Synthesis and Evaluation of Non-peptidic Cysteine Protease Inhibitors of P. falciparum Derived from Etacrynic Acid

A series of etacrynic acid derivatives was synthesized and screened for their in vitro activity against Plasmodium falciparum, as well as their activity against recombinantly expressed falcipain-2 and -3. The two most active compounds of the series displayed IC50 values of 9.0 and 18.8 μM against Plasmodia.


Introduction
Malaria is still a major cause of global human morbidity and mortality, with Plasmodium falciparum being the deadliest among the four Plasmodium species that infect humans. Annually 350-500 million clinical malaria episodes occur, causing more than 1 million deaths each year [1]. The

OPEN ACCESS
increasing spread of P. falciparum due to inadequate vector control, increasing drug resistance, and the lack of effective vaccine strengthens the need for the development of novel drugs for the treatment of malaria [2,3]. New therapeutic strategies are therefore urgently needed. Cysteine proteases of P. falciparum represent attractive antiplasmodial drug targets due to their essential functions for the parasite erythrocytic cycle [4,5,6]. Among them are falcipain-2 and falcipain-3, which are located in the acidic food vacuole of the parasite, play a pivotal role in hemoglobin hydrolysis, and may also participate in erythrocyte rupture [5,7]. Disruption of the hemoglobin degradation pathway is lethal for the parasite. Development of compounds targeting falcipain-2 and falcipain-3 is therefore currently a focus of research. A recent study has further shown that individual cysteine protease inhibitors significantly reduce microgametogenesis in P. falciparum, pointing to a potential role of these inhibitors for transmission blocking strategies [8]. So far several peptidic and peptidomimetic as well as non-peptidic cysteine protease inhibitors have been synthesized and tested against malaria parasites [9][10][11][12]. In the present work a set of non-peptidic cysteine protease inhibitors derived from the loop diuretic etacrynic acid as a lead compound were evaluated for their potency to inhibit recombinantly expressed falcipain-2 and falcipain-3 [13][14][15]. Furthermore, activities against chloroquine-sensitive P. falciparum strain 3D7 and the chloroquine-resistant strain W2 were investigated. Several of these compounds have initially been tested against papain, the prototype cysteine protease of the CAC1 family, and against the SARS coronavirus main protease M pro , as described elsewhere [16,17].

Results and Discussion
In these previous studies compounds possessing an activated double bond revealed covalent, but reversible binding to the cysteine residue of the respective protease. In addition to the structural modifications implemented in previous work [16,17], namely within the compounds 1, 2, 8-21, we now included fluoro-substituted compounds 3-7, several analogues without an activated double bond 25-32, as well as derivatives with polar side chains 22-23, and finally, a biotin-labeled inhibitor 24. In summary, the structure of etacrynic acid was modified as follows (Scheme 1): A: substitution pattern of the aromatic ring, B: esters and amides, C: ortho-position of the double bond containing side chain (cpd. 2), D: removal of the double bond.
The inhibitors were synthesized according to previously described pathways [16,17] which are summarized in Scheme 2. Halogen substituted anisoles were subjected to Friedel-Crafts acylation yielding the corresponding phenolic ketones. Further alkylation of the phenolic hydroxyl functions yielded amides 28, 30 -32, and esters 1, 9 and 29. Introduction of the double bond was performed either by Mannich reaction with TMDM ( 3, 6, 7) or by aldol condensation with formaldehyde ( 4, 8). The latter yielded the free acids 4 and 8 due to concurrent hydrolysis of the ester function. The free acids were coupled to various amides to give the amides 5, 10 -24. Amides 26 and 27 without activated double bond were also synthesized by standard amide coupling methods.
Br-CH 2 -CO 2 R K 2 CO 3 , KI Recombinant falcipain-2 and falcipain-3 were produced as previously described [14,18]. Inhibitory activities against recombinant falcipain-2 and falcipain-3 were evaluated in fluorometric microplate assays using the substrates Cbz-Phe-Arg-AMC and Cbz-Leu-Arg-AMC (AMC, 7-amino-4-methylcoumarin) [19]. The cysteine protease inhibitor E-64 was used as a positive control [20]. The solvent DMSO was used as negative control. Compounds 1-6, 8-11 and 13-28 were tested in vitro against the CQ-sensitive 3D7 P. falciparum strain or the CQ-resistant W2 P. falciparum strain. IC 50 values are shown in Table 1. The in vitro data for the etacrynic acid derivatives are compared to those of the well known drug chloroquine and to E-64. In addition, the cytotoxicity of the inhibitor 23 was studied on human kidney epithelium cell-line 293T, as described previously [21,22], resulting in an IC 50 value of >160 μM. Inspection of the data in Table 1 allows the following conclusions to be drawn: in general, the etacrynic acid derivatives are weak or moderate inhibitors of falcipains and P. falciparum. Nevertheless, some structure-activity relationship can be found: The ethyl ester derivative (9) is more potent than the etacrynic acid (8), both against falcipain-2 and -3. This is in accordance with the results for papain and the SARS-CoV M pro . The dichlorosubstituted compounds (e.g. 9, 20) are better inhibitors than the mono-chloro-(e.g. 1) or fluorosubstituted (6) compounds, with one exception, namely inhibitor 5 (compared to 18). The α,ßunsaturated system appears to be favourable for activity against falcipains when comparing the analogous dichloro-compounds 20 and 27 or compounds 18 and 26. Short voluminous moieties such as the tertbutyl moiety of compounds 5, 28, 29, seem to be advantageous if combined with the fluorosubstituted aromate. In these cases, the activated double bond is apparently not essential for inhibition.
Nearly all etacrynic acid amides show better inhibition properties than the free acid (8). A longer acidic side-chain significantly enhances the activity (23, compared to 20, 21), making 23 the most active inhibitor of falcipains and P. falciparum within the series. However, insertion of an additional acidic group (e.g. 22) diminishes the inhibiting activity. The cytotoxicity/antiplasmodial ratio for the most active compound 23 is >8.5, indicating selectivity against the parasite. As the data against the target enzymes and the parasites do not correlate in all cases (e.g. 14), the question arises whether there are additional or other targets. In order to allow further affinity binding studies the biotinylated dichloro-substituted etacrynic acid amide 24 was included and synthesized according to the methods recently described [17,23] (Scheme 2). Notably, this compound emerged as the most potent inhibitor of falcipains and P. falciparum within the series.

Conclusions
In summary, this paper describes a comprehensive screening of non-peptidic Michael acceptors using etacrynic acid as lead structure. The best inhibition against recombinantly synthesized falcipain-2 and falcipain-3 revealed the compound 24. Moreover, this etacrynic acid amide as well as compound 23 displayed modest antiplasmodial activity in vitro with IC 50 values of 9 and 18.8 μM, respectively, which are in the range of the standard cysteine protease inhibitor E-64. In addition, the high IC 50 value of >160 μM for compound 23 obtained from cytotoxicity assays using the human kidney epithelium cell-line 293T indicates selectivity against the parasite. These results provide basic information for the development of further non-peptidic irreversible cysteine protease inhibitors with etacrynic acid amides as lead compounds. In addition, the good inhibitory properties of 24 allow further affinity binding studies.

General
Melting points were determined in open capillary on a melting point apparatus, model 530, from Büchi, Switzerland. NMR spectra were recorded on an AVANCE 400 MHz spectrometer from Bruker Biospin GmbH, Germany [solvent CDCl 3 (unless otherwise noted); 1 H-NMR, 400.13 MHz; 13 C-NMR, 100.61 MHz]. The optical rotation values were determined on a Perkin-Elmer 241 polarimeter. ESI mass spectra were recorded on an Agilent 1100 ion trap equipped with an Agilent HPLC system. Hydrostatic column chromatography was performed with silica gel 60 (0.063-0.2 mm). All solvents were purified and dried prior to use according to standard literature procedures. Chloroquine, E64 and DMSO (dimethyl sulfoxide) were obtained from Sigma-Aldrich, Deisenhofen, Germany. Alamar Blue ® was purchased from Trinova Biochem, Giessen, Germany. DMEM (Dulbecco's Modified Eagle's Medium) high glucose was delivered from Gibco/Sigma-Aldrich, Deisenhofen, Germany. Hygromycin was obtained from Merck, Darmstadt, Germany.
Method A: Introduction of the double bond. To substances 29, 30 or 32 (1 equiv.) and TMDM (N,N,N',N'-tetramethyldiaminomethane, 20 equiv.), acetic anhydride (20 equiv.) was added slowly. The mixture was heated under reflux at 85 °C. The reaction is followed by 1 H-NMR spectroscopy. After completion of the reaction the mixture was cooled to room temperature and saturated K 2 CO 3 solution was added until gas evolution stopped. The product was extracted with Et 2 O, washed with water and brine and dried with Na 2 SO 4 . The solvent was removed in vacuo and the product purified by column chromatography.
Method B: Syntheses of bromoacetamides. Bromoacetyl bromide (1 equiv.) in absolute CH 2 Cl 2 was cooled to -30 °C. Amine (1 equiv.) and triethylamine (1 equiv.) in absolute CH 2 Cl 2 were added dropwise. After the addition, the reaction mixture was warmed up to room temperature and stirred for 1-2 h further. The solvent was removed in vacuo, the product was dissolved in acetone and filtered off. The bromoacetamides were used directly without further purification.
Method C: Coupling of the bromoacetamide or bromoacetic acid ester with 1-(3-Fluoro-4hydroxyphenyl)butan-1-one. 1-(3-Fluoro-4-hydroxyphenyl)butan-1-one (1 equiv.), K 2 CO 3 (1.5 equiv.) and KI (0.1 equiv.) were refluxed in dry acetone for 1 h. Bromoacetamide (method B) or bromoacetic acid ester (2 equiv.) were added and the mixture heated to reflux for a further 4 h. The mixture was cooled to room temperature, filtered and the solvent was removed in vacuo. The product was dissolved in Et 2 O, washed with brine, KOH solution (5%) and water, and dried with Na 2 SO 4 . The solvent was removed in vacuo and the product was purified by column chromatography.  (4): 1-(3-fluoro-4-hydroxyphenyl)butan-1one. 2-Fluoroanisole (4.00 g, 31.7 mmol) and butyric acid chloride (5.07 g, 47.6 mmol) were dissolved under a N 2 atmosphere in absolute CH 2 Cl 2 (50 mL) and the mixture was cooled to 0-10 °C. AlCl 3 (6.35 g, 47.6 mmol) was added within 30 minutes, and the mixture was stirred for 2-3 h. An additional amount of AlCl 3 (6.35 g, 47.6 mmol) was added, and the mixture was heated under reflux for 2 h. The mixture was poured on ice and acidified with concentrated HCl to pH 1. Tartaric acid was added for complexation of aluminum until the solution was clear. The solution was extracted with Et 2 O and the organic layer was washed with KOH solution (10%) and brine. The organic layer was dried with Na 2 SO 4 and solvent was removed in vacuo. Yield  Ethyl-2-(4-butyryl-2-fluorophenoxy)acetate. 1-(3-Fluoro-4-hydroxyphenyl)butan-1-one (2.00 g, 11.0 mmol), bromoacetic acid ethyl ester (3.67 g, 22.0 mmol), K 2 CO 3 (2.28 g, 16.5 mmol) and KI (183 mg, 1.10 mmol) were refluxed in dry acetone for 5-6 h. The reaction mixture was cooled to room temperature and filtered off. The solvent was removed in vacuo and the product was extracted with

Protein expression and purification
Falcipain-2 and falcipain-3 were recombinantly expressed in E. coli and refolded to active enzyme as previously described [13,24].

Enzyme assays, in-vitro assays
IC 50 values for inhibition of falcipain-2 and falcipain-3 were determined as described previously using the fluorogenic substrates Cbz-Phe-Arg-AMC and Cbz-Leu-Arg-AMC, respectively (AMC, 7amino-4 methylcoumarin) [7,13,14]. As a positive control the well-known cysteine protease inhibitor E-64 [7,25,26] and as a negative control the solvent (DMSO) was used. Dose-dependent effects of compounds on parasite development (P. falciparum strain W2) were quantified by flow cytometry according to a previously published method [27]. First, a screening with inhibitor concentrations of 100, 10 and 1 mM was performed and the percentage activity of infected red blood cells (RBCs) relative to the negative control was determined. Compounds showing concentration dependent inhibition in these assays were selected for determination of IC 50 values (Table 1). Chloroquine [20] and E-64 were used as positive controls, and the solvent DMSO was used as a negative control. For selected compounds assays with the P. falciparum strain 3D7 were performed using a fluorometric assay with Hoechst-33258 [28], or the microculture tetrazolium test measuring parasite lactate dehydrogenase activity [12,29]. The parasites were cultivated in vitro as described previously [30]. Compounds were screened at concentrations between 1 nM and 100 μM. Synchronized ring stages were plated in 96-well plates at a parasitemia of 1%, in the presence of the compounds (dissolved in DMSO). Incubation of parasites with DMSO alone at a concentration of 0.5% v/v was used as a negative control. IC 50 values were calculated by non-linear regression analyses using the programs GaphPad ® Prism 4 and GraFit ® [31,32].

In vitro cytotoxicity assay
The human kidney epithelium-cell line 293T was cultured in DMEM medium supplemented with 20% fetal bovine serum and 1% l-glutamine (200 mM) at 37 °C in a humidified atmosphere containing 5% CO 2 . For the experimental procedures, cells were detached from the flasks with a rubber policeman, washed twice with PBS (phosphate-buffer saline) containing 1 mL trypsin-EDTA (1 x), and suspended at 2 x 10 6 cells mL -1 in complete medium. Inhibitor assays were carried out as described elsewhere except that the cell concentration used in this study was 2 x 10 4 mL -1 well -1 [33].