- Email: [email protected]
Validation of Lucigenin as a Chemiluminescent Probe to Monitor Vascular Superoxide as Well as Basal Vascular Nitric Oxide Production
Biochemical and Biophysical Research Communications 254, 319 –324 (1999) Article ID bbrc.1998.9942, available online at http://www.idealibrary.com on
Validation of Lucigenin as a Chemiluminescent Probe to Monitor Vascular Superoxide as Well as Basal Vascular Nitric Oxide Production Mikhail P. Skatchkov, Daniel Sperling, Ulrich Hink, Alexander Mu¨lsch,* David G. Harrison,† Irene Sindermann, Thomas Meinertz, and Thomas Mu¨nzel 1 Division of Cardiology, Department of Internal Medicine, University Hospital Eppendorf, Hamburg, Germany; *Johann Wolfgang v. Goethe University, Frankfurt, Germany; and †Emory University, Atlanta, Georgia 30322
Received November 9, 1998
Lucigenin has been widely used as a chemiluminescent substrate to monitor vascular superoxide (O 2•2) formation. The validity of lucigenin for detection of O 2•2 has been questioned because O 2•2 is generated by lucigenin itself. It has been shown that the concentration of lucigenin is a critical parameter affecting the validity of this assay. In the present studies we evaluated a reduced concentration of lucigenin (5 mM) as a tool to quantify O 2•2 production in vascular tissue. Lucigenin-induced effects on endothelial function were assessed by isometric tension recording of isolated aortic rings suspended in organ baths. The effects of lucigenin on O 2•2 production were studied using spin trapping and electron spin resonance spectroscopy. Lucigenin at 250 mM but not at 5 mM caused a significant attenuation of endotheliumdependent relaxations to acetylcholine, which was prevented by pretreatment with superoxide dismutase. Spin-trapping studies revealed that lucigenin at 250 mM increased vascular O 2•2 production several fold while 5 mM lucigenin did not stimulate O 2•2 production. Inhibition of NO synthase by N G-momomethyl-L-arginine as well as the removal of the endothelium almost doubled lucigenin-derived chemiluminescence (LDCL), indicating that basal production of endothelium-derived NO depresses the baseline chemiluminescence signal. Thus, lucigenin at a concentration of 5 mM seems to be a sensitive and valid probe for assessing O 2•2 in vascular tissue. It can also be used as an indirect probe to estimate basal vascular NO release. © 1999 Academic Press
1 To whom correspondence should be addressed at Universita¨tskrankenhaus Eppendorf, Abteilung fu¨r Kardiologie, Martinistr. 52, 20246 Hamburg, Germany. Fax: 49-40-4717-3988. E-mail: muenzel@ uke.uni-hamburg.de.
Endothelial dysfunction secondary to reduced NObioactivity has been postulated to play a pathogenic role in the initiation of vascular disease (1). Several experimental and clinical studies have established endothelial dysfunction in diseases such as hypercholesterolemia (2, 3), arterial hypertension (4), chronic smoking (5) and diabetes mellitus (6). Though the precise mechanism underlying endothelial dysfunction remains to be established, one mechanism likely involves inactivation of NO by oxygen derived free radicals (7). The chemiluminescent compound lucigenin has been widely used to assay formation of superoxide anion (O 2•2) in neutrophils (8), macrophages (9), Kupffer cells (10), and in intact vascular tissue (11–13). Traditionally, it was applied in a rather high concentration (250 mM). Under physiological conditions lucigenin exists mainly as a dication (lucigenin 21) (14). Recently it was found that lucigenin in concentrations exceeding 20 mM induces formation of O 2•2 by accepting electrons from the flavin of NO synthase. The transiently formed cation radical (lucigenin 1°) acts as a redox-cycling compound capable to reduce oxygen to O 2•2 (15). More recently, however, Li et al. convincingly demonstrated that O 2•2 production strikingly depended on the lucigenin concentration (16). At 5 mM, lucigenin elicited marked chemiluminescence signals in O 2•2 generating enzyme systems and in isolated cells without stimulating additional O 2•2 production, as shown by electron spin resonance (ESR) studies. The authors recommended spin-trapping experiments to verify the absence of O 2•2 production by lucigenin. Therefore, the first aim of the present study was to test whether or not lucigenin in a concentration of 5 mM is sufficiently sensitive to detect vascular O 2•2 formation. In addition, we used ESR spectroscopy (17) to determine if this concentration of lucigenin is capable of generating O 2•2 in vascular tissue.
319
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Previous studies using lucigenin in a concentration of 250 mM failed to detect changes in the chemiluminescent signal when vessels were exposed to the NO synthase inhibitor N G-nitro-L-arginine (18). This observation is in contradiction to the fact that a significant part of NO derived from endothelial NO synthase or from exogenous NO donors (nitroglycerin, sodium nitroprusside, etc.) is able to react with O 2•2 to form peroxynitrite (19). Indeed, using the cypridina luciferin analogue CLA we recently demonstrated that incubation of isolated blood vessels with N G-monomethyl-Larginine (L-NMMA) or removal of the endothelium markedly increased CLA-chemiluminescence. This finding suggested that endothelium-derived NO affects the O 2•2 induced chemiluminescence signals in vascular tissue (17). Based on these considerations a further objective of the present study was to test whether or not lucigenin in a low concentration (5 mM) can be used to monitor basal NO production. The increase in lucigenin-derived chemiluminescence (LDCL) in response to removal of the endothelium or inhibition of NOS III could be taken as an indirect measure for basal NO formation. Finally, we tested lucigenin applied in low concentrations (5 mM) as a tool to quantify vascular O 2•2 production in vessels from control and hyperlipidemic rabbits and compared these results with those obtained by a recently characterized chemiluminescent O 2•2 detector, CLA, and by spin trapping and electron spin resonance spectroscopy. METHODS Animals and tissue preparation. Studies were performed with isolated aortas from New Zealand White rabbits and hypercholesterolemic Watanabe rabbits. The animals were killed by an overdose of Na pentobarbital The thoracic aorta was isolated and kept in cold Krebs–Henseleit solution of the following composition (mM): NaCl 118.3; KCl 4.69; CaCl 2 1.87; MgSO 4 1.20; K 2HPO 4 1.03; NaHCO 3 25.0; and glucose 11.1; pH 7.4). The aorta was cleaned of adhering adventitial tissue and cut into ring segments of 5 mm length. Organ chamber experiments. Eight ring segments were suspended in individual organ chambers filled with aerated buffer solution (95% O 2; 5% CO 2; 37°C) and changes in isometric tension were recorded with linear force transducers. All experiments were performed in the presence of 10 mM indomethacin to prevent synthesis of prostaglandins. Before the experiments, the rings were gradually stretched and equilibrated at a resting tension of 5 g (rabbit aorta) and 3 g (rat aorta) for at least 1 h. Subsequently, the ring segments were exposed at least three times to 80 mM KCl until a constant contraction was obtained. The vessels were preconstricted with phenylephrine (PE; 10 27 M) to achieve 40 –50% of maximal KClinduced tone and then the endothelium-dependent vasodilator acetylcholine (ACh) was added in cumulative concentrations (1 nM–3 mM). After 30 min incubation with lucigenin (5 or 250 mM) vasodilator responses to ACh were again assessed. To address a potential role of O 22 in lucigenin-induced endothelial dysfunction the concentration response-relationship to ACh was tested 30 min after addition of superoxide dismutase (SOD, 200 U/ml). Detect ion of O 2•2 formation in vascular tissue by LDCL. Vascular O 2•2 formation was measured by recording the chemiluminescence intensity of tissue probes in presence of 5 and 250 mM lucigenin.
Briefly, 5 mm ring segments were placed in a modified Krebs–Hepes buffer and equilibrated for 30 min at 37°C. Scintillation vials containing 2 ml Krebs–Hepes buffer with lucigenin were placed into a scintillation counter (LS 7001, Beckman Instruments Inc., Fullerton, CA) switched to the out-of-coincidence mode. After dark adaptation, background signals were recorded and a vascular segment was added to the vial. Photon counts were then recorded every minute for 8 minutes, and the background was subtracted. The vessels were then dried for 24 h at 90°C and weighed. The results were expressed as counts per min per mg dry weight as described (13).In separate experiments vascular O 22 was assayed using the cypridina luciferin analogue CLA as recently described (17). The assay buffer contained metal chelators such as diethyl-dithiocarbamate (DETC, 100 mM) and deferoxamine (100 mM) to prevent metal-ion-catalyzed reactions increasing the rate of self-oxidation of CLA. DETC was always added to the buffer solution before deferoxamine. Detection of NO in vascular tissue by LDCL quenching. To address the influence of endothelial-derived NO on the chemiluminescence signal associated with the formation of O 22 in some experiments endothelial NO formation was abolished by mechanical removal of the endothelium, or by incubation of vessels for 30 min at 37°C with N G-monomethyl-L-arginine (L-NMMA, 1 mM), an inhibitor of NO synthase. Spin trapping of O 2•2 radicals and electron spin resonance (ESR) studies. To test whether lucigenin elicits O 2•2 formation in vessels the O 2•2 induced generation of nitroxyl radicals was monitored using the spin-trap 1-hydroxy-3-carboxy-2,2,5,-tetramethyl-pyrrolidine hydrochloride (CPH, 1 mM) in the presence of diethylenetriamineaminepentaacetic acid (DTPA, 0.1 mM) as described recently (17). Aortic rings (4 mg) were treated with L-NMMA (1 mM, 30 min) to prevent formation of NO and subsequent oxidation of CPH by peroxynitrite. The vessels were then incubated in Krebs–Henseleit (KH, pH 7,4) solution containing 1 mM CPH with or without lucigenin (5 or 250 mM). Thereafter the vessels were transferred into quartz capillaries (internal diameter 0.8 mm) filled with KH solution and fixed in a dual probe resonator in the cavity of the ESR spectrometer (ER 4105 DR, Bruker, Germany). ESR spectrometer settings were as described recently (17). The concentration of the stable nitroxyl radical CP° generated from O 2•2 and CPH was measured by graphical evaluation of the intensity of the triplet ESR-spectra, using the low field component. Data are presented as percentage increase in the intensity of the ESR-signal after subtraction of the baseline signal originating from autooxidized CPH. Chemicals. CPH and L-NMMA were from Alexis Biochemicals (San Diego, CA). All other chemicals were from Sigma (Deisenhofen, Germany).
RESULTS Effects of high and low concentrations of lucigenin on endothelium-dependent and -independent vasodilation. In PE-constricted rabbit aortic rings the endothelium dependent vasodilator acetylcholine (ACh) induced a dose-dependent relaxation, maximally by 80 6 2% at the highest concentration applied (Table 1). Incubation with 250 mM lucigenin for 30 min markedly decreased the efficacy and potency of ACh (Fig. 1). The lucigenininduced inhibition of vasodilator responses to ACh was markedly improved by SOD (200 U/ml). In contrast, 5 mM lucigenin did not significantly affect vasodilator responses to ACh (Fig. 1). LDCL of vascular tissue exposed to low and high concentrations of lucigenin. In rabbit aortic segments from control animals the chemiluminescence signal
320
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Effects of High- and Low-Dose Lucigenin on EndotheliumDependent and Endothelium-Independent Vasodilation in Rabbit Aorta Acetylcholine
Control 15 mM lucigenin 1250 mM lucigenin 1250 mM lucigenin 1 SOD
ED 50 (2log M)
Max relaxation (%)
27.17 6 0.08 27.09 6 0.07 27.08 6 0.14 27.23 6 0.05
89 6 2 87 6 1 43 6 4* 84 6 4†
Note. Data are presented as means 6 SEM from 8 –10 experiments. SOD, superoxide dismutase, 200 U/ml. * p , 0.05 vs control, †p , 0.05 vs 250 mM lucigenin.
amounted to 1200 6 34 counts/mg/min using 250 mM lucigenin and 478 6 12 counts/mg/min using 5 mM lucigenin (Fig. 2). We also tested the effects of low and high concentrations of lucigenin on CL in vessels from hyperlipidemic rabbits (WHHL). These animals exhibited a significant degree of endothelial dysfunction as indicated by the attenuated vasodilator response to ACh (EC 50 (2log M): WHHL: 6.98 6 0.12 vs control: 7.45 6 0.12. Maximal relaxation: WHHL: 68 6 4% vs control: 92 6 2%; p , 0.05). In aortas of hyperlipidemic animals the luminescence signal was 2660 6 44 counts/mg/min with 250 mM lucigenin and 972 6 121 with 5 mM lucigenin. Therefore, despite a higher luminescence signal with higher concentrated lucigenin the ratio of the signal intensities of control vs hyperlipidemic animals at both concentrations of lucigenin was practically identical (Fig. 2). Similarly, the CLAelicited chemiluminescence was significantly different between control vessels and vessels from hyperlipidemic Watanabe rabbits (Control: 8700 6 1036 counts/ mg/min, n 5 6, WHHL: 25046 6 5155 counts/mg/min, n 5 5, Fig. 4). In accordance with these findings, ESR studies using the spin trap CPH revealed a significant increase in O 2•2 production in vessels from hypercholesterolemic rabbits compared to control vessels (control: 18.6 6 1.1, n 5 5; WHHL: 91.9 6 7.1 arbitrary units, n 5 5, p , 0.05. Fig. 2). Effects of high and low concentrations of lucigenin on the rate of superoxide production in vascular tissue. Incubation of aortic tissue with the spin trap CPH led to time-dependent increase in concentration of CP° radicals as indicated by the increase of intensity of the characteristic ESR triplet signal (17). The rate of CP° radical formation was not affected by 5 mM lucigenin (Fig. 3). In contrast, lucigenin at 250 mM induced a marked acceleration of CP° radical formation by 3.5fold (Fig. 3).
Detection of basal vascular nitric oxide formation by LDCL. In control rabbit aorta removal of the endothelium slightly, but significantly increased LDCL when lucigenin was used in a concentration of 250 mM. Incubation of aortic tissue with L-NMMA failed to increase LDCL significantly under these conditions (Fig. 4). In contrast, in rabbit aortas incubated with 5 mM lucigenin, removal of the endothelium almost doubled LDCL, indicating a significant attenuation of the baseline LDCL signal by endothelium-derived NO. Likewise, blockade of endothelial NO synthase by L-NMMA (100 mM) significantly increased LDCL generated by 5 mM lucigenin (Fig. 4). DISCUSSION In the present study we demonstrate that lucigenin in low concentrations can be used to quantify reliably O 2•2 production in vascular tissue without appreciably increasing superoxide production as assessed by the spin-trap CPH (Fig. 3B). In addition, the increase in LDCL following inhibition of NO production by the NO synthase inhibitor L-NMMA indicates the usefulness of this technique to estimate indirectly basal NO production. The widespread use of lucigenin as an estimate of vascular O 2•2 production was based upon several experimental advantages. These included a stable background chemiluminescence signal, ease of use, and exquisite sensitivity. Additionally, lucigenin does not generate signals in the presence of other reactive oxygen species like peroxynitrite, the hydroxyl radical,
FIG. 1. Effects of low and high concentrations of lucigenin on endothelium-dependent relaxations to ACh in the rabbit aorta. Isolated aortic rings were preconstricted by PE (0.1 mM) and relaxant responses to ACh were assessed by cumulative addition of increasing ACh concentrations, in the absence and presence of 5 and 250 mM lucigenin and 200 U/ml superoxide dismutase (SOD). Data are presented as mean 6 SEM from 4-7 separate experiments. * indicates a significant difference vs control (p , 0.05). † indicates a significant difference vs 250 mM lucigenin.
321
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Assessment of superoxide production in aortic rings from control and hyperlipidemic Watanabe rabbits using either low (5 mM) or high (250 mM) concentrations of lucigenin, the cypridina luciferin analogue (CLA), or the spin trap CPH. Data are presented as mean 6 SEM from 5 to 7 experiments. *p , 0,05 vs control. CL-signals are expressed as chemiluminescence counts/mg dry tissue/min; ESR signal intensities are given in arbitrary units.
singlet oxygen or hydrogen peroxide, unless these are present in extremely high concentrations. Unlike cytochrome c, lucigenin is not affected by peroxynitrite generated in vascular cells (20). Also, unlike chemiluminescence reagents sensitive to O 2•2-dependent oxidation (imidazolo-pyrazines, coelenterazines or cypridina luciferin analogues) the chemiluminescence of lucigenin requires as a primary step the reduction of cationic lucigenin 1° to the unstable lucigenin° radical (14) that emits light. Despite these advantages, the usefulness of LDCL to monitor vascular O 2•2 production has recently been questioned by the observation that lucigenin itself may generate O 2•2 via autooxidation of the lucigenin cation radical (15, 21).Using high concentrations of purified endothelial NOS (0.1 mM) and ESR-spectroscopy Vasquez-Vivar et al. demonstrated a high rate of lucigenin-induced generation of O 2•2 (4 3 10 29 mol/ min), as estimated from the rate of NADPH consumption by NO synthase (15). Interestingly, these data were obtained in the absence of L-arginine, the substrate of NO synthase. Under this condition NO synthase switches from monooxygenase to oxidase activity which generates O 2•2. Therefore, the study of VasquezVivar did not address the rate of O 2•2 formation by NO synthase in the presence of lucigenin, but rather the capacity of lucigenin to accelerate formation of O 2•2 by the NADPH-oxidase activity of NO synthase in its uncoupled state. More recently, Li et al. (16) have found that the likelihood that lucigenin will participate in redox cycling phenomenon such as those described by Vasquez-
Vivar et al. (15) is critically dependent on the concentration of lucigenin used. In this study, it was shown that concentrations of lucigenin less than 10 –20 mM, did not generate radicals in the presence of several flavin containing enzyme systems. A critical question to be addressed is whether O 2•2 measurements in previous studies have been falsified by high concentrations of lucigenin (2, 11–13, 22–24). To clarify this issue we quantified O 2•2 production in vascular tissue in aortic rings from control animals and hyperlipidemic Watanabe rabbits using high and low concentrations of lucigenin and compared these results to those obtained using the chemiluminescent CLA and the O 2•2-sensitive spin trap CPH. Using 5 mM lucigenin, the chemiluminescence signal in control vessels was 2 to 3-fold less than the signal obtained with 250 mM lucigenin. Likewise, the LDCL signal of aortic rings from hyperlipidemic animals was 2-fold higher using 250 mM lucigenin compared to 5 mM lucigenin. Therefore, we can confirm our previous observation that 250 mM lucigenin elicits additional O 2•2 production in vascular tissue (17). However, with all the techniques applied (5 and 250 mM lucigenin, CLA, spin trapping) we consistently observed a 2- to 3-fold lower O 2•2 formation in vessels from control rabbits compared to vessels from hyperlipidemic animals. These data indicate that vascular O 2•2 is increased to the same degree in vessels from control and hyperlipidemic rabbits and that the ratio of vascular O 2•2 production between these two animal groups does not depend on the lucigenin concentration used.
322
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. (A) ESR spectra of a pair of probes (A and B) of vessels placed in a dual probe ESR cavity. (A) Vessels were treated with 5 mM (upper panel) or 250 mM (lower panel) lucigenin, respectively. (B) Vessels were not exposed to lucigenin and served as controls. Both probes (A and B) were subjected simultaneously to magnetic field modulation. The difference in the magnetic field affecting the probes A and B during magnetic field scanning was kept constant at 7.5 Gauss. The left hand side shows spectra obtained at the beginning of the experiment and spectra at the right hand side were obtained 30 minutes later. (B) Summarized results. Bars show the percentage increase in ESR signal intensities of CP° radicals after subtraction of background signals caused by autooxidation of CPH in aortic tissue. In the presence of 250 mM lucigenin the signal increase was considerably higher than in controls or with 5 mM lucigenin. Data are presented as means 6 SEM from five experiments each. Please note the difference in the intensity scale of the upper and lower ESR spectra.
In agreement with previous observations (17) we found that 250 mM lucigenin caused a marked increase in the oxidation rate of the spin trap CPH, e.g., formation of the nitroxyl radical CP°. This was confirmed by isometric tension recordings which demonstrated that 250 mM lucigenin attenuates endothelium-dependent relaxations to acetylcholine. Importantly, this lucigenin-induced endothelial dysfunction was markedly improved by addition of SOD. Due to its negative charge Cu/ZnSOD does not permeate endothelial cell membranes, which are also negatively charged under physiological conditions (11). It is therefore likely that high concentrations of lucigenin interfere with endothelium-dependent relaxations mainly by stimulating the release of O 2•2 into the extracellular space. In contrast to 250 mM lucigenin, 5 mM lucigenin did not affect endothelium-dependent relaxations to ACh. Moreover, by ESR-spectroscopy we could demonstrate that 5 mM lucigenin did not induce formation of the spin trap CP°. Therefore, these results are in line with previous ESRspectroscopic observations that low concentrations of
lucigenin (5 to 20 mM) detect O 2•2 from enzymatic and cellular sources without eliciting additional O 2•2 formation (16). In accordance with our previous results obtained with CLA-dependent chemiluminescence (17) we were able to detect a marked influence of endothelial NO synthase activity on the steady state concentration of O 2•2 as estimated using 5 mM lucigenin. The steadystate concentration of O 2•2 depends on the rate of dismutation of O 2•2 and the rate of reactions of O 2•2 with free radical scavengers such as ascorbate, alphatocopherol, spin trap (CPH), the cation radical lucigenin 1, all of which compete for the reaction with NO. Using isolated vessels we can now demonstrate that inhibition of NO synthase by L-NMMA markedly increased LDCL. This finding indicates that the amount of NO formed in vascular endothelium is sufficiently high to compete for the reaction of O 2•2 with lucigenin 1. Likewise we observed that removal of the endothelium markedly increased LDCL, suggesting that in endothelium-denuded vessels O 2•2 is generated by vas-
323
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES
FIG. 4. Changes in the intensity of lucigenin-derived chemiluminescence of isolated rabbit aortic rings detected with 5 mM (left hand side) and 250 mM lucigenin (right hand side) induced by removal of the endothelium (E 2) (n 5 5) or treatment with the NO synthase inhibitor N G-monomethyl-L-arginine (L-NMMA, n 5 4). Data are presented as mean 6 SEM. *p , 0.05 vs endothelium-intact (E 1) rings exposed to the same lucigenin concentration.
cular smooth muscle cells or other structures in the media/adventitia of the vessel wall. Our data therefore contradict a previous report (18) that LDCL is not significantly influenced by endothelium-derived NO in the rabbit aorta. We can now attribute these negative results to the high rate of O 22 radical formation elicited by 250 mM lucigenin. This additional formation of O 2•2 may completely mask the influence of endothelial NO on the steady state concentration of O 2•2. We therefore conclude that lucigenin in a low concentration [5 mM] is a sensitive and reliable sensor for monitoring O 2•2 production in intact vessels. To estimate absolute rates of O 2•2 formation in vascular tissue experiments should be performed in the presence of L-NMMA to prevent scavenging of O 2•2 by NO. The present data also validate previous result obtained with 250 mM lucigenin, that vascular O 2•2 production is increased in the setting of hypercholesterolemia (2, 25). In addition, the effects of L-NMMA on LDCL may also indicate the usefulness of 5 mM lucigenin to indirectly assess basal vascular NO production. ACKNOWLEDGMENTS This work was supported by the Deutsche Forschungsgemeinschaft, DFG Mu 1079 2/1, and by a vascular biology grant from the William Harvey Institute, London.
1. Cohen, R. (1995) Progr. Cardiovasc Dis. XXXVIII, 105–128. 2. Ohara, Y., Peterson, T. E., and Harrison, D. G. (1993) J. Clin. Invest. 91, 2546 –2551. 3. Ting, H. H., Timini, F. K., Haley, E. A., Roddy, M. A., Ganz, P., and Creager, M. A. (1997) Circulation 1997, 2617–2622. 4. Taddei, S., Virdis, A., Ghiadoni, L., Magagna, A., and Salvetti, A. (1998) Circulation 97, 2222–2229. 5. Heitzer, T., Just, H., and Munzel, T. (1996) Circulation 94, 6 –9. 6. Ting, H. H., Timimi, F. K., Boles, K. S., Creager, S. J., Ganz, P., and Creager, M. A. (1996) J. Clin. Invest. 97, 22–28. 7. Minor, R. L. J., Myers, P. R., Guerra, R. J., Bates, J. N., and Harrison, D. G. (1990) J. Clin. Invest. 86, 2109 –2116. 8. Gyllenhammar, H. (1987) J. Immunol. Methods 97, 209 –213. 9. Milanowski, J., and Dutkiewicz, J. (1990) Allerg. Immunopathol. 18, 211–215. 10. Wang, J. F., Komarov, P., Sies, H., and de-Groot, H. (1992) Hepatology 15, 1112–1116. 11. Munzel, T., Sayegh, H., Freeman, B. A., Tarpey, M. M., and Harrison, D. G. (1995) J. Clin. Invest. 95, 187–194. 12. Rajagopalan, S., Kurz, S., Munzel, T., Tarpey, M., Freeman, B. A., Griendling, K. K., and Harrison, D. G. (1996) J. Clin. Invest. 97, 1916 –1923. 13. Munzel, T., Kurz, S., Rajagopalan, S., Thoenes, M., Berrington, W. R., Thompson, J. A., Freeman, B. A., and Harrison, D. G. (1996) J. Clin. Invest. 98, 1465–1470. 14. Faulkner, K., and Fridovich, I. (1993) Free Radical Biol. Med. 15, 447– 451. 15. Vasquez-Vivar, J., Hogg, N., Pritchard, K. A., Jr., Martasek, P., and Kalyanaraman, B. (1997) FEBS Lett. 403, 127–130. 16. Li, Y., Zhu, H., Kuppusamy, P., Roubaud, V., Zweier, J. L., and Trush, M. A. (1998) J. Biol. Chem. 273, 2015–23. ¨ nggard, E., and Mun17. Skatchkov, M. P., Sperling, D., Hink, U., A zel, T. (1998) Biochem. Biophys. Res. Commun. 248, 382–386. 18. Pagano, P. J., Tornheim, K., and Cohen, R. A. (1993) Am. J. Physiol. 265, H707–H712,. 19. Davidson, C. A., Kaminski, P. M., and Wolin, M. S. (1997) Am. J. Physiol. 273, L437–L444. 20. Davidson, C. A., Kaminski, P. M., and Wolin, M. S. (1997) Nitric Oxide 1, 244 –253. 21. Liochev, S. I., and Fridovich, I. (1997) Arch. Biochem. Biophys. 337, 115–120. 22. Inoue, N., Ohara, Y., Fukai, T., Harrison, D. G., and Nishida, K. (1998) Am. J. Med. Sci. 315, 242–247. 23. Ohara, Y., Peterson, T. E., Zheng, B., Kuo, J. F., and Harrison, D. G. (1994) Arteriosclerosis Thrombosis 14, 1007–1013. 24. Ohara, Y., Peterson, T. E., Sayegh, H. S., Subramanian, R. R., Wilcox, J. N., and Harrison, D. G. (1995) Circulation 92, 898 – 903. 25. White, C. R., Darley-Usmar, V., Berrington, W. R., McAdams, M., Gore, J. Z., Thompson, J. A., Parks, D. A., Tarpey, M. M., and Freeman, B. A. (1996) Proc. Natl. Acad. Sci. USA 93, 8745– 8749.
324
Validation of Lucigenin as a Chemiluminescent Probe to Monitor Vascular Superoxide as Well as Basal Vascular Nitric Oxide Production Mikhail P. Skatchkov, Daniel Sperling, Ulrich Hink, Alexander Mu¨lsch,* David G. Harrison,† Irene Sindermann, Thomas Meinertz, and Thomas Mu¨nzel 1 Division of Cardiology, Department of Internal Medicine, University Hospital Eppendorf, Hamburg, Germany; *Johann Wolfgang v. Goethe University, Frankfurt, Germany; and †Emory University, Atlanta, Georgia 30322
Received November 9, 1998
Lucigenin has been widely used as a chemiluminescent substrate to monitor vascular superoxide (O 2•2) formation. The validity of lucigenin for detection of O 2•2 has been questioned because O 2•2 is generated by lucigenin itself. It has been shown that the concentration of lucigenin is a critical parameter affecting the validity of this assay. In the present studies we evaluated a reduced concentration of lucigenin (5 mM) as a tool to quantify O 2•2 production in vascular tissue. Lucigenin-induced effects on endothelial function were assessed by isometric tension recording of isolated aortic rings suspended in organ baths. The effects of lucigenin on O 2•2 production were studied using spin trapping and electron spin resonance spectroscopy. Lucigenin at 250 mM but not at 5 mM caused a significant attenuation of endotheliumdependent relaxations to acetylcholine, which was prevented by pretreatment with superoxide dismutase. Spin-trapping studies revealed that lucigenin at 250 mM increased vascular O 2•2 production several fold while 5 mM lucigenin did not stimulate O 2•2 production. Inhibition of NO synthase by N G-momomethyl-L-arginine as well as the removal of the endothelium almost doubled lucigenin-derived chemiluminescence (LDCL), indicating that basal production of endothelium-derived NO depresses the baseline chemiluminescence signal. Thus, lucigenin at a concentration of 5 mM seems to be a sensitive and valid probe for assessing O 2•2 in vascular tissue. It can also be used as an indirect probe to estimate basal vascular NO release. © 1999 Academic Press
1 To whom correspondence should be addressed at Universita¨tskrankenhaus Eppendorf, Abteilung fu¨r Kardiologie, Martinistr. 52, 20246 Hamburg, Germany. Fax: 49-40-4717-3988. E-mail: muenzel@ uke.uni-hamburg.de.
Endothelial dysfunction secondary to reduced NObioactivity has been postulated to play a pathogenic role in the initiation of vascular disease (1). Several experimental and clinical studies have established endothelial dysfunction in diseases such as hypercholesterolemia (2, 3), arterial hypertension (4), chronic smoking (5) and diabetes mellitus (6). Though the precise mechanism underlying endothelial dysfunction remains to be established, one mechanism likely involves inactivation of NO by oxygen derived free radicals (7). The chemiluminescent compound lucigenin has been widely used to assay formation of superoxide anion (O 2•2) in neutrophils (8), macrophages (9), Kupffer cells (10), and in intact vascular tissue (11–13). Traditionally, it was applied in a rather high concentration (250 mM). Under physiological conditions lucigenin exists mainly as a dication (lucigenin 21) (14). Recently it was found that lucigenin in concentrations exceeding 20 mM induces formation of O 2•2 by accepting electrons from the flavin of NO synthase. The transiently formed cation radical (lucigenin 1°) acts as a redox-cycling compound capable to reduce oxygen to O 2•2 (15). More recently, however, Li et al. convincingly demonstrated that O 2•2 production strikingly depended on the lucigenin concentration (16). At 5 mM, lucigenin elicited marked chemiluminescence signals in O 2•2 generating enzyme systems and in isolated cells without stimulating additional O 2•2 production, as shown by electron spin resonance (ESR) studies. The authors recommended spin-trapping experiments to verify the absence of O 2•2 production by lucigenin. Therefore, the first aim of the present study was to test whether or not lucigenin in a concentration of 5 mM is sufficiently sensitive to detect vascular O 2•2 formation. In addition, we used ESR spectroscopy (17) to determine if this concentration of lucigenin is capable of generating O 2•2 in vascular tissue.
319
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Previous studies using lucigenin in a concentration of 250 mM failed to detect changes in the chemiluminescent signal when vessels were exposed to the NO synthase inhibitor N G-nitro-L-arginine (18). This observation is in contradiction to the fact that a significant part of NO derived from endothelial NO synthase or from exogenous NO donors (nitroglycerin, sodium nitroprusside, etc.) is able to react with O 2•2 to form peroxynitrite (19). Indeed, using the cypridina luciferin analogue CLA we recently demonstrated that incubation of isolated blood vessels with N G-monomethyl-Larginine (L-NMMA) or removal of the endothelium markedly increased CLA-chemiluminescence. This finding suggested that endothelium-derived NO affects the O 2•2 induced chemiluminescence signals in vascular tissue (17). Based on these considerations a further objective of the present study was to test whether or not lucigenin in a low concentration (5 mM) can be used to monitor basal NO production. The increase in lucigenin-derived chemiluminescence (LDCL) in response to removal of the endothelium or inhibition of NOS III could be taken as an indirect measure for basal NO formation. Finally, we tested lucigenin applied in low concentrations (5 mM) as a tool to quantify vascular O 2•2 production in vessels from control and hyperlipidemic rabbits and compared these results with those obtained by a recently characterized chemiluminescent O 2•2 detector, CLA, and by spin trapping and electron spin resonance spectroscopy. METHODS Animals and tissue preparation. Studies were performed with isolated aortas from New Zealand White rabbits and hypercholesterolemic Watanabe rabbits. The animals were killed by an overdose of Na pentobarbital The thoracic aorta was isolated and kept in cold Krebs–Henseleit solution of the following composition (mM): NaCl 118.3; KCl 4.69; CaCl 2 1.87; MgSO 4 1.20; K 2HPO 4 1.03; NaHCO 3 25.0; and glucose 11.1; pH 7.4). The aorta was cleaned of adhering adventitial tissue and cut into ring segments of 5 mm length. Organ chamber experiments. Eight ring segments were suspended in individual organ chambers filled with aerated buffer solution (95% O 2; 5% CO 2; 37°C) and changes in isometric tension were recorded with linear force transducers. All experiments were performed in the presence of 10 mM indomethacin to prevent synthesis of prostaglandins. Before the experiments, the rings were gradually stretched and equilibrated at a resting tension of 5 g (rabbit aorta) and 3 g (rat aorta) for at least 1 h. Subsequently, the ring segments were exposed at least three times to 80 mM KCl until a constant contraction was obtained. The vessels were preconstricted with phenylephrine (PE; 10 27 M) to achieve 40 –50% of maximal KClinduced tone and then the endothelium-dependent vasodilator acetylcholine (ACh) was added in cumulative concentrations (1 nM–3 mM). After 30 min incubation with lucigenin (5 or 250 mM) vasodilator responses to ACh were again assessed. To address a potential role of O 22 in lucigenin-induced endothelial dysfunction the concentration response-relationship to ACh was tested 30 min after addition of superoxide dismutase (SOD, 200 U/ml). Detect ion of O 2•2 formation in vascular tissue by LDCL. Vascular O 2•2 formation was measured by recording the chemiluminescence intensity of tissue probes in presence of 5 and 250 mM lucigenin.
Briefly, 5 mm ring segments were placed in a modified Krebs–Hepes buffer and equilibrated for 30 min at 37°C. Scintillation vials containing 2 ml Krebs–Hepes buffer with lucigenin were placed into a scintillation counter (LS 7001, Beckman Instruments Inc., Fullerton, CA) switched to the out-of-coincidence mode. After dark adaptation, background signals were recorded and a vascular segment was added to the vial. Photon counts were then recorded every minute for 8 minutes, and the background was subtracted. The vessels were then dried for 24 h at 90°C and weighed. The results were expressed as counts per min per mg dry weight as described (13).In separate experiments vascular O 22 was assayed using the cypridina luciferin analogue CLA as recently described (17). The assay buffer contained metal chelators such as diethyl-dithiocarbamate (DETC, 100 mM) and deferoxamine (100 mM) to prevent metal-ion-catalyzed reactions increasing the rate of self-oxidation of CLA. DETC was always added to the buffer solution before deferoxamine. Detection of NO in vascular tissue by LDCL quenching. To address the influence of endothelial-derived NO on the chemiluminescence signal associated with the formation of O 22 in some experiments endothelial NO formation was abolished by mechanical removal of the endothelium, or by incubation of vessels for 30 min at 37°C with N G-monomethyl-L-arginine (L-NMMA, 1 mM), an inhibitor of NO synthase. Spin trapping of O 2•2 radicals and electron spin resonance (ESR) studies. To test whether lucigenin elicits O 2•2 formation in vessels the O 2•2 induced generation of nitroxyl radicals was monitored using the spin-trap 1-hydroxy-3-carboxy-2,2,5,-tetramethyl-pyrrolidine hydrochloride (CPH, 1 mM) in the presence of diethylenetriamineaminepentaacetic acid (DTPA, 0.1 mM) as described recently (17). Aortic rings (4 mg) were treated with L-NMMA (1 mM, 30 min) to prevent formation of NO and subsequent oxidation of CPH by peroxynitrite. The vessels were then incubated in Krebs–Henseleit (KH, pH 7,4) solution containing 1 mM CPH with or without lucigenin (5 or 250 mM). Thereafter the vessels were transferred into quartz capillaries (internal diameter 0.8 mm) filled with KH solution and fixed in a dual probe resonator in the cavity of the ESR spectrometer (ER 4105 DR, Bruker, Germany). ESR spectrometer settings were as described recently (17). The concentration of the stable nitroxyl radical CP° generated from O 2•2 and CPH was measured by graphical evaluation of the intensity of the triplet ESR-spectra, using the low field component. Data are presented as percentage increase in the intensity of the ESR-signal after subtraction of the baseline signal originating from autooxidized CPH. Chemicals. CPH and L-NMMA were from Alexis Biochemicals (San Diego, CA). All other chemicals were from Sigma (Deisenhofen, Germany).
RESULTS Effects of high and low concentrations of lucigenin on endothelium-dependent and -independent vasodilation. In PE-constricted rabbit aortic rings the endothelium dependent vasodilator acetylcholine (ACh) induced a dose-dependent relaxation, maximally by 80 6 2% at the highest concentration applied (Table 1). Incubation with 250 mM lucigenin for 30 min markedly decreased the efficacy and potency of ACh (Fig. 1). The lucigenininduced inhibition of vasodilator responses to ACh was markedly improved by SOD (200 U/ml). In contrast, 5 mM lucigenin did not significantly affect vasodilator responses to ACh (Fig. 1). LDCL of vascular tissue exposed to low and high concentrations of lucigenin. In rabbit aortic segments from control animals the chemiluminescence signal
320
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Effects of High- and Low-Dose Lucigenin on EndotheliumDependent and Endothelium-Independent Vasodilation in Rabbit Aorta Acetylcholine
Control 15 mM lucigenin 1250 mM lucigenin 1250 mM lucigenin 1 SOD
ED 50 (2log M)
Max relaxation (%)
27.17 6 0.08 27.09 6 0.07 27.08 6 0.14 27.23 6 0.05
89 6 2 87 6 1 43 6 4* 84 6 4†
Note. Data are presented as means 6 SEM from 8 –10 experiments. SOD, superoxide dismutase, 200 U/ml. * p , 0.05 vs control, †p , 0.05 vs 250 mM lucigenin.
amounted to 1200 6 34 counts/mg/min using 250 mM lucigenin and 478 6 12 counts/mg/min using 5 mM lucigenin (Fig. 2). We also tested the effects of low and high concentrations of lucigenin on CL in vessels from hyperlipidemic rabbits (WHHL). These animals exhibited a significant degree of endothelial dysfunction as indicated by the attenuated vasodilator response to ACh (EC 50 (2log M): WHHL: 6.98 6 0.12 vs control: 7.45 6 0.12. Maximal relaxation: WHHL: 68 6 4% vs control: 92 6 2%; p , 0.05). In aortas of hyperlipidemic animals the luminescence signal was 2660 6 44 counts/mg/min with 250 mM lucigenin and 972 6 121 with 5 mM lucigenin. Therefore, despite a higher luminescence signal with higher concentrated lucigenin the ratio of the signal intensities of control vs hyperlipidemic animals at both concentrations of lucigenin was practically identical (Fig. 2). Similarly, the CLAelicited chemiluminescence was significantly different between control vessels and vessels from hyperlipidemic Watanabe rabbits (Control: 8700 6 1036 counts/ mg/min, n 5 6, WHHL: 25046 6 5155 counts/mg/min, n 5 5, Fig. 4). In accordance with these findings, ESR studies using the spin trap CPH revealed a significant increase in O 2•2 production in vessels from hypercholesterolemic rabbits compared to control vessels (control: 18.6 6 1.1, n 5 5; WHHL: 91.9 6 7.1 arbitrary units, n 5 5, p , 0.05. Fig. 2). Effects of high and low concentrations of lucigenin on the rate of superoxide production in vascular tissue. Incubation of aortic tissue with the spin trap CPH led to time-dependent increase in concentration of CP° radicals as indicated by the increase of intensity of the characteristic ESR triplet signal (17). The rate of CP° radical formation was not affected by 5 mM lucigenin (Fig. 3). In contrast, lucigenin at 250 mM induced a marked acceleration of CP° radical formation by 3.5fold (Fig. 3).
Detection of basal vascular nitric oxide formation by LDCL. In control rabbit aorta removal of the endothelium slightly, but significantly increased LDCL when lucigenin was used in a concentration of 250 mM. Incubation of aortic tissue with L-NMMA failed to increase LDCL significantly under these conditions (Fig. 4). In contrast, in rabbit aortas incubated with 5 mM lucigenin, removal of the endothelium almost doubled LDCL, indicating a significant attenuation of the baseline LDCL signal by endothelium-derived NO. Likewise, blockade of endothelial NO synthase by L-NMMA (100 mM) significantly increased LDCL generated by 5 mM lucigenin (Fig. 4). DISCUSSION In the present study we demonstrate that lucigenin in low concentrations can be used to quantify reliably O 2•2 production in vascular tissue without appreciably increasing superoxide production as assessed by the spin-trap CPH (Fig. 3B). In addition, the increase in LDCL following inhibition of NO production by the NO synthase inhibitor L-NMMA indicates the usefulness of this technique to estimate indirectly basal NO production. The widespread use of lucigenin as an estimate of vascular O 2•2 production was based upon several experimental advantages. These included a stable background chemiluminescence signal, ease of use, and exquisite sensitivity. Additionally, lucigenin does not generate signals in the presence of other reactive oxygen species like peroxynitrite, the hydroxyl radical,
FIG. 1. Effects of low and high concentrations of lucigenin on endothelium-dependent relaxations to ACh in the rabbit aorta. Isolated aortic rings were preconstricted by PE (0.1 mM) and relaxant responses to ACh were assessed by cumulative addition of increasing ACh concentrations, in the absence and presence of 5 and 250 mM lucigenin and 200 U/ml superoxide dismutase (SOD). Data are presented as mean 6 SEM from 4-7 separate experiments. * indicates a significant difference vs control (p , 0.05). † indicates a significant difference vs 250 mM lucigenin.
321
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Assessment of superoxide production in aortic rings from control and hyperlipidemic Watanabe rabbits using either low (5 mM) or high (250 mM) concentrations of lucigenin, the cypridina luciferin analogue (CLA), or the spin trap CPH. Data are presented as mean 6 SEM from 5 to 7 experiments. *p , 0,05 vs control. CL-signals are expressed as chemiluminescence counts/mg dry tissue/min; ESR signal intensities are given in arbitrary units.
singlet oxygen or hydrogen peroxide, unless these are present in extremely high concentrations. Unlike cytochrome c, lucigenin is not affected by peroxynitrite generated in vascular cells (20). Also, unlike chemiluminescence reagents sensitive to O 2•2-dependent oxidation (imidazolo-pyrazines, coelenterazines or cypridina luciferin analogues) the chemiluminescence of lucigenin requires as a primary step the reduction of cationic lucigenin 1° to the unstable lucigenin° radical (14) that emits light. Despite these advantages, the usefulness of LDCL to monitor vascular O 2•2 production has recently been questioned by the observation that lucigenin itself may generate O 2•2 via autooxidation of the lucigenin cation radical (15, 21).Using high concentrations of purified endothelial NOS (0.1 mM) and ESR-spectroscopy Vasquez-Vivar et al. demonstrated a high rate of lucigenin-induced generation of O 2•2 (4 3 10 29 mol/ min), as estimated from the rate of NADPH consumption by NO synthase (15). Interestingly, these data were obtained in the absence of L-arginine, the substrate of NO synthase. Under this condition NO synthase switches from monooxygenase to oxidase activity which generates O 2•2. Therefore, the study of VasquezVivar did not address the rate of O 2•2 formation by NO synthase in the presence of lucigenin, but rather the capacity of lucigenin to accelerate formation of O 2•2 by the NADPH-oxidase activity of NO synthase in its uncoupled state. More recently, Li et al. (16) have found that the likelihood that lucigenin will participate in redox cycling phenomenon such as those described by Vasquez-
Vivar et al. (15) is critically dependent on the concentration of lucigenin used. In this study, it was shown that concentrations of lucigenin less than 10 –20 mM, did not generate radicals in the presence of several flavin containing enzyme systems. A critical question to be addressed is whether O 2•2 measurements in previous studies have been falsified by high concentrations of lucigenin (2, 11–13, 22–24). To clarify this issue we quantified O 2•2 production in vascular tissue in aortic rings from control animals and hyperlipidemic Watanabe rabbits using high and low concentrations of lucigenin and compared these results to those obtained using the chemiluminescent CLA and the O 2•2-sensitive spin trap CPH. Using 5 mM lucigenin, the chemiluminescence signal in control vessels was 2 to 3-fold less than the signal obtained with 250 mM lucigenin. Likewise, the LDCL signal of aortic rings from hyperlipidemic animals was 2-fold higher using 250 mM lucigenin compared to 5 mM lucigenin. Therefore, we can confirm our previous observation that 250 mM lucigenin elicits additional O 2•2 production in vascular tissue (17). However, with all the techniques applied (5 and 250 mM lucigenin, CLA, spin trapping) we consistently observed a 2- to 3-fold lower O 2•2 formation in vessels from control rabbits compared to vessels from hyperlipidemic animals. These data indicate that vascular O 2•2 is increased to the same degree in vessels from control and hyperlipidemic rabbits and that the ratio of vascular O 2•2 production between these two animal groups does not depend on the lucigenin concentration used.
322
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. (A) ESR spectra of a pair of probes (A and B) of vessels placed in a dual probe ESR cavity. (A) Vessels were treated with 5 mM (upper panel) or 250 mM (lower panel) lucigenin, respectively. (B) Vessels were not exposed to lucigenin and served as controls. Both probes (A and B) were subjected simultaneously to magnetic field modulation. The difference in the magnetic field affecting the probes A and B during magnetic field scanning was kept constant at 7.5 Gauss. The left hand side shows spectra obtained at the beginning of the experiment and spectra at the right hand side were obtained 30 minutes later. (B) Summarized results. Bars show the percentage increase in ESR signal intensities of CP° radicals after subtraction of background signals caused by autooxidation of CPH in aortic tissue. In the presence of 250 mM lucigenin the signal increase was considerably higher than in controls or with 5 mM lucigenin. Data are presented as means 6 SEM from five experiments each. Please note the difference in the intensity scale of the upper and lower ESR spectra.
In agreement with previous observations (17) we found that 250 mM lucigenin caused a marked increase in the oxidation rate of the spin trap CPH, e.g., formation of the nitroxyl radical CP°. This was confirmed by isometric tension recordings which demonstrated that 250 mM lucigenin attenuates endothelium-dependent relaxations to acetylcholine. Importantly, this lucigenin-induced endothelial dysfunction was markedly improved by addition of SOD. Due to its negative charge Cu/ZnSOD does not permeate endothelial cell membranes, which are also negatively charged under physiological conditions (11). It is therefore likely that high concentrations of lucigenin interfere with endothelium-dependent relaxations mainly by stimulating the release of O 2•2 into the extracellular space. In contrast to 250 mM lucigenin, 5 mM lucigenin did not affect endothelium-dependent relaxations to ACh. Moreover, by ESR-spectroscopy we could demonstrate that 5 mM lucigenin did not induce formation of the spin trap CP°. Therefore, these results are in line with previous ESRspectroscopic observations that low concentrations of
lucigenin (5 to 20 mM) detect O 2•2 from enzymatic and cellular sources without eliciting additional O 2•2 formation (16). In accordance with our previous results obtained with CLA-dependent chemiluminescence (17) we were able to detect a marked influence of endothelial NO synthase activity on the steady state concentration of O 2•2 as estimated using 5 mM lucigenin. The steadystate concentration of O 2•2 depends on the rate of dismutation of O 2•2 and the rate of reactions of O 2•2 with free radical scavengers such as ascorbate, alphatocopherol, spin trap (CPH), the cation radical lucigenin 1, all of which compete for the reaction with NO. Using isolated vessels we can now demonstrate that inhibition of NO synthase by L-NMMA markedly increased LDCL. This finding indicates that the amount of NO formed in vascular endothelium is sufficiently high to compete for the reaction of O 2•2 with lucigenin 1. Likewise we observed that removal of the endothelium markedly increased LDCL, suggesting that in endothelium-denuded vessels O 2•2 is generated by vas-
323
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES
FIG. 4. Changes in the intensity of lucigenin-derived chemiluminescence of isolated rabbit aortic rings detected with 5 mM (left hand side) and 250 mM lucigenin (right hand side) induced by removal of the endothelium (E 2) (n 5 5) or treatment with the NO synthase inhibitor N G-monomethyl-L-arginine (L-NMMA, n 5 4). Data are presented as mean 6 SEM. *p , 0.05 vs endothelium-intact (E 1) rings exposed to the same lucigenin concentration.
cular smooth muscle cells or other structures in the media/adventitia of the vessel wall. Our data therefore contradict a previous report (18) that LDCL is not significantly influenced by endothelium-derived NO in the rabbit aorta. We can now attribute these negative results to the high rate of O 22 radical formation elicited by 250 mM lucigenin. This additional formation of O 2•2 may completely mask the influence of endothelial NO on the steady state concentration of O 2•2. We therefore conclude that lucigenin in a low concentration [5 mM] is a sensitive and reliable sensor for monitoring O 2•2 production in intact vessels. To estimate absolute rates of O 2•2 formation in vascular tissue experiments should be performed in the presence of L-NMMA to prevent scavenging of O 2•2 by NO. The present data also validate previous result obtained with 250 mM lucigenin, that vascular O 2•2 production is increased in the setting of hypercholesterolemia (2, 25). In addition, the effects of L-NMMA on LDCL may also indicate the usefulness of 5 mM lucigenin to indirectly assess basal vascular NO production. ACKNOWLEDGMENTS This work was supported by the Deutsche Forschungsgemeinschaft, DFG Mu 1079 2/1, and by a vascular biology grant from the William Harvey Institute, London.
1. Cohen, R. (1995) Progr. Cardiovasc Dis. XXXVIII, 105–128. 2. Ohara, Y., Peterson, T. E., and Harrison, D. G. (1993) J. Clin. Invest. 91, 2546 –2551. 3. Ting, H. H., Timini, F. K., Haley, E. A., Roddy, M. A., Ganz, P., and Creager, M. A. (1997) Circulation 1997, 2617–2622. 4. Taddei, S., Virdis, A., Ghiadoni, L., Magagna, A., and Salvetti, A. (1998) Circulation 97, 2222–2229. 5. Heitzer, T., Just, H., and Munzel, T. (1996) Circulation 94, 6 –9. 6. Ting, H. H., Timimi, F. K., Boles, K. S., Creager, S. J., Ganz, P., and Creager, M. A. (1996) J. Clin. Invest. 97, 22–28. 7. Minor, R. L. J., Myers, P. R., Guerra, R. J., Bates, J. N., and Harrison, D. G. (1990) J. Clin. Invest. 86, 2109 –2116. 8. Gyllenhammar, H. (1987) J. Immunol. Methods 97, 209 –213. 9. Milanowski, J., and Dutkiewicz, J. (1990) Allerg. Immunopathol. 18, 211–215. 10. Wang, J. F., Komarov, P., Sies, H., and de-Groot, H. (1992) Hepatology 15, 1112–1116. 11. Munzel, T., Sayegh, H., Freeman, B. A., Tarpey, M. M., and Harrison, D. G. (1995) J. Clin. Invest. 95, 187–194. 12. Rajagopalan, S., Kurz, S., Munzel, T., Tarpey, M., Freeman, B. A., Griendling, K. K., and Harrison, D. G. (1996) J. Clin. Invest. 97, 1916 –1923. 13. Munzel, T., Kurz, S., Rajagopalan, S., Thoenes, M., Berrington, W. R., Thompson, J. A., Freeman, B. A., and Harrison, D. G. (1996) J. Clin. Invest. 98, 1465–1470. 14. Faulkner, K., and Fridovich, I. (1993) Free Radical Biol. Med. 15, 447– 451. 15. Vasquez-Vivar, J., Hogg, N., Pritchard, K. A., Jr., Martasek, P., and Kalyanaraman, B. (1997) FEBS Lett. 403, 127–130. 16. Li, Y., Zhu, H., Kuppusamy, P., Roubaud, V., Zweier, J. L., and Trush, M. A. (1998) J. Biol. Chem. 273, 2015–23. ¨ nggard, E., and Mun17. Skatchkov, M. P., Sperling, D., Hink, U., A zel, T. (1998) Biochem. Biophys. Res. Commun. 248, 382–386. 18. Pagano, P. J., Tornheim, K., and Cohen, R. A. (1993) Am. J. Physiol. 265, H707–H712,. 19. Davidson, C. A., Kaminski, P. M., and Wolin, M. S. (1997) Am. J. Physiol. 273, L437–L444. 20. Davidson, C. A., Kaminski, P. M., and Wolin, M. S. (1997) Nitric Oxide 1, 244 –253. 21. Liochev, S. I., and Fridovich, I. (1997) Arch. Biochem. Biophys. 337, 115–120. 22. Inoue, N., Ohara, Y., Fukai, T., Harrison, D. G., and Nishida, K. (1998) Am. J. Med. Sci. 315, 242–247. 23. Ohara, Y., Peterson, T. E., Zheng, B., Kuo, J. F., and Harrison, D. G. (1994) Arteriosclerosis Thrombosis 14, 1007–1013. 24. Ohara, Y., Peterson, T. E., Sayegh, H. S., Subramanian, R. R., Wilcox, J. N., and Harrison, D. G. (1995) Circulation 92, 898 – 903. 25. White, C. R., Darley-Usmar, V., Berrington, W. R., McAdams, M., Gore, J. Z., Thompson, J. A., Parks, D. A., Tarpey, M. M., and Freeman, B. A. (1996) Proc. Natl. Acad. Sci. USA 93, 8745– 8749.
324