Antiplasmodial Activity of Nitroaromatic Compounds: Correlation with Their Reduction Potential and Inhibitory Action on Plasmodium falciparum Glutathione Reductase

With the aim to clarify the mechanism(s) of action of nitroaromatic compounds against the malaria parasite Plasmodium falciparum, we examined the single-electron reduction by P. falciparum ferredoxin:NADP+ oxidoreductase (PfFNR) of a series of nitrofurans and nitrobenzenes (n = 23), and their ability to inhibit P. falciparum glutathione reductase (PfGR). The reactivity of nitroaromatics in PfFNR-catalyzed reactions increased with their single-electron reduction midpoint potential (E17). Nitroaromatic compounds acted as non- or uncompetitive inhibitors towards PfGR with respect to NADPH and glutathione substrates. Using multiparameter regression analysis, we found that the in vitro activity of these compounds against P. falciparum strain FcB1 increased with their E17 values, octanol/water distribution coefficients at pH 7.0 (log D), and their activity as PfGR inhibitors. Our data demonstrate that both factors, the ease of reductive activation and the inhibition of PfGR, are important in the antiplasmodial in vitro activity of nitroaromatics. To the best of our knowledge, this is the first quantitative demonstration of this kind of relationship. No correlation between antiplasmodial activity and ability to inhibit human erythrocyte GR was detected in tested nitroaromatics. Our data suggest that the efficacy of prooxidant antiparasitic agents may be achieved through their combined action, namely inhibition of antioxidant NADPH:disulfide reductases, and the rapid reduction by single-electron transferring dehydrogenases-electrontransferases.


Introduction
The emergence of the resistance of the malaria parasite Plasmodium falciparum to available drugs (e.g., chloroquine or artemisinin [1]) has resulted in the demand for new antimalarial agents and in a better understanding of their mechanisms of action. P. falciparum is particularly vulnerable to oxidative stress, that is, to enhanced generation of reactive oxygen species (ROS), which may be caused by the absence of the antioxidant enzymes catalase and glutathione peroxidase [2].
The antibacterial and antiparasitic activity of nitroaromatic compounds (ArNO 2 ) is well known. In addition to a number of nitroheterocyclic drugs such as nifurtimox and benznidazole that have been used against Chagas disease and sleeping sickness since the 1970s, a new 5-nitroimidazole derivative, fexinidazole, has recently been approved for a treatment against sleeping sickness [3]. Frequently, the therapeutic action of ArNO 2 is attributed to single-electron reduction into their anion radicals (ArNO 2 − ), which in turn undergo redox cycling with the formation of ROS, or to their two/four-electron reduction into hydroxylamines (ArNHOH), able to modify DNA [4][5][6]. The single-electron reduction of ArNO 2 is commonly performed by flavoenzymes dehydrogenases-electrontransferases, which possess natural single-electron acceptors, such as heme-or FeS-proteins [7][8][9][10]. However, there is a relative lack of information about the enzymes responsible for these reactions in parasites. Another point of view is that in trypanosomatids and Leishmania spp., a possible mode of ArNO 2 action is the inhibition of the antioxidant flavoenzyme trypanothione reductase (TR) [11][12][13][14][15][16]. In this case, nitroaromatics also undergo TR-catalyzed redox cycling. In schistosomatids, a possible target of ArNO 2 and other aromatic electron-deficient compounds is thioredoxin glutathione reductase [17,18]. A number of nitrofurans, nitrobenzenes, nitroimidazoles, and 4-nitrobenzothiadiazole were shown to possess in vitro antiplasmodial activity at micromolar or lower concentrations [19][20][21][22][23]; however, the mechanisms of their action remain poorly understood. The activity of a series of nitrobenzenes and nitrofurans roughly increased with their single-electron reduction midpoint potential (redox potential of ArNO 2 /ArNO 2 − couple, E 1 7 ) [19], thus demonstrating a possible relationship between the compound's ease of reductive activation and antiplasmodial activity. The antimalarial activity of nitrothiophenes was also attributed to the formation of ROS [24]. On the other hand, nitroaromatic compounds inhibit antioxidant flavoenzyme glutathione reductase from various sources [19,23,25,26]. Since P. falciparum glutathione reductase (Pf GR) plays a key role in the antioxidant defense of the parasite [2,27,28], it is believed that its inhibitors may act as efficient antiplasmodial agents. Pf GR is a 2 × 55 kD homodimer containing FAD and catalytic disulfide in each subunit, which catalyzes the reduction of glutathione (GSSG) at the expense of NADPH [29]. Human erythrocyte host glutathione reductase (HGR) possesses 45% amino acid sequence identity with Pf GR and its role in the parasite survival is a matter of debate. Both Pf GR and HGR are inhibited by aromatic electron-deficient compounds which were observed to bind at the dimer interface [27,28]. Our previous study demonstrated the absence of relationship between the antiplasmodial activity of nitroaromatic compounds and their efficacy as HGR inhibitors [19]. However, the relationship between the Pf GR inhibition and the antiplasmodial activity of nitroaromatics has not been studied so far.
Extending our previous studies [19,30], here we demonstrate that the in vitro antiplasmodial activity of nitroaromatic compounds partly correlates with their efficacy as Pf GR inhibitors, and partly with their reactivity with single-electron transferring P. falciparum ferredoxin:NADP + oxidoreductase (Pf FNR). Given the data currently available, Pf FNR may act as the most efficient generator of ArNO 2 free radicals in Plasmodium.

Relationship between Antiplasmodial Activity of Nitroaromatic Compounds and Their Single-Electron Reduction Midpoint Potential
The toxicity of nitroaromatic compounds against mammalian cells and bacteria often increases with their single-electron reduction midpoint potential (redox potential of ArNO 2 /ArNO 2 − couple, E 1 7 ). The relationship ∆log IC 50 /∆E 1 7~− 10 V −1 , where IC 50 is the compound concentration for 50% cell survival or, in the case of bacteria or parasites, for 50% growth inhibition, indicates that the main factor of cytotoxicity is redox cycling and oxidative stress [4,8,31]. Indeed, the rates of single-electron reduction of ArNO 2 by flavoenzymes dehydrogenases-electrontransferases, such as NADPH:cytochrome P-450 Molecules 2019, 24, 4509 3 of 15 reductase, ferredoxin:NADP + oxidoreductase, and NO-synthase that initiate their redox cycling, increase with E 1 7 of oxidants, and are relatively insensitive to their structure [7][8][9][10]. In this work, we used a series of nitroaromatic compounds ( Figure 1, Table 1) with available E 1 7 values. For a major part of them, the IC 50 values against the chloroquine-resistant P. falciparum strain FcB1 and the inhibition efficacy against HGR were characterized in a previous work [19]. Among the examined compounds, the representatives of vinylquinoline-substituted nitrofurans (IIIa-IIIh, Figure 1) possess well-promising diverse properties such as inhibition of trypanothione reductase, that is, the potential trypanocidal activity [12] as well as bactericidal and antitumor in vitro activity [32,33], the latter property gaining increasing interest [34]. Nitrobenzenes, nitrofurantoin, and nifuroxime (compounds 1-9,12,14,23, Table 1) were used as model compounds. Table 1 reports the IC 50 values of compounds against P. falciparum strain FcB1, their E 1 7 values, and their calculated octanol/water distribution coefficients (log D). Importantly, the IC 50 values for several nitroaromatic compounds obtained in separate studies were sufficiently close (Table 1). For quantitative analysis, when available, data obtained within the current work were used. The regression analysis of the log IC 50 versus E 1 7 relationship yielded a ratio ∆log IC 50 /∆E 1 7 = −8.37 ± 1.25 V −1 (r 2 = 0.6802) ( Figure S1, Supplementary Materials). The dependence of log IC 50 on log D is poorly expressed (r 2 = 0.2913, Figure S2 Supplementary Materials). However, an introduction of compound log D as a second independent variable resulted in some improvement of the correlation: log IC 50 = −(0.65 ± 0.44) -(7.39 ± 1.34) E 1 7 -(0.12 ± 0.07) log D, (r 2 = 0.7193) (1) survival or, in the case of bacteria or parasites, for 50% growth inhibition, indicates that the main factor of cytotoxicity is redox cycling and oxidative stress [4,8,31]. Indeed, the rates of single-electron reduction of ArNO2 by flavoenzymes dehydrogenases-electrontransferases, such as NADPH:cytochrome P-450 reductase, ferredoxin:NADP + oxidoreductase, and NO-synthase that initiate their redox cycling, increase with E 1 7 of oxidants, and are relatively insensitive to their structure [7][8][9][10].
In this work, we used a series of nitroaromatic compounds ( Figure 1, Table 1) with available E 1 7 values. For a major part of them, the IC50 values against the chloroquine-resistant P. falciparum strain FcB1 and the inhibition efficacy against HGR were characterized in a previous work [19]. Among the examined compounds, the representatives of vinylquinoline-substituted nitrofurans (IIIa-IIIh, Figure  1) possess well-promising diverse properties such as inhibition of trypanothione reductase, that is, the potential trypanocidal activity [12] as well as bactericidal and antitumor in vitro activity [32,33], the latter property gaining increasing interest [34]. Nitrobenzenes, nitrofurantoin, and nifuroxime (compounds 1-9,12,14,23, Table 1) were used as model compounds. Table 1 reports the IC50 values of compounds against P. falciparum strain FcB1, their E 1 7 values, and their calculated octanol/water distribution coefficients (log D). Importantly, the IC50 values for several nitroaromatic compounds obtained in separate studies were sufficiently close ( Table 1). For quantitative analysis, when available, data obtained within the current work were used. The regression analysis of the log IC50 versus E 1 7 relationship yielded a ratio ∆log IC50/∆E 1 7 = −8.37 ± 1.25 V −1 (r 2 = 0.6802) ( Figure S1, Supplementary Materials). The dependence of log IC50 on log D is poorly expressed (r 2 = 0.2913, Figure This shows that the oxidant potency of nitroaromatics and, to some extent, their lipophilicity play definite roles in their antiplasmodial activity.  Table 1. Single-electron reduction midpoint potentials (E 1 7) of nitroaromatic compounds, their concentrations for 50% Plasmodium falciparum growth inhibition (IC50), their calculated octanol/water  Table 1. Single-electron reduction midpoint potentials (E 1 7 ) of nitroaromatic compounds, their concentrations for 50% Plasmodium falciparum growth inhibition (IC 50 ), their calculated octanol/water distribution coefficients at pH 7.0 (log D), and their apparent bimolecular reduction rate constants by Plasmodium falciparum ferredoxin:NADP + oxidoreductase (Pf FNR) (k cat /K m ).

No.
Compound This shows that the oxidant potency of nitroaromatics and, to some extent, their lipophilicity play definite roles in their antiplasmodial activity.

Single-Electron Reduction of Nitroaromatics by PfFNR and PfGR
To the best of our knowledge, the pathways of reduction of nitroaromatic compounds in P. falciparum are not yet well understood. Among flavoenzymes dehydrogenases-electrontransferases that can initiate redox cycling of ArNO 2 , a potential candidate is ferredoxin:NADP + oxidoreductase localized in the apicoplast of the parasites [36,37]. This enzyme plays a significant role in parasite survival, because the functional analysis of P. falciparum genome revealed a high fitness cost of disruption of its gene [38]. Table 1 lists the bimolecular reduction rate constants of ArNO 2 by Pf FNR (k cat /K m ). The k cat values of the reactions were not determined because of a nearly linear dependence of reaction rate on ArNO 2 concentration except for the most reactive oxidant tetryl (k cat = 27.5 ± 2.0 s −1 ). Figure 2 shows the linear relationship between log k cat /K m and E 1 7 of nitroaromatics characterized by the ratio ∆log (k cat /K m )/∆E 1 7 = 11.69 ± 0.73 V −1 (r 2 = 0.9408), which mirrors to some extent the relationship between log IC 50 and E 1 7 (Equation (1)). Pf FNR catalyzes a single-electron reduction of ArNO 2 , as demonstrated by the observation that in the presence of compounds 5,12,14, and 23 (Table 1), the reduction of added cytochrome c takes place at rates that are 140-195% those of NADPH oxidation. Moreover, the reduction of cytochrome c is 15-25% inhibited by 100 U/mL superoxide dismutase. The redox cycling of ArNO 2 is also evident from the consumption of excess O 2 over ArNO 2 during the reaction ( Figure 3A). However, it is also important to note that Pf FNR-catalyzed formation of stable products of ArNO 2 reduction does not start after complete O 2 exhaustion, but takes place at  Table 1. Both yeast and erythrocyte GR catalyze the single-electron reduction of nitroaromatics, although at a low rate [19,25]. Since the formed ArNO2 − undergoes further redox cycling, nitroaromatics are considered as "subversive substrates" for GR. Moreover, there is some evidence that redox-active xenobiotics may be reduced at the NADP(H) binding site of GR [39,40]. It is supposed that the same reactions catalyzed by trypanothione reductase are at least partly responsible for the trypanocidal activity of nitrofurans [11]. Among the examined compounds, tetryl oxidized PfGR most efficiently (kcat = 5.9 ± 0.5 s −1 and kcat/Km = 7.6 ± 0.8 × 10 3 M −1 ·s −1 ), with parameters similar to those of HGR oxidation (kcat ≥ 5.0 s −1 , kcat/Km = 2.0 × 10 3 M −1 ·s −1 [37]). When present, cytochrome c was reduced at rates that are 170-180% those of NADPH oxidation, through a process partly inhibited by superoxide dismutase. . The numbers of compounds correspond to those in Table 1.  Table 1. Both yeast and erythrocyte GR catalyze the single-electron reduction of nitroaromatics, although at a low rate [19,25]. Since the formed ArNO2 − undergoes further redox cycling, nitroaromatics are considered as "subversive substrates" for GR. Moreover, there is some evidence that redox-active xenobiotics may be reduced at the NADP(H) binding site of GR [39,40]. It is supposed that the same reactions catalyzed by trypanothione reductase are at least partly responsible for the trypanocidal activity of nitrofurans [11]. Among the examined compounds, tetryl oxidized PfGR most efficiently (kcat = 5.9 ± 0.5 s −1 and kcat/Km = 7.6 ± 0.8 × 10 3 M −1 ·s −1 ), with parameters similar to those of HGR oxidation (kcat ≥ 5.0 s −1 , kcat/Km = 2.0 × 10 3 M −1 ·s −1 [37]). When present, cytochrome c was reduced at rates that are 170-180% those of NADPH oxidation, through a process partly inhibited by superoxide dismutase.  Both yeast and erythrocyte GR catalyze the single-electron reduction of nitroaromatics, although at a low rate [19,25]. Since the formed ArNO 2 − undergoes further redox cycling, nitroaromatics are considered as "subversive substrates" for GR. Moreover, there is some evidence that redox-active xenobiotics may be reduced at the NADP(H) binding site of GR [39,40]. It is supposed that the same reactions catalyzed by trypanothione reductase are at least partly responsible for the trypanocidal activity of nitrofurans [11]. Among the examined compounds, tetryl oxidized Pf GR most efficiently (k cat = 5.9 ± 0.5 s −1 and k cat /K m = 7.6 ± 0.8 × 10 3 M −1 ·s −1 ), with parameters similar to those of HGR oxidation  [37]). When present, cytochrome c was reduced at rates that are 170-180% those of NADPH oxidation, through a process partly inhibited by superoxide dismutase. 2,4,6-Trinitrotoluene (TNT) (k cat = 0.2 ± 0.05 s −1 , k cat /K m = 150 ± 40 M −1 ·s −1 ), 1,4-dinitrobenzene (k cat = 0.3 ± 0.07 s −1 , k cat /K m = 300 ± 60 M −1 ·s −1 ), and nifuroxime (k cat ≤ 0.06 s −1 , k cat /K m ≤ 110 M −1 ·s −1 ) oxidized Pf GR much more slowly. Other nitroaromatic compounds were even less efficient oxidants of Pf GR. At their saturating concentrations, the rate of NADPH oxidation was almost indistinguishable from the intrinsic NADPH-oxidase activity of the enzyme, 0.07 s −1 . This is in line with the previously reported properties of yeast and erythrocyte GR [19,25]. In conclusion, Pf GR-catalyzed redox cycling of nitroaromatics proceeds with much lower rates than in the analogous reaction of Pf FNR.

Inhibition of P. falciparum Glutathione Reductase by Nitroaromatic Compounds
Next, we analyzed the inhibition of P. falciparum GR by ArNO 2 . Pf GR acts through a "ping-pong" reaction mechanism with separate reductive and oxidative half-reactions [29]. At saturating concentrations of substrates, 100 µM NADPH and 1.0 mM GSSG, the catalytic constant (k cat ) of Pf GR was 138 ± 4.0 s −1 . Nitroaromatic compounds acted on Pf GR as non-or uncompetitive inhibitors with respect to GSSG at fixed NADPH concentration ( Figure 4A,B). Similarly, nitroaromatics acted as uncompetitive inhibitors with respect to NADPH at fixed GSSG concentration ( Figure 5).
These findings are in line with the well-characterized inhibition of yeast and human erythrocyte GR by nitroaromatics [19,25], and Pf GR and HGR by arylisoalloxazines and quinones [27,30,41]. The latter compounds may bind at the interface domain of two subunits of Pf GR in the vicinity of Val-56,56 and Asp-58,58 , which correspond to His-75,75 and Phe-78,78 in HGR [27]. This site is distant from both the NADP(H)-and GSSG-binding regions. The K i values of nitroaromatic compounds of Pf GR determined using GSSG as a variable substrate ( Figure 4A,B), are given in Table 2. The table also contains K i values of nitroaromatic compounds towards HGR determined previously [19] and partly within the present work. Importantly, the K i values for several nitroaromatic compounds obtained in both studies were sufficiently close ( Table 2). It should be noted that significant differences exist between HGR and Pf GR inhibition constants, and that their log values are poorly related (r 2 = 0.3840). This may be attributed to differences in the structure, shape, and flexibility of the intersubunit regions of Pf GR and HGR [24,25]. However, the analysis of the inhibition profile of the two enzymes is beyond the scope of the present work. At their saturating concentrations, the rate of NADPH oxidation was almost indistinguishable from the intrinsic NADPH-oxidase activity of the enzyme, 0.07 s −1 . This is in line with the previously reported properties of yeast and erythrocyte GR [19,25]. In conclusion, PfGR-catalyzed redox cycling of nitroaromatics proceeds with much lower rates than in the analogous reaction of PfFNR.

Inhibition of P. falciparum Glutathione Reductase by Nitroaromatic Compounds
Next, we analyzed the inhibition of P. falciparum GR by ArNO2. PfGR acts through a "ping-pong" reaction mechanism with separate reductive and oxidative half-reactions [29]. At saturating concentrations of substrates, 100 µM NADPH and 1.0 mM GSSG, the catalytic constant (kcat) of PfGR was 138 ± 4.0 s −1 . Nitroaromatic compounds acted on PfGR as non-or uncompetitive inhibitors with respect to GSSG at fixed NADPH concentration ( Figure 4A,B). Similarly, nitroaromatics acted as uncompetitive inhibitors with respect to NADPH at fixed GSSG concentration ( Figure 5).   [19], c taken from [42].
The analysis of antiplasmodial activity of nitroaromatics (Tables 1 and 2) shows that the dependence of log IC 50 on log K i is poorly expressed in the case of Pf GR, being characterized by r 2 = 0.5878 ( Figure S3 Although the log IC 50 of nitroaromatics also increased with their log K i for HGR (Table 2), this dependence was poorly expressed (r 2 = 0.3323). An introduction of log D as a second variable improved it up to r 2 = 0.4543. Finally, the antiplasmodial activity of nitroaromatics was best described by a regression using E 1 7 , log K i for Pf GR, and log D as independent variables: log IC 50 = −(0.27 ± 0.43) -(4.16 ± 1.78) E 1 7 + (0.36 ± 0.14) log K i -(0.15 ± 0.07) log D, (r 2 = 0.7866) On the other hand, the use of log K i for HGR as a variable resulted in a lower regression coefficient, and in an uncertain relationship between log IC 50

Discussion
Our study resolves the debated problem about the mechanisms of antiplasmodial activity of nitroaromatics [19,23,26], and demonstrates that their activity increases both with the ease of their bioreductive activation, expressed as the value of E 1 7 , and the efficiency of inhibition of Pf GR (Equation (3)). These data complement our previous findings on the role of inhibition of Pf GR in the activity of quinones against the same strain [30]. In both cases, the relationship between the activity of compounds and their efficiency as Pf GR inhibitors is revealed using multiparameter regression analysis. On the analogy with Equation (3), the activity of quinones was described by the relationships with ∆log IC 50 /∆log K i = 0.633 -0.763 [30]. These observations point to the importance of Pf GR as potential target for antiplasmodial agents. Apart from protection against the oxidative stress, Pf GR supplies GSH for the glyoxalase system, for the degradation of uncrystallized ferriprotoporphyrin IX, and as a source of reducing equivalents for ribonucleotide synthesis and thioredoxin-dependent antioxidant system [43][44][45][46]. Although no Pf GR knockout data are available in the case of P. falciparum, the functional analysis of its genome revealed a high fitness cost of disruption of the Pf GR gene [38]. On the other hand, the comparison between Equations (3) and (4) points to an insignificant role of HGR inhibition in the antiplasmodial action of nitroaromatics. Although the role of HGR in the survival of P. falciparum is a matter of debate [2,47], our data are in favor of its minor importance for parasite killing. However, this does not discard its role in the protection against an oxidative environment which limits the parasite infection rate, as it is in the case of glucose-6-phospate dehydrogenase deficiency [40]. In this context, one may note that the relationship ∆log IC 50 /∆log K i (Equation (3)) is significantly lower than unity. It may point to a limited role of inhibition of Pf GR in the antiplasmodial activity of nitroaromatics. On the other hand, the disturbance of GSSG/GSH homeostasis under the oxidative or alkylative stress in certain cases may enhance the expression of GR as a compensation mechanism. This phenomenon has been observed in yeast, plants, and mammalian cells [48][49][50][51].
In this context, two other issues related to the mechanism of antiplasmodial activity of ArNO 2 may be discussed. First is the dependence of their IC 50 values on their E 1 7 (Equation (1)), the latter being associated with their redox cycling activity. However, because P. falciparum, during its intraerythrocyte stage, adopts microaerophilic metabolism and relies mainly on anaerobic processes [52,53], the role of the ROS-promoted parasite death should be interpreted with caution. On the other hand, Equation (1) may equally well reflect the rates of formation of DNA-damaging hydroxylamines under the action of single-electron transferring enzymes such as Pf FNR ( Figure 3A,B) which may take place under partly anaerobic conditions. Another point is the role of Pf FNR in the formation of free radicals and/or other reduced forms of ArNO 2 in plasmodia. To the best of our knowledge, nitroreductase activity of P. falciparum flavoenzymes has not been previously evaluated. Some conclusions may be drawn from the enzyme reactivity with a model compound menadione (2-methyl-1,4-naphthoquinone, E 1 7 = −0.20 V) (Table 3), because, as a rule, flavoenzymes reduce quinones much faster than nitroaromatics, or, in exceptional cases, with similar rates ( [7,53], and references therein). Thus, given the data currently available (Table 3), Pf FNR may act as the most efficient generator of ArNO 2 free radicals in plasmodia. P. falciparum thioredoxin reductase may be next to it according to menadione reductase activity (Table 3).
On the other hand, the nitroreductase activity of mitochondrial type II NADH dehydrogenase is expected to be very low because menadione is slowly reducible. To the best of our knowledge, nitroaromatic compounds and soluble quinones have not been previously studied as oxidants of other enzymes of the mitochondrial respiratory chain, namely dihydroorotate dehydrogenase, succinate dehydrogenase, and malate:quinone oxidoreductase. A more general conclusion following from our study is that the efficacy of redox active antiparasitic agents such as quinones, nitroaromatic compounds, isoalloxazines, and aromatic N-oxides [6] may be achieved through two separate types of action, namely inhibition of antioxidant NADPH:disulfide reductases and rapid reduction by flavoenzymes dehydrogenaseselectrontransferases. Currently, these groups of compounds are mainly considered as "subversive substrates" of disulfide reductases such as glutathione reductase, trypanothione reductase, or thioredoxin reductase [11,30,40,54,56]. However, the single-electron reduction of the above compounds by dehydrogenases-electrontransferases of different origin (e.g., mammalian NADPH: cytochrome P-450 reductase, NO-synthase, NADH:ubiquinone reductase, algal FNR, and bacterial flavohemoglobin) is usually faster [7][8][9][10]. Thus, more attention should be given to the studies of these reactions catalyzed by parasite enzymes of this group.

Enzyme Kinetic Studies
All kinetic experiments were carried out spectrophotometrically using a PerkinElmer Lambda 25 UV-VIS spectrophotometer (PerkinElmer, Waltham, MA, USA) in 0.1 M K-phosphate buffer (pH 7.0) containing 1 mM EDTA at 25 • C. The steady-state parameters of reactions, the catalytic constants (k cat(app.) ), and the bimolecular rate constants (or catalytic efficiency constants, k cat /K m ) of the oxidants at fixed concentrations of NADPH correspond to the reciprocal intercepts and slopes of Lineweaver-Burk plots, [E]/v vs. 1/[oxidant], where v is the reaction rate, and [E] is the enzyme concentration. k cat represents the number of molecules of NADPH oxidized by a single active center of the enzyme per second. The rates of Pf FNR-and Pf GR-catalyzed NADPH oxidation in the presence of nitroaromatic compounds or GSSG were determined using the value ∆ε 340 = 6.2 mM −1 ·cm −1 . The rates were corrected for the intrinsic NADPH-oxidase activity of enzymes, which were equal to 0.12 s −1 and 0.07 s −1 for Pf FNR and Pf GR, respectively. In separate experiments, in which 50 µM cytochrome c were included in the reaction mixture, its nitroaromatic-mediated reduction was measured using the value ∆ε 550 = 20 mM −1 ·cm −1 . The kinetic parameters were obtained by the fitting of kinetic data to the parabolic expression using SigmaPlot 2000 version 11.0 (https://systatsoftware.com). The rates of reduction of nitroaromatic compounds (50 µM) by Pf FNR in the absence of external oxygen supply were monitored in the presence of an NADPH-regeneration system (50 µM NADPH, 10 mM glucose-6-phosphate, and 50 U/mL yeast glucose-6-phosphate dehydrogenase) at the specific λ max of absorbance of compounds. In these cases, a sealed spectrophotometer cell was completely filled by the solution containing nitroaromatic compound and NADPH-regeneration system, and the reaction was initiated by the injection of Pf FNR. In parallel, the rate of oxygen consumption was monitored under identical conditions using a Digital Model 10 Clark electrode (Rank Brothers Ltd., Bottisham, UK). In reversible inhibition studies of Pf GR, reaction rates were determined either at fixed NADPH concentration (100 µM) and varied GSSG concentrations (1.0-0.13 mM), or at fixed GSSG concentration (1.0 mM) and varied NADPH concentrations (8-50 µM), and either in the absence or presence of the inhibitor at 4-6 different concentrations. Using tetryl as inhibitor, the reaction rates were corrected for Pf GR-catalyzed NADPH oxidation by tetryl, which was typically less than 0.4% of total reaction rate. Since CB-1954 possessed significant absorbance at 340 nm, the reaction rate of Pf GR was monitored according to GSH-mediated reduction of 5.5'-dithiobis-(2-nitrobenzoic acid) (1.0 mM) using the value ∆ε 412 = 27.2 mM −1 ·cm −1 . The inhibition constants (K i ) were obtained from the Cleland plots, that is, the dependence of 1/k cat on the inhibitor concentration ([I]).

Antiplasmodial In Vitro Activity Studies
The chloroquine-resistant P. falciparum strain FcB1 from Colombia, which is deposited in the Protist collection of Museum National d'Histoire Naturelle, Paris, France, was kindly provided by Dr. H.D. Heidrich (Max-Planck Institut für Biochemie, Martinsried bei München, Germany). P. falciparum FcB1 strain was maintained in continuous culture of human erythrocytes according to [62]. In vitro antiplasmodial activity was determined using a modification of the semiautomatic microdilution technique [63]. Stock solutions of test compounds in DMSO were serially diluted with culture medium and added to asynchronous parasite cultures (1% parasite infected cells and 1% final hematocrit) for 24 h, at 37 • C, prior to the addition of 1.825 MBq of [ 3 H]-hypoxanthine (0.37-1.11 TBq/mmol), for 24 h. The growth inhibition for each compound concentration was determined according to the radioactivity incorporation into the treated culture as compared with that in the control culture. The experiments were repeated in triplicate. ArNO 2 Nitroaromatic compound E 1 7 Single-electron reduction midpoint potential of nitroaromatic compound (redox potential of ArNO 2 /ArNO 2 −. couple) at pH 7.0 HGR Human erythrocyte glutathione reductase IC 50 Compound concentration causing 50% parasite growth inhibition k cat Enzyme catalytic constant k cat /K m Enzyme bimolecular rate constant (catalytic efficiency) K i Enzyme inhibition constant log D Octanol/water distribution coefficient at pH 7.0 Pf FNR P. falciparum ferredoxin:NADP + oxidoreductase Pf GR P. falciparum glutathione reductase ROS Reactive oxygen species TNT 2,4,6-Trinitrotoluene TR Trypanothione reductase