Trypanosoma brucei and Trypanosoma cruzi Tryparedoxin Peroxidases Catalytically Detoxify Peroxynitrite via Oxidation of Fast Reacting Thiols* □ S

Macrophage activation is one of the hallmarks observed in trypanosomiasis, and the parasites must cope with the resulting oxidative burden, which includes the production of peroxynitrite, an unusual peroxo-acid that acts as a strong oxidant and trypanocidal molecule. Cytosolic tryparedoxin peroxidase (cTXNPx) has been recently identified as essential for oxidative defense in trypanosomatids. This peroxiredoxin decomposes per-oxides using tryparedoxin (TXN) as electron donor, which in turn is reduced by dihydrotrypanothione. In this work, we studied the kinetics of the reaction of peroxynitrite with the different thiol-containing components of the cytosolic tryparedoxin peroxidase system in T. brucei ( Tb ) and T. cruzi ( Tc ), namely trypanothione, TXN, and cTXNPx. We found that whereas peroxynitrite reacted with dihydrotrypanothione and Tb TXN at mod-erate rates (7200 and 3500 M (cid:1) 1 s ance 280 as previously Alkylation of cTXNPx thiol groups by NEM was performed by incubation of the enzyme with a 10–20-fold excess of NEM for 5 min. Thiol Measurements— Low molecular weight thiols as well as protein quantitated using the DTNB assay Dihydrorhodamine Oxidation— Stock solutions of dihydrorhodamine Direct Kinetic Studies— The kinetics of peroxynitrite decomposition were studied in a stopped-flow spectrophotometer Applied Photophysics) with a mixing time of 2 ms. Although peroxynitrite decomposition is usually measured at 302 nm M 1 we monitored it at 310 nm ( (cid:3) (cid:2) M (cid:1) 1 (cid:1) 1 ) in avoid interferences by background protein absorption at 302 nm. When the initial rate approach was used (45), the peroxynitrite decomposition at 2–10 ms was fitted to a linear plot. To calculate initial rates of per- oxynitrite decomposition, the slopes were divided by the molar extinc-tion coefficient of peroxynitrite at 310 nm and multiplied by 1.2 (since the absorption at 310 nm derives from peroxynitrite anion, which rep-resents 80% total peroxynitrite concentration at pH 7.4). In pseudo- first-order analysis, apparent rate constants for peroxynitrite decomposition, k obs (s (cid:1) 1 ) values were determined by fitting stopped-flow data to single exponential decays

Trypanosoma brucei and Trypanosoma cruzi are the causative agents of African trypanosomiasis and Chagas disease, respectively, major public health problems affecting millions of people in Africa and Latin America. Both diseases are characterized by an increase in the number of macrophages and the presence of macrophage activation markers (1,2). However, T. brucei is an extracellular parasite, whereas T. cruzi proliferates inside the macrophages and in the cytoplasm of other nucleated cells. Macrophages from T. brucei-and T. cruziinfected mice produce high levels of nitric oxide ( ⅐ NO), which has antiparasitic effects in vitro and in vivo (3)(4)(5). In addition, reactive oxygen intermediates such as superoxide radical (O 2 . ) and hydrogen peroxide (H 2 O 2 ) are synthesized as a result of the oxidative burst by inflammatory cells from T. brucei-and T. cruzi-infected animals (6 -8). Superoxide can be also formed by the parasite itself (e.g. during the generation of the irontyrosyl radical center in the small subunit of ribonucleotide reductase (9), by mitochondrial respiration (10), or by redox cycling of antichagasic drugs (11)). The diffusion-controlled reaction between ⅐ NO and O 2 . leads to the formation of peroxynitrite anion (12), a strong oxidizing and cytotoxic effector molecule against T. cruzi (13,14). Moreover, inflammatory lesions in the central nervous system of mice chronically infected with T. brucei brucei and in the myocardium of acute chagasic rats express type II nitric-oxide synthase and show protein 3-nitrotyrosine immunoreactivity, which has been ascribed to peroxynitrite 1 and/or nitrogen dioxide ( ⅐ NO 2 ) formation (15)(16)(17). Peroxynitrous acid is an unusual peroxo-acid, since it has a low pK a value (6.8 versus 11.6 of the first proton dissociation in hydrogen peroxide, H 2 O 2 ) and a weak O-O bond (bond strength of 90 kJ mol Ϫ1 versus 170 kJ mol Ϫ1 H 2 O 2 ) (18) that makes it an unstable species that decomposes by homolysis (k ϭ 0.9 s Ϫ1 in phosphate buffer, pH 7.4 and 37°C) to yield hydroxyl radical ( ⅐ OH) and ⅐ NO 2 , which either recombine to form nitrate or react with substrates. The short lifetime of peroxynitrous acid and its fast reaction with carbon dioxide (CO 2 ) frequently present in buffers and in biological systems (12) makes biochemical studies more difficult to perform than with H 2 O 2 and organic hydroperoxides. Preferential targets for peroxynitrite in vivo are thiols that can be oxidized both by direct bimolecular reaction and by the reactions with peroxynitrite-derived radicals (12). The direct peroxynitrite-mediated thiol oxidation is a twoelectron oxidation process that leads to the formation of nitrite and the thiol-derived sulfenic acid, which, in the presence of an accessible thiol group, forms a disulfide, resulting in an overall stoichiometry of two thiols oxidized per peroxynitrite (19). The reaction involves peroxynitrous acid and the deprotonated form of the thiol (thiolate, RS Ϫ ). The second order rate constants for the reactions between peroxynitrous acid and low molecular weight thiols at pH 7.4 (ϳ10 2 to 10 4 M Ϫ1 s Ϫ1 ) are inversely related to the thiol pK SH (20). However, there is an increasing number of highly reactive protein thiols that react with peroxynitrite at rates of 10 5 to 10 7 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C (21,22). Among them, the bacterial peroxiredoxin alkylhydroperoxide reductase subunit C serves to catalytically detoxify peroxynitrite (23). It has been postulated that highly reactive cysteines in proteins are located close to positively charged amino acids (24) or at the positive edges of aromatic rings, which promote dissociation of the thiol (pK SH as low as Ͻ5 (23)) (i.e. the local electrostatic environment of cysteine is an important, although not necessarily the only, determinant of its reactivity (25)).
Hydroperoxide detoxification in trypanosomatids is achieved by a series of linked redox pathways that depend on the parasite-specific dithiol dihydrotrypanothione 2 (N(1),N(8)bis(glutathionyl)spermidine) for the supply of reducing equivalents. These pathways differ in subcellular location and substrate specificity (26 -32). Cytosolic tryparedoxin peroxidase (cTXNPx), 3 which is a two-cysteine peroxiredoxin, is essential for hydroperoxide detoxification in T. brucei (33) and for Leishmania chagasi survival from reactive oxygen and nitrogen species formed by infected macrophages (34). Moreover, cTXNPx is induced during H 2 O 2 challenge to T. cruzi (35). The postulated general mechanism of hydroperoxide detoxification by TbcTXNPx, in analogy to tryparedoxin peroxidases from other species (36), thioredoxin peroxidases of yeast (37), bacterial alkylhydroperoxide reductase (38), and some mammalian peroxiredoxins (39), involves an initial Cys-52 oxidation to a sulfenic acid derivative that then reacts with Cys-173 of an inversely oriented identical subunit to form an intersubunit disulfide (32). The same sequence of reactions has been described for bacterial alkylhydroperoxide reductase decomposition of peroxynitrite (23). However, this general mechanism of oxidant detoxification has been challenged recently in the case of L. chagasi cTXNPx (34). In this case, it has been postulated that whereas reactive oxygen species react with Cys-52, peroxynitrite does so with Cys-173.
Cytosolic TXNPx uses tryparedoxin (a thiol-disulfide oxidoreductase classified in the same family as the functional homologue thioredoxin (TXN) (26)) as an electron donor and reduces substrates including hydrogen peroxide and small chain organic hydroperoxides (32). Tryparedoxin is reduced by dihydrotrypanothione, which in turn is regenerated by trypanothione reductase, a homodimeric enzyme containing FAD and a reducible active site disulfide per subunit, at the expense of NADPH. In T. cruzi, exogenous or endogenous peroxynitrite leads to dihydrotrypanothione oxidation to the corresponding disulfide, and it has been postulated that the trypanothione-dependent antioxidant system against peroxynitrite may facilitate the survival of trypanosomes within the oxidative environment of activated macrophages (14).
However, the biochemical mechanism of dihydrotrypanothione oxidation and peroxynitrite detoxification at the cellular level remains unknown. Indeed, since trypanothione reductase, dihydrotrypanothione, tryparedoxin, and tryparedoxin peroxidase are all thiol-containing molecules, they could all in theory react with peroxynitrite. However, the relative importance of these biotargets as reactants of peroxynitrite in vivo would be, in good part, dictated by rate constants and reactant concentrations.
Herein, we present kinetic studies on the reactivity of different components of the tryparedoxin peroxidase antioxidant system in trypanosomes, namely trypanothione, TbTXN, TbcTXNPx, and TccTXNPx with peroxynitrite, with the aim to rationalize the biochemical mechanisms of peroxynitrite detoxification in T. brucei and T. cruzi.
Expression and Purification of T. brucei Tryparedoxin and T. brucei and T. cruzi Cytosolic Tryparedoxin Peroxidases-Proteins were obtained by heterologous expression of the respective genes in Escherichia coli. The gene of TXN of T. brucei brucei, as identified by Lü demann et al. (40), was expressed as an N-terminally His-tagged protein that was purified as described previously (32). Molecular mutants of TbTXN were obtained according to Ref. 32. T. brucei brucei cytosolic TXNPx and mutated forms of the enzyme were also prepared as N-terminally His-tagged proteins as described previously (32). T. cruzi cytosolic TX-NPx was purified as in Ref. 41 with the following modifications. The amplified T. cruzi TXNPx gene was cloned into the pQE-30 vector (Qiagen) between SacI and HindIII. pQE30-TcH6TXNPx in E. coli M15 cells was grown at 30°C with vigorous aeration in LB broth containing 100 g/ml ampicillin and 25 g/ml kanamycin. Expression of recombinant TcH6TXNPx was induced with 0.5 mM isopropyl-␤-D-thiogalactopyranoside when the culture reached A 600 ϭ 0.6. The purification was performed in a 5-ml HiTrap affinity column (Amersham Biosciences) charged with Ni 2ϩ and equilibrated with binding buffer (50 mM sodium phosphate, pH 7.6, containing 10 mM imidazole, 500 mM NaCl) at a flow rate of 3 ml/min. The His-tagged TXNPx was eluted in 50 mM sodium phosphate, pH 7.6, containing 300 mM imidazole, 500 mM NaCl.
Peroxynitrite Synthesis-Peroxynitrite was synthesized in a quenched flow reactor from sodium nitrite and H 2 O 2 under acidic conditions and quantitated as described previously (19). Treating a stock solution of peroxynitrite with granular manganese dioxide eliminated H 2 O 2 remaining from the synthesis. Nitrite (NO 2 Ϫ ) present in samples of peroxynitrite decomposed at acidic pH was typically Ͻ30% of peroxynitrite concentration.
Trypanothione Reduction-Stock solutions of trypanothione disulfide (2.6 mM) were treated with excess sodium borohydride during 30 min and adjusted to pH 2 in order to eliminate excess borohydride. Then samples were adjusted to pH 7.4 and extensively bubbled with argon and stored on ice in the dark. The amount of reduced trypanothione was measured by quantitation of reduced thiol groups, using the DTNB assay (42).
Protein Thiol Reduction and Alkylation-TbTXN, TbcTXNPx, and TccTXNPx were reduced overnight by the addition of a Ͼ10-fold excess of dithiothreitol. Excess reductant was removed immediately before use by passing proteins through a high pressure liquid chromatographyconnected Hitrap column (Amersham Biosciences) with a UV-visible detector at 280 nm and collected manually in rubber-capped tubes. The elution buffer was 100 mM potassium phosphate, pH 7, plus 100 M DTPA, which was extensively degassed before use. Once collected, samples were bubbled for 5 min with argon at 4°C. Protein concentration was measured by the Bradford method, as well as by their absorb-ance at 280 nm as previously described (31). Alkylation of cTXNPx thiol groups by NEM was performed by incubation of the enzyme with a 10 -20-fold excess of NEM for 5 min.
Thiol Measurements-Low molecular weight thiols as well as protein thiols were quantitated using the DTNB assay (42).
Direct Kinetic Studies-The kinetics of peroxynitrite decomposition were studied in a stopped-flow spectrophotometer (SF17MV; Applied Photophysics) with a mixing time of Ͻ2 ms. Although peroxynitrite decomposition is usually measured at 302 nm (⑀ 302 ϭ 1670 M Ϫ1 cm Ϫ1 ) (44), we monitored it at 310 nm (⑀ 300 ϭ 1600 M Ϫ1 cm Ϫ1 ) in order to avoid interferences by background protein absorption at 302 nm. When the initial rate approach was used (45), the peroxynitrite decomposition at 2-10 ms was fitted to a linear plot. To calculate initial rates of peroxynitrite decomposition, the slopes were divided by the molar extinction coefficient of peroxynitrite at 310 nm and multiplied by 1.2 (since the absorption at 310 nm derives from peroxynitrite anion, which represents 80% total peroxynitrite concentration at pH 7.4). In pseudofirst-order analysis, apparent rate constants for peroxynitrite decomposition, k obs (s Ϫ1 ) values were determined by fitting stopped-flow data to single exponential decays with floating end point. Reported values are the average of at least seven separate determinations. Temperature was maintained at 37 Ϯ 0.1°C, and the pH was measured at the outlet.
Competition Kinetic Studies-Mn 3ϩ porphyrins are rapidly oxidized by peroxynitrite (k ϭ 10 5 to 10 7 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C) to the OϭMn 4ϩ derivative, in a process that can be conveniently monitored at the Soret band as the decay of absorbance at 462 nm (46). In the case of Mn 3ϩ -TM-4-PyP, the second order rate constant for its reaction with peroxynitrite was determined previously as 3.7 ϫ 10 6 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C (46). The effect of increasing concentrations of TbcTXNPX on peroxynitrite-mediated Mn 3ϩ -TM-4-PyP oxidation was determined by stopped flow (47). Experiments were performed at 37°C, and the pH was measured at the outlet. Kinetics of peroxynitrite-dependent TbcTX-NPx oxidation in the presence of Mn 3ϩ -TM-4-PyP was estimated by computer-assisted simulation, varying the apparent second order rate constant for the reaction of the enzyme and peroxynitrite at pH 7.4 and 37°C so as to get the best fit to experimental data. In this system, peroxynitrite can (a) decompose to nitrate after proton-catalyzed isomerization (k ϭ 0.9 s Ϫ1 ) (12) (Reaction 1), (b) react with the reduced enzyme (Reaction 2), or (c) react with Mn 3ϩ -TM-4-PyP (Reaction 3).
Alternatively, the kinetics of the competition reaction was studied using a pseudo-first-order approach (48), following the effect of increasing concentrations of the enzyme on the apparent first-order rate constants for peroxynitrite reduction as measured through the oxidation of Mn 3ϩ -TM-4-PyP at 462 nm. Although TbcTXNPx concentrations used were not much greater than peroxynitrite concentrations, computer-assisted simulations indicated that reduced enzyme concentrations would change minimally (ϳ10%) over the time course of the experiment under these conditions. Computer-assisted Simulations-Computer-assisted simulations were performed using the Gepasi program (49).
Statistics-All experiments reported here were repeated and reproduced at different days. Results are expressed as mean values with the corresponding S.D. values. Graphics and data analysis were performed using Sigma Plot.

RESULTS
The Reaction of Peroxynitrite with Trypanothione and T. brucei Tryparedoxin-The addition of increasing excess dihydrotrypanothione concentrations to peroxynitrite led to an increase of the exponential decay of peroxynitrite at pH 7.4 and 37°C. Dihydrotrypanothione reacted with peroxynitrite with an apparent (pH-dependent) second order rate constant of 7200 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C (Fig. 1). In contrast, similar concentrations of trypanothione disulfide did not lead to any increase on the rate of peroxynitrite decomposition, indicating that the thiol groups of the molecule are the responsible for the reactivity (Table I).
Up to 150 M reduced T. brucei tryparedoxin caused only a modest effect on peroxynitrite (24 M) decomposition rate, from 0.89 Ϯ 0.04 to 1.4 Ϯ 0.05 s Ϫ1 , indicating that the second order rate constant between peroxynitrite and the reduced protein is ϳ3500 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C (Table I).
Kinetics of the Reaction of T. brucei Cytosolic Tryparedoxin Peroxidase with Peroxynitrite-Reduced TbcTXNPx increased the rate of peroxynitrite decomposition (Fig. 2). Using an initial rate approach, an apparent second order rate constant of 9 Ϯ 1 ϫ 10 5 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C was determined. The increase in peroxynitrite decomposition rates was abolished by pretreatment of reduced TbcTXNPx with excess NEM, thus pointing to thiol groups as responsible for the fast reactivity of peroxynitrite with the enzyme (data not shown). However, there are six cysteine residues in the TbcTXNPx primary sequence, and at least three of them can be easily titrated with DTNB (see Fig. 5, y axis intercept). In order to unambiguously identify the cysteine residue responsible for the fast reaction between TbcTXNPx and peroxynitrite, we performed kinetic studies using site-directed TbcTXNPx mutated proteins in which Cys-52 or Cys-173 was replaced by serine (C52S or C173S, respectively). C173S TbcTXNPx continued to react fast with peroxynitrite, with an apparent second order rate constant of 3.3 Ϯ 0.4 ϫ 10 5 M Ϫ1 s Ϫ1 , whereas C52S TbcTXNPx had only a marginal effect on the initial rate of peroxynitrite decomposition (Fig. 2), and pseudo-first-order analysis yielded a second order rate constant of 1 ϫ 10 4 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C (Table I).
TbcTXNPx Catalytically Detoxifies Peroxynitrite-In order to test whether peroxynitrite oxidizes TbcTXNPx to a form that can be re-reduced by its natural redundant TbTXN, we performed stopped-flow experiments following peroxynitrite decomposition in the presence of low TbcTXNPx concentrations in the absence or in the presence of reduced TbTXN. Steady-state studies have indicated that tert-butyl hydroperoxide (t-BuOOH)-oxidized TbcTXNPx is reduced by reduced TbTXN with a net forward second order rate constant of 1.1 ϫ 10 5 M Ϫ1 s Ϫ1 at pH 7.6 and 25°C (32). If peroxynitrite oxidizes the enzyme to the same intermediate as t-BuOOH does, it can be expected that the same net second order rate constant will apply for the re-reduction of peroxynitrite-oxidized TbcTXNPx by TbTXN. Then, by computer-assisted simulation studies, we calculated a concentration of TbTXN (Ն70 M) that would allow us to maintain reduced TbcTXNPx (1.5-6 M) concentrations relatively unchanged (ϳ30% change) in the presence of peroxynitrite (18 M) (see Supplemental Material). As has already been mentioned, peroxynitrite does not react very rapidly with reduced TbTXN. TbTXN alone caused only a slight increase in peroxynitrite decomposition rate, whereas the addition of peroxynitrite (18 M) to reduced TbcTXNPx (18 M) produced a rapid initial decrease on peroxynitrite concentration, followed by a slower second phase of peroxynitrite decomposition, that reflected consumption of the enzyme. However, in the presence of both reduced TbcTXNPx (1.5-6 M) and reduced TbTXN (70 M), peroxynitrite had basically an exponential decay that was faster at higher TbcTXNPx concentrations tested (Fig. 3). From the slope of the plot of the apparent first-order rate constants of peroxynitrite decay at each TbcTXNPx concentration (in the presence of TbTXN) versus TbcTXNPx concentration, an apparent second order rate constant for the reaction between peroxynitrite and reduced TbcTXNPx of 7 Ϯ 1 ϫ 10 5 M Ϫ1 s Ϫ1 at pH 7.4 and 37°C was obtained (Fig. 3, inset). This value is very similar to the value obtained by the initial rate approach (9 Ϯ 1 ϫ 10 5 M Ϫ1 s Ϫ1 ; Fig. 2) for the same reaction. In contrast, both C40S TbTXN and C43S TbTXN were unable to sustain a catalytic TbcTXNPx-mediated peroxynitrite decomposition (data not shown).
Kinetics of the Reaction of T. cruzi Cytosolic Tryparedoxin Peroxidase with Peroxynitrite-The peroxynitrite (18 M) decomposition rate was accelerated by reduced TccTXNPx (35-70 M) but not by the NEM-pretreated enzyme. An initial rate approach indicated a second order rate constant of 7.2 Ϯ 0.3 ϫ 10 5 M Ϫ1 s Ϫ1 for the reaction between peroxynitrite and reduced TccTXNPx at pH 7.4 and 37°C (Table I).
Peroxynitrite decomposition in the presence of a heterologous system (41) formed by reduced TccTXNPx (0 -9 M) and reduced TbTXN (70 M) had an exponential behavior, indicating that peroxynitrite-oxidized TccTXNPx could be re-reduced by the heterologous TbTXN. The second order rate constant for the reaction between peroxynitrite and TccTXNPx obtained with the enzyme in turnover was 8.5 Ϯ 1 ϫ 10 5 M Ϫ1 s Ϫ1 , at pH 7.4 and 37°C (Fig. 4), consistent with the value obtained by the initial rate approach.
Peroxynitrite-dependent Oxidation of TbcTXNPx Thiol Groups-The addition of increasing concentrations of peroxynitrite to reduced TbcTXNPx (130 M) led to an oxidation of the enzyme thiol groups. The slope of the curve was ϳ2, indicating  that each peroxynitrite molecule led to the oxidation of two thiol groups of the enzyme (Fig. 5), in agreement with the stoichiometry of the reaction established previously for direct peroxynitrite-mediated oxidation of thiols (19). The addition of predecomposed peroxynitrite (60 M) did not lead to any thiol oxidation in the enzyme (Fig. 5). Even in the presence of physiological concentrations of CO 2 (1.3 mM), there was still significant enzyme thiol oxidation, although it was less than in the absence of CO 2 (Fig. 5).

Inhibition of Peroxynitrite-mediated Mn 3ϩ -TM-4-PyP Oxidation by TbcTXNPx-The fast reaction between peroxynitrite and reduced
TbcTXNPx indicates that the enzyme should inhibit other direct oxidations performed by peroxynitrite. However, peroxynitrite-dependent oxidation of thiols leads to their sulfenic acid derivatives, which are not inert products but typically unstable and reactive, and could promote further oxidations. Therefore, we performed competition experiments in order to determine whether TbcTXNPx could protect other targets from peroxynitrite-dependent oxidation. We chose Mn 3ϩ -TM-4-PyP as a target, since it reacts directly and rapidly with peroxynitrite and can be conveniently followed spectrophotometrically. Thus, the effect of reduced TbcTXNPx on peroxynitrite-mediated Mn 3ϩ -TM-4-PyP oxidation was evaluated. At increasing concentrations of TbcTXNPx, the maximum of Mn 3ϩ -TM-4-PyP oxidation (at 100 -200 ms) decreased, and the apparent first-order rate of peroxynitrite reduction increased (Fig. 6, a-d). Computer-assisted simulations allowed an estimation of the second order rate constant between peroxynitrite and the enzyme as ϳ2 ϫ 10 6 M Ϫ1 s Ϫ1 (Fig. 6). The OϭMn 4ϩ -TM-4-PyP is an unstable species that is readily reduced back to Mn 3ϩ state, the rate of reduction increasing with increasing TbcTXNPx concentration, and significant decay of the OϭMn 4ϩ complex took place over the time scale of its formation (Fig. 6) (i.e. see differences between experimental and simulated data at time Ͼ 100 ms). The observed first-order rate constant k obs for the reduction of peroxynitrite (Reaction 4) is the sum of two contributions (48), one arising from the reaction of peroxynitrite with Mn 3ϩ -TM-4-PyP, which is the y intercept in Fig. 6, inset, and the acceleration in rate arising from the reaction of peroxynitrite with TbTXNPx (slope in Fig. 6, inset). k obs ϭ k͑Mn 3ϩ -TM-4-PyP ϩ ONOO Ϫ ͓͒Mn 3ϩ -TM-4-PyP͔ ϩ k͑TbcTXNPx ϩ ONOO Ϫ ͓͒TbcTXNPx͔ REACTION 4 Linear regression (Fig. 6, inset)  ( 50,51), which react at a diffusion-controlled rate with each other to produce peroxynitrite (52). Peroxynitrite-mediated dihydrorhodamine (DHR) oxidation to rhodamine occurs via radical pathways with ϳ40% yields and has been extensively used to demonstrate and quantitate peroxynitrite formation in biochemical or cellular systems. Exposure of DHR (100 M) to a SIN-1 (0.5 mM)-derived flux of peroxynitrite at pH 7.4 and 25°C led to 0.26 M/min DHR oxidation (Fig. 7, line a), which corresponds to a flux of ϳ0.5 M/min peroxynitrite. The addition of increasing reduced TbcTXNPx (2.5-10 M) resulted in dose-dependent lag phases on peroxynitrite-mediated DHR oxidation, which were subsequently resumed to the same 0.26 M/min DHR oxidation rate (Fig. 7, lines b-e). When the exposure of DHR to SIN-1 was performed in the presence of both reduced TbTXN (6 M) and reduced TbcTXNPx (2.5 M) (Fig. 7, line f), the lag phase was even longer than with either reduced TbcTXNPx (2.5 M) (Fig. 7, line b) or reduced TbTXN (6 M) alone (data not shown), indicating that peroxynitrite-oxidized TbcTXNPx was efficiently re-reduced by TbTXN.

DISCUSSION
Peroxynitrite reacts directly with all of the thiol-containing components of the T. brucei cytosolic tryparedoxin peroxidase system tested (i.e. dihydrotrypanothione, TXN, and cTXNPx) (Figs. 1 and 2, Table I). However, the second order rate constant for the reaction of peroxynitrite and reduced TbcTXNPx is ϳ100 times faster than with dihydrotrypanothione or reduced TbTXN at physiological pH. Cys-52 was identified as the initial peroxynitrite target in the enzyme (Fig. 2), as it is for the reaction with H 2 O 2 and other hydroperoxides (32). The reason for its fast reactivity has been ascribed to its activation by a catalytic triad composed of Cys-52 that is hydrogen-bonded to the hydroxyl group of Thr-49 and electrostatically activated by Arg-128 (24,36), which means that Cys-52 in tryparedoxin peroxidases is most probably deprotonated at physiological pH. Our observations contrast with a recent report on L. chagasi cTXNPx I, which proposed that in this enzyme the Cys-52 residue is essential for detoxifying H 2 O 2 , whereas the Cys-173 residue is essential for peroxynitrite reduction (34). Given the homology between L. chagasi cTXNPx I and TbcTXNPx (53) and the unique redox properties of Cys-52, this apparent discrepancy is intriguing; therefore, the reaction of peroxynitrite with peroxiredoxins of the Leishmania species deserves further investigation.
The addition of peroxynitrite to reduced TbcTXNPx lead to the oxidation of its thiol groups (Fig. 5). Under conditions of excess enzyme, the stoichiometry of the reaction was two thiols oxidized by each peroxynitrite added, consistent with the reported stoichiometry for direct thiol oxidation, which involves a two-electron oxidation of the thiol to its sulfenic derivative, an intermediate detected in the case of peroxynitrite-mediated oxidation of the single thiol group of human serum albumin (54). Peroxynitrite-derived radicals CO 3 . and ⅐ NO 2 oxidize thiols to their corresponding thiyl radicals, which could lead to inactive forms of the enzyme either through the reaction with oxygen and thiyl peroxyl radical formation (RSOO ⅐ ) and a sulfinic acid-containing end product or, in the presence of an accessible thiol group, to a mixed disulfide through the formation of a disulfide radical anion (RSSR៛). However, under the conditions of our experiments, kinetic analysis indicates that, at least initially, one-third of peroxynitrite would be reacting with CO 2 and the other 70% with TbcTXNPx, 4 and therefore the thiol consumption observed in Fig. 5, even in the presence of physiological CO 2 concentrations, is mainly dependent on the direct reactions between the enzyme and peroxynitrite. Tryparedoxin peroxidases are highly abundant enzymes in trypanosomatids, constituting up to 5% of total soluble protein (26,36) and representing about 0.5-1.0 mM active site thiol, which makes this an even more physiological relevant target for peroxynitrite reactivity inside the parasite. If peroxynitrite oxidizes Cys-52 in the enzyme to the same sulfenic acid derivative implicated in hydroperoxide detoxification, it is very likely that the rest of the catalytic process (i.e. the formation of the postulated intersubunit disulfide with Cys-173 and the reduction of the oxidized enzyme by TbTXN) functions in a way similar to that described for other hydroperoxides. This was indeed the case, since although TbTXN did not react very rapidly with peroxynitrite (Table I) and, at the concentration tested in Fig. 3, had only a marginal effect on peroxynitrite decomposition rate, the presence of both TbcTXNPx and TbTXN led to an increase in the rate of exponential peroxynitrite decay, indicating that TbcTXNPx concentrations were maintained relatively constant during the time course of the experiment when TbTXN was also present ( Fig. 3 and Supplemental Mate- 4 The percentage of peroxynitrite that would react with each target is calculated dividing the pseudo-first-order rate constant of peroxynitrite decay in the presence of the target (k CO2 or k TbcTXNPx ), by the overall pseudo-first-order rate constant of peroxynitrite decay in the presence of both targets (k TbTXNPx ϩ k CO2 ) (i.e. k CO2 ϭ k 2 (ONOO Ϫ ϩ CO 2 ) 12 ϫ [CO 2 ] ϭ 4.3 ϫ 10 4 M Ϫ1 s Ϫ1 ϫ 1.3 ϫ 10 Ϫ3 M ϭ 56 s Ϫ1 ; k TbcTXNPx ϭ k 2 (ONOO Ϫ ϩ TbcTXNPx) ϫ [TbcTXNPx] ϭ 8 ϫ 10 5 M Ϫ1 s Ϫ1 ϫ 150 ϫ 10 Ϫ6 M ϭ 120 s Ϫ1 . SCHEME 1. Mechanism of catalytic detoxification of peroxynitrite by cTXNPx.
rial, Fig. 1S). Therefore, TbcTXNPx exposed to peroxynitrite is oxidized to a form that is readily reduced by its natural reductant TbTXN (i.e. TbcTXNPx acts as a tryparedoxin:peroxynitrite oxidoreductase) (Scheme I). The results obtained during DHR oxidation by a SIN-1-derived flux of peroxynitrite, where the lag phase of DHR oxidation was much longer in the presence of both TbcTXNPx and TbTXN than in the presence of TbcTXNPx alone (Fig. 7) confirms the catalytic activity of TbcTXNPx under steady-state conditions. The results obtained using TccTXNPx revealed that this enzyme also decomposes peroxynitrite in a catalytic manner (Fig. 4), and the second order rate constant for the reaction between the enzyme and peroxynitrite was similarly fast (Table I). Moreover, TbcTXNPx had an inhibitory effect on direct (Fig. 6) and radical-dependent (Fig. 7) peroxynitrite-mediated oxidations.
In the case of T. cruzi, macrophage-derived O 2 . and ⅐ NO generate peroxynitrite inside the phagosome, minimizing diffusional restrictions for enacting target molecule reactions inside the parasite during the invasion process. In the case of T. brucei, which is an extracellular parasite, macrophage-derived peroxynitrite is expected to be consumed to some extent in the extracellular space. However, O 2 . is also formed inside T. brucei (e.g. by the tyrosyl radical formed during ribonucleotide reductase turnover), in particular in the proliferative stages of the parasite. Thus, macrophage-derived ⅐ NO, which is a long lived free radical, could reach T. brucei, react with endogenous O 2 .
and form peroxynitrite intracellularly. Therefore, the tryparedoxin:peroxynitrite oxidoreductase activity of T. brucei and T. cruzi cTXNPx reported herein supports its role as an important factor for the survival and proliferation of trypanosomatids in the presence of activated macrophages.