www.elsevier.com/locate/freeradbiomed

Free Radical Biology & M

Original Contribution

Nitrated lipids decompose to nitric oxide and lipid radicals

and cause vasorelaxation

Emersom S. Limaa, Marcelo G. Boninib, Ohara Augustob, Hermes V. Barbeiroc,

Heraldo P. Souzac, Dulcineia S.P. Abdallaa,*

aClinical and Toxicological Analysis Department, Faculty of Pharmaceutical Sciences, University of Sao Paulo, Sao Paulo, BrazilbDepartment of Biochemistry, Chemistry Institute, University of Sao Paulo, Sao Paulo, Brazil

cEmergency Medicine Department, School of Medicine – University of Sao Paulo, Sao Paulo, Brazil

Received 16 November 2004; revised 3 March 2005; accepted 5 April 2005

Available online 27 April 2005

Abstract

Nitric oxide-derived oxidants such as nitrogen dioxide and peroxynitrite have been receiving increasing attention as mediators of nitric oxide

toxicity. Indeed, nitrated and nitrosated compounds have been detected in biological fluids and tissues of healthy subjects and in higher yields in

patients under inflammatory or infectious conditions as a consequence of nitric oxide overproduction. Among them, nitrated lipids have been

detected in vivo. Here, we confirmed and extended previous studies by demonstrating that nitrolinoleate, chlolesteryl nitrolinoleate, and

nitrohydroxylinoleate induce vasorelaxation in a concentration-dependent manner while releasing nitric oxide that was characterized by

chemiluminescence- and EPR-based methodologies. As we first show here, diffusible nitric oxide production is likely to occur by isomerization

of the nitrated lipids to the corresponding nitrite derivatives that decay through homolysis and/or metal ion/ascorbate-assisted reduction. The

homolytic mechanismwas supported by EPR spin-trapping studies with 3,5-dibromo-4-nitrosobenzenesulfonic acid that trapped a lipid-derived

radical during nitrolinoleate decomposition. In addition to provide a mechanism to explain nitric oxide production from nitrated lipids, the

results support their role as endogenous sources of nitric oxide that may play a role in endothelium-independent vasorelaxation.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Lipid peroxidation; Nitrogen dioxide; EPR; Nitrolinoleate; Cholesteryl nitrolinoleate; Vasodilation

Introduction

Nitrated and nitrosated compounds have been shown to

be produced when proteins, thiols, and lipids are exposed

0891-5849/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.freeradbiomed.2005.04.005

Abbreviations: carboxi-PTIO-2-(4-carboxyphenyl)-4,4,5,5-tetramethyli-

midazoline-1-oxyl 3 oxide; ChLNO2, cholesteryl nitrolinoleate; DBNBS,

3,5-dibromo-4-nitrosobenzenesulfonic acid; EPR, electronic paramagnetic

resonance; Hb, rabbit hemoglobin; MGD, N-methylglucamine dithiocarba-

mate; SDS, sodium dodecyl sulfate; HPLC, high-pressure liquid chroma-

tography; LA, linoleate; LC-ESI/MS/MS, liquid chromatography/

electrospray ionization tandem mass spectrometry; LDL, low-density

lipoprotein; LNO2, nitrolinoleate; LONO, nitritelinoleate; LNO2OH, nitro-

hydroxylinoleate; NMR, nuclear magnetic resonance spectroscopy.

* Corresponding author. Av. Prof. Lineu Prestes 580, Butanta, 05508-900,

Sao Paulo, SP, Brasil. Fax: +55 11 3813 2197.

E-mail address: dspa@usp.br (D.S.P. Abdalla).

toSNO/

SNO2 and peroxynitrite, both in vitro and in vivo

[1–3]. Such compounds are potentially involved withSNO

toxicity in a variety of pathophysiological conditions [4].

Indeed, nitration of proteins, DNA, and low molecular

weight compounds has been shown to be a consequence of

oxidative damage associated with nitrosoative stress, a

recurrent event associated with inflammatory and infec-

tious conditions [5].

Oxidized and nitrated lipids that are likely to be produced

in vivo have been shown to be generated during LDL

oxidation by transition metal ions, exposure of lipids toSNO2 or peroxynitrite in vitro and in cell cultures [6–10].

Also relevant, the fast diffusion-controlled reaction between

lipid radicals andSNO may ultimately lead to the production

of a variety of nitrated lipid derivatives [11–13]. This

edicine 39 (2005) 532 – 539

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539 533

reaction participates in the lipid peroxidation chain-breaking

mechanism accounting for the suggestedSNO antioxidant

property [14–16].

Actually, several nitrated compounds have been demon-

strated to exert diverse biological activities by mechanisms

that remain debatable. Nitrolinoleate (LNO2) displays cell-

signaling activities that appear to be anti-inflammatory

leading to inhibition of platelet function and neutrophil

superoxide generation through cAMP-dependent mecha-

nisms [17,18]. Indeed, LNO2 [17,18] and nitrohydroxyar-

achidonate [19] have been shown to exhibit vasorelaxatory

effects in vitro, an observation consistent withSNO pro-

duction. Moreover, LNO2, nitrohydroxylinoleate (LONO2)

[20] and cholesteryl nitrolinoleate (ChLNO2) [21] have been

detected in the human blood plasma and lipoproteins of

normolipidemic and hyperlipidemic subjects evidencing their

formation in vivo.

In this context, this study using mass spectrometry and

chemiluminescence- and EPR-based methodologies clearly

shows that LNO2, LNO2OH, and ChLNO2 spontaneously

decay, producingSNO and probably carbon-centered

radicals at room temperature. Here, we demonstrate

diffusibleSNO production from nitrated lipids based on

several lines of evidence. First, it was possible to confirm

that LNO2 induces vasorelaxation. In addition, a soluble

guanilyl cyclase inhibitor, ODQ, efficiently inhibited LNO2-

induced vasorelaxation. Second, time-dependent nitric oxide

release was observed by chemiluminescence measurements

and ascorbate-stimulated chemiluminescence was shown to

correlate with LNO2 consumption followed by mass

spectrometry unambiguously characterizing LNO2 as source

ofSNO. Third, we provided spectroscopic evidence for

SNO release from nitrated lipids by incubations of nitrated

lipids with MGD2Fe(II) and 2-phenyl-4,4,5,5,-tetramethyli-

midazoline-1-oxyl-3-oxide (carboxy-PTIO) provided expec-

ted EPR-active products ofSNO. Also, a hypothesis about

the mechanisms of decomposition andSNO release by

nitrated lipids is discussed.

Experimental procedures

Materials

2-Phenyl-4,4,5,5,-tetramethylimidazoline-1-oxyl-3-oxide

was purchase from Calbiochem (San Diego, CA). Sodium

[15N] nitrite and nitronium tetrafluoroborate were purchased

from Aldrich Chemical Co (Milwaukee, WI). 2-Propanol

and chromatographic grade methanol were obtained from

Merck (Darmstadt, Germany). All other reagents were from

Sigma Chemical Co (St. Louis, MO). DBNBS was

synthesized from 3,5 dibromosulfanilic acid purchased from

Aldrich Chemical Co. (Milwaukee, WI) as previously

described [22]. Sodium N-methyl-d-glucamine dithiocarba-

mate (MGD) and iron (II) sulfate heptahydrate were obtained

from OMRF spin-trap source (Oklahoma City, OK) and were

mixed in a 2:1 ratio to prepare the iron complex just before

each set of experiments.

Synthesis and characterization of nitrated lipid derivatives

LNO2 and ChLNO2 were obtained from their precursors

linoleate and cholesteryl linoleate, respectively, as previ-

ously described [20,21]. Briefly, the lipids were reacted with

NO2BF4 under agitation and reduced oxygen tensions. Then,

the products were purified, by passing the mixture through a

silica cromatographic column. The fractions were analyzed

by mass spectrometry and those presenting the highest

concentrations of purified LNO2 or ChLNO2 were mixed

and evaporated under vacuum before freezing at �80-C.LNO2OH was prepared according to Lima et al. [20].

Nitration of the hydroperoxylinoleate was achieved by

incubating the lipid with acidified nitrite solutions for 15

min under air. The products were purified by liquid

cromatography and characterized by mass spectrometry as

previously described. The fractions containing the highest

yield of purified LNO2OH were mixed and evaporated under

vacuum. The obtained product was kept frozen at –80-C.

Vasorelaxation of rat aortic rings

Thoracic aortas from male Wistar rats were carefully

removed and freed from all periadventitial tissue. Aortic

rings (ca. 5 mm) were mounted in organ chambers in Krebs-

Henseleit solution (in mmol/L: CaCl2 1.6, MgSO4 1.17,

EDTA 0.026, NaCl 130, NaHCO3 14.9, KCl 4.7, KH2PO4

1.18, glucose 11), at 37-C and were connected to force

transducers (Biopac System TSD 105A, USA). The solution

was bubbled with a carbogenic mixture (95% O2/5% CO2).

The resting tension was set at 1.5 g and after a 60-min

equilibration period. Vessels were precontracted with

noradrenaline 1�10�7 M and thereafter exposed to different

amounts of nitrated lipids synthesized. Control experiments

were performed in which ODQ (a guanylate cyclase

inhibitor, at 100 AM concentration) was added after

noradrenaline. For experiments with endothelium-denuded

arteries, preparations were rubbed in their internal surface

with a cotton-wrapped stick. Cumulative concentration-

effect curves with each nitrated lipid were made for intact or

denuded aortic rings. Control experiments were performed

with the nonnitrated lipids at comparable concentrations.

Determination ofSNO release by chemiluminescense

TheSNO release was measured employing a

SNO

chemiluminescence analyzer (model NOA, Sievers Instru-

ments, Boulder, CO). It was studied in the absence and

presence of theSNO-trapping agents carboxy-PTIO and

rabbit hemoglobin, and the presence of the reducing agent

ascorbate. Briefly, LNO2 or LNO2OH was injected into the

chamber of the home-built apparatus which contained a 2-ml

final volume of phosphate buffer (25 mM, pH 7.4, 0.5% of

Fig. 1. Vasorelaxation effects of nitrated lipids in rat thoracic aortic rings.

The synthesized nitrated lipids induced significant relaxation of endothe-

lium intact precontracted with 1 � 10�7 M noradrenaline and thereafter

exposed to increasing amounts of nitrated lipids. Data are expressed in

percentage of relaxation as mean T SD, n = 5.

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539534

SDS). The release of gaseous compounds was monitored

during at least 8 h at intervals of 15 and 30 min, and 1, 2, 4, 6,

and 8 h. To evaluate the effect ofSNO-trapping agents and

ascorbate on the nitrated lipid-induced chemiluminescence,

LNO2 or LNO2OH was injected into a sealed glass tube (5

ml) containing a 2-ml final volume of phosphate buffer (25

mM, pH 7.4, 0.5% of SDS) in the absence and in the

presence of hemoglobin, carboxy-PTIO, or ascorbate, kept at

25-C and under nitrogen. After 15 min incubation, aliquots

(0.5 ml) of accumulated gaseous materials in the head space

were injected into the detector chamber using a Hamilton

Gastight syringe.

EPR experiments

EPR experiments were performed on a Bruker EMX

spectrometer operating at 9.85 GHz and 100 KHz field

modulation at room temperature (23 T 2-C). To avoid

extensiveSNO oxidation, the experiments with MGD2Fe(II)

and carboxy-PTIO were performed injecting LNO2 and

ChLNO2 samples in sealed flasks containing buffers and the

desired spin trap previously equilibrated with argon.

Typically, a 15-min argon flow was passed through the

buffers before the addition of the nitrated lipids. Stock iron

(II) solutions were prepared by dissolving Fe(II) sulfate in

acidified water kept under argon. Spin-trapping experiments

with DBNBS were performed by incubating the spin

trapping with LNO2 suspensions in phosphate buffer upon

stirring for the indicated times. Aliquots were transferred to

quartz flat cells and the spectra recorded typically 1 min later.

Results

Vasorelaxation of rat aortic rings

Isometric tension analysis revealed that the tested

nitrated lipids (LNO2, LNO2OH, and ChLNO2) were able

to induce endothelium-independent vasorelaxation in a

concentration-dependent manner (Fig. 1), while the vehicle

(ethanol) or the parent lipids (linoleic acid, linoleic acid

hydroperoxide, or cholesteryl linoleate) had no effect (see,

for instance, Fig. 2). Vascular relaxation induced by LNO2

was considerably inhibited by the guanylate cyclase (GC)

inhibitor ODQ (Fig. 2), suggesting the involvement ofSNO

itself or a derivative in the vasomotor actions of nitrated

lipids.

SNO release studies

LNO2, LNO2OH, or LA, as control (ca. 100 nmol), was

injected into a sealed chamber and the kinetics ofSNO

production was monitored as described under Experimental

procedures. From the results shown in Fig. 3 a markedSNO

release could be observed from LNO2 incubated in

phosphate buffer containing 0.5% SDS at early incubation

times. Although somewhat variable, theSNO release

increased over time and seemed to follow a first-order

dependence on LNO2 during the first 60 min, achieving a

plateau thereafter. Differently from LNO2, considerableSNO

release from LNO2OH was observed only at late incubation

times (ca. 2 h). In fact, theSNO production from LNO2OH

seemed to be preceded by an induction phase followed by a

logarithmic phase. In addition to LNO2 and LNO2OH,

ChLNO2 was also shown to releaseSNO (Fig. 4). The

amount ofSNO produced by ChLNO2 (ca. 100 nmol)

incubated in phosphate buffer in the presence of SDS (0.5

%) after 15 min was shown to be similar to that produced by

LNO2 and much higher than that produced by LNO2OH.

MeasuredSNO levels were increased by ascorbate (1 mM)

addition and, as expected, decreased by the addition of nitric

oxide trappers such as carboxy-PTIO (0.5 mM) and rabbit

hemoglobin (30 AM). LA addition to the system did not

result inSNO release (no data shown). Also, as shown in

Fig. 5,SNO release by LNO2 was stimulated by the

reducing agent ascorbate in a concentration-dependent

manner. Furthermore,SNO release (followed by chemilu-

minescence) occurred in parallel with LNO2 disappearance

(followed by mass spectrometry) (Fig. 5).

EPR spin-trapping experiments

To get further proof forSNO release from LNO2 and

ChLNO2, EPR spin-trapping experiments with MGD2Fe(II)

were performed. EPR signal detection was completely

dependent on the presence of the nitrated lipids as shown

by control experiments in which NO2� or decomposed

NO2BF4 (Figs. 6A and B) which did not produce EPR

signals when incubated with MGD2Fe(II). Incubations of

LNO2 (ca. 5 mM) (Figs. 6C and D) or ChLNO2 (no data

shown) in phosphate buffer (0.15 M, pH 7.2) in the

presence of MGD2Fe(II) led to detection of the character-

istic three-line EPR spectrum of the MGD2Fe(II)-nitrosyl

Fig. 2. ODQ effects in the vasorelaxation induced by LNO2. ODQ (100

AM) significantly inhibited vasorelaxation induced by LNO2. ODQ was

added in the organ chambers containing Krebs-Henseleit solution and aortic

rings before LNO2 addition. Linoleate (LA) was used as control. Data are

expressed as mean T SD, n = 3, with * representing P < 0.05 vs. LA or

LNO2 + ODQ addition (t test).

Fig. 4. Chemiluminescence assay of nitric oxide release by nitrated lipids.

Release of superSNO from ca. 100 nmol of LNO2, LNO2OH, and ChLNO2

standards. The effects of ascorbate (1 mM), carboxy-PTIO (0.5 mM), and

hemoglobin (35 AM) in the chemiluminescence produced bySNO release

were also evaluated. Reaction systems and conditions were as described

under Experimental procedures. Points represent mean experimental value T

SD (n = 5).

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539 535

complex (aN = 12.5 G) [23]. Also, when LNO2 was

synthesized with acidic [15N] nitrite the characteristic

doublet of the labeled MGD2Fe nytrosyl EPR spectrum

was detected (Fig. 6, inset). The levels of the MGD2Fe(II)–

nitrosyl complex slightly increased with time (Figs. 6C and

D).These results indicated that the iron (II) present in the

MGD2-containing incubations was accelerating the decom-

position of the nitrated lipids.

In order to follow nitric oxide release by EPR with time,

we tried another spin trap, carboxy-PTIO. This compound, a

relatively stable nitronyl nitroxide that presents a character-

istic five-line EPR spectrum, is reduced by nitric oxide to an

imino nitroxide (carboxy-PTI) (Fig. 7, inset) that displays a

nine-line EPR spectrum [24]. Incubations of carboxy-PTIO

with LNO2 led to a time-dependent decrease of its five-line

EPR signal but the appearance of the carboxy-PTI peaks was

Fig. 3. Determination of direct nitric oxide release by chemiluminescense.

LNO2 or LNO2OH (100 nmol) were injected into the chamber of a home-

built apparatus which contained a 2 ml final volume of phosphate buffer (25

mM, pH 7.4, 0.5% SDS). The release of gaseous compounds was

monitored during at least 8 h, at 15 and 30 min, and 1, 2, 4, 6, and 8 h.

Reaction systems and conditions were as described under Experimental

procedures. Points represent mean value T SD (n = 3).

not observed in parallel (Fig. 7). Low levels of carboxy-PTI

became evident only at incubations longer than 2 h (Fig. 7).

These results suggested that during nitrated lipid decay

radicals other thanSNO, probably linoleate-derived radicals,

were produced. Indeed,SNO itself does not react fast with

most nitroxides nitronyl nitroxides being exceptions [24,25].

Its product nitrogen dioxide reacts fast with most nitroxides

(k ¨ 108 M�1 s�1) but the reversibility of the reaction

usually precludes observation of nitroxide EPR signal decay

in a minutes per hour time scale [25]. In contrast, carbon- and

oxygen-centered radicals react with nitroxides with rate

constants close to the diffusion limit (k ¨ 108 M�1 s�1) andproduce stable EPR-silent products, particularly the first (for

a review, see [26]).

To confirm production of lipid-derived radicals, we

performed experiments with DBNBS that is an efficient

trap of carbon-centered radicals. Incubations of LNO2 in

Fig. 5. Ascorbate-dependent nitric oxide release and LNO2 decomposition

measurements. LNO2 (ca. 100 nmol) was incubated in a 2-ml final volume

of phosphate buffer (25 mM, pH 7.4, 0.5% of SDS) and at increasing

amounts of ascorbate (0, 0.01, 0.1, 1, and 2 AM). LNO2 decomposition andSNO release were measured through LC-ESI/MS/MS and chemilumines-

cence, respectively, as described under Experimental procedures. Points

represent the mean experimental value T SD (n = 3).

Fig. 6. EPR spectra of produced MGD2Fe(II) –nitrosyl complex. Spectra A

and B refer to control experiments in which MGD2Fe (II) was incubated

with 10 mM nitrite (A) or decomposed NO2BF4 at the same concentrations

used in the synthesis of the nitrolipids (B). C and D are spectra of the

MGD2Fe(II) –nitrosyl complex obtained during incubations of MGD2Fe(II)

(0.5 mM) with LNO2 (ca. 5mM) in phosphate buffer (150 mM, pH 7.4)

under argon for 15 (C) or 30 min (D) at room temperature. Inset refers to

incubation of LNO2 synthesized with [15N]nitrite with MGD2Fe(II).

Instrumental conditions: microwave power, 20 mW; time constant, 327.7

ms; scan rate, 0.6 G/s; modulation amplitude, 2.5 G; gain, 2 � 105.

Fig. 7. EPR spectra obtained from incubations containing carboxy-PTIO-EPR s

phosphate buffer (150 mM, pH 7.4) at room temperature for the indicated times. L

conditions: microwave power, 20 mW; time constant, 163.8 ms; scan rate, 0.6 G/s;

between carboxy-PTIO and nitric oxide to produce carboxy-PTI.

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539536

phosphate buffer in the presence of DBNBS (10 mM) led to

the detection of a time-dependent six-line EPR signal which

is consistent with the trapping of a secondary carbon-

centered radical (Fig. 8). The EPR parameters of the

detected spectrum (aN = 14.2 G; aH = 6.9 G) are similar

to those of secondary carbon-centered radical adducts of

DBNBS that contain an adjacent alkyl group and adjacent

electron-deficient carbon such as that of the carboxylate

group [23]. Taken together, these EPR studies demonstrate

that nitrated lipids produce nitric oxide in parallel with lipid-

derived radicals.

Discussion

Nitro and nitroso derivatives of thiols, lipids, and

proteins have been detected in health and disease states

and shown to potentially exert diverse biological roles

through mechanisms that remain obscure [1,27]. More

specifically, LNO2 and ChLNO2 have been detected in the

plasma of normolipidemic subjects [20,21]. Also relevant,

elevated levels of LNO2 have been shown to be produced by

hyperlipidemic patients [20]. Recent studies have demon-

strated that LNO2 is a bioactive compound that induces

vasorelaxation and inhibits platelet activation in vitro [18]

Most of the available data can be explained by assumingSNO production from LNO2. Here, we demonstrate the

pectra of carboxy-PTIO (0.5 mM) incubated with LNO2 (ca. 10 mM) in

abeled peaks refer to the components of carboxy-PTI spectra. Instrumenta

modulation amplitude, 1.0 G; gain, 7.96 � 102. The inset shows the reaction

l

Fig. 9. Proposed mechanism for nitrated lipid decomposition to nitric

oxide. The nitrated lipid (A) suffers isomerization to the corresponding

nitrite derivative (B) which can be reduced to (C) or suffers homolysis to

(D) whose resonance structure (E) is consistent with the lipid-derived

radical trapped here.

Fig. 8. EPR spectra of obtained DBNBS radical adducts. Spectra obtained

during incubations of LNO2 (ca. 10 mM) with DBNBS (10 mM) in

phosphate buffer (150 mM, pH 7.4) at room temperature. (A) Control

experiment in which LNO2 was replaced by linoleate (ca. 10 mM), (B) 10-

min incubation, (C) 20 min-incubation. Instrumental conditions: microwave

power, 20 mW; time constant, 163.8 ms; scan rate, 0.6 G/s; modulation

amplitude, 1.0 G; gain 7.96 � 104.

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539 537

production of diffusibleSNO from nitrated lipids based on

several lines of evidence. First, it was possible to confirm

that LNO2 induces vasorelaxation in a concentration-

dependent (Fig. 1) and endothelium-independent manner

(data not shown). In addition, the soluble guanilyl cyclase

inhibitor ODQ efficiently inhibited LNO2-induced vaso-

relaxation, provingSNO involvement. ChLNO2, which has

been detected in the plasma and lipoprotein fractions of

healthy subjects, and LNO2OH were also capable of

inducing an endothelium-independent vasorelaxation. The

latter compounds were less efficient than LNO2 (Fig. 1) on

promoting vasorelaxation that correlates with the lower

amount ofSNO produced by them as observed by

chemiluminescence experiments (Fig. 4). Actually, in the

case of LNO2, the ascorbate-stimulated chemiluminescence

was shown to correlate with LNO2 consumption followed by

mass spectrometry (Fig. 5), unambiguously characterizing

LNO2 as the source ofSNO. Moreover, the known

SNO

trappers, carboxy-PTIO and hemoglobin, inhibited the

chemiluminescence induced by all the tested nitrated lipids

confirming their decomposition toSNO.

Our results also provide spectroscopic evidence forSNO

release from nitrated lipids (Figs. 6 and 7). Incubations of

LNO2 and ChLNO2 with MGD2Fe(II) (Fig. 6) and of LNO2

with carboxy-PTIO (Fig. 7) provided the expected EPR-

active products ofSNO [23–26]. The time-dependent nitric

oxide release observed by chemiluminescence measure-

ments (Fig. 3) was not observed when the production of

the MGD2Fe(II)–nitrosyl complex was monitored by EPR

(Fig. 6). In the latter case, most of nitric oxide was released

immediately probably because of the presence of iron (II). As

is the case of other reducing agents, including ascorbate (Fig.

4), iron (II) should catalyze the decomposition of nitrated

lipids. This fact argues for the importance of careful analysis

of results obtained in the presence of MGD2Fe(II) as pre-

viously emphasized [28,29]. Here, the use of a second spin

trap, carboxy-PTIO, confirmed the time-dependent nitric

oxide production (Fig. 7) (see, also, Results).

Taken together, the results obtained with both MGD2Fe(II)

and carboxy-PTIO provided important clues to the mecha-

nism by which the nitrated lipids can release nitric oxide.

Indeed, consumption of both carboxy-PTIO and its product,

carboxy-PTI, during LNO2 decomposition (Fig. 7) provided

evidence for a parallel production of lipid-derived radicals

(see Results). This was confirmed by spin-trapping experi-

ments with DBNBS (Fig. 8). The characteristics of the EPR

spectrum obtained in the presence of DBNBS indicate

trapping of a secondary carbon-centered radical with an

adjacent electron-deficient carbon (see Results). Thus,

considering the LNO2 structure, the trapped radical is likely

to be the one shown in Fig. 9 as species (E). Taken together,

the results suggest that nitrated lipids decay through isomer-

ization to the corresponding nitrite derivatives whose

homolysis and/or one-electron reduction lead to nitric oxide

production (Fig. 9). In fact, nitrocompound isomerization to

the corresponding nitrite has been described before for

nitromethane [30], allilic compounds [31,32], and nitro-

glutathione [33]. Under circumstances that favor homolysis

of the nitrite derivatives [34], lipid-derived radicals will be

produced in parallel with nitric oxide (Fig. 9). Relevantly, it

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539538

has been recently reported that aqueous environments

accelerate the decomposition of nitrated lipids to nitric oxide

[35]. Under physiological and pathological conditions lipid

nitration by oxidants such as nitrogen dioxide and perox-

ynitrite is likely to occur. Nitrated lipids may directly or

through transnitrosation reactions act as NO-releasing

agents in processes promoted by aqueous media, adventious

metal ions, and/or reducing agents. Further studies about the

biological roles of nitrated lipids are thus warranted.

Acknowledgments

This work was supported by grants from the Fundacao de

Amparo a Pesquisa do Estado de Sao Paulo (FAPESP),

Conselho Nacional de Desenvolvimento Cientıfico e Tecno-

logico (CNPq), and Programa de Apoio a Nucleos de

Excelencia (PRONEX). The authors thank Dr. Rodrigo L.

O. R. Cunha for helpful discussion.

References

[1] Eiserich, J. P.; Patel, R. P.; O_donnell, V. B. Pathophysiology of nitricoxide and related species: free radical reactions and modification of

biomolecules. Mol. Aspects Med. 19:221–357; 1998.

[2] O_donnell, V. B.; Freeman, B. A. Interactions between nitric oxide

and lipid oxidation pathways: implications for vascular disease. Circ.

Res. 88:12–21; 2001.

[3] O_donnell, V. B.; Eiserich, J. P.; Bloodsworth, A.; Chumley, P. H.;

Kirk, M.; Barnes, S.; Darley-Usmar, V. M.; Freeman, B. A. Nitration

of unsaturated fatty acids by nitric oxide-derived reactive species.

Methods Enzymol. 301:455–471; 1999.

[4] Ridnour, L. A.; Thomas, D. D.; Mancardi, D.; Espey, M. G.;

Miranda, K. M.; Paolocci, N.; Feelisch, M.; Fukuto, J.; Wink, D. A.

The chemistry of nitrosative stress induced by nitric oxide and

reactive nitrogen oxide species. Putting perspective on stressful

biological situations. Biol. Chem. 385:1–10; 2004.

[5] Wink, D. A.; Mitchell, J. B. Chemical biology of nitric oxide:

insights into regulatory, cytotoxic, and cytoprotective mechanisms of

nitric oxide. Free Radic. Biol. Med. 25:434–456; 1998.

[6] Rubbo, H.; Freeman, B. A. Nitric oxide regulation of lipid oxidation

reactions: formation and analysis of nitrogen-containing oxidized

lipid derivatives. Methods Enzymol. 269:385–394; 1996.

[7] Rubbo, H.; Parthasarathy, S.; Barnes, S.; Kalyanaraman, M. K. B.;

Freeman, B. A. Nitric oxide inhibition of lipoxygenase-dependent

liposome and low-density lipoprotein oxidation: termination of ra-

dical chain propagation reactions and formation of nitrogen-contain-

ing oxidized lipid derivatives. Arch. Biochem. Biophys. 324:15–25;

1995.

[8] Hayashi, K.; Noguchi, N.; Niki, E. Action of nitric oxide as an

antioxidant against oxidation of soybean phosphatidylcholine lip-

osomal membranes. FEBS Lett. 370:37–40; 1995.

[9] Yates, M. T.; Lambert, L. E.; Whitten, J. P.; McDonald, L.; Mano,

M.; Ku, G.; Mao, S. J. T. A protective role for nitric oxide in the

oxidative modification of low density lipoproteins by mouse macro-

phages. FEBS Lett. 309:135–138; 1992.

[10] O_donnell, V. B.; Eiserich, J. P.; Chumley, P. H.; Jablonsky, M. J.;

Kirk, M.; Barnes, S.; Darley-Usmar, V. M.; Freeman, B. A. Nitration

of unsaturated fatty acids by nitric oxide-derived reactive nitrogen

species peroxynitrite, nitrous acid, nitrogen dioxide, and nitronium

ion. Chem. Res. Toxicol. 12:83–92; 1999.

[11] Padmaja, S.; Huie, R. E. The reaction of nitric oxide with organic

peroxyl radicals. Biochem. Biophys. Res. Commun. 195:539–544;

1993.

[12] Napolitano, A.; Camera, E.; Picardo, M.; D_ischia, M. Acid-

promoted reactions of ethyl linoleate with nitrite ions: formation

and structural characterization of isomeric nitroalkene, nitrohydroxy,

and novel 3-nitro-1,5-hexadiene and 1,5-dinitro-1, 3-pentadiene

products. J. Org. Chem. 65:4853–4860; 2000.

[13] Napolitano, A.; Camera, E.; Picardo, M.; D_ischia, M. Reactions of

hydro(pero)xy derivatives of polyunsaturated fatty acids/esters with

nitrite ions under acidic conditions. Unusual nitrosative breakdown

of methyl 13-hydro(pero)xyoctadeca-9,11-dienoate to a novel 4-

nitro-2-oximinoalk-3-enal product. J. Org. Chem. 67:1125–1132;

2002.

[14] Bloodsworth, A.; O_donnell, V. B.; Freeman, B. A. Nitric oxide

regulation of free radical- and enzyme-mediated lipid and lipoprotein

oxidation. Arterioscler. Thromb. Vasc. Biol. 28:1707–1715; 2000.

[15] Gutteridge, J. M.; Halliwell, B. Free radicals and antioxidants in the

year 2000. A historical look to the future. Ann. N.Y. Acad. Sci.

899:136–147; 2000.

[16] Rubbo, H.; Radi, R.; Trujillo, M.; Telleri, R.; Kalyanaraman, B.;

Barnes, S.; Kirk, M.; Freeman, B. A. Nitric oxide regulation of

superoxide and peroxynitrite-dependent lipid peroxidation. Forma-

tion of novel nitrogen-containing oxidized lipid derivatives. J. Biol.

Chem. 269:26066–26075; 1994.

[17] Coles, B.; Bloodsworth, A.; Eiserich, J. P.; Coffey, M. J.;

McLoughlin, R. M.; Giddings, J. C.; Lewis, M. J.; Haslam, R. J.;

Freeman, B. A.; O_Donnell, V. B. Nitrolinoleate inhibits platelet

activation by attenuating calcium mobilization and inducing phos-

phorylation of vasodilator-stimulated phosphoprotein through eleva-

tion of cAMP. J. Biol. Chem. 277:5832–5840; 2002.

[18] Lim, D. G.; Sweeney, S.; Bloodsworth, A.; White, C. R.; Chumley,

P. H.; Krishna, N. R.; Schopfer, F.; O’donnell, V. B.; Eiserich, J. P.;

Freeman, B. A. Nitrolinoleate, a nitric oxide-derived mediator of cell

function: synthesis, characterization, and vasomotor activity. Proc.

Natl. Acad. Sci. USA 99:15941–15947; 2002.

[19] Balazy, M.; Iesaki, T.; Park, J. L.; Jiang, H.; Kaminski, P. M.;

Wolin, M. Vicinal nitrohydroxyeicosatrienoic acids: vasodilator

lipids formed by reaction of nitrogen dioxide with arachidonic acid.

J. Pharm. Exp. Ther. 299:1–9; 2001.

[20] Lima, E. S.; Di Mascio, P.; Rubbo, H.; Abdalla, D. S. P. Character-

ization of linoleic acid nitration in human blood plasma by mass

spectrometry. Biochesmistry 41:10717–10722; 2002.

[21] Lima, E. S.; Di Mascio, P.; Abdalla, D. S. Cholesteryl nitrolinoleate, a

nitrated lipid present in human blood plasma and lipoproteins. J. Lipid

Res. 44:1660–1666; 2003.

[22] Kaur, H. A water soluble C-nitroso-aromatic spin-trap-3,5-dibromo-

4-nitrosobenzenesulphonic acid. ‘‘The Perkins spin-trap.’’ Free Radic

Res. 24:409–420; 1996.

[23] Krishna, C. M.; Kondo, T.; Reisz, P. Sonochemistry of aqueous-

solutions of amino-acids and peptides—A spin trapping study.

Radiat. Phys. Chem. 32:121–128; 1988.

[24] Akaike, T.; Yoshida, M.; Miyamoto, Y.; Sato, K.; Kohno, M.;

Sasamoto, K.; Miyazaki, K.; Ueda, S.; Maeda, H. Antagonistic

action of imidazolineoxyl N-oxides against endothelium-derived

relaxing factor/S

NO through a radical reaction. Biochemistry

32:827–832; 1993.

[25] Goldstein, S.; Samuni, A.; Merenyi, G. Reactions of nitric oxide,

peroxynitrite, and carbonate radicals with nitroxides and their corres-

ponding oxoammonium cations. Chem. Res. Toxicol. 17:250–257;

2004.

[26] Kemp, T. J. Kinetic aspects of spin trapping. Prog. React. Kinet.

Mech. 24:287–358; 1999.

[27] Hogg, N. Biological chemistry and clinical potential of S-nitro-

sothiols. Free Radic. Biol. Med. 28:1478–1486; 2000.

[28] Mikoyan, V. D.; Kubrina, L. N.; Serezhenkov, V. A.; Stukan, R. A.;

Vanin, A. F. Complexes of Fe2+ with diethyldithiocarbamate or N-

E.S. Lima et al. / Free Radical Biology & Medicine 39 (2005) 532–539 539

methyl-D-glucamine dithiocarbamate as traps of nitric oxide in ani-

mal tissues: comparative investigations. Biochim. Biophys. Acta

1336:225–234; 1997.

[29] Vanin, A. F.; Huisman, A.; Stroes, E. S.; deRuijter-Heijstek, F. C.;

Rabelink, T. J.; van Faassen, E. E. Antioxidant capacity of

mononitrosyl-iron-dithiocarbamate complexes: implications for NO

trapping. Free Radic. Biol. Med. 30:813–824; 2001.

[30] Hu, W. F.; He, T. J.; Chen, D. M.; Liu, F. C. Theoretical study of

the CH3NO2 unimolecular decomposition potential energy surface.

J. Phys. Chem. A 106:7294–7303; 2002.

[31] Boivin, J.; Kaim, L. E. K.; Kervagoret, J.; Zard, S. Z. A novel

thermal rearrangement of allylic nitro-derivatives into allylic alco-

hols. J. Chem. Soc. Chem. Commun. 15:1006–1008; 1989.

[32] Gonzalez, A. C.; Larson, C. W.; McMillen, D. F.; Golden, D. M.

Mechanism of decomposition of nitroaromatics. Laser-powered

homogeneous pyrolysis of substituted nitrobenzenes. J. Phys. Chem.

89:4809–4815; 1986.

[33] Balazy, M.; Kaminski, P. M.; Mao, K.; Tan, J.; Wolin, M.

S-Nitroglutathione, a product of the reaction between peroxynitrite

and glutathione that generates nitric oxide. J. Biol. Chem. 273:

32009–32015; 1998.

[34] Nicolescu, A. C.; Reynolds, J. N.; Barclay, L. R. C.; Thatcher, G. R. J.

Organic nitrites and NO: inhibition of lipid peroxidation and radical

reactions. Chem. Res. Toxicol. 17:185–196; 2004.

[35] Giles, G.; Schopfer, F.; Baker, P.; Freeman, B.; Lancaster, J. Jr.

Biophysical characterization of nitric oxide release form nitro-

linoleate: implications for signal transduction. Free Radic. Biol.

Med. 37:S26; 2004.