[36] Nitric oxide regulation of lipid oxidation reactions: Formation and analysis of nitrogen-containing oxidized lipid derivatives

[36]

N O REACTIONS WITH LIPID RADICAL SPECIES

385

[36] N i t r i c O x i d e R e g u l a t i o n o f L i p i d O x i d a t i o n R e a c t i o n s : Formation and Analysis of Nitrogen-Containing Oxidized Lipid Derivatives

By H O M E R O

RUBBO and BRUCE A. FREEMAN

Introduction Nitric oxide is a ubiquitous signal transduction molecule and a potent mediator of tissue injury by virtue of its diverse physical properties, which include a low molecular mass, volatility, lipophilicity, and a free radical nature. Nitric oxide undergoes rapid and almost diffusion-limited reactions with other radical species known to mediate oxidant tissue injury, and thus is expected to modulate superoxide (02"-) and lipid radical-mediated reactions potently. The radical-radical reaction of NO. with 02"- to yield peroxynitrite (ONOO-) 1 is more than three times faster than the enzymatic dismutation of Oz'- catalyzed by superoxide dismutases (SOD, ksoD = 2 × 10 9 M -1 sec-1). Similarly, the reaction of NO. with lipid alkoxyl (LO.) and peroxyl (LOO.) radicals is fast (for LOO., k = 1.3 × 109 M -1 sec-1), 2 leading to the expectation that NO. will inhibit less facile radical-mediated lipid chain propagation reactions. It has been convincingly shown that ONOO- is a potent oxidant capable of oxidizing amino acids, lipids, DNA bases, and other biomolecules, with nitrosylation and nitration reactions an important aspect of the reactivity of N O ' . 3-7 While formation of the more potent oxidant ONOO- represents a major potential pathway of oxidative reactivity during periods of enhanced tissue 02"- and NO. production (i.e., inflammation and reperfusion injury), there are numerous examples of NO.-dependent tissue protection during pathological events associated with oxidant stress, suggesting that the array of reactivities and biological actions of NO. is far from understood. For this reason, the

1 R. E. Huie and S. Padmaja, Free Radical Res. Commun. 18, 195 (1993). 2 S. Padmaja and R. E. Huie, Biochem. Biophys. Res. Commun. 195, 539 (1993). 3 R. Radi, J. S. Beckman, K. Bush, and B. A. Freeman, J. Biol. Chem. 266, 4244 (1991). 4 H. Rubbo, A. Denicola, and R. Radi, Arch. Biochem. Biophys. 308, 96 (1994). 5 R. Radi, J. S. Beckman, K. Bush, and B. A. Freeman, Arch. Biochem. Biophys. 228, 481 (1991). 6 j. S. Beckman, T. W. Beckman, J. Chen, P. A. Marshall, and B. A. Freeman, Proc. Natl. Acad. Sci. U.S.A. 87, 1620 (1990). 7 H. Ischiropoulos, L. Zhu, J. Chen, M. Tsai, J. Martin, C. Smith, and J. S. Beckman, Arch. Biochem. Biophys. 298, 431 (1992).

METHODS 1N ENZYMOLOGY,VOL. 269

Copyright© 1996by AcademicPress,Inc. All rights of reproductionin any formreserved.

386

PATHOLOGICAL AND CLINICAL ASPECTS OF NITRIC OXIDE

[36]

reactions of NO- with oxidizing lipids and enzymatic initiators of lipid oxidation were examined. Lipid reactions of NO. are an important area of focus for multiple reasons. First, this reactive species significantly concentrates in lipophilic cell compartments, having a partition coefficient of 6.5 in n-octanol-water. 8 Second, NO. reacts with LO. and LOO. at near diffusion-limited rates, inferring that both lipid peroxidation processes and reactions of lipophilic antioxidants will be influenced by local NO. concentrations. Finally, the central role that NO. plays in vascular diseases, the quantitatively most significant source of morbidity and mortality in western countries, includes important reactivities of NO. both as a signal transduction mediator and toward other free radical species (O2"-, LO-, and L O O ' ) . 9 Thus the issues and methodologies to be addressed herein are designed (1) to reveal the influences of NO. and reactive species commonly associated with oxidant stress on pure lipid, model membrane, and human low-density lipoprotein preparations and (2) to define mechanisms accounting for the protective effects of NO. sometimes observed in pathological events associated with excess production of reactive oxygen species.

Methods

Generation of Reactive Oxygen Species, O N O 0 and NO. Superoxide is generated by xanthine oxidase (XO) using hypoxanthine or acetaldehyde as a substrate, measured by SOD-inhibitable cytochrome c reduction at 550 nm (e = 21,000 M -1 cm-1). Xanthine oxidase activity is determined at 20° by the rate of uric acid production at 295 nm (e = 11,000 M -1 cm-1), or at 308.5 nm (e = 2750 M -1 cm q) when light scattering from liposome suspensions prevents assay of urate formation at 295 nm. Peroxynitrite is synthesized from sodium nitrite and hydrogen peroxide using a quenched-flow reactor as previously described. 3-6 Peroxynitrite concentration is determined spectrophotometrically at 302 nm (e = 1.67 mM -a cm-1), with residual H 2 0 2 eliminated by elution of ONOO on an MnO2 column. Solutions of NO- are prepared by bubbling NO. gas for 30 min into argon-saturated deionized water. Any NO2" present is eliminated

8 T. Malinski, Z. Taha, S. Grunfeld, S. Patton, M. Kapturczak, and P. Tomboulian, Biochem. Biophys. Res. Commun. 193, 1076 (1993). 9 C. R. White, T. A. Brock, L.-Y. Chang, J. Crapo, P. Briscoe, D. Ku, W. A. Bradley, S. H. Gianturco, J. Gore, B. A. Freeman, and M. M. Tarpey, Proc. Natl. Acad. Sci. U.S.A. 9L 1044 (1994).

[36]

N O REACTIONS WITH LIPID RADICAL SPECIES

387

by first bubbling NO. through 5 M NaOH. Nitric oxide production rates and solution concentrations are measured by electrochemical detection using an NO. sensor (Iso-NO; WPI, Inc., Sarasota, FL). S-Nitrosoglutathione (GSNO) is synthesized at 25° by combining equimolar (200 mM) concentrations of reduced glutathione with sodium nitrite in 0.5 M HCI. 1° Stock solutions of S-nitroso-N-acetylpenicillamine (SNAP) are prepared in 100 mM H2SO4. 2,2'-(Hydroxynitrosohydrazono)bisethanamine (SNONOate) is purchased from Cayman Chemicals (Ann Arbor, MI) and used following dilution in potassium phosphate buffer (0.1 M, pH 8.5). Nitric oxide donors are added to the same buffer used in lipid reaction systems to measure rates of NO. release. For example, addition of 100 txM SNAP to air-equilibrated 100 mM potassium phosphate (pH 7.4, 20 °) under standard laboratory lighting conditions results in NO-production for approximately 30 min, achieving a maximum concentration of 8.3/xM in the absence of 02"-. In the case of 800/xM GSNO or 100/zM S-NONOate, constant rates of NO. production are observed for more than 60 min, both achieving steady state concentrations of - 8 IzM in 50 mM potassium phosphate buffer, pH 7.4, in the absence of other reactive target molecules (i.e., 02"- or oxidizing lipid preparations). Detection of Lipid Peroxidation

Human low-density lipoprotein (LDL) is isolated from plasma by differential centrifugation. 11 Egg phosphatidylcholine and 1-palmitoyl-2-arachidonyl-sn-glycero-3-phosphocholine (PC) liposomes are prepared using 6 ml of a 25.4 mM (20 mg ml -a) lipid stock solution in chloroform. Briefly, solvent is removed in vacuo at 45-55 ° and 6 ml of 10 mM potassium phosphate, pH 7.4, is added. The suspension is placed in a 4° water bath and sonicated three times (30 sec each) under argon at 65 W using a sonifier. Liposomes are stored in the dark under argon and used within 24 hr of preparation. Liposomes, LDL, or linolenic acid, suspended in 10 mM potassium phosphate, pH 7.4, is incubated with either xanthine oxidase (XO, 5-10 mU ml-I), O N O O - (1 mM), or soybean lipoxygenase (SLO, 10 or 100 U ml-l), alone and in the presence of NO- generated by addition of 1-100 /~M SNAP, 0.1-1.0 mM GSNO, 20-200 /xM S-NONOate, or by infusion of anaerobically dissolved NO. gas at different rates using a motor-driven microliter syringe. All incubations are continually stirred at 20 ° and lipid peroxidation is measured by formation of conjugated dienes at 234 nm (e = 27,000 M 1 cm-1). Alternatively, aliquots are removed at indicated 10 j. S. Stamler and J. Loscalzo, AnaL Chem. 64, 779 (1992). 1: R. J. Havel, H. A. Eder, and J. H. Bragdon, J. Clin. Invest. 34, 1345 (1955).

388

PATHOLOGICAL AND CLINICAL ASPECTS OF NITRIC OXIDE

[36]

30

20 t~

+

15

5 0

I

I

I

I

I

I

I

0

0.5

1

1.5

2

2.5

3 'NO (gM.min1)

3

2.5

2

1.5

1

0.5

0 O2" (~M. min a)

Fro. 1. The effect of continuous variation of rates of both 02"- and NO. generation on liposome oxidation. PC liposomes (6.6 mg ml 1) were oxidized for 30 rain in stirred incubations containing 50 mM potassium phosphate (pH 7.4), 100 /zM E D T A - F e 3÷ (@) or 100 /IM diethylenetriaminepentaacetic acid (DTPA) (11), 10 mM acetaldehyde, with XO and GSNO concentrations (up to 5 m U m l -1 XO and 660/zM GSNO) calibrated by separate assay to give the noted rates of 02" and NO. production. (©) Liposome oxidation in the absence of XO and GSNO. [Adapted from Rubbo et al. 13 (J. Biol. Chem. 1994) with permission.]

time points for measurement of 2-thiobarbituric acid-reactive products (TBARS) using eM = 150,000 M-1 cm-1 at 532 nm, calculated from reaction with known amounts of malonaldehyde generated by acid hydrolysis of 1,1,3,3-tetramethoxypropane. To prevent further peroxidation of lipid during assay procedures, 0.025% (w/v) butylated hydroxytoluene (BHT) is added to the thiobarbituric acid reagent. Quantitation of hydroperoxide (LOOH) is also determined by N-benzoylleucomethylene blue oxidation (LMB), 12 using e = 4200 M -1 cm -1 as determined from known concentrations of tert-butyl hydroperoxide. Nitric Oxide as a Lipid Prooxidant and Antioxidant

Addition of NO. donors to liposome suspensions exposed to XO, acetaldehyde, and E D T A - F e 3+ stimulates formation of TBA-reactive products at rates of NO. production that are less than or equal to that of O2"-, while lipid oxidation is inhibited at greater rates of NO. production (Fig. 1). 12B. J. Auerbach, J. S. Kiely, and J. A. Cornicelli, Anal. Biochem. 201, 375 (1992).

136]

N O REACTIONS WITH LIPID RADICAL SPECIES

389

160 140 .,tO o "

120 100 80

6o

40 20 0

Lipo LDL

+ SLO

+ "NO

+ "NO + DTPA

FIG. 2. The influence of nitric oxide on lipoxygenase-dependent liposome and human lowdensity lipoprotein oxidation. PC liposomes (1 mg m1-1) and human LDL (0.2 mg ml < ) were oxidized for 30 min in stirred incubations at 20° in 50 mM potassium phosphate containing 0.3 mM EDTA or 100 IzM DTPA at pH 7.4, in the presence of SLO (100 mU ml <) and the infusion of NO. (1/~M min-1). [Adapted from Rubbo et al. 16 (Arch. Biochem. Biophys. 1995) with permission.]

Similarly, NO.-dependent stimulation of XO-mediated lipid peroxidation, followed by inhibition of peroxidation at greater rates of NO. introduction into reaction systems, is observed by the continuous infusion of NO. gas at different rates (0-3/xM rain-t). Peroxynitrite-dependent lipid peroxidation is also inhibited either by infusion of dissolved NO. gas into reaction systems or by addition of NO- donors, t3 Lipoxygenase-induced lipid oxidation is a key source of plasma lipoprotein oxidation and an underlying pathogenic mechanism of atherogenesis. TM Lipoxygenase-dependent lipid and lipoprotein peroxidation is inhibited by NO. derived from GSNO, S-NONOate, or by infusion of micromolar (or less) amounts of pure NO. per minute (Fig. 2). When hydroperoxide (LOOH) species prepared by SLO oxidation of linolenic acid are incubated with bovine serum albumin (BSA), fluorescent products having excitation and emission spectra similar to oxidized LDL are observed (Fig. 3). From this, it has been proposed that after homolytic cleavage of LOOH to the more reactive LOO., a concerted reaction occurs between LOO. and polypeptide amino groups to yield fluorescent adducts without prior LOOH fragmentation to aldehydes or other, more stable products. Is Addition of GSNO to SLO-oxidized linoleic acid prior to addition to BSA completely 13 H. Rubbo, R. Radi, M. Trujillo, R. Telleri, B. Kalyanaraman, S. Barnes, M. Kirk, and B. A. Freeman, J. Biol. Chem. 2A19,26066 (1994). 14 D. Steinberg, S. Parthasarathy, T. E. Carew, J. C. Khoo, and J. L. Witztum, N. Engl. J. Med. 320, 915 (1989). 15j. Fruebis, S. Parthasarathy, and D. Steinberg, Proc. Natl. Acad. Sci. U.S.A. 89,10588 (1992).

390

PATHOLOGICAL AND CLINICAL ASPECTS OF NITRIC OXIDE

[361

70 •

60

50 40 30

20 Q

i,-

10

o 35o

400

450

500

X (nm) FIG. 3. The effect of S-nitrosoglutathione (GSNO) on the generation of fluorescent products. Fluorescence emission spectra were obtained by coincubation of LOOH generated from linoleic acid (0.1 mg m1-1) plus SLO (100 U ml ~) in the absence (--) and presence ( - - - ) of

0.5 mM GSNO and (...) 1 mM GSNO. Bovine serum albumin (0.1 mg ml i) was added after oxidation reactions. [Adapted from Rubbo et al. 16 (Arch. Biochem. Biophys. 1995) with permission.[

inhibits fluorescent product formation (Fig. 3). This inhibition supports the concept that LOO., rather than aldehydic intermediates, is the principal and proximal species responsible for fluorescent adduct formation. The efficient competition of NO. for LOO. reaction with polypeptide amino groups is also inferred, even when NO. is added to preformed lipid hydroperoxides, underscoring the high rate constant of LOO. termination reactions with NO..

Liquid Chromatography-Mass Spectrometry Analysis of Lipid Oxidation Products Liquid chromatography-mass spectroscopic (LC-MS) analysis of lipid oxidation-NO, reaction systems is performed immediately after addition of methanol to give a final concentration of 40% (v/v). Studies are performed on an API III triple-quadrupole mass spectrometer equipped with two Macintosh Quadra 950 computers for data analysis. Lipid oxidation products are separated by reversed-phase H P L C on a 10 cm × 2.1-mm i.d. Aquapore C8 column at a flow rate of 0.2 ml/min using a linear 50-100% (v/v) methanol gradient in 1% (v/v) aqueous acetic acid. The column eluent is split 1 : 1, with 100/~1 min 1 going to the IonSpray interface. Negative or positive ion mass spectra (for linolenic acid or PC liposomes, respectively) are recorded in this mode, with an orifice potential of - 6 0 or +70 V,

[36]

N O REACTIONS WITH LIPID RADICAL SPECIES

391

respectively. MS-MS experiments are carried out by selecting the parent ion with Q1- Collision of the parent ion in Q2 with a mixture of 10% N2-90% Ar results in fragment ions that are separated in Q3. Linolenic acid oxidation induced by XO, purine, and E D T A - F e 3+ yields two principal oxidation products in addition to the [N-H]- ion ( m / z = 227) of the native fatty acid (Table I). These species have a molecular mass of 310 (9- and/or 16-hydroperoxolinolenate) and 342 (9,16-dihydroperoxolinolenate). Lower quantities of 9- and/or 16-hydroxylinolenate are also formed (molecular mass, 294). MS-MS analysis of the species having a molecular mass of 310 yields a loss of 33 mass units, indicating the presence o f - 0 O H . Lipid oxidation reactions simultaneously exposed to NO. reveal molecular compositional characteristics of novel derivatives produced by NO. reaction with oxidizing linolenic acid that are not detectable in the absence of NO. (Table I). These nitrogen-containing products have been identified as nitritolinolenate (molecular mass 323), nitrosoperoxolinolenate (molecular mass 339), hydroxylnitrosoperoxolinolenate (molecular mass 355), and hydroperoxonitrosoperoxolinolenate (molecular mass 371). It is possible that lipid reaction or decomposition pathways for O N O O - involve production of low concentrations of NO. or nitrogen dioxide (NO2"), because minor amounts of the 323 and 339 molecular mass nitrogen-containing oxidized lipid species are detected in ONOO--oxidized linolenate. Further incubation of NO.-containing lipid oxidation systems for 3 hr leads to the loss of apparently reactive or unstable nitrogen-containing oxidized lipid species and a continued generation of peroxidation products (Table I). When PC liposomes oxidized by SLO exhibit inhibition of lipid peroxidation by NO., mass spectral analysis of oxidation products shows concomitant formation of nitrogen-containing oxidized lipid adducts indicative of NO. reaction with oxidized derivatives of P C . 16 In both this instance and for the NO.-dependent formation of lipid radical termination products reported in Table I, it is likely that these unstable species will undergo rapid secondary reactions or rearrangement. Discussion While the reaction of NO. with 02"- yields the much more potent oxidant ONOO-, NO. can also exert direct or indirect antioxidant actions in biological systems undergoing oxidant stress following reaction with metal centers and organic radical intermediates. Indeed, several studies of cell or metal-induced lipoprotein oxidation as well as cell or tissue damage 16 H. Rubbo, S. Parthasarathy, B. Kalyanaraman, S. Barnes, M. Kirk, and B. A. Freeman, Arch. Biochem. Biophys. 324, 15-25 1995.

392

PATHOLOGICAL

AND

CLINICAL

ASPECTS

OF NITRIC

[36]

OXIDE

~"o

0

~'

t~ 0

0

0

~

0

0

~

0

¢)

0c,,)

0

Z 0

V-, <

©

° .,.9, 0 z

© o

oN

('4

Z © Z

~ <,=~,o ,~

111 < :.-/ m

0

i.) <

0

0

0

0

~

~ ,...., ~ ~

~ ~.

o ,'o '~ ~

o

z © k.) Z

z <0

~

"

..o

~o

0

0

~ + + + + + + +

N

u~

[36]

N O REACTIONS WITH LIPID RADICAL SPECIES

393

02 Fe*2

H202~-

O; , . N O

Y L" " f ~

kO0

LOONO

,o.

,ONO

> LON02

LOOH

Fe*2 FIG. 4. Proposed reaction mechanisms of NO' during lipid oxidation events.

during oxidative stress conditions have indicated that either stimulation of endogenous NO. synthesis by induction of nitric oxide synthase gene expression or exogenous supplementation with NO. donors will inhibit oxidant injury at both molecular and functional l e v e l s . 13,17-21 In contrast to these examples, other studies show that the simultaneous production of NO- and 02"-22 or the direct addition of ONOO-9 oxidizes lipoproteins to a potentially atherogenic form. Because ONOO--dependent reactions occur during both early and chronic stages of atherosclerotic disease, as indicated by accumulation of nitrotyrosine in vessel wall proteins, 23 it is concluded from observations presented herein that relative rates of tissue NO. and 02"- production will strongly influence the proatherogenic versus the documented antiatherogenic effects of NO.. We have observed that diverse sources of NO. (SNAP, GSNO, 17 U. Malo-Ranta, S. Yla-Herttuala, T. Metsa-Ketela, O. Jaakkola, E. Moilanen, P. Vuorinen, and T. Nikkari, FEBS Lett. 337, 179 (1994). 18 D. A. Wink, I. Hanbauer, M. C. Krishna, W. DeGraff, J. Gamson, and J. B. Mitchell, Proc. Natl. Acad. Sci. U.S.A. 90, 9813 (1993). 19j. p. Guo, M. R. Siegfried, and A. M. Lefer, Methods Find. Exp. Clin. Pharmacol. 16, 347 (1994). 20 R. Rossaint, H. Gerlach, and K. J. Falke, Eur. J. Anaesthesiol. 11, 43 (1994). 21 I. Kurose, R. Wolf, M. B. Grisham, and D. N. Granger, Circ. Res. 74, 376 (1994). 22 V. M. Darley-Usmar, N. Hogg, V. J. O'Leary, M. T. Wilson, and S. Moncada, Free Radical Res. Commun. 17, 9 (1992). 23 j. S. Beckman, Y. Z. Ye, P. G. Anderson, J. Chen, M. A. Accavitti, M. M. Tarpey, and C. R. White, Biol. Chem. Hoppe-Seyler 375, 81 (1994).

394

P A T H O L O G I C A L A N D C L I N I C A L ASPECTS OF NITRIC O X I D E

[36]

S-NONO, and infusion of anaerobically dissolved NO- gas) result in the NO--mediated inhibition of lipoxygenase-, 02" -, and ONOO--dependent liposomal, lipoprotein, and fatty acid oxidation, yielding nitrogen-containing products. While NO- exerts apparent antioxidant effects by terminating radical chain propagation reactions of alkoxyl and peroxyl radicals, it is important to note that (1) when the ratio of the relative rates of 02" or NO. production in test systems are less than 1, NO. serves as a prooxidant via formation of the potent oxidant ONOO and (2) the products of NO. termination of lipid radical species are unstable and may mediate a different spectrum of as yet undefined target molecule and pathological reactions. Because the reaction of LOO. with o~-tocopherol, a key tissue mediator of lipid peroxidation chain termination reactions, occurs at rates at least three orders of magnitude less than the reaction of LOO- with NO., it is concluded that NO- could act more readily than or in concert with a-tocopherol, lycopene, retinyl derivatives, and/3-carotene as a defense against the formation and reactions of oxidized lipid species. In summary, NO. regulates critical lipid membrane and lipoprotein oxidation events, by (1) contributing to the formation of more potent secondary oxidants from 02"- (i.e., ONOO-), (2) catalyzing the redirection of O£-- and H202-mediated cytotoxic reactions to other oxidative pathways, and (3) termination of lipid radicals to possibly less reactive secondary nitrogen-containing products. Figure 4 shows a proposed mechanism for the reaction of NO. with oxidizing lipids. These novel lipid oxidation adducts are in part organic peroxynitrites and are expected to be produced in vivo under conditions in which diverse inflammatory and pathological processes give rise to similar combinations of reactive species. From the data reported herein, we also conclude that (1) the relative rates of production and steady state concentrations of 02"- and NO., (2) the cellular and anatomical sites of production of 02"- and NO., and (3) the dominant operative mechanisms of oxidant damage in tissues at the time of 02"- and NO. production will profoundly influence expression of the differential oxidant injury-enhancing and protective effects of NO.. Full understanding of the physiological roles of NO., coupled with detailed insight into NO. regulation of oxygen radical-dependent reactions, will yield a more rational basis for the use of NO. donors and inhibitors of NO. synthases for therapeutic purposes.