Nitrogen dioxide oxidizes mitochondrial cytochrome c

Abstract

We previously reported that high micromolar concentrations of nitric oxide were able to oxidize mitochondrial cytochrome c at physiological pH, producing nitroxyl anion (Sharpe and Cooper, 1998 Biochem. J. 332, 9-19). However, the subsequent re-evaluation of the redox potential of the NO/NO(-) couple suggests that this reaction is thermodynamically unfavored. We now show that the oxidation is oxygen-concentration dependent and non stoichiometric. We conclude that the effect is due to an oxidant species produced during the aerobic decay of nitric oxide to nitrite and nitrate. The species is most probably nitrogen dioxide, NO(2)(•) a well-known biologically active oxidant. A simple kinetic model of NO autoxidation is able to explain the extent of cytochrome c oxidation assuming a rate constant of 3×10(6)M(-1)s(-1) for the reaction of NO(2)(•) with ferrocytochrome c. The importance of NO(2)(•) was confirmed by the addition of scavengers such as urate and ferrocyanide. These convert NO(2)(•) into products (urate radical and ferricyanide) that rapidly oxidize cytochrome c and hence greatly enhance the extent of oxidation observed. The present study does not support the previous hypothesis that NO and cytochrome c can generate appreciable amounts of nitroxyl ions (NO(-) or HNO) or of peroxynitrite.

Graphical abstract

Fig. 1

Cytochrome c oxidation by proliNO under aerobic conditions. 40 μM proliNO was added to 10 μM ferrocytochrome c at a temperature of 30 °C and a pH of 7.4, buffered with 100 mM potassium phosphate, 100 μM DTPA. Cytochrome c reduction was calculated as described under Materials and methods. Means ± SD (n = 4).

Fig. 2

Spectroscopic changes in cytochrome c treated with proliNO under aerobic and anaerobic conditions. Conditions were as for Fig. 1. (a) Spectra are shown for 0, 5, 15, and 30 min after the aerobic addition of 40 μM proliNO followed by the addition of 10 μM K₃Fe(CN)₆. Dashed lines indicate the spectrum post ferricyanide addition. Subsequent addition of sodium dithionite (0.03% final concentration) produced essentially the same spectrum as that of the initial ferrocytochrome c. (b) Same as (a) but under anaerobic conditions. (c) Time courses from (a) and (b); (d) same as (c) but replacing 100 mM potassium phosphate with 100 mM sodium Hepes.

Fig. 3

Parallel measurements of oxygen and NO concentrations and the concomitant changes in the redox state of cytochrome c. The change in redox state of 10 μM ferrocytochrome c was monitored after the addition of a single 40 μM aliquot of proliNO. The medium was a 100 mM potassium phosphate, 100 μM DTPA, pH 7.4, buffer, at 30 °C. (a) Time courses of NO and O₂ concentrations and (inset) cytochrome c redox state under aerobic conditions. (b) Same as (a) but under anaerobic conditions.

Fig. 4

Effects of excess oxygen upon ferrocytochrome c oxidation induced by nitric oxide. Each experiment involved two additions of 40 μM proliNO to 10 μM ferrocytochrome c. The oxygen concentration in the solution was increased by flushing the cuvette with O₂ gas before a second addition of proliNO as indicated. Other conditions were as for Fig. 1 (phosphate buffer).

Fig. 5

Effect of urate upon the NO-induced oxidation of ferrocytochrome c. Each experiment involved addition of 40 μM proliNO to aerobic 10 μM ferrocytochrome c. Other conditions were as for Fig. 1 (phosphate buffer). (Trace a) 1.0 mM urate and a second addition of proliNO added at the indicated times. (Trace b) 1.0 mM urate present from start of experiment.

Fig. 6

Aerobic titration of ferrocytochrome c with proliNO in the presence and absence of urate. The fractional oxidation of the cytochrome is plotted after successive additions of proliNO to 10 μM ferrocytochrome c under standard aerobic conditions (as in Fig. 1) and in the presence or absence of 1.0 mM urate. The plot shows micromolar cytochrome c oxidized against microequivalents of total NO added. Stoichiometries were calculated from the linear slopes.

Fig. 7

Pathways of nitric oxide autoxidation in an aqueous medium. c²⁺/c³⁺, ferric/ferrous cytochrome c; radical species indicated by superscript dots (^•). Rate constants were taken from the literature as follows: NO autoxidation pathway values are from Lancaster and references therein; urate oxidation by NO₂^• from Simic and Jovanovin and Ford et al. ; the rates for NO binding to oxidized cytochrome c were from Sharpe and Cooper ; the rate of NO₂^• oxidation of cytochrome c was calculated to best fit the data in this paper (see Figs. 8a and 8c); the rate of urate radical oxidation of cytochrome c and its uncatalyzed decay to an unspecified product were chosen to be consistent with the data relating to the enhanced oxidation of cytochrome c in its presence (see Figs. 8b and 8c)—note that the data allow considerable leeway in the absolute choice of these rate constants though the ratio of the two is more constrained.

Fig. 8

Modeling of cytochrome c oxidation by nitrogen dioxide. Simulations were carried out using the rate constants in Fig. 7 unless otherwise indicated. (a) The effect of the addition of 78 µM NO to 10 mM cytochrome c, varying the rate constant (M^− 1 s^− 1) for the NO₂^• oxidation of cytochrome c as indicated. (b) Same as (a) with varying urate concentrations as indicated. (c) The computed concentration of ferricytochrome c formed after simulations of repeated boluses of NO in the absence and presence of 1 mM urate. The simulation was allowed to run for 50 min before the next addition of NO. Lines indicate linear regression (with slopes indicated) to all the points in the control data and the linear part of the urate curve, i.e., when the total [NO] < 25 µM. (d) The effects of the addition of 0.5 mM NO to 800 mM ferrocytochrome c on the concentrations of NO and ferricytochrome c (left axis). The ratio of [NO] removed to ferricytochrome c formed is indicated on the right axis. A starting oxygen concentration of 200 µM was used for all simulations. Unless otherwise indicated all other species were set to zero concentration at the start of the simulation.