Iron and redox cycling. Do's and don'ts.

A major form of toxicity arises from the ability of iron to redox cycle, that is, to accept an electron from a reducing compound and to pass it on to H2O2 (the Fenton reaction). In order to do so, iron must be suitably complexed to avoid formation of Fe2O3. The ligands determine the electrode potential; this information should be known before experiments are carried out. Only one-electron transfer reactions are likely to be significant; thus two-electron potentials should not be used to determine whether an iron(III) complex can be reduced or oxidized. Ascorbate is the relevant reducing agent in blood serum, which means that iron toxicity in this compartment arises from the ascorbate-driven Fenton reaction. In the cytosol, an iron(II)-glutathione complex is likely to be the low-molecular weight iron complex involved in toxicity. When physiologically relevant concentrations are used the window of redox opportunity ranges from +0.1 V to +0.9 V. The electrode potential for non-transferrin-bound iron in the form of iron citrate is close to 0 V and the reduction of iron(III) citrate by ascorbate is slow. The clinically utilised chelators desferrioxamine, deferiprone and deferasirox in each case render iron complexes with large negative electrode potentials, thus being effective in preventing iron redox cycling and the associated toxicity resulting from such activity. There is still uncertainty about the product of the Fenton reaction, HO• or FeO2+.

The Haber-Weiss cycle--70 years later.

The chain reactions HO* + H2O2 --> H2O + O2*- + H+ and O2*- + H+ + H2O2 --> O2 + HO* + H2O, commonly known as the Haber-Weiss cycle, were first mentioned by Haber and Willstätter in 1931. George showed in 1947 that the second reaction is insignificant in comparison to the fast dismutation of superoxide, and this finding appears to have been accepted by Weiss in 1949. In 1970, the Haber-Weiss reaction was revived by Beauchamp and Fridovich to explain the toxicity of superoxide. During the 1970s various groups determined that the rate constant for this reaction is of the order of 1 M(-1) s(-1) or less, which confirmed George's conclusion. The reaction of superoxide with hydrogen peroxide was dropped from the scheme of oxygen toxicity, and superoxide became the source of hydrogen peroxide, which yields hydroxyl radicals via the Fenton reaction, Fe2+ + H2O2 --> Fe3+ + HO- + HO*. In 1994, Kahn and Kasha resurrected the Haber-Weiss reaction again, but this time the oxygen was believed to be in the singlet (1delta(g)) state. As toxicity arises not from a Fenton-catalysed Haber-Weiss reaction, but from the Fenton reaction, the Haber-Weiss reaction should not be mentioned anymore.

Oxidation of NADH by chloramines and chloramides and its activation by iodide and by tertiary amines.

Irreversible oxidation of reduced nicotinamide nucleotides by neutrophil-derived halogen oxidants (HOCl, chloramines, HOBr, etc.) is likely to be a highly lethal process, because of the essential role of NAD(P)H in important cell functions such as mitochondrial electron transport, and control of the cellular thiol redox state by NADPH-dependent glutathione reductase. Chloramines (chloramine-T, NH(2)Cl, etc.) and N-chloramides (N-chlorinated cyclopeptides) react with NADH to generate the same products as HOCl, i.e., pyridine chlorohydrins, as judged from characteristic changes in the NADH absorption spectrum. Compared with the fast oxidation of NADH by HOCl, k approximately 3 x 10(5) M(-1) s(-1) at pH 7.2, the oxidation by chloramines is about five orders of magnitude slower; that by chloramides is about four orders of magnitude slower. Apparent rate constants for oxidation of NADH by chloramines increase with increasing proton or buffer concentration, consistent with general acid catalysis, but oxidation by chloramides proceeds with pH-independent kinetics. In presence of iodide the oxidation of NADH by chloramines or chloramides is faster by at least two orders of magnitude; this is due to reaction of iodide with the N-halogen to give HOI/I(2), the most reactive and selective oxidant for NADH among HOX species. Quinuclidine derivatives (QN) like 3-chloroquinuclidine and quinine are capable of catalyzing the irreversible degradation of NADH by HOCl and by chloramines; QN(+)Cl, the chain carrier of the catalytic cycle, is even more reactive toward NADH than HOCl/ClO(-) at physiological pH. Oxidation of NADH by NH(2)Br proceeds by fast, but complex, biphasic kinetics. A compilation of rate constants for interactions of reactive halogen species with various substrates is presented and the concept of selective reactivity of N-halogens is discussed.

On the oxidation of cytochrome c by hypohalous acids.

Oxidation of cytochrome c, a key protein in mitochondrial electron transport and a mediator of apoptotic cell death, by reactive halogen species (HOX, X2), i.e., metabolites of activated neutrophils, was investigated by stopped-flow. The fast initial reactions between FeIIIcytc and HOX species, with rate constants (at pH 7.6) of k > 3 x 10(6) M(-1) s(-1) for HOBr, k > 3 x 10(5) M(-1) s(-1) for HOCl, and k = (6.1+/-0.3) x 10(2) M(-1) s(-1) for HOI, are followed by slower intramolecular processes. All HOX species lead to a blue shift of the Soret absorption band and loss of the 695-nm absorption band, which is an indicator for the intact iron to Met-80 bond, and of the reducibility of FeIIIcytc. All HOX species do, in fact, persistently impair the ability of FeIIIcytc to act as electron acceptor, e.g., in reaction with ascorbate or O2*-. I2 selectively oxidizes the iron center of FeIIcytc, with a stoichiometry of 2 per I2, and with k(FeIIcytc + I2) approximately 4.6 x 10(4) M(-1) s(-1) and k(FeIIcytc + I2*-) = (2.9+/-0.4) x 10(8) M(-1) s(-1). Oxidation of FeIIcytc by HOX species is not selectively directed toward the iron center; HOBr and HOCl are considered to react primarily by N-halogenation of side chain amino groups, and HOI mainly by sulfoxidation. There is some evidence for the generation of HO* radicals upon reaction of HOCl with FeIIcytc. Chloramines (e.g., NH2Cl), bromamine (NH2Br), and cyclo-Gly2 chloramide oxidize FeIIcytc slowly and unselectively, but iodide efficiently catalyzes reactions of these N-halogens to yield fast selective oxidation of the iron center; this is due to generation of I2 by reaction of I- with the N-halogen and recycling of I- by reaction of I2 with FeIIcytc. Iodide also catalyzes methionine sulfoxidation and thiol oxidation by NH2Cl. The possible biological relevance of these findings is discussed.

Gibbs energy of formation of peroxynitrite.

A standard Gibbs energy of formation of 16.6 kcal mol(-)(1) has been reported for peroxynitrite [Merényi, G., and Lind, J. (1998) Chem. Res. Toxicol. 11, 243-246]. This value is based on the rate constants for the forward and backward rate constants of the equilibrium O2*- + NO* if ONOO(-). A rate constant of 0.017 s(-)(1) for the backward rate constant was determined by observing the formation of C(NO(2))(3)(-) when peroxynitrite was mixed with C(NO(2))(4). However, a similar rate constant is also observed in the presence of NO(*), which indicates that formation of C(NO(2))(3)(-) is due to a process other than the reduction of C(NO(2))(4) by O2*-. Additionally, copper(II) nitrilotriacetate enhances the decay of ONOO(-) at pH 9.3, without reduction of copper(II). The preferred thermodynamic values are therefore as follows: Delta(f)H degrees (ONOO(-)) = -10 +/- 2 kcal mol(-)(1), Delta(f)G degrees (ONOO(-)) = 14 +/- 3 kcal mol(-)(1), S degrees (ONOO(-)) = 31 eu, and E degrees '(ONOOH/NO(2)(*), H(2)O) = 1.6 V at pH 7 [Koppenol, W. H., and Kissner, R. (1998) Chem. Res. Toxicol. 11, 87-90].

100 years of peroxynitrite chemistry and 11 years of peroxynitrite biochemistry.

The paper on the unusual properties of a mixture of hydrogen peroxide and nitrous acid by Baeyer and Villiger from 1901 can be regarded as the first report on peroxynitrite. In 1990, Beckman and co-workers suggested that peroxynitrite, formed from the reaction of superoxide with nitrogen monoxide, could be a transient oxidizing species in vivo, a report that revolutionized investigations in the field of oxidative stress.

Peroxynitrite does not decompose to singlet oxygen ((1)Delta (g)O(2)) andnitroxyl (NO(-)).

According to Khan et al. [Khan, A. U., Kovacic, D., Kolbanovskiy, A., Desai, M., Frenkel, K. & Geacintov, N. E. (2000) Proc. Natl. Acad. Sci. USA 97, 2984-2989], peroxynitrite (ONOO(-)) decomposes after protonation to singlet oxygen ((1)Delta(g)O(2)) and singlet oxonitrate (nitroxyl, (1)NO(-)) in high yield. They claimed to have observed nitrosyl hemoglobin from the reaction of NO(-) with methemoglobin; however, contamination with hydrogen peroxide gave rise to ferryl hemoglobin, the spectrum of which was mistakenly assigned to nitrosyl hemoglobin. We have carried out UV-visible and EPR experiments with methemoglobin and hydrogen peroxide-free peroxynitrite and find that no NO(-) is formed. With this peroxynitrite preparation, no light emission from singlet oxygen at 1270 nm is observed, nor is singlet oxygen chemically trapped; however, singlet oxygen was trapped when hydrogen peroxide was also present, as previously described [Di Mascio, P., Bechara, E. J. H., Medeiros, M. H. G., Briviba, K. & Sies, H. (1994) FEBS Lett. 355, 287-289]. Quantum mechanical and thermodynamic calculations show that formation of the postulated intermediate, a cyclic form of peroxynitrous acid (trioxazetidine), and the products (1)NO(-) and (1)Delta(g)O(2) requires Gibbs energies of ca. +415 kJ .mol(-1) and ca. +180 kJ.mol(-1), respectively. Our results show that the results of Khan et al. are best explained by interference from contaminating hydrogen peroxide left from the synthesis of peroxynitrite.

On the irreversible destruction of reduced nicotinamide nucleotides by hypohalous acids.

Degradation of the reduced pyridine nucleotides NMNH and NADH by HOCl involves two distinct stages: a fast reaction, k = 4.2 x 10(5) M(-1) s(-1), leads to generation of stable pyridine products (Py/Cl) with a strong absorption band at 275 nm (epsilon = 12.4 x 10(3) M(-1) cm(-1) in the case of NMNH); secondarily, a subsequent reaction of HOCl, k = 3.9 x 10(3) M(-1) s(-1), leads to a complete loss of the aromatic absorption band of the pyridine ring. HOBr and HOI(I(2)) react similarly. Apparent rate constants of the primary reactions of HOX species with NMNH at pH 7.2 increase in the order HOCl (3 x 10(5) M(-1) s(-1)) < HOBr( approximately 4 x 10(6) M(-1) s(-1)) < HOI(I(2))( approximately 6.5 x 10(7) M(-1) s(-1)). HOBr reacts fast also with the primary product Py/Br, k approximately 9 x 10(5) M(-1) s(-1), while the reactions of HOI and I(2) with Py/I are slower, approximately 1.4 x 10(3) M(-1) s(-1) and >6 x 10(3) M(-1) s(-1), respectively. Halogenation of the amide group of NMN(+) by HOX species is many orders of magnitude slower than oxidation of NMNH. Taurine inhibits HOCl-induced oxidation of NADH, but HOBr-induced oxidation is not inhibited because the taurine monobromamine rapidly oxidizes NADH, and oxidation by HOI(I(2)) is not inhibited because taurine is inert toward HOI(I(2)). Also sulfur compounds (GSH, GSSG, and methionine) are less efficient in protecting NADH against oxidation by HOBr and HOI(I(2)) than against oxidation by HOCl. The results suggest that reactions of HOBr and HOI(I(2)) in a cellular environment are much more selectively directed toward irreversible oxidation of NADH than reactions of HOCl. It is noteworthy that the rather inert N-chloramines react with iodide to generate HOI(I(2)), i.e., the most reactive and selective oxidant of reduced pyridine nucleotides. NMR investigations show that the primary stable products of the reaction between NMNH and HOCl are various isomeric chlorohydrins originating from a nonstereospecific electrophilic addition of HOCl to the C5&dbond;C6 double bond of the pyridine ring. The primary products (Py/X) of NMNH all exhibit similar absorption bands around 275 nm and are hence likely to result from analogous addition of HOX to the C5&dbond;C6 bond of the pyridine ring. Since the Py/X species are stable and inert toward endogeneous reductants like ascorbate and GSH, they may generally be useful markers for assessing the contribution of hypohalous acids to inflammatory injury.

The quantitative oxidation of methionine to methionine sulfoxide by peroxynitrite.

Both peroxynitrous acid and peroxynitrite react with methionine, k(acid) = (1.7 +/- 0.1) x 10(3) M(-1) s(-1) and k(anion) = 8.6 +/- 0.2 M(-1) s(-1), respectively, and with N-acetylmethionine k(acid) = (2.8 +/- 0.1) x 10(3) M(-1) s(-1) and k(anion) = 10.0 +/- 0.1 M(-1) s(-1), respectively, to form sulfoxides. In contrast to the results of Pryor et al. (1994, Proc. Natl. Acad. Sci. USA 91, 11173-11177), a linear correlation between k(obs) and [met] was obtained. Surprisingly, for every two sulfoxides and nitrites formed, one peroxynitrite is converted to nitrate. Thus, methionine also catalyzes the isomerization of peroxynitrite to nitrate. Neither the pH nor the concentration of methionine affected the distribution of the yields of nitrite, nitrate, and methionine sulfoxide, which were the only products detected. No products other than nitrite, nitrate, and methioninesulfoxide could be detected. The reactions of methionine and N-acetylmethionine with peroxynitrous acid and peroxynitrite are simple bimolecular reactions that do not involve an activated form of peroxynitrous acid or of peroxynitrite. Nitrite, produced together with methionine sulfoxide, or present as a contamination in the peroxynitrite preparation, is not innocuous, but oxidizes methionine by one electron, which leads to the formation of methional and ethylene.

The preparation of apo-Cu,Zn superoxide dismutase by ion-exchange chromatography on iminodiacetic acid-sepharose.

The superoxide dismutases (EC 1.15.1.1) are a family of enzymes that catalyze the dismutation of superoxide radical anion to dioxygen and hydrogen peroxide. The active site contains a critical metal ion such as manganese, iron, or copper. The copper-containing protein also has one zinc ion bound per subunit. The standard method used to remove the metal ions from Cu,Zn superoxide dismutase has been to exhaustively dialyze the protein against chelating agents at low pH. We have developed a new method where the protein is bound to ion-exchange medium based on iminodiacetic acid immobilized on Sepharose. The bound protein is treated with a buffer containing edta at pH 3.5 to remove metal ions; the buffer is then exchanged for acetate buffer to remove edta, after which the protein is eluted by a salt gradient. An advantage of this method is that a single chromatography step is sufficient to produce apo protein. Results are shown for both human and bovine dimeric Cu,Zn superoxide dismutase and the monomeric Escherichia coli Cu,Zn superoxide dismutase. In every case, the metals were removed efficiently.

Mechanism of reaction of myeloperoxidase with nitrite.

Myeloperoxidase (MPO) is a major neutrophil protein and may be involved in the nitration of tyrosine residues observed in a wide range of inflammatory diseases that involve neutrophils and macrophage activation. In order to clarify if nitrite could be a physiological substrate of myeloperoxidase, we investigated the reactions of the ferric enzyme and its redox intermediates, compound I and compound II, with nitrite under pre-steady state conditions by using sequential mixing stopped-flow analysis in the pH range 4-8. At 15 degrees C the rate of formation of the low spin MPO-nitrite complex is (2.5 +/- 0.2) x 10(4) m(-1) s(-1) at pH 7 and (2.2 +/- 0.7) x 10(6) m(-1) s(-1) at pH 5. The dissociation constant of nitrite bound to the native enzyme is 2.3 +/- 0.1 mm at pH 7 and 31.3 +/- 0.5 micrometer at pH 5. Nitrite is oxidized by two one-electron steps in the MPO peroxidase cycle. The second-order rate constant of reduction of compound I to compound II at 15 degrees C is (2.0 +/- 0.2) x 10(6) m(-1) s(-1) at pH 7 and (1.1 +/- 0.2) x 10(7) m(-1) s(-1) at pH 5. The rate constant of reduction of compound II to the ferric native enzyme at 15 degrees C is (5.5 +/- 0.1) x 10(2) m(-1) s(-1) at pH 7 and (8.9 +/- 1.6) x 10(4) m(-1) s(-1) at pH 5. pH dependence studies suggest that both complex formation between the ferric enzyme and nitrite and nitrite oxidation by compounds I and II are controlled by a residue with a pK(a) of (4.3 +/- 0.3). Protonation of this group (which is most likely the distal histidine) is necessary for optimum nitrite binding and oxidation.

Conformation of peroxynitrite: determination by crystallographic analysis.

Peroxynitrite is an inorganic toxin of biological importance. It is formed in vivo from the diffusion-limited reaction of nitrogen monoxide with superoxide. Due to the partial double bond between the nitrogen and the first peroxide oxygen, peroxynitrite can occur in two conformations, cis and trans. The synthesis of tetramethylammonium peroxynitrite in ammonia [Bohle, D. S., et al. (1994) J. Am. Chem. Soc. 116, 7423-7424] yields small crystals if the ammonia is left to evaporate slowly. X-ray structure analysis shows that peroxynitrite crystallizes in the cis form, relative to the N-O bond. Crystal twinning or disorder prevents the determination of accurate bond lengths and bond angles. However, a nearly flat (torsion angle of 22 degrees ) molecule with O=N, N-O, and O-O bond lengths of 1.16, 1.35, and 1.41 A, respectively, would fit the observed electron density best. The space group of tetramethylammonium peroxynitrite is Pmmn (59), and it has the following unit cell dimensions: a = 7.1778(11) A, b = 8.6893(13) A, and c = 5.7266(9) A.

Kinetic study of the reaction of glutathione peroxidase with peroxynitrite.

Glutathione peroxidases and their mimics, e.g., ebselen or diaryl tellurides, efficiently reduce peroxynitrite/peroxynitrous acid (ONOO-/ONOOH) to nitrite and protect against oxidation and nitration reactions. Here, we report the second-order rate constant for the reaction of the reduced form of glutathione peroxidase (GPx) with peroxynitrite as (8.0 +/- 0.8) x 10(6) M-1 s-1 (per GPx tetramer) at pH 7.4 and 25 degreesC. The rate constant for oxidized GPx is about 10 times lower, (0.7 +/- 0.2) x 10(6) M-1 s-1. On a selenium basis, the rate constant for reduced GPx is similar to that obtained previously for ebselen. The data support the conclusion that GPx can exhibit a biological function by acting as a peroxynitrite reductase.

The basic chemistry of nitrogen monoxide and peroxynitrite.

After a discussion of the physical chemistry of nitrogen monoxide, such as solubility (1.55 mM at 37 degrees C and an ionic strength of 0.15 M) and diffusion constant (4.8 x 10(-5) cm2s(-1)), several reactions that can acts as sinks are discussed, namely the reaction with dioxygen, with thiols and with superoxide. Of these, the latter reaction leads to a powerful oxidant, peroxynitrite. The thermodynamic and kinetic properties of this molecule are also reviewed.

The reaction of peroxynitrite with zeaxanthin.

The oxygenated carotenoids zeaxanthin and lutein, found in the macular area of the retina, may offer protection against or repair of oxidative damage associated with the degenerative diseases of aging. Since both superoxide and nitrogen monoxide, which react to form peroxynitrite, are found in the retina, we studied the reaction of peroxynitrite with zeaxanthin in liposomes. Zeaxanthin was easily incorporated into liposomes constructed from the fully saturated lipid L-alpha-dimyristoyl-phosphatidylcholine (C14:0) and from egg lecithin, and its absorbance spectrum in liposomes strongly resembles in shape and amplitude that of zeaxanthin dissolved in methanol. The reaction between peroxynitrite and zeaxanthin is first-order in both substrates. The pH profile indicates that the reaction with zeaxanthin involves peroxynitrous acid and not the conjugate anion. We hypothesize that zeaxanthin plays a major role in protection of macular tissue from oxidative damage.

Can O=NOOH undergo homolysis?

Recent thermodynamic calculations of Merenyi and Lind [(1997) Chem. Res. Toxicol. 10, 1216-1220] suggest that O=NOOH can undergo homolysis to form the hydroxyl radical and nitrogen dioxide. This result is based in part on our statement that the enthalpy of ionization of O=NOOH is close to zero [Koppenol et al. (1992) Chem. Res. Toxicol. 5, 834-842]. As the ionization of O=NOOH is sensitive to the milieu and the rate of isomerization (to nitrate) to the total concentration of O=NOOH and O=NOO- [Kissner et al. (1997) Chem. Res. Toxicol. 10, 1285-1292], we reinvestigated the temperature dependence of the ionization constant and determined a deltaHo of 4+/-2 kcal mol(-1). This results in a standard Gibbs energy of homolysis of 16 kcal mol(-1) and a rate of homolysis of 1 x 10(-2) s[-1]. Given the uncertainty in the Gibbs energy of homolysis, upper and lower rates are 1 x 10(-4) and 0.6 s(-1), slower than the rate of isomerization, 1.2 s(-1) at 25 degrees C. The recombination of the homolysis products NO2. and HO. is known to lead to mainly peroxynitrous acid. If one assumes that a few percent of the recombinations lead to nitrate instead, then the rate of homolysis must be much higher than the rate of isomerization. We conclude therefore that homolysis is unlikely.