trypanothione reductase inhibitors against Trypanosoma brucei and Trypanosoma cruzi.

CHEMMEDCHEM FULL PAPERS DOI: 10.1002/cmdc.201300399

Toward the Development of Dual-Targeted Glyceraldehyde-3-phosphate Dehydrogenase/ Trypanothione Reductase Inhibitors against Trypanosoma brucei and Trypanosoma cruzi Federica Belluti,[a] Elisa Uliassi,[a] Giacomo Veronesi,[a] Christian Bergamini,[a] Marcel Kaiser,[b, c] Reto Brun,[b, c] Angelo Viola,[a] Romana Fato,[a] Paul A. M. Michels,[d] R. Luise Krauth-Siegel,[e] Andrea Cavalli,[a, f] and Maria Laura Bolognesi*[a] A significant improvement in the treatment of trypanosomiases has been achieved with the recent development of nifurtimox–eflornithine combination therapy (NECT). As an alternative to drug combinations and as a means to overcome most of the antitrypanosomatid drug discovery challenges, a multitarget drug design strategy has been envisaged. To begin testing this hypothesis, we designed and developed a series of quinone–coumarin hybrids against glyceraldehyde-3-phosphate dehydrogenase/trypanothione reductase (GAPDH/TR). These enzymes belong to metabolic pathways that are vital to Trypanosoma brucei and Trypanosoma cruzi, and have thus been considered promising drug targets. The synthesized molecules were characterized for their dual-target antitrypanoso-

mal profile, both in enzyme assays and in in vitro parasite cultures. The merged derivative 2-{[3-(3-dimethylaminopropoxy)2-oxo-2H-chromen-7-yl]oxy}anthracene-1,4-dione (10) showed an IC50 value of 5.4 mm against TbGAPDH and a concomitant Ki value of 2.32 mm against TcTR. Notably, 2-{4-[6-(2-dimethylaminoethoxy)-2-oxo-2H-chromen-3-yl]phenoxy}anthracene-1,4dione (compound 6) displayed a remarkable EC50 value for T. brucei parasites (0.026 mm) combined with a very low cytotoxicity toward mammalian L6 cells (7.95 mm). This promising low toxicity of compound 6 might be at least partially due to the fact that it does not interfere with human glutathione reductase.

Introduction The trypanosomiases are very serious so-called neglected tropical diseases (NTD).[1] In sub-Saharan Africa, Trypanosoma brucei causes sleeping sickness or human African trypanosomiasis (HAT),[2] and, primarily in Latin America, Trypanosoma cruzi causes Chagas disease.[3] Both trypanosomiases are difficult to [a] Prof. Dr. F. Belluti, E. Uliassi, G. Veronesi, Dr. C. Bergamini, A. Viola, Prof. Dr. R. Fato, Prof. Dr. A. Cavalli, Prof. Dr. M. L. Bolognesi Department of Pharmacy & Biotechnology, University of Bologna Via Belmeloro 6 / Via Irnerio 48, 40126 Bologna (Italy) E-mail: [email protected] [b] Dr. M. Kaiser, Prof. Dr. R. Brun Swiss Tropical & Public Health Institute Socinstrasse 57, 4002 Basel (Switzerland) [c] Dr. M. Kaiser, Prof. Dr. R. Brun University of Basel, Petersplatz 1, 4003 Basel (Switzerland) [d] Prof. Dr. P. A. M. Michels Institute of Structural & Molecular Biology School of Biological Sciences, University of Edinburgh Mayfield Road, Edinburgh EH9 3JR (UK) [e] Prof. Dr. R. L. Krauth-Siegel Biochemie-Zentrum, Universitt Heidelberg Im Neuenheimer Feld 328, 69120 Heidelberg (Germany) [f] Prof. Dr. A. Cavalli Department of Drug Discovery & Development Istituto Italiano di Tecnologia, Via Morego 30, 16163 Genova (Italy) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.201300399.

2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

treat, with drugs and vaccines not yet available.[4] Current insufficient antiparasite chemotherapy regimens rely mostly on single-target drugs that are often plagued by toxic side effects, lack of efficacy, and development of resistance. A significant improvement in the treatment of HAT has been recently achieved by the introduction of the first combination regimen, based on the co-administration of oral nifurtimox and of a decreased dose of intravenous eflornithine (referred to as nifurtimox–eflornithine combination therapy, or NECT).[5] Thanks to its improved efficacy, shortened duration, and diminished side effects, it represents a major advance in terms of making the antitrypanosomatid treatment safer, less expensive, and easier to administer.[6] As an alternative to drug combinations, an innovative pharmacological strategy in the field could involve designing multitarget ligands[7] (MTL), that is, single small molecules that can modulate multiple vital targets in parasites’ metabolic pathways.[8] As for combinations, a multitarget approach improves the chances of better efficacy and preventing the development of drug-resistant parasites.[8] Moreover, it would be advantageous over a combined administration of individual drugs given the simplification of the therapeutic regimen and the fact that it would also obviate possible drug–drug interactions.[7, 8] This latter aspect is especially relevant for young children, and malnourished, immunocompromised, or debilitated ChemMedChem 0000, 00, 1 – 13

&1&

These are not the final page numbers! ÞÞ

CHEMMEDCHEM FULL PAPERS patients which are most commonly affected by NTD in developing countries.[9] However, although a multitarget drug discovery strategy offers advantages in principle, it still represents a challenging task for medicinal chemists.[10] In this study, we aimed to develop dual-targeted inhibitors against T. brucei and T. cruzi glyceraldehyde-3-phosphate dehydrogenase/trypanothione reductase (TR/GAPDH). These two enzymes have been identified as validated targets for antitrypanosomatid drug discovery.[11] The fact that the bloodstream form of African trypanosomes relies on glycolysis as the sole source of energy supply[12] and the unusual compartmentalization of this pathway inside glycosomes,[13] make the development of selective GAPDH inhibitors (which should only exhibit minimal affinity for the human enzyme) particularly attractive.[11b] GAPDH catalyzes the oxidative phosphorylation of d-glyceraldehyde 3-phosphate to 1,3-bisphosphoglycerate. Another metabolic peculiarity is that trypanosomatid protists possess a unique thiol metabolism.[14] The discovery that an unusual low-molecular-weight thiol, namely N1,N8-bisglutathionylspermidine (or trypanothione), rather than glutathione, is the major redox reactive metabolite in trypanosomatids[15] has elevated the enzymes involved in its metabolism to paradigms of antiparasitic drug targets.[16] In this pathway, TR,[17] an NADPHdependent disulfide oxidoreductase, is responsible for reducing trypanothione disulfide (TS2) to dihydrotrypanothione [T(SH)2] and for maintaining a reducing intracellular milieu. In the search for TR/GAPDH inhibitors, we considered the previously identified hit compound B6 (1 in Table 1) as a suita-

www.chemmedchem.org ble starting point.[18] Remarkably, it was shown in retrospect to inhibit GAPDH[19] and TR[20] in the micromolar range. Furthermore, it exhibited trypanocidal potency in the nanomolar range (EC50 value of 80 nm against T. brucei rhodesiense) and a promising selectivity index (SI of 74), as assessed in experiments using in vitro cultured parasites and human cell lines.[18] Based on these findings, we designed and synthesized eight quinone–coumarin hybrid compounds and characterized them for their multitarget antitrypanosomal profile, both in enzyme assays and in in vitro cultures of parasites.

Results and Discussion Chemistry

The synthesis of the two series of quinone–coumarin hybrids (3–6 and 7–10) was accomplished as illustrated in Schemes 1– 3. To obtain the desired coumarin intermediates 11, 16, and 18, we used two different synthetic approaches. In detail, the reaction of 2,5-dihydroxybenzaldehyde and sodium 2-(4-methoxyphenyl)acetate in the presence of acetic anhydride and triethylamine gave, upon alkaline hydrolysis, 6-hydroxy-3-(4methoxyphenyl)-2H-chromen-2-one (11) in good yield[21] (Scheme 1). Following the same conditions, reaction of 2,4-dihydroxybenzaldehyde and sodium 2-(4-hydroxyphenyl)acetate acetic furnished 7-hydroxy-3-(4-hydroxyphenyl)-2H-chromen-2one (16, Scheme 2). Reaction of 2-hydroxy-4-methoxybenzaldehyde and N-acetylglycine, in the presence of acetic anhydride and sodium acetate,[22] provided the N-(7-methoxy-2-oxoTable 1. Antitrypanosomal activity against T. b. rhodesiense (Tbr) and T. cruzi (Tc), and cytotoxicity in a rat skele2H-chromen-3-yl)acetamide (17) tal myoblast cell line (L6) of derivatives 3–10 and reference compounds.[a] that was subjected to acidic hydrolysis with methanolic HCl, to obtain the desired 3-hydroxy-7methoxy-2H-chromen-2-one[23] (18, Scheme 3). The hydroxy derivatives 11 and 18 were then reacted with the selected wchloroalkyldimethylamines in N,N-dimethylformamide (DMF) and in the presence of Cs2CO3 Compd Ar n EC50 [mm] Selectivity Index as the base,[24] to obtain comTbr Tc L6 Tbr Tc pounds 12, 13, 19, and 20. The 5.92[18] 74.0[18] 4.70 1 – – 0.08 1.26[18] methoxy group was then re– 0.05 4.66[18] 1.00[18] 20.0[18] 0.21 2 C4H4 3 – 2 0.95 1.26 4.14 4.32 3.26 moved by treatment with BBr3 4 – – 2.84 26.3 13.42 4.72 0.51 or HBr solution (33 % in AcOH) 5 – 1 2.51 1.47 9.62 3.83 6.55 in dichloromethane as solvent, 1 0.026 31.23 7.95 301.8 0.25 6 C4H4 to afford the phenol com7 – 1 4.56 8.46 4.27 0.93 0.50 1 0.14 29.86 4.5 32.03 0.15 8 C4H4 pounds 14, 15 and 21, 22, re9 – 2 9.18 43.15 13.43 1.46 0.31 spectively. Finally, the quinone [b] 2 0.25 ND 1.30 5.23 – 10 C4H4 moieties were introduced by [20] [20] [20] – 18.3 1830.0 – melarsoprol 0.01 a nucleophilic substitution reac> 384[20] – > 225[20] benznidazole – 1.70[20] podophyllotoxin – – 0.017[20] – – tion, firstly treating the phenol intermediates 14–16 and 21, 22 [a] IC50 values represent the concentration of a compound that causes 50 % growth inhibition and are the mean of two independent determinations that varied by less than a factor of two. The experimental error is with Cs2CO3 or K2CO3 and then within 50 %. SI = (IC50 for L6)/(IC50 for the respective protozoan parasite).[b] Not determined due to cytotoxicwith the selected bromoquiity in L6 cells. none (2-bromoanthraquinone


ChemMedChem 0000, 00, 1 – 13

&2&


CHEMMEDCHEM FULL PAPERS

www.chemmedchem.org or 2-bromonaphthoquinone), to afford the final compounds 3– 10.

Design rationale and biological studies

Scheme 1. Synthesis of compounds 3, 5, and 6. Reagents and conditions: a) Ac2O, anhydrous NEt3, 180 8C, 18 h, then K2CO3, RT, 6 h; b) w-chloroalkyldimethylamine hydrochloride, Cs2CO3, DMF, N2, 110 8C, 8 h; c) BBr3, CH2Cl2, 70 8C, 1 h, then RT, 6 h; d) 2-bromoquinone, Cs2CO3, DMF, RT, 3 h.

Scheme 2. Synthesis of compound 4. Reagents and conditions: a) Ac2O, anhydrous NEt3, 180 8C, 18 h, then K2CO3, RT, 6 h; b) 2-bromonaphthoquinone, Cs2CO3, DMF, RT, 3 h.

Scheme 3. Synthesis of compounds 7–10. Reagents and conditions: a) Ac2O, AcONa, 120 8C, 5 h; b) MeOH, HCl 3 n, 120 8C, 10 h; c) w-chloroalkyldimethylamine hydrochloride, Cs2CO3, DMF, N2, 110 8C, 8 h; d) HBr solution (33 % in AcOH), 80 8C, 14 h; e) 2-bromoquinone, K2CO3, DMF, RT, 3 h.


Our strategy to discover novel dual inhibitors started from the finding that 1 inhibited both targets at similar concentration ranges.[19, 20] As a first step, we calculated IC50 values of 7.2 and 9.0 mm, for GAPDH and TR, respectively. Although the inhibitory potencies were in the micromolar range, a balanced affinity against the targets of interest was obtained, which is a prerequisite in multitarget drug discovery. With these data in hand, we looked for a strategy to improve the potencies of the starting hit through chemical optimization. On the basis of the criteria of the so-called “rule of three” it is evident that 1 can be truly considered a fragment.[25] In fact, it has a molecular weight of 250.25 Da, fewer than three hydrogen bond acceptors/donors, a clog P value of 2.92, a polar surface area (PSA) of 43.37 2, and ligand efficiencies for GAPDH and TR of 0.31 and 0.24, respectively. Recently, a fragment-based approach[7] has been considered very effective in multitarget drug discovery.[26] In a typical embodiment, once a fragment binding to two proteins has been identified, a library of analogues with additional functionality could be made, with the goal of increasing the affinity toward both targets while concomitantly maintaining balanced activities. Following this strategy, 1 was firstly investigated by docking simulation and its resulting predicted mode of association with GAPDH and TR binding sites is shown in Figure 1. In detail, 1 binds to GAPDH at the active site with the catalytic Cys166 properly positioned to possibly give a nucleophilic attack to the C2 carbon of the naphthoquinone ring. This would allow it to act as a covalent inhibitor of GAPDH as previously hypothesized[19] and herein partially supported by biochemical experiments (see below). Further interactions with His194 and Cys166 backbone contributed to stabilize 1 into the binding pocket of GAPDH. Docking simulations of 1 with TR showed that this fragment could be accommodated into the wide TR catalytic pocket. In particular, one of the two carbonyls of the naphthoquinone ring established a hydrogen bond interaction with the Leu399 backbone, while Phe396 and Lys21 of TR interacted with 1 by means of p–p and cation–p interactions. Remarkably, these interactions could allow our compounds to display selectivity over human glutathione reductase (hGR).[27] These initial docking simulations showed that 1 could bind both enzymes by establishing favorable interactions with active site amino acids, in support of our strategy to use 1 as starting fragment for GAPDH/TR dual-inhibitor discovery. Starting from the molecular information, we expanded 1 through a framework combination approach with the intention to develop more potent, bigger, dual-target inhibitors. Among the reported inhibitors of GAPDH, chalepin (IC50 64 mm), a coumarin-containing natural product extracted from the plant Pilocarpus spicatus, is the one with potential druglike properties. In addition, the crystallographic structure of the chalepin–TcGAPDH complex has been available since 2002.[28] ChemMedChem 0000, 00, 1 – 13

&3&



www.chemmedchem.org

Figure 1. Docking pose of 1 at the a) TbGAPDH (2X0N) and b) TbTR (2WPF) binding sites.

Regarding a rational drug design of an inhibitor of TR, a distinctive feature of this enzyme is its extremely wide and negatively charged binding site for TS2. This has been exploited for obtaining selectivity against hGR, the closest related counterpart in the mammalian host, which has a positively charged binding site for glutathione disulfide (GSSG).[17] Building on these considerations, we expanded the structure of 1 by incorporating a coumarin and an amino functionality. So far, two subsets of compounds were generated through merging or fusing approaches,[7] as depicted in Figure 2. In detail, in compounds 3–6, the coumarin framework has been linked at position 4 of the phenoxy ring of 1; in 3 and 5, an amino function has been introduced at position 6 of the coumarin through ethoxy or propoxy linkers, whereas the introduction of a second quinone moiety led to 4. In addition, to enlarge the chemical diversity and based on the antitrypanosomal activity of the higher homologue 2 (Figure 2),[18] the anthraquinone 6 was also designed. Similarly, by integrating the selected moiet-

Figure 2. Design strategy leading to the fused and merged quinone–coumarin hybrids 3–10 (see Table 1 for individual structures).


ies to a greater extent, the subset 7–10, in which the benzene ring of 1 and 2 and that of coumarin overlaps, was obtained. The designed 3–10 were then synthesized, and their activity profile evaluated toward the two enzymes and in whole cell parasite assays. First, we tested the inhibitory potency of 3–10 against TbGAPDH using a spectrophotometric method, as described by Wiggers et al.[29] with minor modifications (see the Experimental Section). In brief, TbGAPDH activity was measured by following NAD reduction at 340 nm using GAP (glyceraldehyde 3-phosphate) as a substrate. Compounds 3–10 were assayed at a concentration of 10 mm, and the percentages of residual enzymatic activity measured, in comparison with 1 and 2 as parent compounds. As shown in Figure 3, all of the newly developed hybrids (either belonging to the merged or fused series) displayed remaining activities ranging from 10 to 50 %, which in most cases were higher than those observed when adding 1 or 2. The most active compound was the merged derivative 10, which has been analyzed at different concentrations (ranging from 1 to 50 mm, data not shown), obtaining an IC50 value of 5.4 mm, only slightly higher than that of 1 (IC50 = 3.3 mm). When analyzing the reported data, the structure–activity relationship (SAR) appeared quite “flat”, as the major changes made to the structure

Figure 3. Effect of compounds 1–10 (at 10 mm) on TbGAPDH activity. The activities were measured spectrophotometrically, and data are the means SD of triplicate determinations.

ChemMedChem 0000, 00, 1 – 13

&4&


CHEMMEDCHEM FULL PAPERS of 1 did not result in significant activity differences. The “flat” SAR within the compound series might point toward a very similar binding mode. One possible explanation may be that, in analogy with what we observed for 1,[19] the 2-aryloxynaphthoquinone moieties of 3–10 could react with Cys166 of the GAPDH active site by forming a covalent bond. We hypothesized that 1 may react with the thiol group of Cys166, and a reaction at C2 with the phenate displacement as leaving group may take place.[19] Indeed, by analyzing the structures of our hybrids, it is evident that they might undergo the same reaction as they also carry a phenoxy leaving group. To support this hypothesis, we exploited the peculiar spectral features of our derivatives. They contain a fluorophore (the coumarin scaffold; likely the leaving group of the covalent interaction), although it displays only a very low fluorescence due to the quenching effect of the quinone moiety. Nonetheless, the nucleophilic attack on the quinone by the thiol group, causing formation of a thioether bond, should result in a dissociation of the coumarin moiety that then becomes more fluorescent. Indeed, an increase in fluorescence could be detected in spectrofluorimetric experiments, by following fluorescence emission at 485 nm (with excitation at 335 nm). We treated compound 10 with molecules carrying a thiol group (for example, dithiothreitol (DTT) and N-acetylcysteine (NAC)), then we compared its fluorescence emission spectra with that of 3-(3aminopropoxy)-7-hydroxycoumarin 22 (the actual leaving group). Indeed, Figure 4 a shows the fluorescence emission spectra of compound 10 (curve 1 from the bottom) and fragment 22 (curve 4 from the bottom), whereas curves 2 and 3 refer to the emission of 10 treated with DTT and NAC, respectively. These latter curves are clearly similar to those obtained for the coumarin fragment alone. The same behavior could be observed when treating 10 with GAPDH (Figure 4 b), supporting the idea that a nucleophilic substitution reaction between 10 and a thiol nucleophile could occur. In parallel, 3–10 were studied in comparison with 1 and 2 in the TR enzymatic assay, as described previously.[20] As evident from the graph of Figure 5 a, the merged hybrids 7, 8, and 10 proved to be more effective inhibitors of TR than 2, and nearly as potent as 1. In fact, at a fixed inhibitor concentration of 10 mm, and either 100 or 40 mm TS2, the remaining activity ranged from 25 to 62 % for the merged hybrids, and between 23 and 65 % for compounds 1 and 2. In contrast, at 10 mm, the fused series did not display any significant activity. Thus, we studied the effect of 3–6 at concentrations of 50–100 mm (Figure 5 b). This activity trend was difficult to rationalize, as purposely addressed docking studies were not informative in this respect (data not shown). Taken together, the most active of the newly synthesized compounds was 10, which was also the most powerful GAPDH inhibitor. Notably, no difference in the degree of inhibition at the two TS2 concentrations was observed for 7–10, 3, or 5, suggesting an inhibition mechanism independent from the substrate concentration. Indeed, the kinetic analysis revealed that 10 inhibits TR noncompetitively with a Ki value of 2.7 mm (Figure 6). Thus, in principle, TR and GAPDH are modulated by 10 at the same concentration range. 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemmedchem.org

Figure 4. a) Comparison of the emission scan (measured as excitation of 335 nm) of compound 10 in the absence (first curve from bottom) or presence of thiols DTT and NAC (second and third curves, respectively), and that of the coumarin fragment 22 at an equimolar concentration. The fluorescence maxima of 10 treated with the thiol nucleophiles were similar to that of 22, suggesting that a nucleophilic displacement with the release of the coumarin leaving group might occur. b) Emission scans of 10 in the presence of GAPDH.

To assess the TR/GR selectivity, the compounds were studied at a fixed concentration against both enzymes (40 mm for 3–6 and 5 mm for 7–10) and the results graphically depicted in Figure 7. Despite the presence in 3 and 5–10 of the protonatable amino group as a putative determinant of selectivity, we could not verify the desired profile. In fact, all compounds, except 6, were more effective toward human GR than trypanosomal TR. The same applied for 4, which does not carry the amino functionality. Intriguingly, for compound 6 only a marginal inhibition of GR was observed at a concentration of 40 mm. Concerning the mechanism of inhibition, we speculated that it may be similar to that proposed for other quinone inhibitors.[20, 30] These ligands are known to easily shuttle between hydroquinone and quinone forms. It is therefore feasible that in addition to the observed noncompetitive inhibition, 10 may, ChemMedChem 0000, 00, 1 – 13

&5&



www.chemmedchem.org

Figure 6. Lineweaver–Burk plots for the inhibition of TcTR by 10. The assays contained 100 mm NADPH and two fixed concentrations of inhibitor as indicated in the graph, and the TS2 concentration was varied.

Figure 5. Inhibition of TR by a) 7–10, 1 and 2, and b) 3–6. The activity was measured following the NADPH consumption at 340 nm as described in the Experimental Section. The assays contained a fixed concentration of 100 or 40 mm TS2 and the indicated inhibitor concentration. The controls contained the same amount of DMSO used to dissolve the compounds. The remaining percent activity refers to that of the enzyme in the presence of DMSO.

at least in part, act as a redox cycler. In this process, TR reduces the quinone to a semiquinone (one-electron reduction) or to a hydroquinone (two-electron reduction), causing futile consumption of NADPH. The semiquinone formed can then react with molecular oxygen, producing superoxide anion radicals. The combination of TR inhibition and redox cycling might be beneficial, as it should lead to stronger trypanocidal properties.[20, 30, 31] Of course, such

a redox cycling might also apply to human GR and in this case it would lead to an unwanted cytotoxicity; however, the intrinsic oxidase activity of human GR is much lower than that of TR.[32] Thus, our derivatives were studied for their ability to induce the intrinsic oxidase activity of TR and GR by following the consumption of NADPH in the absence of the respective disulfide substrate (Figure 8). The oxidase activity of TR was increased up to nearly a 100-fold in the presence of 3, 5, and 9, which all share a naphthoquinone moiety. Encouragingly, the oxidase activity of GR remained much lower than that of TR. The other derivatives were much less effective as subversive substrates for both enzymes, with 6 being completely inactive for GR. This behavior is very promising in terms of selectivity against the host. To detect the generation of radical species in the reaction of TR with 3, 5, and 9, the oxidase assay was coupled to the oneelectron reduction of oxidized cytochrome c (Cyt c–Fe3 + ). By following the absorption increase at 550 nm, the single-electron reduction of the heme protein was detected (Table 2). To substantiate the putative generation of superoxide radicals in

Table 2. Detection of radical species in the reaction of trypanothione reductase (TR) with compounds 3, 5, and 9 using an oxidase assay.[a] Compd 3 5 9

NADPH cons. [U mg1] 1.93 0.13 1.24 0.06 0.55 0.06

Cyt c–Fe3 + red. [U mg1] SOD + SOD 3.07 0.24 1.86 0.06 0.80 0.05

0.47 0.04 0.36 0.01 0.23 0.02

NADPH ox./ Cyt c red.[b]

1 e red.[c] [%]

Inh. (SOD)[d] [%]

1:1.6 1:1.5 1:1.45

79.5 75 73

84.5 81 71

[a] The oxidase activity of TR in the presence of compounds 3, 5, and 9 was assessed by measuring the consumption of NADPH and by coupling the reaction to cytochrome c reduction. The latter reaction was carried out in the absence or presence of superoxide dismutase (SOD) to distinguish between a direct and superoxide-anion-mediated single-electron reduction (1 e red.) of Cyt c–Fe3 + . Data represent the mean SD of a triplicate determination. [b] Ratio of NADPH oxidation/Cyt c reduction (NADPH ox./Cyt c red.); stoichiometry = 1:2. [c] Single-electron reduction (1 e red.) is expressed as a percentage of overall reduction. [d] Percent inhibition by SOD.


ChemMedChem 0000, 00, 1 – 13

&6&



www.chemmedchem.org

Figure 8. Increase of NADPH oxidation by hGR (white columns) and TcTR (gray columns) in the absence of the natural substrates, but in the presence of compounds 3 and 5–10 at 40 mm acting as subversive substrates. The activity of the enzymes, evaluated by following NADPH consumption at 340 nm, is expressed in U mg1 and compared with their intrinsic oxidase activity (IOA), measured in the presence of the same amount of DMSO used to dissolve the compounds. Data are the means SD of triplicate determinations.

Figure 7. Comparison of the inhibition of hGR (black columns) and TcTR (gray columns) by a) compounds 7–10 at 5 mm, and b) compounds 3–6 at 40 mm. The activity of both enzymes was measured by following NADPH consumption at 340 nm as described in the Experimental Section. The assays for TR activity contained a fixed concentration of 100 mm TS2. The assays for hGR activity contained a fixed concentration of 1 mm GSSG. The controls contained the same amount of DMSO used to dissolve the compounds. The remaining percent activity refers to that of the enzyme in the presence of DMSO. Data are the means SD of triplicate determinations.

the overall reaction, superoxide dismutase (SOD) was added to the assay, as previously reported.[30] In the presence of SOD, the rate of Cyt c–Fe3 + reduction was lowered by 71 to 85 % as expected if superoxide anions are formed, because the dismutation of superoxide radicals by SOD proceeds much faster than the corresponding reduction to molecular oxygen by Cyt c–Fe3 + . The remaining activity indicates that Cyt c–Fe3 + is 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

partially reduced directly by the quinone radical, as also previously observed for menadione-based subversive substrates of TR.[30] Despite the moderate in vitro potency of these compounds against the two enzymes, we were quite encouraged by experiments performed with intact T. brucei and T. cruzi parasites and mammalian cells. We evaluated the activity of 3–10 against T. b. rhodesiense (bloodstream stage, STIB 900 strain) and T. cruzi (intracellular amastigote stage, Tulahuen C2C4 strain) parasites, in comparison to the reference drugs melarsoprol and benznidazole. Furthermore, their cytotoxicity against mammalian cells was assessed using the L6 rat skeletal myoblast cell line. All the newly synthesized compounds displayed moderate to high activity (Table 1). They were generally more efficient against T. brucei than T. cruzi. The most effective compounds were 6, 8, and 10, with 6 inhibiting the proliferation of African trypanosomes at a concentration similar to that of the drug melarsoprol. Remarkably, against T. cruzi, 3 and 5 displayed EC50 values that were even lower than that of the corresponding reference drug benznidazole. As 10 was the most active compound in the enzymatic assays, whereas 6 was one of the weakest inhibitors of GAPDH/TR, it clearly emerged that the in vitro whole cell data did not follow a trend similar to that of the in vitro enzyme data, with decreases in the IC50 values not being always mirrored by corresponding decreases in the EC50 values. The promising data obtained for 6 could suggest that this compound has additional off-targets, or is selectively concentrated/metabolically activated in the parasite, or a combination of these. Interestingly, 6 displayed a very low cytotoxicity toward L6 cells (7.95 mm) relative to its EC50 value for T. brucei parasites (0.026 mm). Thanks to this profile, 6 ChemMedChem 0000, 00, 1 – 13

&7&


CHEMMEDCHEM FULL PAPERS showed a remarkable SI > 300. This compound fulfills the “hit criteria” of activity and selectivity against T. b. rhodesiense (IC50 < 0.2 mg mL1; SI > 100) as defined by the drug-screening network of TDR,[33] and is eligible for further in vivo investigation.

Conclusions In NTD, a multitarget drug discovery approach may lead to more effective, affordable, and sustainable medicines. Toward this goal, we designed and synthesized novel GAPDH/TR dual inhibitors by expanding fragment 1. From a practical perspective, the obtained data reveal that against GAPDH, the most potent derivative 10 showed an IC50 value of 5.3 mm, making it among the most effective trypanosomal GAPDH inhibitors developed so far. Compound 10 showed a concomitant Ki of 2.32 mm against TR, which is encouraging in terms of dual activity. The molecules were then tested in whole parasite growth assays and displayed remarkable EC50 values against the trypanosomes. Compounds 3, 5, and 6 showed activities similar to that of the reference drugs benznidazole and melarsoprol. However, the lack of correlation between GAPDH/TR inhibition and the trypanocidal effects for most of the derivatives, suggested that other targets might be involved. Interestingly, we discovered that 6 was not very toxic against the host cell (at levels > 300-fold the EC50 value against the T. brucei parasite). The specificity of this derivative was much higher when compared with the parent molecules 1 and 2 and with all the newly synthetized hybrids. In fact, a general cytotoxicity was the major concern for the drug-likeness of several quinone-based hybrids.[18, 20] We speculate that the lower toxicity of 6 might be, at least in part, attributable to the fact that it does not interfere with human GR, neither acting as an inhibitor nor as a subversive substrate. All in all, these data are promising and warrant further investigation of 6 as a lead candidate against African and American trypanosomiases. From a conceptual perspective, the data collected in this work emphasize that obtaining a multitarget ligand is still a difficult endeavor for medicinal chemists, and some might even assert impossible.[10] However, in our opinion, the high possible reward of these molecules may make their development a risk well worth taking.

Experimental Section

www.chemmedchem.org dicated with element symbols, analytical results obtained for those elements are within 0.4 % of the theoretical values. The IR spectra were recorded on a Jasco FT/IR-4100. Chromatographic separations were performed on silica gel columns by a flash method (Kieselgel 40, 0.040–0.063 mm, Merck). Reactions were followed by thin layer chromatography (TLC) on pre-coated silica gel plates (Merck Silica Gel 60 F254) and then visualized with a UV lamp. Compounds were named following IUPAC rules as applied by Beilstein-Institute AutoNom (version 2.1), a PC-integrated software package for systematic names in organic chemistry. 6-Hydroxy-3-(4-methoxyphenyl)-2H-chromen-2-one (11): A mixture of 2,5-dihydroxybenzaldehyde (1.38 g, 10 mmol), sodium 2-(4methoxyphenyl)acetate (1.88 g, 10 mmol), freshly distilled acetic anhydride (3 mL), and anhydrous triethylamine (3 mL) was heated at 180 8C for 18 h. After cooling, a K2CO3 saturated aqueous solution (10 mL) was added and the resulting mixture was stirred at RT for 6 h. The reaction mixture was then acidified and the precipitate was filtered off and crystallized from EtOH to afford a yellow solid (2.15 g, 80 %), mp: 132 8C. 1H NMR (CDCl3, 300 MHz): d = 3.82 (s, 3 H), 6.96–6.99 (m, 4 H), 7.25 (d, J = 8.8 Hz, 1 H), 7.69–7.71 (m, 3 H), 7.87 ppm (br, 1 H). 7-Hydroxy-3-(4-hydroxyphenyl)-2H-chromen-2-one (16): A mixture of 2,4-dihydroxybenzaldehyde (1.38 g, 10 mmol), sodium 2-(4hydroxyphenyl)acetate (1.74 g, 10 mmol), freshly anhydrous acetic anhydride (3 mL), and anhydrous triethylamine (3 mL) was heated at 180 8C for 18 h. After cooling, a K2CO3 saturated aqueous solution (10 mL) was added and the resulting mixture was stirred at RT for 6 h. The mixture was then acidified and the precipitate was filtered off and crystallized from EtOH to afford a yellow solid (1.27 g, 50 %); mp: 184 8C. 1H NMR ([D6]DMSO, 300 MHz): d = 6.96 (d, J = 8.8 Hz, 4 H), 7.25 (d, J = 8.8 Hz, 1 H), 7.69 (d, J = 8.8 Hz, 2 H), 7.70 (s, 1 H), 7.87 (br, 1 H, OH), 8.15 ppm (br, 1 H, OH). N-(7-Methoxy-2-oxo-2H-chromen-3-yl)acetamide (17): A mixture of 2-hydroxy-4-methoxybenzaldehyde (7.00 g, 46 mmol), N-acetylglycine (5.39 g, 46 mmol), freshly distilled acetic anhydride (21.7 mL, 230 mmol), and sodium acetate (15.10 g, 184 mmol) was heated at 120 8C for 5 h. After cooling, a mixture of ice/water (100 mL) was added and the precipitate was filtered off and crystallized from EtOH to afford a yellow solid (2.84 g, 26.5 %); mp: 230–231 8C. 1H NMR (CDCl3, 300 MHz): d = 2.22 (s, 3 H), 3.85 (s, 3 H), 6.91 (d, J = 2.8 Hz, 1 H), 6.95 (dd, J = 2.8, 8.8 Hz, 1 H), 7.40 (d, J = 8.8 Hz, 1 H), 8.03 (br, 1 H), 8.66 ppm (s, 1 H). 3-Hydroxy-7-methoxy-2H-chromen-2-one (18): 17 (1.2 g, 5.14 mmol) was dissolved in MeOH (36 mL) and 3N HCl (200 mL), the solution was heated at 110 8C for 10 h, and the precipitate was filtered off, obtaining a yellow solid (0.75 g, 74 %). 1H NMR (CDCl3, 300 MHz): d = 3.85 (s, 3 H), 6.87 (d, J = 2.8 Hz, 1 H), 6.89 (dd, J = 2.8, 8.8 Hz, 1 H), 8.04 (s, 1 H), 7.33 (d, J = 8.8 Hz, 1 H), 7.94 ppm (br, 1 H).

Chemistry High-grade commercial products were used as starting materials, unless otherwise specified in this section. Solvents were of analytical grade. Melting points were determined in open glass capillaries, using a Bchi apparatus and are uncorrected. 1H NMR and 13C NMR spectra were recorded on Varian Gemini spectrometers and chemical shifts are reported as parts per million (ppm d value) relative to the peak for tetramethylsilane (TMS) as internal standard. Standard abbreviations indicating spin multiplicities are given as follows: s (singlet), d (doublet), t (triplet), br (broad), q (quartet), or m (multiplet). Mass spectra were recorded on a Waters ZQ 4000 apparatus operating in electrospray mode (ES). Wherever analyses are only in 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

General procedures for compounds 12, 13, 19, and 20 CsCO3 (0.65 g, 1.8 mmol) was added to a solution of hydroxycoumarin derivative (0.50 g, 1.8 mmol) in DMF (5 mL), and the mixture was stirred at RT and under N2 atmosphere for 30 min. Subsequently, a mixture of w-chloroalkyldimethylamine hydrochloride (2.0 mmol), CsCO3 (0.65 g, 1.8 mmol) in DMF (5 mL) was added. The resulting reaction mixture was heated at 110 8C, under N2 atmosphere, for 8 h. On cooling, water was added (100 mL) and the resulting solid was collected by filtration. The crude product was purified by flash column chromatography on silica gel using a mixture of CH2Cl2/MeOH (95:05) as eluent. ChemMedChem 0000, 00, 1 – 13

&8&


CHEMMEDCHEM FULL PAPERS 6-(2-(Dimethylaminoethoxy)-7-hydroxy-3-(4-methoxyphenyl)-2Hchromen-2-one (12): Starting from 11 and 2-chloro-N,N-dimethylethanamine hydrochloride (0.29 g), compound 12 was obtained (0.48 g, 75 %); mp: 122.2 8C. 1H NMR (CDCl3, 300 MHz): d = 3.10 (s, 6 H), 3.81 (t, J = 6.4 Hz, 2 H), 4.59 (t, J = 6.4 Hz, 2 H), 6.96–6.99 (m, 4 H), 7.25 (d, J = 8.8 Hz, 1 H), 7.69 (d, J = 8.8 Hz, 2 H), 7.72 ppm (s, 1 H). 6-(3-(Dimethylaminopropoxy)-3-(4-methoxyphenyl)-2H-chromen-2-one (13): Starting from 11 and 3-chloro-N,N-dimethylpropan-1-amine hydrochloride (0.31 g), compound 13 was obtained (0.45 g, 71 %), mp: 129.3 8C. 1H NMR (CDCl3, 300 MHz): d = 1.95– 2.05 (m, 2 H), 2.29 (s, 6 H), 2.50 (t, J = 6.4 Hz, 2 H), 3.85 (s, 3 H), 4.06 (t, J = 6.4 Hz, 2 H), 6.96–7.06 (m, 4 H), 7.25 (d, J = 8.8 Hz, 1 H), 7.64 (s, 1 H), 7.66 ppm (d, J = 8.8 Hz, 2 H). 3-(2-Dimethylaminoethoxy)-7-methoxy-2H-chromen-2-one (19): Starting from 18 (0.34 g) and 2-chloro-N,N-dimethylethanamine hydrochloride (0.29 g), compound 19 was obtained (0.46 g, 97 %), mp: 121–123 8C. 1H NMR (CDCl3, 300 MHz): d = 2.80 (t, J = 5.8 Hz, 2 H), 2.25 (s, 6 H), 3.85 (s, 3 H), 4.09 (t, J = 5.8 Hz, 2 H), 6.79 (d, J = 2.8 Hz, 1 H), 6.82 (dd, J = 2.8, 8.8 Hz, 1 H), 6.86 (s, 1 H), 7.28 (d, J = 8.8 Hz, 1 H), 7.94 (br, 1 H), 8.33 ppm (s, 1 H). 3-(3-Dimethylaminopropoxy)-7-methoxy-2H-chromen-2-one (20): Starting from 18 (0.34 g) and 3-chloro-N,N-dimethylpropan-1amine hydrochloride (0.31 g), compound 20 was obtained (0.46 g, 97 %), mp: 125–126 8C. 1H NMR (CDCl3, 300 MHz): d = 2.00–2.04 (m, 2 H), 2.22 (s, 6 H), 2.45 (t, J = 5.8 Hz, 2 H), 3.79 (s, 3 H), 4.02 (t, J = 5.8 Hz, 2 H), 6.74 (d, J = 2.8 Hz, 1 H), 6.78 (dd, J = 2.8, 8.8 Hz, 1 H), 6.82 (s, 1 H), 7.22 ppm (d, J = 8.8 Hz, 1 H).

General procedure for methyl ether cleavage Procedure A (compounds 14, 15): A 1.0 m CH2Cl2 solution of BBr3 (2.6 mL, 2.6 mmol) was slowly added to a solution of the methoxy derivative (0.48 g, 1.3 mmol) in dry CH2Cl2 at 70 8C and under N2 atmosphere. The reaction mixture was then stirred at the same temperature for 1 h and then at RT for 6 h. The resulting solution was added to a mixture of ice and aqueous NaHCO3 (10 mL) and extracted with EtOAc (3 50 mL). The solvent was dried (Na2SO4) and evaporated under reduced pressure. The crude residue was purified by chromatography on silica gel using a mixture of CH2Cl2/MeOH/NH4OH (90:9.9: 0.1) as eluent. Procedure B (compounds 21, 22): The methoxy derivative (1.3 mmol) was dissolved in HBr solution (33 % in AcOH, 28 mL) and the solution was heated at 80 8C for 14 h. The reaction mixture was allowed to stand overnight, ethyl ether was then added and the solid formed was collected by vacuum filtration. The crude residue was purified by chromatography on silica gel using a mixture of CH2Cl2/EtOH/NH4OH (90:9.9: 0.1) as eluent. 6-(2-(Dimethylaminoethoxy)-3-(4-hydroxyphenyl)-2H-chromen-2one (14): Following procedure A and starting from 12 (0.48 g, 1.3 mmol), compound 14 was obtained (0.28 g, 66 %), mp: 122– 123 8C. 1H NMR (CD3COCD3, 400 MHz): d = 3.10 (s, 6 H), 3.79–3.85 (m, 2 H), 4.55–4.62 (m, 2 H), 6.96 (d, J = 8.8 Hz, 2 H), 7.29–7.45 (m, 3 H), 7.62 (d, J = 8.8 Hz, 2 H), 8.94 ppm (br, 1 H). 6-(3-(Dimethylaminopropoxy)-3-(4-hydroxyphenyl)-2H-chromen2-one (15): Following procedure A and starting from 13 (0.45 g, 1.3 mmol), compound 14 was obtained (0.32 g, 72 %), mp: 129– 130 8C. 1H NMR (CD3COCD3, 400 MHz): d = 1.96–2.08 (m, 2 H), 2.39 (s, 6 H), 2.47 (t, J = 6.4 Hz, 2 H), 4.06 (t, J = 6.4 Hz, 2 H), 6.96 (d, J = 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemmedchem.org 8.8 Hz, 2 H), 7.21–7.30 (m, 2 H), 7.45 (s, 1 H), 7.67 (d, J = 8.8 Hz, 2 H), 8.03 ppm (br, 1 H). 3-(2-Dimethylaminoethoxy)-7-hydroxy-2H-chromen-2-one (21): Following procedure B and starting from 19 (0.34 g), compound 21 was obtained (0.12 g, 40 %), mp: 143–145 8C. 1H NMR (CD3OD, 400 MHz): d = 2.47 (s, 6 H), 2.93 (t, J = 5.4 Hz, 2 H), 4.17 (t, J = 5.4 Hz, 2 H), 6.71 (d, J = 2.6 Hz, 1 H), 6.82 (dd, J = 2.4, 8.4 Hz, 1 H), 7.24 (s, 1 H), 7.38 ppm (d, J = 8.4 Hz, 1 H). 3-(3-Dimethylaminopropoxy)-7-hydroxy-2H-chromen-2-one (22): Following procedure B and starting from 20 (0.34 g), compound 21 was obtained (0.12 g, 40 %), mp: 143–145 8C. 1H NMR (CD3OD, 400 MHz): d = 2.22–2.05 (m, 2 H), 2.37 (s, 6 H), 2.63 (t, J = 5.4 Hz, 2 H), 4.02 (t, J = 5.4 Hz, 2 H), 7.00 (d, J = 2.6 Hz, 1 H), 6.73 (dd, J = 2.4, 8.4 Hz, 1 H), 7.11 (s, 1 H), 7.31 ppm (d, J = 8.4 Hz, 1 H).

General procedure for the synthesis of compounds 3–10 The phenol derivative (14–16, 20, 21) (0.1 g, 0.29 mmol) was dissolved in anhydrous DMF (7 mL) and a base (Cs2CO3 or K2CO3) was added, the mixture was stirred at RT for 1 h. Then, 2-bromoanthracene-1,4-dione or 2-bromonaphthalene-1,4-dione was added and the reaction mixture was stirred at RT for 3 h. The resulting solution was put into an ice/water mixture (15 mL) and the solid formed was collected by vacuum filtration. The crude was purified by flash column chromatography using a mixture of CH2Cl2/MeOH (95:5) as eluent. 2-{4-[6-(3-Dimethylaminopropoxy)-2-oxo-2H-chromen-3-yl]phenoxy}naphthalene-1,4-dione (3): Starting from 15, Cs2CO3 (0.10 g, 0.29 mmol), and 2-bromonaphthalene-1,4-dione (0.07 g, 0.29 mmol), compound 3 was obtained (0.07 g, 65 %), mp: 129– 130 8C. 1H NMR (CDCl3, 400 MHz): d = 1.96–2.24 (m, 2 H), 2.29 (s, 6 H), 2.45 (t, J = 6.4 Hz, 2 H), 4.11 (t, J = 6.4 Hz, 2 H), 5.30 (s, 1 H), 6.05 (s, 1 H), 6.96 (d, J = 8.8 Hz, 2 H), 7.21–7.30 (m, 2 H), 7.45 (s, 1 H), 7.67 (d, J = 8.8 Hz, 2 H), 7.75–7.85 (m, 2 H), 8.01–8.06 (m, 1 H), 8.09– 8.12 ppm (m, 1 H); 13C NMR (CDCl3, 400 MHz): d = 27.51, 45.74, 57.91, 68.54, 111.27, 112.22, 113.32, 113.45, 118.01, 122.55, 126.66, 127.84, 133.48, 135.20, 140.75, 150.22, 152.32, 156.88, 157.33, 160.77, 179.10, 181.66 ppm; IR (Nujol) n˜ = 1722, 1680, 1645, 1595, 1584 cm1; MS (ESI + ) m/z: 496 [M + H] + . 2-{4-[6-(2-Dimethylaminoethoxy)-2-oxo-2H-chromen-3-yl]phenoxy}naphthalene-1,4-dione (5): Starting from 14, Cs2CO3 (0.10 g, 0.29 mmol), and 2-bromonaphthalene-1,4-dione (0.07 g, 0.29 mmol), compound 5 was obtained (0.1 g, 89 %), mp: 127– 129 8C. 1H NMR (CDCl3, 400 MHz): d = 2.37 (s, 6 H), 2.78 (t, J = 6.4 Hz, 2 H), 4.12 (t, J = 6.4 Hz, 2 H), 5.35 (s, 1 H), 6.06 (s, 1 H), 7.03 (d, J = 2.8 Hz, 1 H), 7.15–7.28 (m, 4 H), 7.65–7.88 (m, 4 H) 8.02–8.13 (m, 1 H), 8.18–8.23 ppm (m, 1 H); 13C NMR (CDCl3, 400 MHz): d = 45.74, 59.91, 70.12, 111.33, 112.01, 113.27, 113.45, 118.84, 122.55, 126.64, 127.22, 133.95, 135.80, 140.43, 151.72, 155.72, 160.18, 177.33, 161.71, 178.90, 183.61 ppm; IR (Nujol) n˜ = 1719, 1677, 1636, 1589, 1570 cm1; MS (ESI + ) m/z: 482 [M + H] + . 2-{4-[6-(2-Dimethylaminoethoxy)-2-oxo-2H-chromen-3-yl]phenoxy}anthracene-1,4-dione (6): Starting from 14, Cs2CO3 (0.10 g, 0.29 mmol), and 2-bromoanthracene-1,4-dione (0.09 g, 0.29 mmol), compound 6 was obtained (0.11 g, 71 %), mp: 158–160 8C. 1H NMR (CDCl3, 400 MHz): d = 2.40 (s, 6 H), 2.65 (t, J = 6.4 Hz, 2 H), 4.10 (t, J = 6.4 Hz, 2 H), 6.06 (s, 1 H), 6.99 (d, J = 1.8 Hz, 1 H), 7.10 (dd, J = 2.8, 8.8 Hz, 1 H), 7.22 (d, J = 8.8 Hz, 2 H), 7.260(s, 1 H), 7.26 (d, J = 8.8 Hz, 1 H), 7.75 (m, 4 H), 7.82 (d, J = 8.8 Hz, 2 H), 8.06–8.08 (m, 1 H), 8.19– 8.21 ppm (m, 1 H); 13C NMR (CDCl3, 400 MHz): d = 45.45, 59.33, 70.41, 11.35, 112.55, 113.47, 117.68, 118.99, 122.22, 124.68, 129.33, ChemMedChem 0000, 00, 1 – 13

&9&


CHEMMEDCHEM FULL PAPERS 130.77, 131.73, 131.99, 137.25, 138.46, 140.32, 149.85, 151.32, 158.66, 157.64, 160.81, 178.36, 186.54 ppm; IR (Nujol) n˜ = 1711, 1659, 1625, 1597, 1574 cm1; MS (ESI + ) m/z: 532 [M + H] + . 2-{4-[7-(1,4-Dioxo-1,4-dihydronaphthalen-2-yloxy)-2-oxo-2Hchromen-3-yl]phenoxy}naphthalene-1,4-dione (4): Starting from 16, Cs2CO3 (0.20 g, 0.56 mmol), and 2-bromonaphthalene-1,4-dione (0.14 g, 0.56 mmol), compound 4 was obtained (0.055 g, 35 %), mp: 171–173 8C. 1H NMR (CD3COCD3, 400 MHz): d = 6.01(s, 1 H), 6.09 (s, 1 H), 7.05 (dd, J = 8.4 Hz e 1.8 Hz, 1 H), 7.12 (d, J = 1.8 Hz, 1 H), 7.61 (d, J = 8.4 Hz, 1 H), 7.71–7.83 (m, 8 H), 8.02–8.13 (m, 2 H), 8.18–8.23 ppm (m, 2 H); 13C NMR (CD3COCD3, 400 MHz): d = 103.11, 112.22, 115.99, 117.62, 118.41, 122.98, 124.32, 126.42, 127.88, 127.95, 130.54, 131.21, 132.72, 133.69, 135.41, 139.71, 154.22, 156.97, 157.11, 157.82, 160.41, 178.25, 183.45 ppm; IR (Nujol): n˜ = 1734, 1701, 1682, 1673, 1656, 1585, 1565 cm1; MS (ESI + ) m/z: 567 [M + H] + . 2-{[3-(2-Dimethylaminoethoxy)-2-oxo-2H-chromen-7-yl]oxy}naphthalene-1,4-dione (7): Starting from 21 (0.072 g), K2CO3 (0.04 g, 0.29 mmol), and 2-bromoanthracene-1,4-dione (0.09 g, 0.29 mmol), compound 7 was obtained (0.013 g, 11 %), mp: 164– 165 8C. 1H NMR (CDCl3, 400 MHz): d = 2.36 (s, 6 H), 2.84 (t, J = 6.0 Hz, 2 H), 4.15 (t, J = 6.0 Hz, 2 H), 6.04 (s, 1 H), 6.88 (s, 1 H), 7.04 (dd, J = 2.4 and 8.4 Hz, 1 H), 7.12 (d, J = 2.4 Hz, 1 H), 7.47 (d, J = 8.4 Hz, 1 H), 7.77–8.06 (m, 4 H), 8.07–8.09 (m, 1 H), 8.20–8.21 ppm (m, 1 H); 13 C NMR (CDCl3): d = 46.23, 57.70, 68.16, 109.53, 113.35, 114.54, 117.85, 118.32, 126.56, 127.18, 128.15, 131.23, 132.08, 133.96, 134.83, 144.29, 150.66, 152.80, 159.94, 168.21, 176.73, 184.85 ppm; IR (Nujol) n˜ = 1721, 1676, 1651, 1593, 1574 cm1; MS (ESI + ) m/z: 406 [M + H] + . 2-{[3-(2-Dimethylaminoethoxy)-2-oxo-2H-chromen-7-yl]oxy}anthracene-1,4-dione (8): Starting from 21 (0.072 g), K2CO3 (0.04 g, 0.29 mmol), and 2-bromoanthracene-1,4-dione (0.09 g, 0.29 mmol), compound 8 was obtained (0.030 g, 29 %), mp: 181–183 8C. 1 H NMR (CDCl3, 400 MHz): d = 2.33 (s, 6 H), 2.87 (t, J = 6.0 Hz, 2 H), 4.17 (t, J = 6.0 Hz, 2 H), 6.16 (s, 1 H), 6.90 (s, 1 H), 7.09 (dd, J = 2.4, 8.4 Hz, 1 H), 7.16 (d, J = 2.4 Hz, 1 H), 7.49 (d, J = 8.4 Hz, 1 H), 7.70– 7.72 (m, 2 H), 8.06–8.12 (m, 2 H), 8.60 (s, 1 H), 8.75 ppm (s, 1 H); 13 C NMR (CDCl3, 400 MHz): d = 46.23, 58.70, 68.11, 109.54, 113.37, 114.34, 117.65, 118.29, 126.56, 127.12, 128.21, 131.57, 131.97, 131.99, 132.08, 133.96, 134.83, 144.33, 150.51, 152.80, 159.82, 168.32, 176.73, 184.79 ppm; IR (Nujol) n˜ = 1728, 1676, 1645, 1594, 1583 cm1; MS (ESI + ) m/z: 456 [M + H] + . 2-{[3-(3-Dimethylaminopropoxy)-2-oxo-2H-chromen-7-yl]oxy}naphthalene-1,4-dione (9): Starting from 22 (0.072 g), K2CO3 (0.04 g, 0.29 mmol), and 2-bromoanthracene-1,4-dione (0.09 g, 0.29 mmol) compound 7 was obtained (0.059 g, 50 %), mp: 165– 167 8C. 1H NMR (CDCl3, 400 MHz): d = 3.00–2.12 (m, 2 H), 2.11 (s, 6 H), 2.50 (t, J = 6.0 Hz, 2 H), 4.08 (t, J = 6.0 Hz, 2 H), 6.02 (s, 1 H), 6.87 (s, 1 H), 6.93 (dd, J = 2.4 and 8.4 Hz, 1 H), 7.05 (d, J = 2.4 Hz, 1 H), 7.43 (d, J = 8.4 Hz, 1 H), 7.77–7.88 (m, 4 H), 8.00–8.10 (m, 1 H), 8.17– 8.19 ppm (m, 1 H); 13C NMR (CDCl3, 400 MHz): d = 20.20, 46.23, 57.70, 68.00, 109.63, 113.67, 114.12, 117.34, 118.87, 126.12, 127.43, 128.26, 131.78, 132.00, 134.17, 134.27, 144.46, 150.76, 152.92, 160.24, 163.91, 176.76, 184.88 ppm; IR (Nujol) n˜ = 1716, 1659, 1625, 1591, 1573 cm1; MS (ESI + ) m/z: 420 [M + H] + . 2-{[3-(3-Dimethylaminopropoxy)-2-oxo-2H-chromen-7-yl]oxy}anthracene-1,4-dione (10): Starting from 22 (0.072 g), K2CO3 (0.04 g, 0.29 mmol), and 2-bromoanthracene-1,4-dione (0.09 g, 0.29 mmol), compound 8 was obtained (0.076 g, 74 %), mp: 196– 197 8C. 1H NMR (CDCl3, 400 MHz): d = 2.01–2.12 (m, 2 H), 2.29 (s, 6 H), 2.53 (t, J = 6.0 Hz, 2 H), 4.13 (t, J = 6.0 Hz, 2 H), 6.16 (s, 1 H), 6.91


www.chemmedchem.org (s, 1 H), 7.09 (dd, J = 2.4 and 8.4 Hz, 1 H), 7.11 (d, J = 2.4 Hz, 1 H), 7.48 (d, J = 8.4 Hz, 1 H), 7.72–7.74 (m, 2 H), 8.06–8.12 (m, 2 H), 8.61 (s, 1 H), 8.77 ppm (s, 1 H); 13C NMR (CDCl3, 400 MHz): d = 20.25, 46.13, 58.34, 68.56, 109.72, 113.96, 114.11, 117.22, 118.67, 126.56, 127.75, 128.69, 131.43, 131.51, 131.55, 131.84, 131.99, 132.30, 133.76, 134.43, 137.12, 137.66, 144.33, 150.50, 152.85, 159.82, 168.32, 176.57, 184.79 ppm; IR (Nujol) n˜ = 1720, 1656, 1636, 1590, 1575 cm1; MS (ESI + ) m/z: 470 [M + H] + .

Computational studies The three-dimensional structures of TbGAPDH (PDB ID: 2X0N) and TbTR (PDB ID: 2WPF) were downloaded from the Protein Data Bank.[34] Then, the structures were prepared using the Protein Preparation Wizard[35] workflow as follows: assigning bond orders, adding hydrogen atoms, filling in missing loops or side chains, deleting water, cofactors, ligands, metals, and ions. Following this step, a geometric optimization of the whole system to a RMSD of 0.3 ﬀ was performed using OPLS-2005 force field. The 3D molecular structures of compounds were built with the Schrçdinger software. The energy minimization was carried out using the OPLS-2005 force field. Then, all the compounds were prepared by LigPrep module 2.4.[36] Docking simulations were performed with Glide 5.6[37] in the extra-precision mode within a region centered on Cys166 of GAPDH and Pro398 of TR. The default grid size was adopted from the Glide program and the scaling factor for protein van der Waals radii was set as 0.8 in the receptor grid generation. No constraints were applied for all the docking studies. For each compound, a maximum of ten poses were saved after the docking process. Finally, structures with lowest Glide XP scores were selected for further analysis.

Biological assays Kinetic analysis of GAPDH Recombinant TbGAPDH was expressed and purified in the Fato laboratory, FaBiT Department, University of Bologna, as previously described.[19] TbGAPDH activity was assayed spectrophotometrically by following NAD + reduction at 340 nm in triethanolamine (TEA) buffer (10 mm TEA, 1.7 mm NaHCO3, 100 mm KCl, 5 mm MgSO4, and 1 mm EDTA) pH 7.6 at 25 8C, as reported by Wiggers et al.[29] The reaction mixtures contained TEA Buffer, pH 7.6, TbGAPDH (ranging from 200 to 300 mg of protein), 400 mm NAD + , 500 mm KH2PO4, 300 mm GAP and varying concentrations of inhibitors in a total volume of 1 mL. Stock solutions of the inhibitors were prepared in DMSO. Specific activity was expressed as mmol min1 mg1 of purified enzyme and inhibition data were presented as percentage of activity compared with the controls without inhibitor. Experiments were carried out in triplicate.

Fluorescence experiments Fluorescence experiments were carried out with a Fluorimeter Jasco FP 777 at 25 8C, in TEA buffer, using 10 mm of compound 10 and of the fragment 3-(3-aminopropoxy)-7-hydroxycoumarin, in the presence and absence of 50 mm DTT and N-acetylcysteine (NAC). Wavelength values of excitation (335 nm) and emission (485 nm) were acquired experimentally in the specific conditions of buffer and temperature described in the previous paragraph. In a different set of experiments 0.030 mg mL1 of purified TbGAPDH ChemMedChem 0000, 00, 1 – 13

&10&



www.chemmedchem.org

was added to 10 mm of compound 10 dissolved in TEA buffer. Fluorescence spectra were recorded every minute for 15 min.

Kinetic analysis of trypanothione reductase and glutathione reductase The pET3a-TbTR plasmid was a kind gift of Dr. Alan Fairlamb (Dundee, UK).[38] Both recombinant TcTR and TbTR were prepared following the procedure described by Sullivan and Walsh.[39] A sample of recombinant human GR (hGR) was a kind gift of Dr. Heiner Schirmer (Heidelberg, Germany). Trypanothione disulfide (TS2) was generated enzymatically as described previously.[40] TR activity was measured in a total volume of 1 mL of 40 mm HEPES, 1 mm EDTA, pH 7.5, containing 100 mm NADPH and varying concentrations of TS2 and/or inhibitors as well as ~ 10 mU TR. The absorption decrease due to NADPH oxidation was followed at 340 nm and 25 8C. Stock solutions of the inhibitors were prepared in DMSO. The Ki value was determined using Equation (1). Ki ¼

½I Vmax Vmax ðobsÞ

1

Rat skeletal myoblasts (L6 cells) and the standard compound podophyllotoxin were used for the assays, which were performed in 96well microtiter plates. Each well contained 100 mL of RPMI 1640 medium supplemented and 10 % fetal bovine serum, and 4 104 L6 cells. Serial drug dilutions of seven threefold dilution steps were prepared. The assay was run and evaluated following the same protocol as used for T. b. rhodesiense.

ð1Þ

The ability of the compounds to induce the oxidase activity of TR was measured in the presence of 100 mm NADPH, 40 mm of the compounds (or an equivalent volume of 10 mL of DMSO for blanks), and 1.29 U of TR in a total volume of 1 mL. Under these conditions the spontaneous NADPH oxidation resulted in an absorption decrease of 0.0026 min. This value was subtracted from the enzyme-catalyzed rate. In the second type of oxidase assays, the reaction mixture also contained 25 mm of Cyt c (Fe3 +). The absorption increase at 550 nm due to the single-electron reduction of Cyt c was followed. Finally, 67 mg of SOD was added to the assay mixture to prevent cytochrome c reduction by superoxide anions. The activity of hGR was measured in 1 mL of 20.5 mm KH2PO4, 26.5 mm K2HPO4, EDTA 1 mm, KCl 200 mm, pH 7.5 containing saturating substrate concentrations of 100 mm NADPH and 1 mm GSSG; 50 mU and 17 U GR were used for the inhibition and redox cycling experiments, respectively.

Trypanosome growth inhibition assays The in vitro growth inhibition activity of compounds against T. b. rhodesiense and T. cruzi as well as the cytotoxicity against rat skeletal myoblasts (L6 cells) were determined as described previously.[41] Briefly, T. b. rhodesiense, STIB 900 strain, and the standard drug, melarsoprol, were used for the assay. Minimum Essential Medium (50 mL) with 15 % heat inactivated horse serum was added to each well of a 96-well microtiter plate. Serial drug dilutions of seven threefold dilution steps were prepared. Then 104 bloodstream forms in 50 mL were added to each well and the plate was incubated at 37 8C under a 5 % CO2 atmosphere for 72 h. Alamar Blue (10 mL) was then added to each well and incubation continued for a further 2–4 h. Then, the plates were read with a Spectramax Gemini XS microplate fluorimeter (Molecular Devices Cooperation, Sunnyvale, CA, USA) using an excitation wavelength of 536 nm and an emission wavelength of 588 nm. Data were analyzed using the microplate reader software Softmax Pro (Molecular Devices Cooperation, Sunnyvale, CA, USA). 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

For assays of T. cruzi growth inhibition the Tulahuen strain C2C4 containing the b-galactosidase (Lac Z) gene and the standard drug benznidazole were used. Rat skeletal myoblasts (L6 cells) were seeded in 96-well microtiter plates containing RPMI 1640 medium (100 mL) with 10 % FBS. After 24 h the medium was removed and replaced by 100 mL medium containing 5000 trypomastigote forms per well. After 48 h, the medium was removed from the wells and replaced by 100 mL fresh medium with or without a serial drug dilution of seven threefold dilution steps. After 96 h of incubation, the substrate CPRG/Nonidet (50 mL) was added to all wells. A color reaction developed within 2–6 h and could be read photometrically at 540 nm. Data were transferred into the graphic program Softmax Pro (Molecular Devices), which calculated EC50 values.

Acknowledgements This research was supported by grants from the University of Bologna (M.L.B.). G.V. thanks the COST Action CM-0801 “New Drugs for Neglected Diseases” of the EC for a grant that enabled him to work for two months in the laboratory of R.L.K.-S. at the Biochemistry Center (BZH) in Heidelberg. Keywords: framework combinations · glyceraldehyde-3phosphate dehydrogenase · multitarget ligands · neglected tropical diseases · trypanothione reductase [1] M. P. Barrett, R. J. S. Burchmore, A. Stich, J. O. Lazzari, A. C. Frasch, J. J. Cazzulo, S. Krishna, Lancet 2003, 362, 1469 – 1480. [2] P. G. E. Kennedy, Lancet Neurol. 2013, 12, 186 – 194. [3] F.-X. Lescure, G. Le Loup, H. Freilij, M. Develoux, L. Paris, L. Brutus, G. Pialoux, Lancet Infect. Dis. 2010, 10, 556 – 570. [4] P. J. Hotez, B. Pecoul, PLoS Neglected Trop. Dis. 2010, 4, e718. [5] G. Priotto, S. Kasparian, W. Mutombo, D. Ngouama, S. Ghorashian, U. Arnold, S. Ghabri, E. Baudin, V. Buard, S. Kazadi-Kyanza, M. Ilunga, W. Mutangala, G. Pohlig, C. Schmid, U. Karunakara, E. Torreele, V. Kande, Lancet 2009, 374, 56 – 64. [6] a) A. R. Renslo, J. H. McKerrow, Nat. Chem. Biol. 2006, 2, 701 – 710; b) M. L. Bolognesi, Curr. Top Med. Chem. 2011, 11, 2824 – 2833. [7] R. Morphy, Z. Rankovic, Drug Discovery Today 2007, 12, 156 – 160. [8] a) A. Cavalli, M. L. Bolognesi, J. Med. Chem. 2009, 52, 7339 – 7359; b) P. M. Njogu, K. Chibale, Curr. Med. Chem. 2013, 20, 1715 – 1742. [9] M. L. Bolognesi, Curr. Med. Chem. 2013, 20, 1639 – 1645. [10] B. L. Roth, ACS Med. Chem. Lett. 2013, 4, 316 – 318. [11] a) M. A. Comini, L. Floh in Trypanosomatid Diseases, Wiley-VCH, Weinheim, 2013, pp. 167 – 199; b) M. Gualdrn-Lpez, P. A. M. Michels, W. QuiÇones, A. J. C ceres, L. Avil n, J.-L. Concepcin in Trypanosomatid Diseases, Wiley-VCH, Weinheim, 2013, pp. 121 – 151. [12] A. F. Coley, H. C. Dodson, M. T. Morris, J. C. Morris, Mol. Biol. Int. 2011, 2011, Article ID 123702. [13] P. A. M. Michels, F. Bringaud, M. Herman, V. Hannaert, Biochim. Biophys. Acta Mol. Cell Res. 2006, 1763, 1463 – 1477. [14] R. L. Krauth-Siegel, M. A. Comini, T. Schlecker, Subcell. Biochem. 2007, 44, 231 – 251. [15] R. L. Krauth-Siegel, M. A. Comini, Biochim. Biophys. Acta Gen. Subj. 2008, 1780, 1236 – 1248. [16] A. Schmidt, R. L. Krauth-Siegel, Curr. Top Med. Chem. 2002, 2, 1239 – 1259.

ChemMedChem 0000, 00, 1 – 13

&11&


CHEMMEDCHEM FULL PAPERS [17] R. L. Krauth-Siegel, O. Inhoff, Parasitol. Res. 2003, 90, S77 – S85. [18] M. L. Bolognesi, F. Lizzi, R. Perozzo, R. Brun, A. Cavalli, Bioorg. Med. Chem. Lett. 2008, 18, 2272 – 2276. [19] S. Pieretti, J. R. Haanstra, M. Mazet, R. Perozzo, C. Bergamini, F. Prati, R. Fato, G. Lenaz, G. Capranico, R. Brun, PLoS Neglected Trop. Dis. 2013, 7, e2012. [20] F. Lizzi, G. Veronesi, F. Belluti, C. Bergamini, A. Lopez-Sanchez, M. Kaiser, R. Brun, R. L. Krauth-Siegel, D. G. Hall, L. Rivas, M. L. Bolognesi, J. Med. Chem. 2012, 55, 10490 – 10500. [21] S. Rizzo, A. Cavalli, L. Ceccarini, M. Bartolini, F. Belluti, A. Bisi, V. Andrisano, M. Recanatini, A. Rampa, ChemMedChem 2009, 4, 670 – 679. [22] A. A. Kudale, J. Kendall, C. C. Warford, N. D. Wilkins, G. J. Bodwell, Tetrahedron Lett. 2007, 48, 5077 – 5080. [23] K. Takaishi, M. Izumi, N. Baba, K. Kawazu, S. Nakajima, Bioorg. Med. Chem. Lett. 2008, 18, 5614 – 5617. [24] A. Cavalli, F. Lizzi, S. Bongarzone, R. Brun, R. L. Krauth-Siegel, M. L. Bolognesi, Bioorg. Med. Chem. Lett. 2009, 19, 3031 – 3035. [25] D. C. Rees, M. Congreve, C. W. Murray, R. Carr, Nat. Rev. Drug Discovery 2004, 3, 660 – 672. [26] a) G. Bottegoni, A. D. Favia, M. Recanatini, A. Cavalli, Drug Discovery Today 2012, 17, 23 – 34; b) X. H. Ma, Z. Shi, C. Tan, Y. Jiang, M. L. Go, B. C. Low, Y. Z. Chen, Pharm. Res. 2010, 27, 739 – 749. [27] K. V. Santhosh, D. V. Kumar S. K. Venkatesan, V. K. Dubey, Sci. World J. 2012, 2012, Article ID 963658 . [28] F. Pav¼o, M. S. Castilho, M. T. Pupo, R. L. A. Dias, A. G. Correa, J. B. Fernandes, M. F. G. F. da Silva, J. Mafezoli, P. C. Vieira, G. Oliva, FEBS Lett. 2002, 520, 13 – 17. [29] H. J. Wiggers, J. Cheleski, A. Zottis, G. Oliva, A. D. Andricopulo, C. A. Montanari, Anal. Biochem. 2007, 370, 107 – 114.


www.chemmedchem.org [30] L. Salmon-Chemin, E. Buisine, V. Yardley, S. Kohler, M. A. Debreu, V. Landry, C. Sergheraert, S. L. Croft, R. L. Krauth-Siegel, E. Davioud-Charvet, J. Med. Chem. 2001, 44, 548 – 565. [31] B. Stump, M. Kaiser, R. Brun, R. L. Krauth-Siegel, F. Diederich, ChemMedChem 2007, 2, 1708 – 1712. [32] K. Blumenstiel, R. Schçneck, V. Yardley, S. L. Croft, R. L. Krauth-Siegel, Biochem. Pharmacol. 1999, 58, 1791 – 1799. [33] R. Pink, A. Hudson, M.-A. Mouries, M. Bendig, Nat. Rev. Drug Discovery 2005, 4, 727 – 740. [34] RCSB Protein Data Bank, http://pdb.org/pdb/home/home.do (accessed December 13, 2013). [35] Schrçdinger Suite 2010 Protein Preparation Wizard: Epik version 2.1, Impact version 5.6, and Prime version 2.2, Schrçdinger LLC, New York, NY (USA), 2010. [36] LigPrep, Schrçdinger LLC, New York, NY (USA), 2010. [37] Glide, Schrçdinger LLC, New York, NY (USA), 2010. [38] D. C. Jones, A. Ariza, W.-H. A. Chow, S. L. Oza, A. H. Fairlamb, Mol. Biochem. Parasitol. 2010, 169, 12 – 19. [39] F. X. Sullivan, C. T. Walsh, Mol. Biochem. Parasitol. 1991, 44, 145 – 147. [40] M. A. Comini, N. Dirdjaja, M. Kaschel, R. L. Krauth-Siegel, Int. J. Parasitol. 2009, 39, 1059 – 1062. [41] I. Orhan, B. S¸ener, M. Kaiser, R. Brun, D. Tasdemir, Mar. Drugs 2010, 8, 47 – 58.

Received: October 9, 2013 Revised: December 5, 2013 Published online on && &&, 0000

ChemMedChem 0000, 00, 1 – 13

&12&


FULL PAPERS Mind the GAPDH/TR: The trypanosomiases are very serious yet neglected tropical parasitic diseases that are difficult to treat, as effective drugs and vaccines are not yet available. Starting with a quinone–coumarin hybrid scaffold, the design and biological evaluation of single small molecules that can modulate multiple vital targets in the parasites’ metabolic pathways is presented.


F. Belluti, E. Uliassi, G. Veronesi, C. Bergamini, M. Kaiser, R. Brun, A. Viola, R. Fato, P. A. M. Michels, R. L. Krauth-Siegel, A. Cavalli, M. L. Bolognesi* && – && Toward the Development of DualTargeted Glyceraldehyde-3-phosphate Dehydrogenase/Trypanothione Reductase Inhibitors against Trypanosoma brucei and Trypanosoma cruzi

ChemMedChem 0000, 00, 1 – 13

&13&


Subcellular distribution of trypanothione reductase in bloodstream and procyclic forms of Trypanosoma brucei.

Inhibitors of Trypanosoma brucei trypanothione reductase: comparative molecular field analysis modeling and structural basis for selective inhibition.

Cloning, sequencing, overproduction and purification of trypanothione reductase from Trypanosoma cruzi.

Nanomolar Inhibitors of Trypanosoma brucei RNA Triphosphatase.

Activity of Bisnaphthalimidopropyl Derivatives against Trypanosoma brucei.

De Novo Genome Assembly Shows Genome Wide Similarity between Trypanosoma brucei brucei and Trypanosoma brucei rhodesiense.

Nitric oxide-mediated cytostatic activity on Trypanosoma brucei gambiense and Trypanosoma brucei brucei.

SUMOylation in Trypanosoma brucei.

Localization of serum resistance-associated protein in Trypanosoma brucei rhodesiense and transgenic Trypanosoma brucei brucei.

Novel pyrrolobenzoxaboroles: design, synthesis, and biological evaluation against Trypanosoma brucei.

Characterisation of melarsen-resistant Trypanosoma brucei brucei with respect to cross-resistance to other drugs and trypanothione metabolism.

R-Configuration of 4-Aminopyridyl-Based Inhibitors of CYP51 Confers Superior Efficacy Against Trypanosoma cruzi.

Trypanosoma cruzi but not Trypanosoma brucei fails to induce a chemiluminescent signal in a macrophage hybridoma cell line.

DNA methylation in Trypanosoma cruzi.

The Oral Antimalarial Drug Tafenoquine Shows Activity against Trypanosoma brucei.

Diminazene aceturate residues in the tissues of healthy, Trypanosoma congolense and Trypanosoma brucei brucei infected dogs.

Fetal IgG specificities against Trypanosoma cruzi antigens in infected newborns.

Trypanosoma evansi is alike to Trypanosoma brucei brucei in the subcellular localisation of glycolytic enzymes.

Cultivation of bloodstream Trypanosoma brucei.

Trypanosoma cruzi, organ transplant related.

Lipid composition of Trypanosoma cruzi.

Clathrin expression in Trypanosoma cruzi.

Evaluation of DL-alpha-difluoromethylornithine against susceptible and drug-resistant Trypanosoma brucei brucei.

Trypanosoma cruzi-Host Cell Interaction.