Free Radical Biology & Medicine 42 (2007) 1193–1200www.elsevier.com/locate/freeradbiomed

Original Contribution

Mn porphyrin-based superoxide dismutase (SOD) mimic, MnIIITE-2-PyP5+,targets mouse heart mitochondria

Ivan Spasojević a,⁎, Yumin Chen b, Teresa J. Noel b, Yiqun Yu a, Marsha P. Cole b,1, Lichun Zhang a,Yunfeng Zhao b,2, Daret K. St. Clair b, Ines Batinić-Haberle c,⁎

a Department of Medicine, Duke University Medical Center, Durham, NC 27710, USAb Graduate Center for Toxicology, University of Kentucky, Lexington, KY 40536, USA

c Department of Radiation Oncology, Duke University Medical Center, 231 Nanaline H. Duke, Box 3711, Durham, NC 27710, USA

Received 14 September 2006; revised 8 January 2007; accepted 9 January 2007Available online 13 January 2007

Abstract

The Mn(III) meso-tetrakis(N-ethylpyridinium-2-yl)porphyrin, MnIIITE-2-PyP5+ (AEOL-10113) has proven effective in treating oxidativestress-induced conditions including cancer, radiation damage, diabetes, and central nervous system trauma. The ortho cationic pyridyl nitrogens ofMnTE-2-PyP5+ are essential for its high antioxidant potency. The exceptional ability of MnIIITE-2-PyP5+ to dismute O2·− parallels its ability toreduce ONOO− and CO3

−. Decreasing levels of these species are considered its predominant mode of action, which may also involve redoxregulation of signaling pathways. Recently, Ferrer-Sueta at al. (Free Radic. Biol. Med. 41:503–512; 2006) showed, with submitochondrialparticles, that ≥3 μM MnIIITE-2-PyP5+ was able to protect components of the mitochondrial electron transport chain from peroxynitrite-mediateddamage. Our study complements their data in showing, for the first time that micromolar mitochondrial concentrations of MnIIITE-2-PyP5+ areobtainable in vivo. For this study we have developed a new and sensitive method for MnIIITE-2-PyP5+ determination in tissues. The method isbased on the exchange of porphyrin Mn2+ with Zn2+, followed by the HPLC/fluorescence detection of ZnIITE-2-PyP4+. At 4 and 7 h after a single10 mg/kg intraperitoneal administration of MnIIITE-2-PyP5+, the mice (8 in total) were anesthetized and perfused with saline. Mitochondria werethen isolated by the method of Mela and Seitz (Methods Enzymol. 55:39–46; 1979). We found MnIIITE-2-PyP5+ localized in heart mitochondria to2.95 ng/mg protein. Given the average value of mitochondrial volume of 0.6 μL/mg protein, the calculated MnIIITE-2-PyP5+ concentration is5.1 μM, which is sufficient to protect mitochondria from oxidative damage. This study establishes, for the first time, that MnIIITE-2-PyP5+, ahighly charged metalloporphyrin, is capable of entering mitochondria in vivo at levels sufficient to exert there its antioxidant action; such a resultencourages its development as a prospective therapeutic agent.© 2007 Elsevier Inc. All rights reserved.

Keywords: MnIIITE-2-PyP5+, AEOL-10113; SOD mimic targeting mitochondria; HPLC/fluorescence detection of Mn porphyrin in vivo; Heart mitochondria; Mnporphyrin/Zn porphyrin exchange

Abbreviations: SOD, superoxide dismutase; NHE, normal hydrogen electrode; MnP, any Mn(III/II) porphyrin; para isomer, MnIIITM-4-PyP5+, Mn(III) meso-tetrakis(N-methylpyridinium-4-yl)porphyrin; ortho isomer; MnIIITE-2-PyP5+, Mn(III) meso-tetrakis(N-ethylpyridinium-2-yl)porphyrin (AEOL-10113); MnIIITBAP3−(alsoknown as MnTCPP3−), Mn(III) meso-tetrakis(4-carboxylatophenyl)porphyrin; ZnIITE-2-PyP4+, Zn meso-tetrakis(N-ethylpyridinium-2-yl)porphyrin; ZnIITnBu-2-PyP4+,Zn meso-tetrakis(N-n-butylpyridinium-2-yl)porphyrin; H2Talkyl-2-PyP

4+ meso-tetrakis(N-alkylpyridinium-2-yl)porphyrin, alkyl being methyl (M), ethyl (E), n-butyl(nBu) and n-hexyl (nHex); meso, is indicating substitution in 5,10,15,and 20 positions on porphyrin core; HIF-1, hypoxia inducible factor-1, NF-κB, nuclear factor-κB;AP-1, activator protein-1; BSA, bovine serum albumin; TFA, trifluoroacetic acid; PBS, phosphate-buffered saline.⁎ Corresponding authors. I. Spasojević is to be contacted at fax: +1 684 9094. I. Batinic-Haberle, fax: +1 919 684 8885.E-mail addresses: spaso001@mc.duke.edu (I. Spasojević), ibatinic@duke.edu (I. Batinić-Haberle).

1 Present address: Department of Medicine-Pharmacology, University of Pittsburg, BSTWR E1314 Pittsburgh, PA 15260, USA.2 Present address: Department of Pharmacology, Toxicology & Neuroscience, Louisiana State University Health Science Center, Shreveport, LA 71130, USA.

0891-5849/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.freeradbiomed.2007.01.019

1194 I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

Introduction

Due to the key role of mitochondria in health and diseases,compounds able to enter the mitochondria were actively sought[1–4]. Murphy’s group has been developing mitochondria-targeted ubiquinones [1,2]. These compounds have a lipophilictriphenylphosphonium cation attached to coenzyme Q by theway of an alkyl linker. Kalyanaraman’s group showed thatcarboxyproxyl nitroxide linked to triphenylphosphonium ioncan inhibit peroxide-induced oxidative damage and apoptosis[3]. More recently Asayama et al. [4] prepared a MnIIITM-4-PyP5+ derivatized with mitochondrial targeting peptide.

Since it was designed [5], MnIIITE-2-PyP5+ has beenextensively studied in vitro [5–18], and successfully utilizedin animal models of different diseases [19–29]. Its high efficacyarises from its favorable, biologically compatible metal-centered redox potential (E1/2 = +228 mV vs NHE). Such apotential, which is similar to the potential of superoxidedismutases, allows it to perform both steps of the catalysis ofO2U− dismutation at nearly identical, high rates [8]. This positive

potential is predominantly due to the presence of the positivelycharged ortho pyridyl nitrogens that exert a strong electron-withdrawing effect on the Mn. In addition, the ortho positivecharges (forming a funnel) provide electrostatic facilitation forthe reaction of MnIIITE-2-PyP5+ with negatively chargedreactive species such as O2

U−, ONOO−, and CO3U− [11]. We

have shown that a monocationic parent Mn(III) porphyrin,MnIIIBr8T-2-PyP

+, possessing identical E1/2 asMnIIITE-2-PyP5+,is more than 100-fold less potent a catalyst of O2

U− dis-mutation [11]. The ability of MnIIITE-2-PyP5+ to dismute O2

U−parallels its ability to scavenge ONOO− and CO3

U− [10,15]. Asa consequence of highly positive E1/2 for MnIIIP/MnIIP redoxcouple, MnIIITE-2-PyP5+ is easily reducible by low-molecular-weight cellular reductants such as ascorbic acid, tetrahydro-biopterin, and glutathione [5]. Thus scavenging of reactivespecies by MnIIITE-2-PyP5+ is likely to be coupled to cellularreductants.

Although essential for its action, the multiply cationicnature of MnTE-2-PyP5+ may decrease its distribution withinthe cell and its organelles. Recently, Ferrer-Sueta et al. haveelegantly shown with submitochondrial particles that ≥3 μMMnIIITE-2-PyP5+ is preferable to other Mn(III) porphyrins suchas MnIIITM-4-PyP5+ and MnIIITBAP3−, due to its easyreducibility by Complex I and Complex II of the mitochondrialelectron transport chain, as well as by flavoenzymes such asxanthine oxidase and glucose oxidase [14]. The reducedporphyrin, MnIITE-2-PyP4+, in turn effectively reduces perox-ynitrite, superoxide, and carbonate radical [14]. The catalyticcycle of MnIIITE-2-PyP5+ protects not only Complex II but alsoother components of the mitochondrial electron transport chain[14]. Our study was intended to complement the work of Ferrer-Sueta et al. [14] by showing that mitochondrial micromolarlevels of MnIIITE-2-PyP5+ are attainable in vivo. Such a findingjustifies further optimization of these compounds as potentialtherapeutics.

Thus far no sensitive method for the in vivo detection ofMn porphyrins has been reported. The HPLC/UV/Vis method

used for the in vivo determination of the anionic porphyrin,MnIIITBAP3− (also known MnIIITCPP3−), is not sensitiveenough [30,31]. We have previously developed a HPLC/UV/Vis method for the separation of atropoisomers of cationicporphyrins, H2Talkyl-2-PyP

4+ and its Zn and Mn complexes(alkyl being methyl, ethyl butyl, and hexyl) from aqueoussolutions [32]. Using 1H NMR and X-ray we were able toidentify the individual atropoisomers (αααα, αααβ, ααββ,and αβαβ) separated by HPLC [32]. Yet, UV/Vis detection isnot sensitive enough for the low in vivo levels of porphyrins.For this study we have developed a very sensitive HPLC/fluorescence method which is based on the exchange of theporphyrin Mn2+ site with Zn2+, followed by the fluorescencedetection of ZnIITE-2-PyP4+.

Experimental: general

Porphyrins

MnIIITE-2-PyP5+ (454 nm, log ε = 5.14) [5] and ZnTE-2-PyP5+ (425.5 nm, log ε = 5.46) [33] were prepared as describedpreviously.

ZnIITnBu-2-PyP4+

ZnIITnBu-2-PyP4+ was used as an internal standard and wasprepared as was its ethyl analogue, ZnIITE-2-PyP4+. Elementalanalysis for ZnIITnBu-2-PyPCl4 × 12.5 H2O: Calculated: C,52.65; H, 6.70; N, 8.77; Cl, 11.10. Found: C, 52.65; C; H, 6.07;N, 8.76; Cl, 11.21. The log values of molar absorptivities ofZnIITnBu-2-PyP5+ were 4.46 (263.5 nm), 4.52 (327.0 nm), 5.64(426.0 nm), 4.43 (557.0 nm), and 3.88 (593.5 nm).

Sodium L-ascorbate, mannitol, bovine serum albumin(BSA), sucrose, and ethyleneglycol-bis(-aminoethylether)-N′,N, N′, N′-tetraacetic acid (EGTA), and ZnSO4 × 7 H2O werefrom Sigma and Zn(CH3COO)2 × 2 H2O from J. T. Baker.Acetonitrile and trifluoroacetic acid (TFA) were from FisherScientific and triethylamine from Pierce. Methanol (anhydrous,absolute) was from Mallinckrodt. Glacial acetic acid was fromEM Science. Phosphate-buffered saline (50 mM Na phosphate,0.9% NaCl, pH 7.4) (PBS) was from Gibco. Anti-MnSOD wasfrom Upstate (Lake Placid, NY), anti-Lamin A from Santa Cruz(Santa Cruz, CA), and anti-β-actin from Sigma (St. Louis, MO).

Uv/Vis spectroscopy

Uv/Vis was done on a Shimadzu 2501 PC UV/Visspectrophotometer.

HPLC

Equipment: Waters 2695 HPLC system (pump, autosampler,column oven) and a Waters, Model 2475 fluorescence detectorset on Gain 100, λexc = 425 nm and λabs = 656 nm [34–36].Column: YMC-Pack, ODS-AM, C18 column (3 μm particlesize, 120 A pore size, 150 × 4.6 mm) at 45°C. Elution gradient at1.5 mL/min: 0–15–20 min, 100% A –80% A –100% A,

Fig. 1. The spectral change associated with MnIIIP reduction, followed by Zn2+

for Mn2+ exchange at 6 μM total porphyrin concentration, 8.3 mM ascorbic acid,and 0.16 M Zn acetate in acetonitrile-free HPLC solvent A, pH 6.2, 25°C.

1195I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

followed by 5 min column conditioning. Solvent A: 95%aqueous (deionized water, 20 mM triethylamine, pH 2.7,adjusted with concentrated TFA) and 5% acetonitrile. SolventB: acetonitrile. Autosampler temperature: 4°C. Injectionvolume: 80 μL.

Methods

Transformation of the MnIIITE-2-PyP5+ into its fluorescentform

Mn(III) porphyrins are nonfluorescent [37]. Yet when theMn3+ is replaced by a redox inactive metal, such as zinc, themetalloporphyrin becomes fluorescent [33]. This fact was usedas a basis in the development of a highly sensitive method forthe detection of minute quantities of MnIIITE-2-PyP5+ in themice heart mitochondria homogenates.

Optimization of the Mn2+ to Zn2+ metal exchange

The MnIIITE-2-PyP5+ is a stable complex, and even in98% sulfuric acid no observable loss of Mn occurred within24 h [16]. However, the reduced Mn(II) porphyrin is a labilecomplex, and in the presence of excess zinc a metalexchange occurs, leading to the formation of ZnIITE-2-PyP4+

(Scheme 1). The pH most favorable for this exchange was6.2. Lower pH causes the protonation of ascorbic acid(pKa = 4.04) [38], and thus slows the reduction of Mn3+ toMn2+. It also causes the demetallation of the Zn porphyrin[39]. Higher pH causes the precipitation of Zn hydroxospecies (pKa = 10.6 for the ZnII(H2O)2

2+===>ZnII(H2O)(OH)+ + H+) [40]. Therefore, the fastest exchange and bestchromatography were obtained in acetonitrile-free HPLCsolvent A whose acidity was adjusted with diluted trifluor-oacetic acid to pH 6.2. The exchange was significantly fasterwith Zn acetate than with Zn sulfate. Briefly, the 6 μMMnIIITE-2-PyP5+ was reduced aerobically by 8.3 mMascorbate to MnIITE-2-PyP4+ in acetonitrile-free solvent A,pH 6.2, as followed by the Soret band shift from 454 to438 nm [41]. Subsequently, 0.16 M Zn acetate was addedand the metal exchange was followed on a Shimadzu 2501 PCUV/Vis spectrophotometer as the disappearance of MnIITE-2-PyP4+ Soret band at 438 nm, and appearance of 425.5 nm

Scheme 1. The exchange of porphyrin Mn2+ with Zn2+, under conditions shown belowMnTE-2-PyP5+ levels in vivo.

ZnIITE-2-PyP4+ Soret band [33] (Fig. 1). The exchange wascompleted within either 14 h at 25°C or 1 h at 50°C.

MnIIITE-2-PyP5+ treatment/heart mitochondria isolation

The University of Kentucky Medical Center ResearchAnimal Facility has a continuously accredited program fromAAALAC International. All experiments using animals wereperformed according to the approved protocol for humane careand use of animals. Eleven C57BL/6 mice were used for thestudy. Three animals were injected with saline only. Eight micewere injected intraperitoneally with 10 mg/kg of MnIIITE-2-PyP5+ in saline. This dose was chosen based on a number of invivo experiments where this and similar doses were used[21,22,25,26]. The mice weighed from 20 to 25 g; thus thevolumes of porphyrin solution or saline injected rangedaccordingly from 200 to 250 μL. Two animals were euthanizedwith high-dose pentobarbital at 4 h, and another nine animals at7 h after the injection of either porphyrin or saline. Beforeharvesting tissue, the animals were perfused with saline to avoidartifacts related to the retention of blood in the heart.

Heart mitochondria was isolated as described previously byMela and Seitz [42,43]. Briefly, the hearts were collected, rinsedin ice-cold isolation buffer (0.225 M manitol, 0.075 M sucrose,

, was used as a basis for the HPLC/fluorometric method for the determination of

Fig. 3. The representative HPLC chromatogram of 33.3 nM ZnIITE-2-PyP4+.The 37.5 nM ZnIITnBu-2-PyP4+ was the internal standard. The individualatropoisomers were assigned based on our previous HPLC/1H NMR/X-raystudy [32]. The “α”denotes an N-alkyl above the porphyrin plane, and “β”below the plane [32].

1196 I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

1 mM EGTA, pH 7.4), and cut into small pieces. The hearttissue was washed three times with the isolation buffer toremove any residual blood, and was homogenized at 500 rpmwith a chilled Teflon pestle in a glass cylinder with 10 strokes.The homogenate was centrifuged at 480g at 4°C for 5 min in aSorval SS 34 rotor. The resulting supernatant was filteredthrough a double-layered cheesecloth, and was centrifuged at7700g at 4°C for 10 min. Supernatant was saved to check forleakage from mitochondria using MnSOD, a mitochondrialmatrix enzyme, as an indicator by Western blotting. The pelletwas rinsed with 0.5 mL of the isolation buffer with gentleshaking to remove the “fluffy layer” (damaged mitochondria)on top of the pellet. The wall of the centrifuge tube was cleanedwith cotton swabs to remove lipids. The pellet was washed bygentle resuspension in 3 mL isolation buffer using the smoothsurface of a glass rod, and centrifuged at 7700g at 4°C for10 min. The supernatant was saved to again check for leakagefrom the mitochondria. The washing was repeated once more.The resulting mitochondria were collected for further analysis.The purity of mitochondria was examined using Lamin A (anuclear protein) and β-actin (a cytoskeletal protein) asindicators by Western blotting and is shown in Fig. 2. Theprotein levels were determined by colorimetric assay (Bio-Rad,Richmond, CA).

MnIIITE-2-PyP5+ extraction/analysis

Heart mitochondrial homogenate (100 μL) was transferredinto a 2-mL polypropylene screw-cap vial. Then, 200 μL ofdeionized water and 300 μL of 1% acetic acid in methanol wereadded and mixed by vortexing for 30 s. After incubation for30 min and again vortexing for 30 s, the homogenate wascentrifuged 5 min at 16,000g to separate proteins. The 400 μLof the supernatant was pipetted into a 5-mL polypropylene tube(10 × 50 mm). Solvent was completely removed in a SavantSpeed-Vac evaporator at 40°C within 1 h. The dry residue wasdissolved in a 100 μL of deionized water containing 20 mMtriethylamine, the acidity of which was adjusted with dilutedTFA to pH 6.2, followed by two cycles of 30 s vortexing and

Fig. 2. Integrity and purity of isolated mitochondria. Mitochondria were purifiedby centrifuging heart homogenate on a mannitol-sucrose gradient twice. Westernblot probing for Lamin A (a nuclear protein) and β-actin (a cytoskeletal protein)indicated minimum contamination from nonmitochondrial fractions, andWestern blot probing against MnSOD indicated that there was no leaking ofthe purified mitochondria. Lane 1, supernatant from centrifugation of wholeheart homogenate; lane 2, supernatant from first washing; lane 3, supernatantfrom second washing; lane 4, purified mitochondria.

centrifugation for 2 min at 2000g. The amount of 80 μL of thesupernatant was transferred into a 2-mL polypropylene screw-cap tube, and 30 μL of 1.0 M Zn acetate in water and 10 μL offreshly prepared 0.11 M sodium ascorbate in water were added.The solution was left either overnight at 25°C or for 1 h at 50°C.Then, 20 μL of 2.7 M TFA and 20 μL of 100 nM ZnIITnBu-2-PyP4+ were added, followed by vortexing for 30 s andcentrifugation for 5 min at 16,000g. The supernatant (110 μL)was transferred to an HPLC autosampler polypropylene vialequipped with a silicone/PTFE septum.

All four atropoisomers of ZnIITE-2-PyP5+ were present inthe chromatogram at the abundance ratio reported previously[32] (Fig. 3). ZnIITnBu-2-PyP4+ was used as an internalstandard, and its atropisomers were also present in thechromatogram at the expected abundance ratio (Fig. 3) [32].The assignment of the individual atropoisomers, αααα, αααβ,ααββ, and αβαβ, of both Zn porphyrins was done based onour previous HPLC/1H NMR/X-ray study [32].

Calibration curve

A set of serially diluted standard samples of MnIIITE-2-PyP5+ from 1.25 ng/mL to100 ng/mL either in mitochondrialhomogenate that had 3 mg protein/mL (Fig. 4) or in 3 mgBSA/mL PBS (Fig. 4, inset) was used to construct thecalibration curve. With BSA/PBS the internal standardZnTnBu-2-PyP4+ was added at 375 nM concentration, whilewith mitochondrial homogenate 10-fold less ZnTnBu-2-PyP4+

(37.5 nM) was used, giving rise to a 10-fold difference inresponse. Response was calculated as the ratio betweenstandard peak area and internal standard peak area.

We compared curves obtained when using either αααβisomer (most abundant) peak area or total area under peaks of allfour isomers and obtained identical slopes. However, by usingonly αααβ peak our lowest limit of quantification (LLD) isbetter because quantification of the least abundant αβαβ isomer

Fig. 4. Calibration curves for the determination of MnTE-2-PyP5+ levels in vivo.The curves were prepared by diluting MnTE-2-PyP5+ into either mitochondrialhomogenate or into BSA/PBS (inset) whose protein concentration was 3 mg/mLhomogenate. With BSA/PBS the internal standard was added at 375 nMconcentration, while with mitochondrial homogenate 10-fold less ZnTnBu-2-PyP4+ (37.5 nM) was used. Thus there is a 10-fold difference in otherwiseidentical slopes (0.00186 and 0.0191). BSA/PBS can be therefore used insteadof mitochondrial homogenate, as valid and less costly alternative calibrationmedium for determination of Mn porphyrin levels in vivo.

1197I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

at those concentration values is not possible (LLD would belimited by the area of the least abundant isomer). In ourexperience [32] we do not see changes in atropoisomer abun-dance distribution after administration/extraction of the com-pound to/from animals.

Recovery of MnIIITE-2-PyP5+ (the overall efficacy ofextraction from mitochondria plus Mn to Zn exchange) was98%, and was determined in the following way. A knownamount of MnIIITE-2-PyP5+ was added to mitochondrialhomogenate (3 mg/mL protein), and the procedure of extrac-tion/metal exchange was followed. The response from thisexperiment was compared to the response of the sample where acorresponding amount of ZnIITE-2-PyP4+ (equal to 100% yieldof MnIIITE-2-PyP5+) was diluted into mobile phase.

Results and discussion

Mitochondria are thought to be the major source ofintracellular reactive species under normal and severalpathological conditions. Thus injury to mitochondria con-tributes to a number of human pathologies. Not surprisingly,therapeutics targeting mitochondria have been intensivelysought.

MnIIITE-2-PyP5+ and its mode of action/s

Based on structure–activity relationships between thecatalytic rate constant for O2

U− dismutation and the E1/2 for theMIIIP/MIIP redox couple [5], we have developed several potentcatalytic Mn porphyrin-based antioxidants, MnIIITE-2-PyP5+

being the best characterized. We have shown that MnIIITE-2-PyP5+ is remarkably effective both in vitro [5–18] and in vivo[19–29] as a consequence of its several readily accessible

oxidation states (+2, +3, +4, and +5) [5,15,17]. It catalyticallydismutes superoxide with log kcat of 7.76 [5], whereas thelog kcat for SOD enzyme is between 9.3 and 8.84 [44–49].Due to the ready reducibility of MnIIITE-2-PyP5+, theelimination of superoxide is likely coupled to the reduction ofMnIIITE-2-PyP5+ with cellular reductants [15,17]. Radi’s groupstudied the peroxynitrite-related chemistry of MnIIITE-2-PyP5+

showing that it is among the fastest reductants for peroxynitriteand for carbonate radical [10,14,15]. The elimination ofperoxynitrite is coupled to the reduction of both MnIIITE-2-PyP5+ and O=MnIVTE-2-PyP4+ by cellular reductants [15].Given the rapid reaction between nitric oxide and superoxide(k = 2 × 1010 M−1 s−1) [50], it is highly likely that eliminationof peroxynitrite may be the predominant mode of action ofMnIIITE-2-PyP5+ [14,15,51].

Since the observation that micromolar MnIIITE-2-PyP5+

fully protects SOD-deficient Escherichia coli against aerobicinhibition of growth [5,19], the compound has beensuccessfully used in in vivo studies where oxidative stressis a putative factor [20–28]. The most remarkable data wereobtained on cancer [20,21], radioprotection [25], and diabetes[22,26,52,53]. Recent data suggest that MnIIITE-2-PyP5+

affects signaling pathways through inactivation of transcrip-tion factors such as AP-1 [20], HIF-1 [21], and NF-κB[29,52,53]. Two different mechanisms seem to be involved.The first relates to the ability of MnIIITE-2-PyP5+ to redox-modulate signaling pathways through decreasing levels orreactive oxygen and nitrogen species involved in theactivation of transcription factors [20,21]. The secondmechanism relates to the inhibition of NF-κB, by oxidationof exposed cysteine thiol groups on its p50 subunit upontranslocation into the nucleus [52,53]. Based on the results ofFerrer-Sueta et al. O=MnIVTE-2-PyP4+, formed from theoxidation of MnIIITE-2-PyP5+ by peroxynitrite (or lessefficiently by H2O2), can in turn rapidly oxidize glutathione[15,17] and thus presumably protein cysteines as well. It islikely that both mechanisms may occur in parallel.

Recently, Ferrer-Sueta et al. have shown, with submito-chondrial particles, that ≥3 μM MnIIITE-2-PyP5+ protectsperoxynitrite, superoxide, or carbonate radical-sensitive tar-gets in mitochondria, while being reduced by complexes I andII of mitochondrial electron transport with NADH orsuccinate as electron sources [14]. This study is intended tocomplement their research by showing that despite significanthydrophilicity as a result of its high positive charge, which isessential for its antioxidant potency, MnIIITE-2-PyP5+ isindeed able to get into the mouse heart mitochondria,following a single intraperitoneal administration, and at levelshigh enough to protect mitochondrial respiration fromoxidative damage.

In vivo detection of MnIIITE-2-PyP5+

Despite the efficacy of MnIIITE-2-PyP5+ in a number ofanimal studies, a sensitive method for its in vivo determina-tion has been lacking. Studies, indicating the inhibition oftranscription factors [20,21,29,52], the mimicking of MnSOD

1198 I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

[20], and the oxidation of p50 subunit of NF-κB [52], implythat MnIIITE-2-PyP5+ distributes within the cell interior, andpresumably into organelles as well. For the purpose of thiswork we have developed a new and sensitive method basedon the reduction of MnIIITE-2-PyP5+ by ascorbate, followedby the exchange of Mn2+ with Zn2+, followed by HPLC/fluorescence detection of ZnIITE-2-PyP4+ (Scheme 1, Fig. 1).Given the high fluorescence of ZnIITE-2-PyP4+, we wereable to detect porphyrin levels as low as 0.5 ng/mg ofprotein. Calibration curves obtained by diluting MnIIITE-2-PyP5+ either into mitochondrial homogenate (Fig. 4) or intoBSA/PBS (Fig. 4, inset) were linear in the range from 1.25(+/–3%) to 100 (+/–1%) ng/mL of homogenate. There wasa 10-fold difference in otherwise essentially identical slopesdue to the 10-fold difference in the amount of internalstandard used (0.00186 with BSA/PBS and 0.0191 withmitochondrial homogenate) (Fig. 4). Thus BSA/PBS can beused instead of mitochondrial homogenate as a valid andmore convenient alternative calibration medium for Mnporphyrin determination. The same slope was also obtainedwith plasma as a calibration medium.

The ortho isomers of MnIIITE-2-PyP5+, ZnIITE-2-PyP4+,and their parent metal-free ligand, H2TE-2-PyP

4+, haveatropoisomers, which we have previously characterized byHPLC/UV/Vis/NMR and X-ray methods [32]. We observedthat the abundance ratio of four isomers of ZnIITE-2-PyP4+

and their retention times on HPLC were the same inprocessed heart homogenates or in aqueous solution [32],indicating the absence of strong interactions of any of thefour atropoisomers with cellular components (Fig. 3 and Ref.[32]). Our data also imply that the Mn porphyrin does notundergo oxidative degradation or any other modification invivo. The method is suitable for the analysis of a variety ofbiological samples [54]. It may further be useful for thestudies of other cationic Mn(III) porphyrins that we havebeen developing [8]. We are currently using the method for

Table 1Levels of MnIIITE-2-PyP5+ in C57BL/6 mouse heart mitochondria

Mouse Drug Total protein(10 mg/kg ip) (mg/mL homogenate

Mice were sacrificed 7 h after drug administration1 MnIIITE-2-PyP5+ 2.8202 MnIIITE-2-PyP5+ 2.5433 MnIIITE-2-PyP5+ 2.6824 MnIIITE-2-PyP5+ 2.8335 MnIIITE-2-PyP5+ 1.7946 MnIIITE-2-PyP5+ 1.824

Mice were sacrificed 4 h after drug administration7 MnIIITE-2-PyP5+ 1.2238 MnIIITE-2-PyP5+ 1.909

Control mice9 Saline 3.14710 Saline 3.19311 Saline 2.940

The mean value was calculated to be (2.95±0.56) ng/mg protein.

the ongoing pharmacokinetic study of MnIIITE-2-PyP5+ inmice.

MnIIITE-2-PyP5+ levels in mitochondria

Our data show that MnIIITE-2-PyP5+ localizes in mouseheart mitochondria after a single 10 mg/kg intraperitonealdose. The (2.95 ± 0.56) ng/mg protein of MnIIITE-2-PyP5+

was found (Table 1). The same levels were obtained at 4 or7 h after the administration. The mitochondrial volumereported elsewhere varied greatly from essentially beingnonexistent [55,56] to as much as 1.2 μL/mg protein [55–62]. (The upper limit of 1.2 μL/mg protein was previouslyutilized by Radi et al. [60] and Quijano et al. [63] to calculateconcentrations of mitochondrial MnSOD and cytochrome c tobe 20 and 400 μM, respectively). Based on an average valueof mitochondrial volume of 0.6 μL/mg protein, the calculatedMnIIITE-2-PyP5+ concentration in mitochondria is 5.1 μM.Ferrer-Sueta et al. have shown that MnIIITE-2-PyP5+ was ableto divert peroxynitrite from inactivation of succinate dehy-drogenase and succinate oxidase equally efficiently at 3 μM(81%), 5 μM (85%), or 10 μM (83%), but not at 1 μM (28%)[14]. Thus based on their data and our findings, the 5.1 μMconcentration of MnIIITE-2-PyP5+, achieved in mitochondriaafter single administration, is high enough to protectmitochondrial respiration against peroxynitrite-mediated oxi-dative damage [5,10,14,15]. Additionally, at such levelsMnIIITE-2-PyP5+ can compete for ONOO− with CO2, andcan additionally intercept a substantial amount of CO3

U−.This work answers affirmatively an important question as to

whether MnIIITE-2-PyP5+ and similar, highly charged metallo-porphyrins are capable of entering mitochondrion at levels highenough to exert there their presumed antioxidant action.Moreover it indicates that having been taken into mitochondriain vivo the Mn porphyrin resists washing out during isolation ofthe mitochondria.

MnIIITE-2-PyP5+ MnIIITE-2-PyP5+

) (ng/mL homogenate) (ng/mg protein)

6.55 2.326.75 2.657.82 2.918.15 2.883.83 2.137.17 3.93

4.28 3.506.29 3.29

0.00 0.000.00 0.000.00 0.00

1199I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

Conclusions

The new and sensitive method developed herein enables usto show that, despite its hydrophilicity, i.e., overall high positivecharge, which is essential for its in vivo antioxidant efficacy,MnIIITE-2-PyP5+ targets mitochondria after only a single doseof 10 mg/kg, achieving micromolar levels that are sufficient toexert protection of the mitochondrial respiration from oxidativedamage. Therefore, along with the report of Ferrer-Sueta et al.[14] this study justifies further improvement of Mn(III)porphyrins as therapeutics for protecting mitochondria fromoxidative damage.

Acknowledgments

Ivan Spasojević acknowledges NIH/NCI Duke Comprehen-sive Cancer Center Core Grant (5-P30-CA14236-29). InesBatinić-Haberle appreciates financial support from NIH- IR21-ESO/3682, W19A167798-01, and NIH U19 AI67798-01.Yumin Chen, Marsha Cole, Yunfeng Zhao, Teresa Noel, andDaret St. Clair are thankful to the support of NIH Grants CA49797, CA 73599, and CA 94853. Authors are also thankful toRafael Radi and Gerardo Ferrer-Sueta for helpful discussionsand to Irwin Fridovich for critical reading of the manuscript andfor his continuous support.

References

[1] James, A. M.; Cocheme, H. M.; Smith, R. A. J.; Murphy, M. P. Interactionsof mitochondria-targeted and untargeted ubiquinones with the mitochon-drial respiratory chain and reactive oxygen species. J. Biol. Chem.28:21295–21312; 2005.

[2] Smith, R. A. J.; Porteous, C. M.; Ganes, A. M.; Murphy, M. P. Deliveryof bioactive molecules to mitochondria. Proc. Natl. Acad. Sci. USA100:5407–5412; 2003.

[3] Dhanasekaran, A.; Kotamraju, S.; Karunakaran, C.; Kalivendi, S. V.;Thomas, S.; Joseph, J.; Kalyanaraman, J. J. B. Mitochondria superoxidedismutase mimetic inhibits peroxide-induced oxidative damage andapoptosis: role of mitochondrial superoxide. Free Radic. Biol. Med.39:567–583; 2005.

[4] Asayama, S.; Kawamura, E.; Nagaoka, S.; Kawakami, H. Design ofmanganese porphyrin modified with mitochondrial signal peptide for anew antioxidant. Mol. Pharm. 3:468–470; 2006.

[5] Batinic-Haberle, I.; Benov, L.; Spasojevic, I.; Hambright, P.; Crumbliss,A. L.; Fridovich, I. The relationship between redox potentials, protondissociation constants of pyrrolic nitrogen, and in vitro and in vivosuperoxide dismutase activities of manganese(III) and iron(III) cationic andanionic porphyrins. Inorg. Chem. 38:4011–4022; 1999.

[6] Spasojevic, I.; Batinic-Haberle, I. Manganese(III) complexes withporphyrins and related compounds as scavengers of superoxide. Inorg.Chim. Acta 317:230–242; 2001.

[7] Kachadourian, R.; Batinic-Haberle, I.; Fridovich, I. Syntheses andSOD-like activities of partially chlorinated manganese(III)-5,10,15,20-tetrakis(N-methylpyridinium-2-yl)porphyrin. Inorg. Chem. 38:391–396;1999.

[8] Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Bondurant, B.; Okado-Matsumoto, A.; Fridovich, I.; Vujaskovic, Z.; Dewhirst, M. W. New PEG-ylated Mn(III) porphyrins approaching catalytic activity of SOD enzyme.Dalton Trans. 617–624; 2006.

[9] Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Hambright, P.; Neta, P.;Okado-Matsumoto, A.; Fridovich, I. New class of potent catalysts of O2·−

dismutation. Mn(III) methoxyethylpyridyl- and methoxyethylimidazolyl-porphyrins. J. Chem. Soc. Dalton Trans. 1696–1702; 2004.

[10] Ferrer-Sueta, G.; Vitturi, D.; Batinic-Haberle, I.; Fridovich, I.; Goldstein, S.;Czapski, G.; Radi, R. Reactions of manganese porphyrins with peroxyni-trite and carbonate radical anion. J. Biol. Chem. 278: 27432–27438; 2003.

[11] Spasojevic, I.; Batinic-Haberle, I.; Reboucas, J. S.; Idemori, Y. M.;Fridovich, I. Electrostatic contribution in the catalysis of O2·− dismutationby superoxide dismutase mimics. J. Biol. Chem. 278:6831–6837; 2003.

[12] Batinic-Haberle, I.; Spasojevic, I.; Stevens, R. D.; Hambright, P.;Fridovich, I. Manganese(III) meso tetrakis ortho N-alkylpyridylporphy-rins. synthesis, characterization and catalysis of O2·− dismutation. J. Chem.Soc., Dalton Trans. 2689–2696; 2002.

[13] Batinic-Haberle, I. Manganese porphyrins and related compounds asmimics of superoxide dismutase. Methods Enzymol. 349:223–233; 2001.

[14] Ferrer-Sueta, G.; Hanninbal, L.; Batinic-Haberle, I.; Radi, R. Reduction ofmanganese porphyrins by flavoenzymes and submitochondrial particles: acatalytic cycle for the reduction of peroxynitrite. Free Radic. Biol. Med.41:503–512; 2006.

[15] Ferrer-Sueta, G.; Batinic-Haberle, I.; Spasojevic, I.; Fridovich, I.Peroxynitrite scavanging by manganese(III) meso-tetrakis-(N-methylpyr-idyl)porphyrins. Chem. Res. Toxicol. 12:442–449; 1999.

[16] Batinic-Haberle, I.; Benov, L.; Spasojevic, I.; Fridovich, I. The ortho effectmakes manganese (III) meso-tetrakis(N-methylpyridinium-2-yl)porphyrin(MnTM-2-PyP) a powerful and potentially useful superoxide dismutasemimic. J. Biol. Chem. 273:24521–24528; 1998.

[17] Batinic-Haberle, I.; Spasojevic, I.; Fridovich, I. Tetrahydrobiopterinrapidly reduces SOD mimic, Mn(III) tetrakis(N-ethylpyridinium-2-yl)porphyrin. Free Radic. Biol. Med. 37:367–374; 2004.

[18] Day, B. J.; Batinic-Haberle, I.; Crapo, J. D. Metalloporphyrins are potentinhibitors of lipid peroxidation. Free Radic. Biol. Med. 26:730–736; 1999.

[19] Okado-Matsumoto, A.; Batinic-Haberle, I.; Fridovich, I. Complementationof SOD deficient Escherichia coli by manganese porphyrin mimics ofsuperoxide dismutase. Free Radic. Biol. Med. 37:401–410; 2004.

[20] Zhao, Y.; Chaiswing, L.; Oberley, T. D.; Batinic-Haberle, I.; St. Clair, W.;Epstein, C. J.; St. Clair, D. A mechanism-based antioxidant approach forthe reduction of skin carcinogenesis. Cancer Res. 65:1401–1405; 2005.

[21] Moeller, B. J.; Batinic-Haberle, I.; Spasojevic, I.; Rabbani, Z. N.; Anscher,M. S.; Vujaskovic, Z.; Dewhirst, M. W. Effects of a catalyticmetalloporphyrin antioxidant on tumor radioresponsiveness. Int. J.Radic. Oncol. Biol. Phys. 63:545–552; 2005.

[22] Benov, L.; Batinic-Haberle, I. A manganese porphyrin SOD mimicsuppresses oxidative stress and extends the life span of streptozotocin-diabetic rats. Free Radic. Res. 38:81–88; 2005.

[23] Sheng, H.; Spasojevic, I.; Warner, D. S.; Batinic-Haberle, I. Mouse spinalcord compression injury is ameliorated by intrathecal manganese(III)porphyrin. Neurosci. Lett. 366:220–225; 2004.

[24] Sheng, H.; Enghild, J.; Bowler, R.; Patel, M.; Calvi, C. L.; Batinic-Haberle,I.; Day, B. J.; Pearlstein, R. D.; Crapo, J. D.; Warner, D. S. Effects ofmetalloporphyrin catalytic antioxidants in experimental brain ischemia.Free. Radic. Biol. Med. 33:947–961; 2002.

[25] Vujaskovic, Z.; Batinic-Haberle, I.; Rabbani, Z. N.; Feng, Q. -F.;Kang, S. K.; Spasojevic, I.; Samulski, T. V.; Fridovich, I.; Dewhirst,M. W.; Anscher, M. S. A Small molecular weight catalytic metallopor-phyrin antioxidant with superoxide dismutase (SOD) mimetic propertiesprotects lungs from radiation-induced injury. Free. Radic. Biol. Med.33:857–863; 2002.

[26] Piganelli, J. D.; Flores, S. C.; Cruz, C.; Koepp, J.; Young, R.; Bradley, B.;Kachadourian, R.; Batinic-Haberle, I.; Haskins, K. A Metalloporphyrinsuperoxide dismutase mimetic (SOD mimetic) inhibits autoimunediabetes. Diabetes 51:347–355; 2002.

[27] Mackensen, G. B.; Patel, M.; Sheng, H.; Calvi, C. L.; Batinic-Haberle, I.;Day, B. J.; Liang, L. P.; Fridovich, I.; Crapo, J. D.; Pearlstein, R. D.;Warner, D. S. Neuroprotection from delayed post-ischemic administrationof a metalloporphyrin catalytic antioxidant in the rat. J. Neurosci.21:4582–4592; 2001.

[28] Aslan, M.; Ryan, T. M.; Adler, B.; Townes, T. M.; Parks, D. A.; Thompson,J. A.; Tousson, A.; Gladwin, M. T.; Tarpey, M.; Patel, R. P.; Batinic-Haberle, I.; White, C. R.; Freeman, B. A. Oxygen radical inhibition of

1200 I. Spasojević et al. / Free Radical Biology & Medicine 42 (2007) 1193–1200

nitric-oxide dependent vascular function in sickle cell disease. Proc. Natl.Acad. Sci. USA. 98:15215–15220; 2001.

[29] Bottino, R.; Balamurugan, A. N.; Bertera, S.; Pietropaolo, M.; Trucco, M.;Piganelli, J. D. Preservation of human islet cell functional mass by anti-oxidative action of a novel SOD mimic compound. Diabetes 51:2561–2567; 2002.

[30] Li, Q.-Y.; Pedersen, C.; Day, B. J.; Patel, M. Dependence of excitotoxicityneurodegeneration on mitochondrial aconitase inactivation. J. Neurochem.78:746–755; 2001.

[31] Oury, T. D.; Thakker, K.; Menache, M.; Chang, L.-Y.; Crapo, J. D.; Day, B.J. Attenuation of bleomycin-induced pulmonary fibrosis by a catalyticantioxidant metalloporphyrin. Am. J. Respir. Cell. Mol. Biol. 25:164–169;2001.

[32] Spasojevic, I.; Menzeleev, R.; White, P. S.; Fridovich, I. RotationalIsomers of N-alkylpyridylporphyrins and their metal complexes. HPLCseparation, 1H NMR and X-ray structural characterization, electrochem-istry and catalysis of O2·− disproportionation. Inorg. Chem. 41:5874–5881; 2002.

[33] Benov, L.; Batinic-Haberle, I.; Spasojevic, I.; Fridovich, I. Isomeric N-alkylpyridylporphyrins and their Zn(II) complexes: inactive as SODmimics but are powerful photosenzitizers. Arch. Biochem. Biophys.402:159–165; 2002.

[34] Kalyanasundaram, K. Photochemistry of water-soluble porphyrins:comparative study of isomeric tetrapyridyl- and tetrakis(N-methylpyridi-niumyl)porphyrins. Inorg. Chem. 23:2453–2459; 1984.

[35] Vergeldt, F. J.; Koehorst, R. B. M.; Van Hoek, A.; Schaafsma, T. J.Intramolecular interactions in the ground and excited state of tetrakis(N-methylpyridyl)porphyrins. J. Phys. Chem. 99:4397–4405; 1995.

[36] Quimby, D. J.; Longo, F. R. Luminescence studies of several tetra-rylporphyrins and their Zn derivatives. J. Am. Chem. Soc. 97:5111–5117;1975.

[37] Harriman, A.; Porter, G. Photochemistry of manganese porphyrins. Part 4.Photosensitized reductions of quinines. J. Chem. Soc. Farady II 76:1429–1441; 1980.

[38] Taqui Khan, M. M.; Martell, A. E. The kinetics of reaction of iron(III)chelates of aminopolycarboxylic acids with ascorbic acid. J. Am. Chem.Soc. 90:3386–3389; 1968.

[39] Hambright, P.; Batinic-Haberle, I.; Spasojevic, I. Meso tetrakis ortho, metaand para N-alkylpyridylporphyrins: kinetics of copper(II) and zinc(II)incorporation and zinc porphyrin demetalation. J. Porphyrins Phthalo-cyanines 7:139–146; 2003.

[40] Kragten, J. Atlas of Metal-Ligand Equilibria in Aqueous Solution.Halstead, New York; 1978.

[41] Spasojevic, I.; Batinic-Haberle, I.; Fridovich, I. Nitrosylation of manga-nese(II) tetrakis(N-ethylpyridinium-2-yl)porphyrin. Nitric Oxide: Biol.Chem. 4:526–533; 2000.

[42] Mela, L.; Seitz, S. Isolation of mitochondria with emphasis on heartmitochondria from small amounts of tissue. Methods Enzymol.55:39–46; 1979.

[43] Nedergaard, J.; Cannon, B. Overview-preparation and properties ofmitochondria from different sources. Methods Enzymol. 55:3–28; 1979.

[44] Lawrence, G. D.; Sawyer, D. T. Potentiometric titrations and oxidation-reduction potentials of manganese and copper-zinc superoxide dismutases.Biochemistry 18:3045–3050; 1979.

[45] Barrette W. C. Jr.; Sawyer, D. T.; Free, J. A.; Asada, K. Potentiometrictitrations and oxidation-reduction potentials of several iron superoxidedismutases. Biochemistry 22:624–627; 1983.

[46] Ellerby, R. M.; Cabelli, D. E.; Graden, J. A.; Valentine, J. S. Copper-zincsuperoxide dismutase: why not pH-dependent? J. Am. Chem. Soc.118:6556–6561; 1996.

[47] Vance, C. K.; Miller, A. F. A simple proposal that can explain the inactivityof metal-substituted superoxide dismutases. J. Am. Chem. Soc. 120:461–467; 1998.

[48] Michel, E.; Nauser, T.; Sutter, B.; Bounds, P. L.; Koppenol, W. H. Kineticproperties of Cu,Zn-superoxide dismutase as a function of metal content.Arch. Biochem. Biophys 439:234–240; 2005.

[49] Goldstein, S.; Fridovich, I.; Czapski, G. Kinetic properties of Cu,Zn-superoxide dismutase as a function of metal content—Order restored. FreeRadic. Biol. Med. 41:937–941; 2006.

[50] Nauser, T.; Koppenol, W. H. The rate constant of the reaction of superoxideand nitrogen monoxide: approaching the diffusion limit. J. Phys. Chem. A106:4084–4086; 2002.

[51] Ferrer-Sueta, G.; Quijano, C.; Alvarez, B.; Radi, R. Reactions ofmanganese porphyrins and manganese-superoxide dismutase with perox-ynitrite. Methods Enzymol. 349:23–27; 2002.

[52] Tse, H.; Milton, M. J.; Piganelli, J. D. Mechanistic analysis of theimmunomodulatory effects of a catalytic antioxidant on antigen-presentingcells: implication for their use in targeting oxidation/reduction reactions ininnate immunity. Free Radic. Biol. Med. 36:233–247; 2004.

[53] Bottino, R.; Balamurugan, A. N.; Tse, H.; Thirunavukkarasu, C.; Ge, X.;Profozich, J.; Milton, M.; Ziegenfuss, A.; Trucco, M.; Piganelli, J. D.Response of human islets to isolation stress and the effect of antioxidanttreatment. Diabetes 53:2559–2568; 2004.

[54] Spasojevic, I.; Batinic-Haberle, I.; Yumin, C.; Noel, T.; Lichun, Z.; St.Clair, D.K. MnTE-2-PyP5+ pharmacokinetics. In preparation.

[55] Nedergaard, J.; Cannon, B. Apparent masking of [3H]GDP binding in ratbrown-fat mitochondria is due to mitochondrial swelling. Eur. J. Biochem.164:681–686; 1987.

[56] Cohen, N. S.; Cheung, C.-W.; Raijman, L. Measurements of mitochondrialvolumes are affected by the amount of mitochondria used in determina-tions. Biochem. J. 245:375–379; 1987.

[57] Halestrap, A. P.; Quinlan, P. T. The intramitochondrial volume measuredusing sucrose as an extramitochondrial marker overestimates the truematrix volume determined with manitol. Biochem. J. 214:387–393;1983.

[58] Halestrap, A. P.; Dunlop, J. L. Intramitochondrial regulation of fatty acidb-oxidation occurs between flavoprotein and ubiquinone. Biochem. J. 239:559–565; 1986.

[59] Beavis, A. D.; Brannan, R. D.; Garlid, K. D. Swelling and contraction ofmitochondrila matrix: I. A structural interpretation of the relationshipbetween light scattering and matrix volume. J. Biol. Chem. 260:13424–13433; 1985.

[60] Radi, R.; Turrens, J. F.; Freeman, B. A. Cytochrome c-catalyzed membranelipid peroxidation by hydrogen peroxide. Arch. Biochem. Biophys.288:118–125; 1991.

[61] Halestrap, A. P.; Quinlan, P. T.; Whipps, D. E.; Armston, A. E. Regualtionof the mitochondrial matrix volume in vivo and in vitro. The role ofcalcium. Biochem. J. 236:779–787; 1986.

[62] Rickwood, D.; Wilson, M. T.; Darley-Usmar, V. M. Mitochondria. IRLPress, Oxford; 1987.

[63] Quijano, C.; Hernandez-Saavedra, D.; Castro, L.; McCord, J. M.; Freeman,B. A.; Radi, R. Reaction of peroxynitrite with Mn-superoxide dismutase.Role of the metal center in decomposition kinetics and nitration. J. Biol.Chem. 276:11631–11638; 2001.