Biochemical Journal

Research article

Inhibition of the chlorinating activity of myeloperoxidase by tempol: revisiting the kinetics and mechanisms

Raphael F. Queiroz, Sandra M. Vaz, Ohara Augusto

Abstract

The nitroxide tempol (4-hydroxy-2,2,6,6-tetramethyl piperidine-1-oxyl) reduces tissue injury in animal models of inflammation by mechanisms that are not completely understood. MPO (myeloperoxidase), which plays a fundamental role in oxidant production by neutrophils, is an important target for anti-inflammatory action. By amplifying the oxidative potential of H2O2, MPO produces hypochlorous acid and radicals through the oxidizing intermediates MPO-I [MPO-porphyrin•+-Fe(IV)=O] and MPO-II [MPO-porphyrin-Fe(IV)=O]. Previously, we reported that tempol reacts with MPO-I and MPO-II with second-order rate constants similar to those of tyrosine. However, we noticed that tempol inhibits the chlorinating activity of MPO, in contrast with tyrosine. Thus we studied the inhibition of MPO-mediated taurine chlorination by tempol at pH 7.4 and re-determined the kinetic constants of the reactions of tempol with MPO-I (k=3.5×105 M−1·s−1) and MPO-II, the kinetics of which indicated a binding interaction (K=2.0×10−5 M; k=3.6×10−2 s−1). Also, we showed that tempol reacts extremely slowly with hypochlorous acid (k=0.29 and 0.054 M−1·s−1 at pH 5.4 and 7.4 respectively). The results demonstrated that tempol acts mostly as a reversible inhibitor of MPO by trapping it as MPO-II and the MPO-II–tempol complex, which are not within the chlorinating cycle. After turnover, a minor fraction of MPO is irreversibly inactivated, probably due to its reaction with the oxammonium cation resulting from tempol oxidation. Kinetic modelling indicated that taurine reacts with enzyme-bound hypochlorous acid. Our investigation complements a comprehensive study reported while the present study was underway [Rees, Bottle, Fairfull-Smith, Malle, Whitelock and Davies (2009) Biochem. J. 421, 79–86].

  • chlorinating activity
  • kinetics
  • myeloperoxidase
  • myeloperoxidase inhibition
  • nitroxides
  • tempol

INTRODUCTION

Nitroxides are stable free radicals that have been extensively used as biophysical tools and are receiving increased attention as potential therapeutic agents because of their pronounced antioxidant properties and low toxicity [1,2]. Under physiological conditions, nitroxides (TPNO) almost act as catalytic antioxidants because they react with diverse biological oxidants and reductants during recycling through the oxammonium cation (TPNO+) and hydroxylamine-derivative (TPNOH) respectively. After several redox cycles, nitroxides are eventually consumed by recombination reactions with specific radicals, such as tyrosyl and thiyl radicals, and/or by metabolism [36]. Tempol (4-hydroxy-2,2,6,6-tetramethyl piperidine-1-oxyl) is a highly investigated nitroxide in vivo and has been shown to effectively reduce tissue injury in animal models of inflammation [7]. The mechanisms of protection are not completely understood, but the antioxidant and anti-inflammatory actions of nitroxides are thought to be related [1,810].

An important target for anti-inflammatory compounds is the haemprotein MPO (myeloperoxidase; EC 1.11.1.7) that plays a fundamental role in oxidant production by neutrophils. MPO amplifies the oxidative potential of its co-substrate H2O2 through the oxidizing enzyme intermediates MPO-I [MPO-porphyrin•+-Fe(IV)=O] and MPO-II [MPO-porphyrin-Fe(IV)=O]. MPO-I contains two more oxidizing equivalents than the resting enzyme [MPO-porphyrin-Fe(III)] and is produced by the rapid reaction between MPO and H2O2 (eqn 1). MPO-II is produced by the reduction of MPO-I (eqn 4) and is a less potent oxidant than MPO-I (MPO-I/MPO-II, Eo=1.35 V; MPO-II/MPO, Eo=0.97 V) (reviewed in [11,12]). Embedded Image (1) Embedded Image (2) Embedded Image (3) Embedded Image (4)

Among the reactions catalysed by MPO under physiological conditions, the two-electron oxidation of chloride to hypochlorous acid (eqns 1 and 2) has received the most attention because chloride is present at high physiological concentrations and because hypochlorous acid possesses microbicidal activity. The discovery of protein tyrosine nitration in tissues and biological fluids has brought attention to the peroxidase activity of MPO, which oxidizes many biotargets by one-electron mechanisms through the intermediacy of MPO-I and MPO-II (eqns 1, 3 and 4) [11,12]. Nitrite is one of the endogenous substrates of MPO and is oxidized to nitrogen dioxide, particularly at acidic pH [1315]. The latter oxidizes and nitrates biomolecules, whereas hypochlorous acid oxidizes and chlorinates biomolecules. In addition to being an integral part of the innate immune defense system, emerging evidence has indicated that the oxidants produced by MPO (eqns 1–4) are key mediators of tissue damage associated with many inflammatory diseases such as atherosclerosis, asthma, rheumatoid arthritis, cystic fibrosis and some cancers [11,12,16,17]. Thus there is considerable interest in compounds that inhibit MPO and ameliorate the adverse effects of MPO-derived oxidants.

Nitroxides have been shown to reduce damage produced by MPO-derived oxidants in vitro and in cell systems [6,14,18]. For instance, nitroxides prevent the cellular damage promoted by MPO/H2O2/phenol in HL-60 cells [6], and tempol inhibits protein nitration mediated by MPO/H2O2/nitrite at acidic pH [14]. In the latter study, on the basis of the kinetic and spectroscopic data, we concluded that the inhibitory effect of tempol was mainly due to its rapid reaction with nitrogen dioxide to produce the oxammonium cation, which is then recycled back to tempol by reacting with H2O2 and superoxide radicals to produce oxygen and regenerate nitrite [14]. This conclusion is applicable to MPO-mediated protein nitration at acidic pH because nitrite (1 mM) reacts with MPO-I and MPO-II very rapidly under these conditions (kMPO-I=1.1×107 M−1·s−1 and kMPO-II=8.9×104 M−1·s−1) [13] and overcompetes the reaction between tempol (10 μM) and MPO-I and II [14].

However, we found that tempol also acted as a MPO inhibitor as it decreased MPO-mediated taurine chlorination [19]. This process results from the oxidation of chloride by MPO-I and the production of enzyme-bound or free hypochlorous acid (eqns 1 and 2), which chlorinates taurine. The inhibitory effect of tempol on taurine chlorination was not consistent with the second-order rate constants of the reaction between tempol and MPO-II that we previously determined [14]. Thus we studied the inhibition of MPO-mediated taurine chloramine formation by tempol at pH 7.4 and re-determined the second-order constants of the reactions of tempol with MPO-I and MPO-II. As our studies were underway, Rees et al. [18] reported that nitroxides inhibit MPO-mediated hypochlorous formation monitored by methionine and protein oxidation in vitro and in neutrophils. They proposed that MPO-II accumulation was responsible for the mechanism of tempol inhibition. In the present study, this mechanism is extended and placed into a quantitative framework by kinetic studies and computer simulations.

EXPERIMENTAL

Materials

All chemicals were purchased from Sigma–Aldrich, Merck or Fisher and were analytical grade or better. MPO from human leukocytes was obtained from Planta Natural Products (purity index A430/A280=0.85) or from Alexis Biochemicals (purity index A430/A280=0.65). Bovine erythrocyte Sod1 was purchased from Roche Applied Science and bovine liver catalase was from Boehringer. The MPO concentration was determined spectrophotometrically (ϵ430=8.9×104 M−1·cm−1 per haem) [20]. Hypochlorous acid was distilled from a commercial solution and kept at −80°C until use. Stock solutions of hypochlorous acid were prepared immediately before use, and the concentrations were determined spectrophotometrically (ϵ290=3.5×102 M−1·cm−1) [21]. The solutions of H2O2 were prepared before use, and the concentrations were determined spectrophotometrically by reaction with horseradish peroxidase to produce compound I (Δϵ403=5.5×104 M−1·cm−1) [22]. Tempol was obtained from Sigma–Aldrich and its concentration was determined spectrophotometrically (ϵ240=1.44×103 M−1·cm−1) [23]. All solutions were prepared with water purified in a Millipore Milli-Q system. All buffers were treated with Chelex-100 to remove trace amounts of metal ion contaminants prior to use.

Chlorinating activity of MPO

The reaction mixtures contained MPO (15 or 500 nM per haem), H2O2 (50 μM), chloride (100 mM), taurine (15 mM) and tempol (at the specified concentrations) in 50 mM phosphate buffer, pH 7.4. The reactions were incubated at 30 or 37°C, and the chlorinating activity was monitored by H2O2 consumption and taurine chloramine formation. H2O2 consumption was monitored amperometrically with a H2O2-selective electrode coupled to an Apollo 4000 potentiostat (World Precision Instruments). Taurine chloramine formation was quantified spectrophotometrically after reaction with 2-nitro-5-thiobenzoic acid (ϵ412=1.41×104 M−1·cm−1) [24]. The reactions were started by the addition of H2O2. Aliquots were removed at different times and the reaction was stopped by the addition of 100 μg/ml catalase.

UV–visible spectroscopy

The changes in the UV–visible spectra of MPO were monitored using a Shimadzu UV-2550 spectrophotometer.

Transient-state kinetics

The transient-state kinetic studies were performed with a stopped-flow spectrophotometer (Applied Photophysics SX-18MV) at 25.0±0.5°C using the sequential mixing mode. For the reaction of MPO-I with tempol, MPO (0.6 μM) was premixed with H2O2 (12 μM) in 50 mM phosphate buffer, pH 7.4. After a delay of 20 ms, the MPO-I formed was allowed to react with various concentrations of tempol which were in excess of at least 10-fold compared with MPO-I to ensure pseudo-first-order kinetics [2527]. The formation of MPO-II was monitored at 456 nm, which is the isosbestic point between MPO and MPO-I. In the reaction of MPO-II with tempol, MPO (0.6 μM) was premixed with H2O2 (6 μM) and homovanillic acid (0.55 μM) in 50 mM phosphate buffer, pH 7.4 [28]. After a delay time of 40 s, the MPO-II formed was allowed to react with various concentrations of tempol, which were at least 10-fold in excess of MPO-II. Alternatively, MPO (0.6 μM) was premixed with H2O2 (12 μM) [26]. After a delay of 2 s, the MPO-II formed was allowed to react with various concentrations of tempol. The conversion of MPO-II into the native enzyme was monitored at 456 nm. In all cases, the kobs values were determined using the single curve-fit equation of the instrument software. Three to six determinations of kobs were performed for each substrate concentration. The apparent second-order rate constants were calculated from the slopes using linear least-squares regression analysis.

For the reaction between tempol and MPO-II, the plot of kobs against the concentration of tempol showed saturation behaviour. Thus the data were fitted to a rectangular hyperbolic function using non-linear regression analyses to calculate the kinetic constants (see the Results section) [26].

EPR experiments

The EPR spectra were recorded at room temperature (25±2°C) on a Bruker EMX instrument. The reaction mixtures containing MPO were transferred to flat cells, and the spectra were scanned over time. The following instrumental conditions were used: microwave power, 20 mW; time constant, 327.7 ms; sweep time, 335.5 s; modulation amplitude, 1.0 G; and receiver gain, 2.36×105. To study the reaction between tempol and hypochlorous acid, stopped-flow EPR was employed. The standard cavity of the EPR instrument was replaced with a Bruker ER4117 D-MTV dielectric mixing resonator, which had a 9 mm distance between the mixing cell and the resonator centre [29]. Tempol (5 mM) and hypochlorous acid (5 mM) solutions in 50 mM acetate buffer, pH 5.4, or 50 mM phosphate buffer, pH 7.4, were transferred to 10 ml plastic syringes mounted on a syringe infusion pump (Harvard apparatus pump 22) and mixed with a flow rate of 0.6 ml/min until the flow was stopped. The magnetic field was fixed at the maximum of tempol central peak. The following instrumental conditions were used: microwave power, 20 mW; time constant, 327.7 ms; conversion time, 2621.44 ms; modulation amplitude, 1.0 G; and receiver gain, 2.36×102.

Kinetic simulations

The kinetic simulations were performed with the Gepasi 3.30 software (http://www.gepasi.org) [30].

Statistical analysis

All data are expressed as the means±S.D. Statistical significance was calculated using the Student's t test or, in the case of multiple comparisons, with one-way ANOVA with Tukey post-test using the GraphPad 4.00 software. The IC50 values were determined by fitting a rectangular hyperbola to dose–response curves using non-linear regression.

RESULTS

Inhibition of MPO-mediated taurine chlorination by tempol

The chlorinating activity of MPO (15 nM) in the presence of chloride (100 mM), taurine (15 mM) and H2O2 (50 μM) was monitored by H2O2 consumption (Figure 1A) and taurine chloramine formation (Figures 1B and 1C). The consumption of H2O2 produced taurine chloramine, as expected from one H2O2 producing one enzyme-bound or free hypochlorous acid (eqns 1 and 2), which produces one taurine chloramine in the presence of excess of taurine [24]. Tempol inhibited both H2O2 consumption and taurine chlorination in concentration- and time-dependent manners (Figure 1). These results suggested that the nitroxide inhibits the enzymatic system rather than reacting with free hypochlorous acid. Control experiments also showed that tempol (50 μM) does not inhibit taurine chlorination promoted by hypochlorous acid (50 μM) at pH 7.4 (results not shown).

Figure 1 Inhibition of MPO-mediated taurine chlorination by tempol

MPO-mediated chlorinating activity monitored by H2O2 consumption (A) and taurine chloramine formation (BD). In (A), (C) and (D), the reactions contained MPO (15 nM), H2O2 (50 μM), taurine (15 mM), sodium chloride (100 mM), DTPA (diethylenetriaminepenta-acetic acid; 0.1 mM) and the specified concentrations of tempol in phosphate buffer, pH 7.4. In (B), the reactions contained MPO (15 nM), H2O2 (50 μM), taurine (15 mM), sodium chloride (100 mM) and DTPA (0.1 mM) in phosphate buffer, pH 7.4, in the absence of tempol (○); in the presence of tempol (50 μM) (□); and in the presence of tempol (50 μM) and tyrosine (50 μM) (Δ). In (C), the% of inhibition of taurine chloramine formation by the specified tempol concentration compared with the control (absence of tempol) are plotted after 15 min (closed bars) and 30 min (open bars) of incubation. In (D), the data for the 30-min incubation in (C) were re-plotted to calculate the IC50 value, which was determined by fitting a rectangular hyperbola to the dose–response curves using non-linear regression. All incubations were performed at 37°C and the MPO used was purchased from Alexis. Taurine chloramine formation was quantified by reaction with 2-nitro-5-thiobenzoic acid as described in the Experimental section. The values shown are the means±S.D. obtained in three independent experiments; *P≤0.001 (B) (one-way ANOVA with Tukey post-test for multiple comparisons) and (C) (Student's t test).

Previously, hypochlorous acid has been shown to oxidize nitroxides to the corresponding oxammonium cations in acidic media [31,32], but the kinetics of the reaction have not been investigated. In the present study, we employed stopped-flow EPR to follow the decay of the tempol (5 mM) signal upon mixing with hypochlorous acid (5 mM) (Figure 2A). The decay of the EPR signal of tempol was slow at pH 5.4 and even slower at pH 7.4 (Figure 2A). The corresponding second-order rate constants were calculated by the initial rate approach as 0.29±0.01 and 0.054±0.002 M−1·s−1 at pH 5.4 and 7.4 respectively (Figure 2A). Even at acidic pH, tempol reacted extremely slowly with hypochlorous acid, indicating that tempol cannot act as a scavenger of the oxidant.

Figure 2 Tempol is consumed by hypochlorous acid but is marginally consumed during MPO-mediated chlorination

(A) Decay of the central peak of the EPR signal of tempol with time upon mixing tempol (5 mM) and hypochlorous acid (5 mM) solutions in acetate buffer, pH 5.4, or phosphate buffer, pH 7.4, at 25°C. The spectra shown are representative of the tempol spectrum at pH 5.4 at 0 min (left spectrum) and 15 min (right spectrum). (B) The EPR signal of tempol at different incubation times in a reaction mixture containing MPO (15 nM), H2O2 (50 μM), taurine (15 mM), NaCl (100 mM), DTPA (diethylenetriaminepenta-acetic acid; 0.1 mM) and tempol (1 μM) in phosphate buffer, pH 7.4. The reaction was performed at 37°C, and the MPO used was purchased from Alexis.

In the case of MPO-mediated taurine chlorination, tempol acted almost catalytically because its EPR spectrum maintained practically the same intensity throughout the incubation time (Figure 2B). Because the inhibitory effect of tempol was time-dependent, its IC50 value is also time-dependent. Indeed, IC50 values of 0.8, 1.8 and 3.5 μM were calculated depending on if the percentage inhibition considered is that of the initial rate of H2O2 consumption or that of the yield of taurine chloramine at 15 and 30 min respectively (see, for instance, Figure 1D).

The time-dependence of the inhibitory effect of tempol (Figure 1) suggested that the nitroxide was acting as a reversible inhibitor that removed MPO from its chlorinating cycle. This occurs when MPO is trapped as MPO-II, because the inhibitor is readily oxidized by MPO-I but reacts slowly with MPO-II [11,12]. Tyrosine reverses this kind of inhibition because it reacts quickly with MPO-II (k=1.6×104 M−1·s−1) [27] and re-establishes MPO turnover in the chlorinating cycle (eqns 1 and 2). Tyrosine abrogated the inhibitory effect of tempol (Figure 1C), which further argues that tempol removes MPO from the chlorination cycle (see also below). However, this conclusion was not consistent with the second-order rate constant that we had previously determined for the reaction of tempol with MPO-II (k=2.1×104 M−1·s−1) [14], which was comparable with that determined for tyrosine [27]. Recognizing some problems in our previous work, such as the use of low concentrations of MPO (0.2 μM) and tempol (2-10 μM) and a commercial MPO whose purity index was not the best available (A430/A280=0.65), we decided to re-evaluate the second-order rate constants of the reactions of tempol with MPO-I and MPO-II with highly purified MPO (A430/A280=0.85) and higher initial concentrations of the enzyme (0.6 μM). As these experiments were in progress, Rees et al. [18] reported that nitroxides inhibit MPO-mediated hypochlorous formation monitored by methionine oxidation and proposed that MPO-II accumulation was responsible for the inhibitory mechanism. Although we obtained similar results, revisiting the kinetics was relevant to refine the proposed mechanism.

Re-evaluation of the kinetics of the tempol reactions with MPO-I and MPO-II

Transient-state kinetic experiments were performed in a stopped-flow spectrophotometer in the sequential mixing mode as described in the Experimental section [2527]. The reaction of tempol (2–200 μM) with MPO-I, which was formed by pre-incubation of MPO (0.6 μM) with H2O2 (12 μM) for 20 ms was followed by MPO-II formation at 456 nm. Typical pseudo-first-order kinetics was observed for each tempol concentration (Figure 3, inset). Plotting the obtained kobs values against the tempol concentration provided a straight line, the slope of which provided the second-order rate constant of tempol reaction with MPO-I (k5=(3.5±0.5)×105 M−1·s−1, pH 7.4, 25°C) (Figure 3; Table 1, eqn 5). This value confirmed that tempol is a good substrate for MPO-I [14].

Figure 3 Determination of the second-order rate constant of the reaction of tempol with MPO-I

Pseudo first-order rate constants (kobs) were determined upon pre-mixing of MPO (0.6 μM) with H2O2 (12 μM) in 50 mM phosphate buffer, pH 7.4, at 25°C. After a delay of 20 ms, the MPO-I formed was allowed to react with the indicated concentrations of tempol in 0.1 M phosphate buffer, pH 7.4. The reaction time course was monitored at 456 nm, and the inset shows typical traces obtained in the absence and presence of tempol (50 μM) (black lines) and curve fit (grey line). The MPO used was purchased from Planta Natural Products. The apparent second-order rate constant was calculated from the slope using linear least-square regression analysis. The values shown are the means±S.D. obtained from three independent experiments.

View this table:
Table 1 Reactions and rate constants discussed in the text and employed in kinetic modelling

The entries in bold were used to model the experimental data with the Gepasi 3.30 software. Tau, taurine; Tau-Cl, taurine chloramines.

In the case of MPO-II, the reaction with tempol did not follow simple second-order kinetics (Figure 4A). At each tempol concentration (6–200 μM), the MPO-II decay at 456 nm showed typical pseudo-first-order kinetics after an initial lag phase (Figure 4A, inset). The slow phase is due to enzyme cycling, because the formation of MPO-II requires excess H2O2 or excess H2O2 plus homovanillic acid [28]. All of the kinetic data shown in Figure 4 were obtained with MPO-II produced from pre-incubation of MPO (0.6 μM) with H2O2 (6 μM) and homovanillic acid (0.55 μM), but essentially the same results are obtained when MPO-II was produced by pre-incubation of MPO with H2O2 (see the Experimental section). In both cases, the plot of kobs against tempol was a rectangular hyperbola (Figure 4A). This saturation behaviour indicates that a binding interaction between tempol and MPO-II is followed by enzyme reduction to MPO and tempol oxidation. This process can be described by the equations below (eqns 6 and 7), which permit calculation of the kinetic constants involved [26]. Embedded Image (5) Embedded Image (6) Embedded Image (7) Embedded Image (8) Embedded Image (9)

Figure 4 Determination of the kinetic constants of the reaction of tempol with MPO-II

(A) Pseudo first-order rate constants (kobs) were determined upon pre-mixing MPO (0.6 μM) with H2O2 (6 μM) and homovanillic acid (0.55 μM) in 50 mM phosphate buffer, pH 7.4, at 25°C. After a delay of 40 s, the MPO-II formed was allowed to react with the indicated concentrations of tempol in 0.1 M phosphate buffer, pH 7.4. The reaction time course was monitored at 456 nm, and the inset shows typical traces obtained in the absence or presence of tempol (50 μM) (black lines) and curve fit (grey line). The MPO used was purchased from Plant Natural Products. The plot is nonlinear least-squares fit of the data. (B) Re-plot of (A) in the low tempol concentration range permitted calculation of the apparent second-order rate constant from the slope. The values shown are the means±S.D. obtained from three independent experiments.

The values of k7 [(3.6±0.2)×10−2 s−1] and K [(2.04±0.1)×10−5 M] were calculated directly from eqn (8) using a nonlinear least-square fit to the data (Figure 4A). At low tempol concentrations, the plot of kobs against tempol concentration was linear (Figure 4B), and an apparent second-order rate constant value was obtained from the slope [k6=(1.00±0.03)×102 M−1·s−1]. By substituting the calculated values of k6 and K in eqn (9), the k−6 value was calculated as (2.04±0.02)×10−2 s−1.

It should be noted that in our previous study [14], the employed concentrations of tempol were in the linear portion of the plot of kobs against tempol concentration (Figure 5A), anticipating a second-order constant at approximately the value determined in the present study (~1.0×102 M−1·s−1). However, the value previously determined was approximately 10-fold higher, as was the case of the reaction between tempol and MPO-I (Figure 3) [14]. Some of the differences in these values may be attributed to errors associated with the previously employed experimental conditions, including the low concentration of a less pure MPO preparation. Other artifacts could also have contributed to the differences, but we have not been able to trace these. Nevertheless, we are confident in the values reported in the present study because they were determined several times and are consistent with the inhibitory effect of tempol on the chlorinating activity of MPO.

Figure 5 MPO intermediates and products monitored by UV–visible absorption

Variation of the VIS spectrum of MPO with time in reactions containing MPO (0.5 μM), H2O2 (50 μM), chloride (0.1 M), taurine (15 mM) and DTPA (diethylenetriaminepenta-acetic acid, 0.1 mM) in 50 mM phosphate buffer, pH 7.4, in the absence (A) and presence (B) of 50 μM tempol. (C) The same as (B) followed at 430 nm to monitor the recovery of MPO after successive additions of H2O2 (50 μM). (D) Spectrum of MPO (0.5 μM) before and 5 min after the addition of the oxammonium cation of tempo (TPNO+) (1 μM). The reactions were performed at 30°C, and the MPO used was purchased from Plant Natural Products.

Mechanism of inhibition

To determine the mechanistic details of tempol inhibition of the chlorinating activity of MPO, we performed experiments with higher concentrations of MPO to monitor the MPO intermediates during enzyme turn-over (Figures 5 and 6A). In reactions containing MPO (0.5 μM), H2O2 (50 μM), chloride (0.1 M) and taurine (15 mM), H2O2 was consumed in less than 1s in the absence of tempol and in about 7 min in the presence of 50 μM tempol (Figure 6B). Practically no changes in the UV–visible spectrum of MPO were detected in the absence of tempol because the reaction was too fast to be monitored in a normal spectrophotometer (Figure 5A). Accordingly, H2O2 was rapidly consumed to produce a quantitative yield of taurine chloramine (results not shown), whereas native MPO was recovered to a great extent (approximately 98%) (Figure 5A).

Figure 6 Kinetic modelling of the inhibition of MPO-mediated taurine chlorination by tempol

Time-dependent formation of MPO intermediates at 456 nm (A), H2O2 consumption (B), taurine chloramine formation (C) and tempol consumption (D) in reactions containing MPO (0.5 μM), H2O2 (50 μM), chloride (0.1 M), taurine (15 mM), DTPA (diethylenetriaminepenta-acetic acid, 0.1 mM) and tempol (50 μM) in 50 mM phosphate buffer, pH 7.4; tempol was absent where indicated. Black lines and symbols correspond to the experimental data, and grey lines correspond to the kinetic modelling with the Gepasi 3.30 software employing the reactions and kinetic constants shown in bold in Table 1. The incubations were performed at 30°C, and the MPO used was purchased from Planta Natural Products.

In the presence of tempol, MPO intermediates, whose UV–visible spectra are practically indistinguishable from that of MPO-II (A625/A456=0.17) [33], accumulated during steady-state and quickly decayed back to MPO upon H2O2 depletion (Figures 5B and 6A). Addition of superoxide dismutase (250 units/ml) did not alter steady-state establishment or termination (results not shown). MPO was partially recovered after turnover (approximately 82%) (Figure 5B), suggesting that part of the produced oxammonium cation reacts with MPO [31,32]. This was confirmed by successive additions of 50 μM H2O2 to spent reaction mixtures because MPO concentration decreased after each successive cycle, and the reaction became slower (Figure 5C). In the absence of tempol, MPO recycled three times with minimum haem loss (<10%) (results not shown). Further confirmation of the side reaction was obtained by showing that addition of the pre-formed oxammonium cation of tempo (1 μM) to MPO (0.5 μM) promoted changes in the UV–visible spectrum of the enzyme (Figure 5D). This result is in agreement with a previous study showing that the oxammonium cation of tempo reacts with the haem and amino acid residues of myoglobin [32]. Tempo is a tempol analogue, the oxammonium cation of which is more stable and more amenable to study than that of tempol [34]. The reactions of the oxammonium cation of tempol with MPO can also explain why tyrosine did not completely reverse the inhibitory effect of tempol (Figure 1B). However, this reaction is a marginal process in the inhibition of the chlorinating activity of MPO by tempol, becoming significant only after several catalytic cycles (Figure 5C). Thus, MPO inhibition by tempol is mostly a reversible process that results from the accumulation of MPO intermediates that generally decay back to MPO upon H2O2 consumption (Figures 5B, 5C and 6A). In agreement, the experimental data (Figure 6, black lines or symbols) were reasonably modelled by kinetic simulation with the Gepasi 3.30 software (Figure 6, grey lines) [30] considering the experimental conditions, the kinetic constants determined in the present study for the reactions of tempol with MPO-I and MPO-II, and pertinent reactions and kinetic constants available in the literature (Table 1, reactions in bold) [25,3539].

DISCUSSION

Previously, a comprehensive study by Rees et al. [18] demonstrated that a range of cyclic aliphatic and aromatic nitroxides inhibit MPO chlorinating activity monitored by methionine oxidation in a structure-dependent manner. Positively charged and neutral nitroxides were the most effective and displayed IC50 values in the low micromolar range. All of the data indicated that effective nitroxides are oxidized rapidly by MPO-I in the presence of physiological concentrations of chloride to yield MPO-II. The latter accumulated because it reacted slowly with the nitroxides. The reactivity of the nitroxides towards MPO-I was considered to be a key factor in the efficiency of nitroxides as inhibitors of the chlorinating activity of the enzyme. Tempol was included among the nitroxides studied by Rees et al. [18], and the results reported in the present study on tempol inhibition of the chlorinating activity of MPO monitored by taurine chlorination (Figures 1, 2, 5 and 6) agreed with most of their data and conclusions. The contribution of the present study was to provide kinetic data (Figures 3 and 4) to refine the inhibitory mechanism, which is important for understanding the actions of nitroxides in inflammatory environments.

The present study showed that tempol reacts with hypochlorous acid with a very low second-order rate constant even at acidic pH (k=0.29±0.01 M−1·s−1, pH=5.4) (Figure 2). Thus, nitroxides cannot act as hypochlorous acid scavengers unless they are designed to possess chemical groups that react with the oxidant, such as thiols, thioethers and amines [11]. This is different from the high reactivity of nitroxides towards other biologically relevant reactive species, such as superoxide anion, hydroxyl radical, nitrogen dioxide and carbonate radical [13,39].

The kinetics of the reactions of tempol with MPO-I and MPO-II were re-examined under optimized conditions, confirming that tempol reacts quickly with MPO-I [k=(3.5±0.5)×105 M−1·s−1] (Figure 3). The second-order rate constant we had previously determined was over-estimated, probably due to errors associated with the experimental conditions used (see Results) [14]. On the other hand, the reaction of tempol with MPO-II was found to be more complex than previously considered [14,18], displaying saturation kinetics (Figure 4). This response is unusual for MPO-II-mediated oxidations, but has been previously reported for acetaminophen [40], ascorbate and iodide [26] and attributed to a simple binding interaction between them and MPO-II. Similarly, a binding interaction between MPO-II and tempol should occur before electron transfer to produce MPO and the oxammonium cation of tempol [K=(2.04±0.1)×10−5 M; k7=(3.6±0.1)×10−2 s−1] (eqns 6 and 7). It is intriguing that saturation kinetics towards MPO-II is presented by substrates as diverse structurally and chemically as tempol (Eo=0.84 V) [41], acetaminophen (Eo=0.71 V) [42], ascorbate (Eo=0.28 V) [43] and iodide (Eo=0.44 V) [44] (see Supplementary Table S1 at http://www.BiochemJ.org/bj/439/bj4390423add.htm). Such a behaviour supports the notion that oxidation of substrates by MPO-II is constrained as much by the one-electron oxidation potential of the substrates as by other characteristics that remain incompletely understood [26,45,46].

The [MPO-II–tempol] complex suggested by kinetics could not be proved by spectrophotometry as its UV–visible spectrum is practically indistinguishable from that of MPO-II. The latter has an A625/A456 ratio of 0.17 [33], whereas that of the chlorinating incubation in the presence of tempol under steady-state conditions was approximately 0.20 (Figure 5B). Nevertheless, the [MPO-II–tempol] complex, which slowly decays to both MPO-II and tempol (k−6=2.0×10−2 s−1) and MPO and TPNO+ (k7=3.6×10−2 s−1) (Figure 4, Table 1), appears to be a crucial player in the mechanism by which tempol inhibits the chlorinating activity of MPO. This is supported by the reasonable agreement obtained between experimental (Figure 6, black lines and symbols) and modelled changes in the substrate and product concentrations with time during MPO-mediated taurine chlorination in the presence of tempol (Figure 6, Table 1, reactions in bold) [25,3539]. Likewise, the results obtained with low MPO concentration (Figure 1) were also simulated within reasonable limitations (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/439/bj4390423add.htm and also below). It is worth noting that the steady-state concentration of the MPO intermediates with an absorption maximum at 456 nm was modelled as almost equal concentrations of MPO-II and the [MPO-II–tempol] complex (Figure 6A). Modelling of the tempol concentration shows that the changes are too small to be detected experimentally (Figure 6D). However, part of the produced oxammonium cation may react with MPO [32], because the enzyme is not completely recovered after turnover (Figure 5B and 5C). In agreement, an equimolar concentration of tyrosine was not able to completely reverse the inhibitory effect of tempol (Figure 1C).

The reactions employed to model the experimental data are shown in bold type in Table 1 and should be commented upon because the production of free hypochlorous acid during MPO-mediated chlorination of small substrates, such as taurine and methionine, remains debatable [11,38,47,48]. Methionine was employed to study the inhibition of nitroxides by Rees et al. [18], whereas taurine was employed here to examine the specific effect of tempol. In both cases, the inhibitory mechanism in the presence of physiological concentrations of chloride (0.1 M) can be modelled within reasonable limitations if we assume that the substrates are chlorinated by the [MPO-I–Cl] complex (Table 1, reaction 2d; Figure 6). Other assumptions result in a chlorinating processes that would be over in less than 1 s under the experimental conditions of Figure 6 even in the presence of tempol (Supplementary Figures S2 and S3 at http://www.BiochemJ.org/bj/439/bj4390423add.htm). For instance, Supplementary Figure S2 shows the modelling obtained on the assumption that MPO-I reacts with chloride in one step to produce free hypochlorous acid (eqn 2), which then reacts with substrates (Table 1, reaction 2c) [35]. Supplementary Figure S3 shows the modelling on the assumption that the [MPO-I–Cl] complex (Table 1, reaction 2a) decays to free hypochlorous acid, which chlorinates substrates [36]. These simulations disagree completely with the experimental data reported here (Figure 6) and that reported for methionine oxidation by Rees et al. (in Figure 2 [18]). Thus the inhibitory effect of tempol on MPO-mediated chlorination of taurine (the present study) and oxidation of methionine [18] supports the idea that these substrates are modified before hypochlorous acid leaves the active site [11,38,47,48].

The modelling of the experimental data (Figure 6 and Supplementary Figure S1) was not perfect and did not change much on inclusion of other reactions of MPO intermediates, such as the reactions of MPO-I and MPO-II with superoxide anion [18,49]. Thus the differences between experimental data and modelling are probably due to uncertainties in the studies of MPO-catalysed reactions. For instance, the MPO concentration is usually expressed in haem contents, which does not take specific activity into account. This does not influence determination of second-order constants by the pseudo-first-order approach, but would influence kinetic modelling. In addition, the reaction of MPO-I with chloride depends on several factors, including the chloride concentration, which has led to differences in the kinetic constants reported by different groups [35,50]. These groups solved the discrepancy in a joint study where the chloride concentrations were varied up to 6 mM and provided the equilibrium constant used in the present study (Table 1; reaction 2a) [36] because it modelled our data better (Figure 6 and Supplementary Figure S1). Nevertheless, it is not trivial to model reported data on inhibition of the chlorinating activity of MPO [5153], while taking into account the two-step process involving k2a, k−2a and k2b (Table 1) [36] or the widely employed overall second-order rate constant of 2.5×104 M−1·s−1 (eqn 2) [35].These uncertainties are likely to be clarified in the future. At this point, the modelling of our experimental data within reasonable limitations (Figure 6 and Supplementary Figure S1) argues that the reactions shown in bold in Table 1 play a major role in the inhibitory effect of tempol.

In conclusion, our studies complemented the work of Rees et al. [18] by demonstrating that tempol does not scavenge hypochlorous acid, but acts mainly as a reversible inhibitor of MPO by trapping the enzyme as MPO-II and the [MPO-II–tempol] complex, which do not participate in the chlorinating cycle. Although this kind of inhibition may be abrogated in inflammatory environments by endogenous substrates that reduce MPO-II back to MPO, nitroxides display a range of biological activities that may preserve their inhibitory effects in vivo [17]. For instance, the superoxide dismutase mimetic activity of nitroxides has been proposed to preserve their ability to inhibit MPO-mediated oxidations in activated neutrophil cultures [18]. Thus nitroxides may act as both enzyme modulators and antioxidants in vivo [1,3,810,18]. These possibilities should be further explored in view of the therapeutic potential of nitroxides to protect against inflammatory injury.

AUTHOR CONTRIBUTION

Raphael Queiroz designed and performed the research and analysed the data. Sandra Vaz performed the initial experiments. Ohara Augusto designed the research, analysed the data and wrote the paper.

FUNDING

This work was supported by grants from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) e Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). The authors are members of the INCT de Processos Redox em Biomedicina-Redoxoma (FAPESP/CNPq/CAPES) [grant numbers 2008/57721-3 and 2008/573530].

Abbreviations: MPO, myeloperoxidase; MPO-I, MPO-porphyrin•+-Fe(IV)=O; MPO-II, MPO-porphyrin-Fe(IV)=O; tempol, 4-hydroxy-2,2,6,6-tetramethyl piperidine-1-oxyl

References

View Abstract