, 69–137. 20. Buettner, G. R. (1993) Arch. Biochem. Biophys. 300, 535–543. 21. Hughes, M. N., and Nicklin, H. G. (1971) J. Chem. Soc. A, 164– 168. 22. Hearse, D. J., Kusama, Y., and Bernier, M. (1989) Circ. Res. 65, 146–153. 23. Vandeplassche, G., Bernier, M., Thone´, F., Borgers, M., Kusama, Y., and Hearse, D. J. (1990) J. Mol. Cell. Cardiol. 22, 287–301.
/ m4606$9221
11-21-95 13:39:34
arca
373
24. Tarr, M., and Valenzeno, D. P. (1991) J. Mol. Cell. Cardiol. 23, 639–649. 25. Coetzee, W. A., Chikawa, H., and Hearse, D. J. (1994) Am. J. Physiol. H 266, 909–919. 26. Radi, R., Beckman, J. S., Bush, K. M., and Freeman, B. C. (1991) J. Biol. Chem. 266, 4244–4250. 27. Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. M., and Freeman, B. C. (1990) Proc. Natl. Acad. Sci. USA 87, 1620– 1624. 28. Felix, C. C., Reszka, K., and Sealy, R. C. (1983) Photochem. Photobiol. 37, 141–147. 29. Mauzerall, D., and Fehr, G. (1964) Biochem. Biophys. Acta 79, 430–432. 30. Mason, R. P. (1979) Reviews in Biochemical Toxicology, I (Hodgson, E., Bend, O. R., and Philpot, R. M., Eds.), p. 151, Elsevier/ North-Holland, New York.
AP: Archives
Vol. 324, No. 2, December 20, pp. 367–373, 1995
Interaction of Nitric Oxide with Photoexcited Rose Bengal: Evidence for One-Electron Reduction of Nitric Oxide to Nitroxyl Anion Ravinder Jit Singh,1 Neil Hogg, and B. Kalyanaraman2 Biophysics Research Institute, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, Wisconsin 53226-0509
Received July 24, 1995, and in revised form September 15, 1995
The interaction of nitric oxide (•NO) with Rose Bengal (RB) in the presence of electron donors was investigated. Upon illumination of a mixture of RB and •NO with visible light, an enhancement in the rate of •NO consumption was observed that increased with increasing RB concentration. In the presence of electron donors (NADH, glutathione, or ascorbate), the rates of • NO depletion increased further. NADH enhanced •NO depletion to a greater extent than either glutathione or ascorbate. Photoactivated RB under anaerobic conditions reacts with NADH to form the RB anion radical (RB•0), which has a characteristic visible absorption band centered at 418 nm. Rose Bengal anion radical disproportionates to give RB and a colorless reduced form of RB, RBH0. The net result of this process is the photobleaching of RB. The presence of •NO during irradiation of RB and NADH introduced a lag time into the kinetics of RB photobleaching. The length of this lag time was proportional to the concentration of •NO. A similar lag time, which was also dependent on the • NO concentration, was observed in the kinetics of formation of RB•0. The three-line electron spin resonance (ESR) spectrum of RB•0, with an intensity ratio 1:2:1, was obtained during irradiation of RB and NADH under anaerobic conditions. •NO introduced a concentration-dependent lag time into the kinetics of the appearance of this ESR signal. We propose that •NO oxidizes RB•0 to regenerate RB and thus inhibit photobleaching until •NO is consumed. This reaction predicts the formation of NO0, the one-electron reduced form of •NO. Nitrous oxide, a characteristic dimerization product of NO0, was detected by gas chromatography. This evidence indicates the occurrence 1 Visiting Fellow, Department of Chemistry, Guru Nanak Dev University, Amristar, India. 2 To whom correspondence should be addressed. Fax: 414-2668515. E-mail: [email protected].
of a Type I mechanism between photoactivated RB and • NO. q 1995 Academic Press, Inc. Key Words: nitric oxide; Rose Bengal; photosensitization; nitrous oxide; nitroxyl anion.
Nitric oxide (•NO),3 a physiologically relevant free radical that is involved in inflammation (1), neurotransmission (2), and the maintenance of vascular tone (3), has also been implicated in the mechanisms of many pathologies including atherosclerosis (4) and ischemia/reperfusion injury (5). The contradictory behavior of •NO has been rationalized in terms of reactions of one-electron-reduced and the one-electron-oxidized forms of •NO, i.e., nitroxyl anion (NO0) and nitrosonium cation (NO/), respectively (6). The impaired release of endothelial-dependent relaxing factor (EDRF), known to be •NO, has been implicated in the vasoconstrictive effects of photodynamic therapy (PDT) (7). The interaction of •NO with PDT sensitizers may be responsible for the effects of PDT on the vasculature. RB has been widely used as a photodynamic sensitizer due to a large molar extinction coefficient and high triplet quantum yield (8). Both the Type II (1O2 mechanism) and Type I (electron-transfer mechanism) chemistry of RB have been thoroughly investigated in aerobic media (9). RB in the triplet state interacts with triplet oxygen to form singlet oxygen with a quantum yield of 0.76. In the presence of a reduc3 Abbreviations used: EDRF, endothelium-derived relaxing factor; ESR, electron spin resonance; GSH, glutathione (reduced form); GSNO, nitrosoglutathione; HNO, nitrosyl hydride; NADH, b-nicotinamide adenine dinucleotide (reduced form); •NO, nitric oxide; N2O, nitrous oxide; NO0, nitroxyl anion; 0OONO, peroxynitrite; PBS, phosphate-buffered saline; PDT, photodynamic therapy; RB, Rose Bengal; RB•0, Rose Bengal anion radical; RBH0, reduced form of RB.
367
0003-9861/95 $12.00 Copyright q 1995 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ m4606$9221
11-21-95 13:39:34
arca
AP: Archives
368
SINGH, HOGG, AND KALYANARAMAN
ing agent, electron transfer to triplet RB occurs leading to the formation of the RB anion radical (RB•0), which reacts with oxygen to give superoxide (9). •NO has been widely used as a physical quencher for the excited singlet and triplet states of photosensitizers in photochemical reactions (10). We report here that •NO can also react with RB•0 to regenerate RB and prevent photobleaching. Concomitantly, •NO is reduced to NO0 (via a Type I mechanism), which subsequently dimerizes to form nitrous oxide (N2O) (11). MATERIALS AND METHODS Materials. RB was obtained from Aldrich Chemical Co. (Milwaukee, WI) and was used without further purification. b-Nicotinamide adenine dinucleotide (reduced form) (NADH), glutathione (reduced form) (GSH), and sodium ascorbate were purchased from Sigma Chemical Co. (St. Louis, MO). •NO gas was supplied by Matheson Gas (Joliet, IL). Argon and N2O were obtained from AIRCO (Murray Hill, NJ). Solutions of •NO were prepared after purging through a solution of NaOH to remove higher oxides of •NO and quantified as reported earlier (12). Irradiation procedures. Before incidence on the reaction mixture, light was passed through a copper sulfate (100 g/liter) solution and a long pass filter of l ú 408 nm. For •NO electrode measurements, samples were purged with argon in a thermostated YSI (Yellow Springs, OH) oxygen-electrode chamber modified by the addition of a glass window and irradiated with a 300-W Hg arc lamp source. The kinetics of the reaction of •NO with RB in the presence of light were studied by immersing the •NO electrode (World Precision, Instruments, Inc., Sarasota, FL) through an air-tight seal into the thermostated chamber. Reagents were added through the seal. Spectrophotometric studies were performed using a Hewlett–Packard diode array spectrophotometer. Solution containing RB and NADH in phosphate-buffered saline (PBS) (20 mM, 145 mM NaCl, pH 7.4) in a sealed cuvette was degassed and purged with argon in the dark. Appropriate concentrations of •NO solution were added to the sealed cuvette through a septum, and the changes in the visible spectra during continuous irradiation were monitored. N2O measurements. Nitrous oxide was detected using a Varian 3700 gas chromatograph equipped with a Porapak Q column (6* 1 1 89) and a thermal conductivity detector operating at 307C with a flow rate of 5 ml/min (13). Peaks were identified by comparison with the retention time of authentic N2O. The solution containing RB and NADH in an 8-ml sealed glass vial was degassed and purged with argon gas in the dark. •NO solution was added to this mixture and the glass vial was irradiated for 15 min. A 100-ml portion of the head space gas was injected using an air-tight Hamilton syringe. Quantification of N2O production was performed by determining the total amount of N2O (i.e., N in the following equation) in the gas and aqueous phases using the equation N Å Cg(Vg / aV1), where Cg equals the concentration of N2O in the gas phase, Vg is the volume of the gas phase, V1 is the volume of the liquid phase, and a (0.544 at 257C) is the Bunsen absorption coefficient of N2O (14). Kinetic simulations. Kinetic simulations were performed using software written by Frank Neese, Faculta¨t fu¨r Biologie, Universita¨t Konstanz, Konstanz, Germany. ESR measurements. ESR spectra were recorded on a Varian E109 spectrometer operating at 9.5 GHz and employing 100 kHz field modulation. Samples were prepared in the nitrogen glove box and taken up in a 100-ml capillary, which was sealed at both ends with
/ m4606$9221
11-21-95 13:39:34
arca
FIG. 1. Effect of RB concentration on the rate of decay of •NO. The solution containing RB (0–10 mM) and •NO (20 mM) in deoxygenated PBS was irradiated with light (l ú 408 nm, I0 Å 100 W/m2) and the rate of consumption of •NO was monitored with an •NO electrode. The initial rate was plotted as a function of RB concentration. Inset: (a) The kinetics of decay in the presence of light alone, and (b) the effect of addition of RB (10 mM) to the same solution.
Miniseal (Baxter, IL). A 4-mm quartz tube containing the capillary was irradiated inside the ESR cavity using a monochromator.
RESULTS
Figure 1 shows the effect of RB on the stability of NO as monitored by a •NO electrode. The background rate of •NO decay under anaerobic conditions at 207C was not affected by light (l ú 408 nm, I0 Å 100 W/m2) or by RB in the dark (data not shown). However, RB (10 mM) in the presence of light greatly increased the rate of •NO decay (Fig. 1, inset). The increased rate of • NO decay was a linear function of RB concentration in the range of 0.2 to 3.0 mM and reached a plateau at ú4 mM (Fig. 1). NADH (70 mM) in the presence of RB (1.5 mM) and light increased the rate of •NO depletion from 0.02 to 0.7 mM s01. The rate of decay of •NO was also increased in the presence of GSH (70 mM) and ascorbate (200 mM) to 0.06 and 0.38 mM s01, respectively. This is analogous to the effects observed with RB and oxygen in the presence of reducing agents (15). The characteristic time-dependent optical absorbance changes that occur upon photoactivation of RB in the presence of •NO and NADH are shown in Fig. 2. RB has an absorbance of 550 nm (e Å 76,000 01 M cm01) (9). During irradiation, this absorbance peak decays with time and can be used to monitor photo•
AP: Archives
PHOTOACTIVATION OF NITRIC OXIDE
369
spectrum with a 1:2:1 intensity ratio (Fig. 5, inset). Based on data in the literature, this spectrum was analyzed in terms of an electron interacting with two equivalent protons (a1H Å a8H Å 3.1 G) (9). The formation of the radical was monitored by fixing the magnetic field at the peak of the center line (Fig. 5, inset). As shown, •NO introduced a lag time to the kinetics of the appearance of this radical (Fig. 5) in a manner identical to that observed by optical spectroscopy (Fig. 4). The N2O, formed from dimerization of NO0, was detected by gas chromatography (Fig. 6). Irradiation of RB (100 mM) and NADH (1 mM) in the presence of •NO (350 mM) gave a yield of 180 { 10 nmol of N2O after 15 min. This indicates that approximately 36% of the added •NO has been converted to NO0. Irradiation of either RB or NADH in the presence of •NO gave a much smaller yield of N2O (21.0 { 1.9 nmol) (Fig. 6). DISCUSSION FIG. 2. Changes in the characteristic absorption peaks for RB, RB•0, and NADH on irradiation. The solution containing RB (8 mM), NADH (80 mM), and •NO (16 mM) in degassed PBS was continuously irradiated with light (l ú 408 nm, I0 Å 30 W/m2) and the changes in the uv/visible absorption spectrum were monitored every 5 s. The figure shows a smoothed-surface representation of the data, and lines do not represent individual scans.
bleaching. The peak at 340 nm is largely due to the presence of NADH, and the changes in absorbance reflect NADH consumption. However, the many interfering absorbances in this region make such changes difficult to quantify. During the course of irradiation, a third peak, centered at 418 nm, forms and decays with time. This peak has been attributed previously to RB•0 (9). The extinction coefficient for this species has been determined to be 37,600 M01 cm01 by pulse radiolysis (16). Ultraviolet-visible absorption spectrophotometry can be used to monitor continuously the changes in RB and RB•0 during irradiation of RB and NADH. We have used this method to investigate the effects of •NO on the kinetics of this reaction. Figure 3A shows the effect of 0–80 mM •NO on the kinetics of RB photobleaching in the presence of NADH. The addition of •NO introduced a lag time into the kinetics of RB photobleaching. The length of this lag time was dependent on the amount of •NO added. The kinetic traces of RB•0 formation and decay from the experiments shown in Fig. 3A are shown in Fig. 4A. •NO introduced a lag time into the kinetics of formation of RB•0, the length of which was proportional to the concentration of •NO. The corresponding lag times apparent in Figs. 3A and 4A were similar in magnitude. The reaction between RB•0 and •NO was also monitored by ESR spectroscopy. Under anaerobic conditions, RB•0 exhibits a characteristic three-line ESR
/ m4606$9221
11-21-95 13:39:34
arca
The data presented here are consistent with the following mechanism (Eqs. [1] to [7]). Visible radiation
FIG. 3. Effect of •NO concentration on the kinetics of RB bleaching. (A) RB (8 mM), •NO (0–80 mM), and NADH (80 mM) were incubated in PBS. Upon continuous irradiation (l ú 408 nm, I0 Å 30 W/m2), the changes in absorbance at 550 nm were recorded as a function of time. (B) Kinetic simulation of the data in A using the mechanism defined in the text.
AP: Archives
370
SINGH, HOGG, AND KALYANARAMAN
k8
NO0 / NO0 / 2H/ r N2O / H2O
[8]
In the presence of NADH, triplet RB is reduced to RB•0 generating NAD• (Eq. [2], k2 Å 4000 M01 s01). NAD• may also react with triplet RB generating RB•0 and NAD/ (Eq. [3], k3 Å 9.6 1 108 M01 s01). In the absence of •NO, RB•0 disproportionates to give RB and RBH0, the colorless end product (Eq. [4], k4 Å 3.1 1 104 M01 s01). However, in the presence of •NO, RB•0 reacts with • NO forming RB and NO0 (Eq. [5]). NAD• formed in Eq. [2] may either disproportionate (k6 Å 1 1 108 M01 s01) or react with •NO. N2O is subsequently formed from the rapid disproportionation of NO0 as shown in Eq. [8] (k8 Å 2 1 109 M01 s01) (17). Kinetic simulations of the above mechanism (see Appendix), using published rate constants, are shown in Figs. 3B and 4B. The rate constants k1 , k5 , and k7 were allowed to vary to give the best fit to the experimental data. The simulations shown in Figs. 3B and 4B correspond to the experimental data in Figs. 3A and 4A. There is good agreement between the simulated and
FIG. 4. Effect of •NO concentration on the kinetics of the formation of RB•0. (A) RB (8 mM), •NO (0–80 mM), and NADH (80 mM) were incubated in PBS. Upon continuous irradiation (l ú 408 nm, I0 Å 30 W/m2), the changes in absorbance at 418 nm were recorded as a function of time. (B) Kinetic simulation of the data in A using the mechanism defined in the text.
excites ground-state RB to the triplet state (Eq. [1]). The rate for this reaction will depend on the intensity of light incident on the solution and on the optical characteristics of RB. k1
RB r 3RB k2
RB / NADH r RB•0 / NAD• / H/
3
k3
NAD• / RB r NAD/ / RB•0 k4
RB•0 / RB•0 / H/ r RB / RBH0 k5
RB•0 / •NO r NO0 / RB k6
NAD• / NAD• / H/ r NADH / NAD/ k7
NAD• / •NO r NAD/ / NO0
/ m4606$9221
11-21-95 13:39:34
[1] [2] [3] [4] [5] [6] [7]
arca
FIG. 5. Effect of •NO on the ESR signal of RB•0 formation. RB (50 mM) was incubated with NADH (250 mM) with various concentrations of •NO (0–1000 mM) under anaerobic conditions in PBS. The change in the intensity of the central field line of RB•0 was monitored. The sample was irradiated with monochromatic light (l Å 550 nm) (I0 Å 40 W/m2) inside the ESR cavity. Spectrometer conditions: modulation amplitude, 4 G; time constant, 0.128 s; microwave power, 10 mW. (Inset) Structure and ESR spectrum of RB•0. ESR spectrum was obtained upon irradiation of RB (50 mM) and NADH (250 mM) under nitrogen. Spectrometer conditions: modulation amplitude, 1 G; time constant, 0.128 s; microwave power, 5 mW.
AP: Archives
371
PHOTOACTIVATION OF NITRIC OXIDE
The photosensitized energy transfer between triplet RB and nitrosoglutathione (GSNO) has been proposed to account for the enhanced release of •NO from GSNO (Eqs. [11] and [12]). RB* / GSNO r RB / GSNO* GSNO* r GS• / •NO
FIG. 6. N2O formation from the reaction between •NO, NADH, and RB. The anaerobic solution containing RB (100 mM) and NADH (1 mM) in the presence of •NO (350 mM) was irradiated with light (l ú 408 nm, I Å 50 W/m2) for 15 min after which N2O formed in the head space was measured by gas chromatography.
experimental data at all concentrations of •NO. Differences in the exact shapes of the curves are apparent; however, it is conceivable that the above mechanism is a simplification of the system. However, we believe that the above mechanism models all of the salient features of the observed reaction. From these kinetic simulations, the rate constant for the reaction between •NO and RB•0 (k5) was determined to be 3.8 1 104 M01 s01. This is almost equal to the rate constant for the disproportionation of RB•0 and smaller than the rate constant for the corresponding reaction with oxygen. The rate constant between NAD• and NO• was determined to be 1.3 1 108 M01 s01. Nitric oxide was also consumed by the irradiation of RB in the absence of reducing agents (Fig. 1), albeit at a slower rate. The most likely explanation for this is that triplet RB is known to react with ground-state RB to give RB•0 and RB•/ (Eq. [9]). Reaction of •NO with either or both of these species would lead to •NO consumption. Direct energy transfer between RB and •NO (analogous to the Type II mechanism) may also lead to consumption of •NO, although the fate of the excitedstate NO• remains uncertain. RB / 1RB r RB•0 / RB•/
3
[9]
H/
/ m4606$9221
11-21-95 13:39:34
[10]
arca
[12]
Both electron-transfer and energy-transfer reactions seem plausible during RB-sensitized activation of •NO (Scheme 1). The one-electron reduction potential of RB/ RB•0 is negative (00.51 V) (16). NO0, being isoelectronic with molecular oxygen, can exist in the singlet (1NO0) or in the triplet (3NO0) form. The one-electron reduction potential (E7 relative to NHE) for the •NO/ 3 NO0 couple is /0.39 V and E7 for the •NO/1NO0 couple is 00.35 V (19). Therefore, the reduction of •NO to NO0 by RB•0 is thermodynamically feasible (20). The consequences of NO0 production in vivo remain unknown. It can bind to ferric iron complexes (especially heme iron) forming nitrosyl product or react with oxygen (21) to form peroxynitrite, a potent biological oxidant. Scheme 1 summarizes the possible Type I and Type II reaction mechanisms for the photoactivation of • NO and nitrosothiols in the presence of RB and indicates the potential formation of NO0 and peroxynitrite. Hearse and co-workers have developed a model for inducing sudden and controlled photodynamic oxidative stress in isolated hearts and cardiac myocytes (22). They have shown that photoactivation of RB leads to rapid electrophysiological changes, arrhythmias, and progressive decline in coronary flow in aerobic isolated rat and rabbit hearts. The reactive oxygen species (singlet oxygen or superoxide) were suggested to be responsible for the damage to the endothelium and/or vascular smooth muscle. Photoactivation of RB inhibits Na– Ca exchange current in myocytes possibly through alterations in sulfhydryl groups (23–25).
Photosensitized reactions between RB and nitrosothiol have been reported previously (18). RB•0 has been shown to reduce nitroxide to hydroxylamine via the Type I mechanism (Eq. [10]). RB•0 / úN{O• r RB / úN{OH
[11]
SCHEME I
AP: Archives
372
SINGH, HOGG, AND KALYANARAMAN
These pathophysiological effects could also be explained by the intermediacy of 0OONO. Peroxynitrite has been shown to oxidize the protein sulfhydryl group to disulfides (26). •NO is constantly generated in the vasculature by the constitutive endothelial •NO synthase, which maintains the vascular tone (3). In the present study, we have shown that •NO could be photoreduced to NO0, which can also form 0OONO on reaction with oxygen. Alternatively, 0OONO could be • formed by the reaction between O•0 2 and NO (27). The pathophysiological changes observed during photoactivation of RB in isolated hearts are consistent with the rapid depletion of EDRF vis a vis •NO present in the endothelium. As mentioned previously, PDT has been shown to impair the release of EDRF in endothelium (7). We postulate that reactions shown in Scheme I could also account for the pathophysiological changes in the myocardium caused by photoactivation of RB. The present findings may also be extended to other photochemical and biochemical systems. Radical anions are formed during photoreduction of sensitizers such as hematoporphyrin, uroporphyrin, etc. (28, 29). It is conceivable that •NO can be photoactivated to peroxynitrite via a similar mechanism (Scheme 1) in the presence of these photosensitizers. Radical anions are also formed as obligatory intermediates during reductive activation of xenobiotics (e.g., adriamycin, paraquat) (30). The toxicity of these compounds has been linked to their ability to form superoxide by redox activation. It is plausible that •NO may undergo a similar type of reduction to NO0 and 0OONO during redox activation of such xenobiotics. In conclusion, we have shown that RB, in the presence of light and reducing agents, can reduce •NO to NO0 via a Type I mechanism. Under these conditions, N2O, a diagnostic product of NO0 formation, has been detected. APPENDIX
The following differential equations were constructed from Eqs. [1]–[7]. This set of differential equations is not analytically soluble. Therefore, we have used numerical methods to calculate the kinetics of each species. d[RB] Å 0k1[RB] / k4[RB•0]2 dt / k5[RB•0][•NO] 0 k3[RB][NAD•] [1] d[3RB] Å k1[RB] 0 k2[3RB][NADH] dt
[2]
d[NADH] Å 0k2[3RB][NADH] / k6[NAD•][NAD•] [3] dt
/ m4606$9221
11-21-95 13:39:34
arca
d[RB•0] Å /k2[3RB][NADH] / k3[RB][NAD•] dt 0 2k4[RB•0]2 0 k5[RB•0][•NO]
[4]
•
d[NAD ] Å /k2[3RB][NADH] 0 k3[RB][NAD•] dt 0 2k6[NAD•]2 0 k7[NAD•][•NO]
[5]
/
d[NAD ] Å /k3[3RB][NAD•] dt / k6[NAD•]2 / k7[NAD•][•NO]
[6]
d[RBH0] Å /k4[RB•0]2 dt
[7]
d[•NO] Å 0k5[RB•0][•NO] 0 k7[NAD•][•NO] dt
[8]
d[NO0] Å /k5[RB•0][•NO] / k7[NAD•][•NO] dt
[9]
ACKNOWLEDGMENTS The research has been supported by NIH Grants CA49089, HL45048, and RR01008.
REFERENCES 1. Hibbs, J. B., Jr., Taintor, R. R., and Vavrin, Z. (1987) Science 235, 550–565. 2. Knowles, R. G., Placios, M., Palmer, R. M. J., and Moncada, S. (1989) Proc. Natl. Acad. Sci. USA 86, 5129–5162. 3. Palmer, R. M. J., Ferrige, A. G., and Moncada, S. (1987) Nature 327, 524–526. 4. Cook, J. P., and Tsao, P. S. (1994) Atheroscler. Thromb. 5, 653– 654. 5. Konorev, E. A., Tarpey, M. M., Joseph, J., Baker, J. E., and Kalyanaraman, B. (1995) J. Pharmacol. Exp. Ther. 274, 200– 206. 6. Stamler, J. S., Singel, D. J., and Loscalzo, J. (1992) Science 258, 1898–1902. 7. Gilissen, M. J., Merbel-de Wit, L. E. A., Star, W. M., Koster, J. F., and Sluiter, W. (1993) Cancer Res. 53, 2548–2552. 8. Neckers, D. C. (1989) J. Photochem. Photobiol. 47, 1–29. 9. Sarna, T., Zajac, J., Bowman, M. K., and Truscott, T. G. (1991) J. Photochem. Photobiol. 60A, 295–310. 10. Kusuhara, S., and Hardwick, R. (1964) J. Chem. Phys. 41, 2386– 2389. 11. Fukuto, J. M., Chaing, K., Hszieh, R., Wong, P., and Chandhuri, G. (1992) J. Pharmacol. Exp. Ther. 263, 546–551. 12. Hogg, N., Singh, R. J., Joseph, J., Neese, F., and Kalyanaraman, B. (1995) Free Rad. Res. Commun. 22, 47–56. 13. Fukuto, J. M., Wallace, G. C., Hsziaeh, R., and Chaudhuri, G. (1992) Biochem. Biopharmacol. 43, 607–613. 14. Pufhal, R. A., Wishnok, J. S., and Marletta, M. A. (1995) Biochemistry 34, 1930–1941. 15. Lee, P. C., and Rodgers, M. A. (1987) Photochem. Photobiol. 45, 79–87.
AP: Archives
PHOTOACTIVATION OF NITRIC OXIDE 16. Lambert, C., Sarna, T., and Truscott, T. G. (1990) J. Chem. Soc. Faraday Trans. 86, 3879–3882. 17. Bazylinski, D. A., and Hollocher, T. C. (1985) Inorg. Chem. 24, 4285–4288. 18. Singh, R. J., Hogg, N., Joseph, J., and Kalyanaraman, B. (1995) FEBS Lett. 360, 47–51. 19. Stanburg, D. M. (1989) Adv. Inorg. Chem. 33, 69–137. 20. Buettner, G. R. (1993) Arch. Biochem. Biophys. 300, 535–543. 21. Hughes, M. N., and Nicklin, H. G. (1971) J. Chem. Soc. A, 164– 168. 22. Hearse, D. J., Kusama, Y., and Bernier, M. (1989) Circ. Res. 65, 146–153. 23. Vandeplassche, G., Bernier, M., Thone´, F., Borgers, M., Kusama, Y., and Hearse, D. J. (1990) J. Mol. Cell. Cardiol. 22, 287–301.
/ m4606$9221
11-21-95 13:39:34
arca
373
24. Tarr, M., and Valenzeno, D. P. (1991) J. Mol. Cell. Cardiol. 23, 639–649. 25. Coetzee, W. A., Chikawa, H., and Hearse, D. J. (1994) Am. J. Physiol. H 266, 909–919. 26. Radi, R., Beckman, J. S., Bush, K. M., and Freeman, B. C. (1991) J. Biol. Chem. 266, 4244–4250. 27. Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. M., and Freeman, B. C. (1990) Proc. Natl. Acad. Sci. USA 87, 1620– 1624. 28. Felix, C. C., Reszka, K., and Sealy, R. C. (1983) Photochem. Photobiol. 37, 141–147. 29. Mauzerall, D., and Fehr, G. (1964) Biochem. Biophys. Acta 79, 430–432. 30. Mason, R. P. (1979) Reviews in Biochemical Toxicology, I (Hodgson, E., Bend, O. R., and Philpot, R. M., Eds.), p. 151, Elsevier/ North-Holland, New York.
AP: Archives