Toxicology 208 (2005) 207–212

Redox interactions of nitric oxide with dopamineand its derivatives

Fernando Antunesa,∗, Carla Nunesb, Joao Laranjinhab, Enrique Cadenasc

a Instituto de Investiga¸cao Cient´ıfica Bento da Rocha Cabral and Grupo de Bioquimica dos Oxidantes e Antioxidantes,Centro de Quimica e Bioquimica, Universidade de Lisboa, Portugal

b Faculty of Pharmacy and Center for Neuroscience and Cell Biology, University of Coimbra, Portugalc Department of Molecular Pharmacology and Toxicology, University Southern California, Los Angeles, CA, USA

Available online 4 January 2005

Abstract

Nitric oxide (•NO) is a ubiquitous diffusible messenger in the central nervous system.•NO and derived nitrogen speciesmay interact with catecholamines, thus, modifying not only its regulatory actions but also producing oxidants and free radicalsthat are likely to trigger toxic pathways in the nervous system. Oxidative pathways and chain oxidation reactions triggered bycatecholamines may be broken by ascorbate and glutathione, of which there is ample supply in the brain. At the subcellular level,mitochondria and cytosolic dopamine storage vesicles are likely to provide site-specific settings for•NO and catecholaminesinteractions. Thus, a complex picture emerges in which the steady- state levels of the individual reactants, the rate constants ofthe reactions involved, the oxygen tension, and the compartmentalization of reactions determine the biological significance oft • • lm n with then©

K

1

aids

enti-dis-

res ofn-toh at-

por-ivegen-s

0

he redox interactions betweenNO and dopamine metabolism in the brain. The physiological relevance ofNO-driven chemicaodifications of dopamine and its derivatives and the ensuing free radical production are discussed in connectioeurodegeneration inherent in Parkinson’s disease.2004 Elsevier Ireland Ltd. All rights reserved.

eywords:Parkinson’s disease; DOPAC; Mitochondria; Storage vesicles; Oxidative stress; Free radicals

. Introduction

Parkinson’s disease is one of the most prevalentge-associated neurodegenerative disorders character-

zed by the degeneration of neuromelanin-pigmentedopaminergic neurons in thepars compactaof thesub-tantia nigra. Although the mechanisms underlying

∗ Corresponding author.E-mail address:fantunes@fc.ul.pt (F. Antunes).

this disease are unknown, efforts towards the idfication of the fundamental causes of Parkinsons’sease are based on the recognition of unique featudopaminergic cells (Ebadi and Sharma, 2003; Madel et al., 2003): (1) the restriction of cell deaththe dopamine-producing neurons has focused muctention on the dopamine molecule itself as an imtant factor; (2) mitochondrial dysfunction and oxidatstress appear to be two major contributors to deeration of dopaminegic neurons; (3)•NO increase

300-483X/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.tox.2004.11.033

208 F. Antunes et al. / Toxicology 208 (2005) 207–212

during the progress of Parkinson’s diseases as a result ofinflammation-like processes and fosters dopamine de-pletion; the ensuing neurotoxicity is averted by nNOSinhibitors; and (4)•NO participates in the etiologyof the disease as indicated by the protective effectagainst 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine(MPTP)-induced dopaminergic degeneration observedboth in a mutant mice lacking the inducible nitric oxidesynthase (iNOS) (Klivenyi et al., 2000) and in a mutantmice lacking the nNOS gene.

Despite the involvement of•NO in catecholaminemetabolism, the chemical interaction of dopamine andits metabolites with•NO constitutes a source of neuro-toxic molecules, which may contribute in the cellularprocess leading to neurodegeneration. In this work, wereview some recent data on the redox interaction be-tween•NO and dopamine and its derivatives as wellas the implications for the mitochondrial damage andneurodegeneration associated with Parkinsons’s dis-ease caused by these interactions.

1.1. Redox chemistry of NO with dopamine andderivatives

•NO does not react directly with dopamine in anaer-obic conditions (Laranjinha and Cadenas, 2002; Rettoriet al., 2002), but in the presence of O2 dopamine is read-ily oxidized. The species thought responsible for thisreaction is the product of the trimolecular reaction of O2w • • • -t ,w• d,N ton for-m dtf ion(

2

•

(4)

(5)

Concurring with dopamine oxidation, 3,4-dihydroxyphenylacetic acid (DOPAC), a majormetabolite of dopamine resulting from two ox-idative steps catalysed by monoamine oxidaseand aldehyde dehydrogenase in mitochondria, and6-hydroxydopamine, a sympathetic neurotoxin, origi-nated from the oxidation of dopamine, for very lowoxygen tensions undergo one-electron oxidation by•NO forming the semiquinone and the nitroxyl anion(NO−) (reaction(5)).

At the low •NO concentrations found in biologi-cal systems electron abstraction is expected to be themain dopamine oxidation pathway. The semiquinoneformed can undergo either an oxidative or a reductivedecay. Reaction with oxygen yields the quinone andthe superoxide anion (O2•−) (reaction(6)); the lattercan continue the oxidative chain by reacting with•NO,at diffusion-controlled rates (k= 1.9× 1010 M−1 s−1)and resulting in peroxynitrite (ONOO−) formation(reaction(7)). Protonation of ONOO− to peroxyni-trous acid (ONOOH) involves a cage reaction (re-a • •c ineo ac-t oft ac-c te org s,2 nes tion(

•

O

ith NO (reaction(1)), NO2, or at high NO concenrations (in the mM range) N2O3 (dinitrogen trioxide)hich is the product of the reaction between•NO2 with

NO (reaction(2)). Concerning the products forme2O3—being a strong nitrosating agent—leadsitrosation with the ensuing 6-nitrosodopamineation (reaction(3)), which slowly is converte

o 6-nitrodopamine, whereas•NO2 leads to theormation of the semiquinone radical (react4)).

•NO + O2 → 2 •NO2 (1)

NO + •NO2 → N2O3 (2)

(3)

ction (8)) containing both HO and NO2, whichan promote the one-electron oxidation of dopamr any of its derivatives to the semiquinone (re

ion (9)). Alternatively, the one-electron reductionhe semiquinone species to the quinol can beomplished by cell reductants, such as ascorbalutathione (reaction(10)) (Laranjinha and Cadena002). If the local concentration of the semiquinopecies is high, disproportionation can occur (reac11)).

(6)

NO + O2•− → ONOO− (7)

NOO− + H+ → ONOOH ↔ [HO• . . . NO2•]

(8)

F. Antunes et al. / Toxicology 208 (2005) 207–212 209

(9)

(10)

(11)

Besides the formation of dopamine oxidation prod-ucts, which by themselves can be harmful for thecell (see below), the interaction between•NO anddopamine can be an important source of oxidants, suchas O2

•− and ONOO−, or lead to the oxidation of an-tioxidants, such as glutathione or ascorbate. Overall,these can constitute an important source of oxidativestress in the cell.

Taking in account that at the low•NO concentrationfound in vivo the trimolecular reaction of•NO withO2 is slow, the•NO-derived oxidant that is expectedto be more relevant for the oxidation of dopamine invivo is ONOO−. However, for those dopamine-derivedspecies that can be directly oxidized by•NO, this re-action is potentially relevant, for it can evade the con-trol by antioxidant systems, such as superoxide dis-mutase (SOD). In fact, it has been shown that SOD,by removing O2

•−, protects dopamine from oxidationby •NO and inhibits the formation of ONOO− underaerobic conditions (Riobo et al., 2002). Interestingly,Cu,Zn-SOD enhances semiquinone formation in anaer-o tiono nef -d eticc m-psN ofC

1c

en-v thep a-t ase( ne;

(b) the electrophilic character of quinones derived fromdopamine oxidation that can directly damage severalcellular components including the mitochondrion; and(c) dopamine chain oxidation that generates reactiveoxygen species in a self-propagated cycle driven byO2

•−.Concerning the first pathway, the levels of MAO in

dopaminergic neurons are controversial, and probablyonly low amounts of MAO are present in dopaminer-gic cells (Hida et al., 1999). Moreover, independentlyof the levels of MAO in dopaminergic neurons, if thecytosolic dopamine concentration is taken into account(≈1�M) the reaction catalysed by MAO, which hasa KM ≈ 650�M (Hauptmann et al., 1996), will beseverely limited by the low dopamine concentration.It can, thus, be estimated that the production of H2O2by this pathway is not significant when compared withother cellular sources of H2O2, such as mitochondrialrespiration (Antunes et al., 2002).

The potential harmful biological actions of elec-trophilic species, such as quinones, is well documented(Brunmark and Cadenas, 1989), but the mechanisms ofdopamine oxidation leading to electrophilic quinonesas well as their quantitative importance are difficult toaccess. Concerning the initial step of dopamine oxi-dation either O2•− or ONOO− are likely candidates.Species like the hydroxyl radical or•NO2 may not beformed in high enough amounts in order to be quan-titatively important. A criterion to distinguish amongthese species on their relative importance for dopamineo inei ines izedd dentpp idicp ectd ofd tedf the

biosis by a mechanism that may involve oxidaf NO− back to•NO, hence increasing semiquino

ormation (Laranjinha and Cadenas, 2002). Thus, uner such conditions Cu,Zn-SOD imposes a kinontrol of •NO -mediated DOPAC oxidation accolished within a cycle involving the•NO→ NO− tran-ition during DOPAC oxidation (reaction(5)) and theO− → •NO transition coupled to the reductionu2+ in superoxide dismutase.

.2. Biological relevance of NO and dopaminehemical interactions

Dopamine damaging effects to the cell can beisaged by the following general pathways: (a)roduction of H2O2 during the oxidative deamin

ion of dopamine catalysed by monamine oxidMAO) present in the outer mitochondrial membra

xidation is the intracellular localization of dopamn dopaminergic cells. The final steps of dopamynthesis take place in the cytosol; once synthesopamine is rapidly transported via an ATP-depenrocess into storage vesicles (Flatmark, 2000). An im-ortant characteristic of these vesicles is their acH (approximately 5.5) which is thought to protopamine from oxidation, since the anion formopamine is the oxidizable form, while the protona

orm is stabilized to oxidation. Nevertheless, due to

210 F. Antunes et al. / Toxicology 208 (2005) 207–212

much higher concentration of dopamine in the storagevesicles (>1 mM) compared with that in the cytosol(Brautigam et al., 1985), the concentration of the anionform of dopamine in the vesicles is expected to be ap-proximately seventeen-fold higher than in the cytosol,in spite of the low pH in the vesicles (Antunes et al.,2002).

Thus, the oxidation of dopamine should be analyzednot only in the cytosol but as well in the storage vesicles.In the vesicles, the formation of superoxide is expect tobe slow, and its migration from the cytosol to these vesi-cles is not probable to be significant because superox-ide does not cross biomembranes, and so all superoxidethat could cross the membrane would be in the form ofthe hydroperoxyl (Gus’kova et al., 1984), its protonatedform. On the other hand, there is evidence of physi-cal association between storage vesicles and mitochon-dria through tubular channels (Carmichael and Smith,1978), and for the mitochondrial origin of the storagevesicles (Issidorides et al., 1996). So, the possibilityof superoxide diffusing directly from mitochondria,one of the main cellular sites concerning superoxideproduction, to the vesicles should not be discarded.Concerning•NO, it can cross biomembranes freely(Subczynski et al., 1996), implying that the membraneof the vesicles does not provide protection against NOformed extracellularly or intracellularly. Once insidethese vesicles,•NO could react with the low amountsof superoxide and form ONOO−, which could initi-ate the oxidation of dopamine. Subsequently, this pro-c medb ineo cals thep mineo ingp

iss per-o lledb i-ts res-e ox-i ryc 6(a ox-

idation in the cytosol can assume some relevance if thestorage is impaired leading to increased cytosolic levels(Lotharius and Brundin, 2002).

In addition to dopamine oxidation, nitration by•NO-derived nitrogen oxides (most likely N2O3) has beenreported to occur in vitro when•NO gas was bubbledinto an aerobic solution of the catecholamine (Daveu etal., 1997). However, the high•NO concentration (up tomM) required turns questionable its physiological rel-evance even under inflammatory conditions (i.e., fol-lowing expression of inducible iNOS in microglia andastrocytes) in which•NO concentration is supposed toincrease.

Associated with dopamine oxidation there are well-defined cellular effects that eventually could lead todopaminergic cell death. A central theme in this re-gard is mitochondrial damage. Dopamine quinonescan directly damage the mitochondrion by impairingthe respiratory chain among other effects (Bermanand Hastings, 1999; Ben-Shachar et al., 1995). More-over, dopamine can potentiate the•NO-damaging ef-fect on the respiratory chain (Antunes et al., 2002),causing apoptotic death in neurons by an early in-sult to mitochondria (Wallace et al., 1997; Bermanand Hastings, 1999). An impaired mitochondrial res-piration will cause a decrease ATP production, whichis absolutely necessary for the storage of dopamine inthe vesicles. This may cause an increase in the cytoso-lic levels of dopamine, where dopamine oxidation ismore feasible than in the storage vesicles (Lotharius

n-t totake,ech-ycleul-ay

kelyn-

nec-ite-

-trixase

81

ess could proceed in a self-propagated cycle fory reactions(6)–(9). In the storage vesicles, dopamxidation could be expected to be a significant loource of superoxide radicals. Independently, ofrecise mechanism there is now evidence of dopaxidation occurring in the storage vesicles, includrocesses mediated by•NO (Antunes et al., 2002).

Concerning the cytosol, O2•− can be present in thite at high levels, namely through the leakage of suxide from the respiratory chain in a process controy VDAC (Han et al., 2003). Therefore, the direct in

iation of dopamine oxidation by O2•− is a plausiblecenario. This oxidation could be increased in the pnce of•NO via two pathways: (a) increase of super

de production caused by•NO acting on the respiratohain in an antimycin-like effect (Poderoso et al., 199;b) •NO interaction with superoxide forming ONOO−,nd thus, initiating dopamine oxidation. Dopamine

and Brundin, 2002), further increasing the mitochodrial damage. In this context, it is also relevannote that dopamine quinones directly impair the upof dopamine into the storage vesicles (Terland et al.1997). So, independent of the precise molecular manism involved, the basis for a self-propagated cwhich amplifies the initial damage are set, leadingtimately to the death of the dopaminergic cell. It malso be added that DOPAC potentiates•NO-inducedcell death in PC12 cells by mechanisms that are lito involve impairment of mitochondrial respiration (upublished observations). This observation is in contion with the notion that mitochondria provides a sspecific setting for the oxidation of DOPAC by•NO(Laranjinha and Cadenas, 2002): First, DOPAC synthesis is expected to occur in the mitochondrial mavia the activity of unspecific aldehyde dehydrogen(Ambroziak and Pietruszko, 1991; Tank et al., 19);

F. Antunes et al. / Toxicology 208 (2005) 207–212 211

second, the steady-state level of•NO and O2•−, in the

mitochondrial matrix is expected to be high (Boveris etal., 2000), contributing to peroxynitrite formation (re-action(7)). Under these circumstances, the oxidationof DOPAC by ONOO− may be an important source ofDOPAC semiquinone and ensuing redox transitions ofthis species.

Overall, it may be surmised that the redox chemistrydescribed here involving, on the one hand,•NO and de-rived nitrogen species and, on the other hand, dopamineand DOPAC, may be a source of signals that positivelyreinforce cellular pathways leading to cell death associ-ated with Parkinson’s disease. Nevertheless, one shouldalso mention that glutathione and ascorbate, two an-tioxidants abundant in the brain in the mM range, areable to reduceo-semiquinones (Laranjinha and Cade-nas, 2002) to the parent phenolic compounds (reac-tion (10)). Thus, by favouring the reductive decay ofo-semiquinones, GSH and ascorbate prevent the forma-tion of superoxide radical (reaction(6)) and the subse-quent production of toxic species (reactions(7)–(9)).Finally, compartmentalization of reactions, either invesicles or mitochondria, provides site-specific settingsthat facilitate oxidation of catecholamines and ensuingproduction of free radicals. A complex picture emergeswhere the steady-state levels of the individual reactants,the rate constants of the reactions involved, the oxygentension and the compartmentalization of reactions willset the boundaries for dopamine and its metabolites as

• eth,

-nd42m

dro-, di-.

Antunes, F., Han, D., Rettori, D., Cadenas, E., 2002. Mitochondrialdamage by nitric oxide is potentiated by dopamine in PC12 cells.Biochim. Biophys. Acta 1556, 233–238.

Ben-Shachar, D., Zuk, R., Glinka, Y., 1995. Dopamine neurotoxi-city: inhibition of mitochondrial respiration. J. Neurochem. 64,718–723.

Berman, S.B., Hastings, T.G., 1999. Dopamine oxidation alters mi-tochondrial respiration and induces permeability transition inbrain mitochondria: implications for Parkinson’s disease. J. Neu-rochem. 73, 1127–1137.

Boveris, A., Costa, L.E., Poderoso, J.J., Carreras, M.C., Cadenas,E., 2000. Regulation of mitochondrial respiration by oxygen andnitric oxide. Ann. N. Y. Acad. Sci. 899, 121–135.

Brautigam, M., Kittner, B., Herken, H., 1985. Evaluation ofneurotropic drug actions on tyrosine hydroxylase activityand dopamine metabolism in clonal cell lines. Arzneimit-telforschung. 35, 277–284.

Brunmark, A., Cadenas, E., 1989. Redox and addition chemistry ofquinoid compounds and its biological implications. Free Radic.Biol. Med. 7, 435–477.

Carmichael, S.W., Smith, D.J., 1978. Direct connections betweenmitochondria and catecholamine-storage vesicles demonstratedby high voltage electron microscopy in rat adrenal medulla. CellTissue Res. 191, 421–432.

Daveu, C., Servy, C., Dendane, M., Marin, P., Ducrocq, C., 1997.Oxidation and nitration of catecholamines by nitrogen oxidesderived from nitric oxide. Nitric Oxide 1, 234–243.

Ebadi, M., Sharma, S.K., 2003. Peroxynitrite and mitochondrial dys-function in the pathogenesis of Parkinson’s disease. Antioxid.Redox Signal. 5, 319–335.

Flatmark, T., 2000. Catecholamine biosynthesis and physiologicalregulation in neuroendocrine cells. Acta Physiol. Scand. 168,1–17.

Gus’kova, R.A., Ivanov, I.I., Akhobadze, V.V., Rubin, A.R., 1984.Permeability of bilayer lipid membranes for superoxide (O2

•−)

H 003.e su-78,

H . Thes ox-ys.

H y ofmin-

I ctureurons.

K Reif,n-

265–

L itric892–

radicals. Biochim. Biophys. Acta 778, 579–583.an, D., Antunes, F., Canali, R., Rettori, D., Cadenas, E., 2

Voltage-dependent anion channels control the release of thperoxide anion from mitochondria to cytosol. J. Biol. Chem. 25557–5563.

auptmann, N., Grimsby, J., Shih, J.C., Cadenas, E., 1996metabolism of tyramine by monoamine oxidase A/B causeidative damage to mitochondrial DNA. Arch. Biochem. Bioph335, 295–304.

ida, T., Hasegawa, Y., Arai, R., 1999. Histochemical studdopamine-degrading monoamine oxidase activity in dopaergic neurons of rat brain. Brain Res. 842, 491–495.

ssidorides, M.R., Kriho, V., Pappas, G.D., 1996. The fine struof large dense-core organelles in human locus coeruleus neNeurol. Res. 18, 57–63.

livenyi, P., Andreassen, O.A., Ferrante, R.J., Lancelot, E.,D., Beal, M.F., 2000. Inhibition of neuronal nitric oxide sythase protects against MPTP toxicity. Neuroreport. 11, 11268.

aranjinha, J., Cadenas, E., 2002. Oxidation of DOPAC by noxide. Effect of superoxide dismutase. J. Neurochem. 81,900.

potential targets forNO and derived oxidants in thbrain, affecting the physiological chemistry of bo•NO and catecholamines.

Acknowledgements

Supported by grants from Fundac¸ao Ciencia e Tecnologia and FEDER (POCTI/2001/BCI/42365 aPOCTI/2001/BCI/42245) and grant 5R01ES0113from NIH. FA acknowledges financial support froFCT – Portugal (BPD/11487/2002).

References

Ambroziak, W., Pietruszko, R., 1991. Human aldehyde dehygenase. Activity with aldehyde metabolites of monoaminesamines, and polyamines. J. Biol. Chem. 266, 13011–13018

212 F. Antunes et al. / Toxicology 208 (2005) 207–212

Lotharius, J., Brundin, P., 2002. Pathogenesis of Parkinson’s disease:dopamine, vesicles and alpha-synuclein. Nat. Rev. Neurosci. 3,932–942.

Mandel, S., Grunblatt, E., Riederer, P., Gerlach, M., Levites, Y.,Youdim, M.B., 2003. Neuroprotective strategies in Parkinson’sdisease: an update on progress. CNS Drugs 17 (10), 729–762.

Poderoso, J.J., Carreras, M.C., Lisdero, C., Riobo, N., Schopfer, F.,Boveris, A., 1996. Nitric oxide inhibits electron transfer and in-creases superoxide radical production in rat heart mitochondriaand submitochondrial particles. Arch. Biochem. Biophys. 328,85–92.

Rettori, D., Tang, Y., Dias Jr., L.C., Cadenas, E., 2002. Pathways ofdopamine oxidation mediated by nitric oxide. Free Radic. Biol.Med. 33, 685–690.

Riobo, N.A., Schopfer, F.J., Boveris, A.D., Cadenas, E., Poderoso,J.J., 2002. The reaction of nitric oxide with 6-hydroxydopamine:implications for Parkinson’s disease. Free Radic. Biol. Med. 32,115–121.

Subczynski, W.K., Lomnicka, M., Hyde, J.S., 1996. Permeability ofnitric oxide through lipid bilayer membranes. Free Radic. Res.24, 343–349.

Tank, A.W., Weiner, H., Thurman, J.A., 1981. Enzymology and sub-cellular localization of aldehyde oxidation in rat liver. Oxida-tion of 3,4-dihydroxyphenylacetaldehyde derived fromdopamineto 3,4-dihydroxyphenylacetic acid. Biochem. Pharmacol. 30,3265–3275.

Terland, O., Flatmark, T., Tangeras, A., Gronberg, M., 1997.Dopamine oxidation generates an oxidative stress mediated bydopamine semiquinone and unrelated to reactive oxygen species.J. Mol. Cell. Cardiol. 29, 1731–1738.

Wallace, D.C., Lott, M.T., Brown, M.D., 1997. Mitochondrial defectsin neurodegenartive diseases and aging. In: Beal, M.F., Howel,N., Bodis-Wollner, I. (Eds.), Mitochondria and Free Radicalsin Neurodegenerative Diseases. Wiley-Liss, Inc., New York, pp.277–289.