Tandem Synthesis of Furanaphthoquinones via Enamines and Evaluation of Their Antiparasitic Effects against Trypanosoma cruzi

Furanaphthoquinones are well known in medicinal chemistry for exhibiting relevant structural heterogeneity and bioactivities. In this work, it was synthesized a series of furanaphthoquinones through a tandem reaction between lawsone (8) and cyclic ketones in the presence of morpholine. This strategy provides an efficient and general method for synthesizing furanaphthoquinones with activity against the epimastigote form of Trypanosoma cruzi (T. cruzi), the parasite that causes Chagas disease. Compound 9b was the better prototype, and it exhibited high potency for causing parasite death, showed reduced acute toxicity towards mammalian cells, and was capable of rupturing the epimastigote plasma membrane and acting on sterol 14α-demethylase (CYP51). Additionally, 9b reduced trypomastigote viability by 99% after 24 h. Candidate 9b demonstrated the best and most promising profile when bound to CYP51.


Introduction
Medicinal chemistry has significantly contributed to the search for new drugs that can improve the quality of life of patients affected by various diseases. However, socalled neglected tropical diseases lack clinically effective drugs, and more research needs to be conducted to find new prototypes that can be transformed into drugs. Today, more than one billion people, one-sixth of the world's population, are suffering from neglected tropical diseases. 1 Chagas disease, caused by the protozoan T. cruzi, is a neglected tropical disease with high prevalence and significant morbidity and mortality. The treatment of Chagas disease is still based on only two drugs, nifurtimox and benznidazole, both of which have limited efficacy in the late chronic phase, cause frequent side effects, and have evolved drug resistance.
In this context, naphthoquinones are considered promising structures in the field of medicinal chemistry, as their antitrypanosomal potential has been reported. 1 These compounds are associated with various biochemical processes in microorganisms such as fungi and bacteria and higher plants and animals. These compounds are involved in respiratory chain electron transport, energy production, coagulation, and blood cells. Naphthoquinone structural diversity and its similarities to endogenous molecules demonstrate that these compounds can treat and cure diseases caused by pathogenic microorganisms. [1][2][3][4] Research groups have attempted to promote structural modifications in the naphthoquinone core in search of compounds with a better spectrum of action against T. cruzi. Thus, some naphthoquinones, such as 1-3, and their derivatives, such as 4, have been reported in the literature [5][6][7][8][9] to have activity against T. cruzi (Scheme 1a). Continuing our interest in the synthesis of naphthoquinones that can be used against T. cruzi, the etiologic agent of Chagas disease, we decided to study the reaction between lawsone (5a) and cyclic ketones 8 in the presence of morpholine to form furanaphthoquinones 9 via in situ formation of intermediate enamine (Scheme 1c). 1,4-Naphthoquinones can react with enamine 6 to produce furanaphthoquinone 7 (Scheme 1b). [10][11][12][13][14] However, enamines must be prepared in advance. This work describes a tandem reaction involving lawsone (5a), ketones (8), and morpholine, with the in situ formation of enamines, to produce furanaphthoquinones 9; the trypanocidal activity of the new compounds against the epimastigote forms of T. cruzi was also evaluated.

Chemistry
Analytical grade solvents (Biograde, São Paulo, Brazil) were used. Column chromatography was performed using silica gel (SilicaFlash® P60 0.040-0.063 mm, Silicycle, São Paulo, Brazil). Melting points (mp) were obtained on a Thermo Scientific 9100 apparatus (Waltham, USA) and were uncorrected. Infrared spectra (IR) were recorded on a Shimadzu IR Prestige-21 FTIR spectrometer (Kyoto, Japan) in KBr tablet (Sigma-Aldrich, São Paulo, Brazil). The reactor used to perform the reactions in the closed vessel was a Berghof model BR-300 (Eningen, Germany). 1 H and 13 C nuclear magnetic resonance (NMR) spectra were recorded at room temperature using a VNMRSYS-500 (Palo Alto, USA) or a Varian MR 300 instrument (Palo Alto, USA), using the solvents indicated, with tetramethylsilane (TMS) as the internal standard. Chemical shifts (d) are given in ppm, and coupling constants (J) are given in hertz (Hz). High-resolution mass spectra (HRMS) were recorded on a MICROMASS Q-TOF mass spectrometer (Milford, USA).
Then, the solvent was removed under reduced pressure, and the residue was purified by column chromatography on silica gel using a mixture of hexane:acetone as the gradient. General procedure for preparation of 11a-11c To a 50 mL round bottom flask containing 10 (0.34 mmol), 10 mL of concentrated sulfuric acid was carefully added dropwise. The mixture was stirred at room temperature for 10 min and then poured onto ice, forming a precipitate in an aqueous medium. The suspension was extracted with ethyl acetate, the organic layer was dried with anhydrous sodium sulfate, and the solvent was removed under reduced pressure. The residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent.  General procedure for preparation of 12a-12c A 50 mL round bottom flask containing a mixture of 9a-9c (0.6 mmol), acetic anhydride (106 mmol), pyridine (3 drops) and Zn 0 (0.6 mmol) was incubated with stirring at room temperature for 4 days. Then, the mixture was filtered, and the solution was extracted with dichloromethane, washed with saturated sodium bicarbonate solution, and dried with anhydrous sodium sulfate. The solvent was then removed under reduced pressure. General procedure for preparation of 13a-13c In a 600 mL high-pressure reactor (Berghof, Model BR-300), a solution in ethanol (200 mL) of 9 (4 mmol) and Pd/C (10 mol%) was stirred under 27 psi pressure of hydrogen for 10 min. Then, the mixture was filtered, and the solvent was removed under reduced pressure. The residue was purified by column chromatography on silica gel using hexane/ethyl acetate as the eluent.

Cultures of parasites
Epimastigote forms of T. cruzi (strain Y) were maintained in liver infusion tryptose (LIT) medium at 28 °C and through weekly passages. The mice were infected with trypomastigotes of T. cruzi obtained after metacyclogenesis in LIT medium at 28 °C to obtain the necessary parasite number for animal infection. The infected cells were maintained at 34 °C with CO 2 . Parasites purified from the blood were resuspended in Dulbecco's modified Eagle's medium (DMEM) supplemented with 20% fetal bovine serum (FBS).

T. cruzi (Y strain -bloodstream trypomastigotes)
Trypomastigotes were obtained from the blood samples of infected albino Swiss mice at the peak of parasitemia. The purified parasites were resuspended in DMEM (Sigma-Aldrich, São Paulo, Brazil) supplemented with 10% FBS as reported previously. 8,9 Vero cells were cultured in Roswell Park Memorial Institute (RPMI) medium supplemented with 5% FBS, 100 UI mL -1 antibiotic mixture, and 10 µg mL -1 streptomycin at 37 °C in an atmosphere of 5% CO 2 until the cells reached 90% confluence. The cell monolayer was infected with trypomastigotes (Y strain, 10 parasites per cell). After 24 h, the supernatant medium was collected, and Vero cells and amastigotes were removed by centrifugation at 1000 g for 5 min. Trypomastigotes were collected by centrifugation at 1600 g for 10 min.
Trypanocidal activity in trypomastigotes and amastigotes (Y strain) was determined using 1 × 10 6 cells mL -1 in 96-well plates and incubation at 37 °C for 24 h in the presence of the corresponding drug. Trypomastigote viability was determined by counting in a Neubauer chamber.
Cell metabolism activity test (Redox) in vitro using the resazurin reduction technique Peritoneal macrophages derived from Swiss Webster strain mice were obtained through peritoneal lavage. The cell suspension contained in the supernatant was centrifuged at 1500 rpm for 5 min. The pellet was resuspended in 1 mL RPMI medium and counted in a Neubauer chamber. A total of 4 × 10 5 macrophages well -1 were seeded in triplicate in 96-well plates with RPMI medium supplemented with 10% FBS. These cells were incubated in an oven at 37 °C in a humidified atmosphere of 5% CO 2 for 24 h. After the cells adhered for 24 h, all the media was removed, and the compounds were added to plates in SFP-free (Shahidi Ferguson Perfringens free) RPMI medium at 6 concentrations equal to 0.01, 0.1, 1, 25, 50, and 100 µM using a final volume of 200 µL well -1 . As a negative control, the wells were maintained only with cells and culture medium. It was used 0.5% Triton X-100 (Sigma-Aldrich, São Paulo, Brazil) as a positive control. The plates were incubated again for 24 h, and after that interval, the colorimetric assay was performed by reducing resazurin (RZ, Sigma-Aldrich, São Paulo, Brazil). A volume of 100 µL of the supernatant was removed from each well, and 20 µL of RZ (0.15 mg mL -1 ) was added. The plates were incubated again for 4 h, and the readings were obtained in a spectrophotometer at 575 and 595 nm using SoftMax Pro software (version 5.1). 15 Data analysis was performed using GraphPad Prism 5. 16 In vitro cell toxicity assay using the lactate dehydrogenase (LDH) release technique After the treatment mentioned above in peritoneal macrophages, a volume of 50 µL was collected from each well and transferred to a new plate. A substrate mix prepared according to the manufacturer's specifications (Promega, Wisconsin, USA) was added and incubated for 30 min. After this time, the reaction stop reagent was added, and immediately afterwards, the absorbance was read in a spectrophotometer at 490 nm using SoftMax Pro software (version 5.1). 15

Determination of trypanocidal action
The T. cruzi culture incubated for 24 h in LIT medium was added to a 96-well plate at 1 × 10 6 cells per 200 µL. It was added a 1 μM concentration of each compound and 20 μL of 0.2 µM propidium iodide solution for 30 min. The fluorescence was quantified as arbitrary fluorescence units of non-viable cells in a SpectraMax M4 spectrophotometer (Molecular Devices, California, United States of America) at 565-605 nm wavelengths. Triton X-100 (Sigma-Aldrich, São Paulo, Brazil) was used as a trypanocidal control, and the culture without compound addition was used as a negative control. All treatments were performed in triplicate.

Spectroscopic CYP51 TC binding assay
Spectrophotometric titration was used for binding assays. A volume of 3 mL of 50 mM Tris-HCl (pH 7.5) and 10% glycerol was used in a UV-visible (UV-Vis) scanning spectrophotometer (Varian, Palo Alto, United States of America). The CYP51Tc (sterol 14α-demethylase from T. cruzi) concentration used was 0.1 μM. Stock solutions of 9a, 9b, 9c, 11a and 11b were solubilized in dimethyl sulfoxide (DMSO, Sigma-Aldrich, São Paulo, Brazil) at 20 μM. Titrations were performed using a 3.5 mL quartz cuvette with a path length of 1 cm, and the inhibitor was added to 300 μL aliquots. As a negative control, DMSO was added in the same amounts, and the difference spectra were measured. KD (dissociation constant) values recorded for data titration of points were fitted to quadratic hyperbola using GraphPad Prism software (GraphPad Software Inc.), 16 as follows: A obs = (Amax/2 × Et){(S + Et + KD) − [(S + Et + KD)2 − 4 × S × Et]0.5}, where A obs is the absorption change determined at any inhibitor concentration, Amax is the maximal absorption shift recorded at saturation, KD is the dissociation constant for the inhibitor-enzyme complex, Et is the total enzyme concentration used, and S is the inhibitor concentration.

Molecular modeling of ligands
The compounds of the series 9-11 were initially built using the program Avogadro. 17,18 The geometry optimizations of molecules were then performed using the MMFF94 (Merck molecular force field 94). The optimized conformers were submitted to refinement calculations of the geometry optimization using the semiempirical parametric method PM6 performed by the MOPAC2016 program. 19

Molecular docking
The program Molegro Virtual Docker (MVD) 20 was used to perform molecular docking (CLC Bio, 8200, Aarhus, Denmark). The MolDock score algorithm [GRID] with a grid resolution of 0.30 Å was selected as the score function, and the partial charges were assigned according to the MVD charge scheme. The MolDock Optimizer algorithm was used with a search space of 32 Å around the T. cruzi protein CYP51 structure. The CYP51 protein structure (PDB code: 4CK9) was retrieved from the Protein Data Bank (PDB). Molecular docking of the MC series compounds was performed using the same parameter set (runs = 100, population size = 50, max interactions = 2000, scaling factor = 0.50, and crossover rate = 0.90). The ligand poses were selected based on MolDock score values. Both the Molegro Virtual Docker 18 and PyMOL 21 were used to visualize and analyze the optimized molecular complexes.

Protein/ligand binding enthalpy
Molecular complexes of MC series compounds and T. cruzi 14-α-lanosterol demethylase (CYP51) obtained by docking and separate optimized ligands with protein before docking were applied to MOZYME calculation using a dielectric constant of 78.4. The calculations were performed using the MOPAC2016 program to indicate the molecular system heat of formation (ΔH f ). [17][18][19][20][21][22][23] Protein CYP51 and heme group atom coordinates were fixed. For ligand molecules, geometry optimization was freely allowed.
The possible mechanism of this reaction is stopped by the nucleophilic attack of enamine formed in situ, leading to intermediate II after oxidation of quinol and elimination of morpholine, resulting in furan-1,4-naphthoquinones 9a-9c. This mechanism had already been reported by Kobayashi et al. 11 in a reaction between 2-hydroxy-1,4-naphthoquinones and previously synthesized enamines (Scheme 3).
The strategy used to improve the solubility of quinones in the biological environment, and therefore for them to be formulated to reach the biological target, is to transform them into diacetylated derivatives. Reichstein et al. 24 showed that metabolically labile furan-1,4-naphthoquinones were designed as prodrugs and were only slightly less active than their parent naphthoquinone counterparts in suppressing keratinocyte hyperproliferation in the micromolar range. In addition, Ma et al. 25 reported the preclinical evaluation of prodrug derivatives of β-lapachone encapsulated in biocompatible and biodegradable poly(ethylene glycol)-β-poly(D-lactic acid) (PEG-β-PLA) micelles. These formulations were used for the treatment of non-small cell lung cancers (NSCLC) that overexpress nicotinamide adenine dinucleotide phosphate (NAD(P)H):quinone oxidoreductase 1 (NQO1) and demonstrated efficiencies > 95% with significantly reduced hemolysis and methemoglobinemia that currently limits ARQ761 formulations.
To promote small changes in the furanaphthoquinone nucleus that transform the compounds into more soluble substances in biological media, reductive acetylation with acetic anhydride, zinc and pyridine was performed. As a result, acetylated derivatives 12a-12c were obtained with excellent yields that ranged from 92 to 97% (Scheme 4). On the other hand, to investigate and compare the biological activity against T. cruzi, the catalytic hydrogenation of 9a-9c to obtain 13a-13c was performed in a reactor at 27 psi, and yields were low (30-35%) (Scheme 3). Finally, the structures of the new compounds were determined with 1 H NMR, 13 C NMR, and IR analysis in combination with mass spectrometry (see Supplementary Information section).

Biological assays
Cell viability All naphthoquinone derivatives described were evaluated in vitro against T. cruzi. In the LDH release assay, the activity of this intracellular enzyme occurs when there is plasma membrane damage and subsequent cytoplasmic content leakage. In this scenario, after plasma membrane rupture, the cell dies from necrosis. Compounds 9a-9c, 11a-11c, 12a-12c, and 13a-13c at a concentration of 10 μM did not cause cellular toxicity. However, 10a-10c caused toxicity after treatment for 24 h (Figure 1).

Trypanocide assays
The resazurin reduction assay was performed on epimastigotes of T. cruzi from the Y strain, which were treated for 72 h. Derivatives 9a-9c and 11a-11c at a concentration of 10 µM reduced parasite viability compared with non-treated parasites (CN). None of the treatments yielded activity higher than that of the positive controls, 0.05% Triton X-100 (CP) and 100 µM benznidazole (BZ, Sigma-Aldrich, São Paulo, Brazil). However, 9a, 9b, and 11b exerted similar effects to BZ. All six derivatives reduced parasite viability by more than 75% compared with CN ( Figure 2).
The cytotoxicity concentration (CC 50 ) estimated for the samples was determined after the treatments, and the concentrations ranged from 0.001 to 100 µM for 24 h. The  The results obtained indicate that analogs with high potency affect parasite viability. Derivatives 9b and 9c were the most potent compared with other compounds. The IC 50 (half maximal inhibitory concentration) values for these analogs were in the range of 30 µM. These substances were tested against trypomastigote forms (Table 1). Treatment with 1 µM reduced trypomastigote viability compared to benznidazole. 9b and 9c produced the highest effects, reducing the parasite numbers by 99 and 95%. However, 11a exhibited an effect similar to benznidazole. Additionally, 9a and 11b had worse effects than benznidazole.

Selectivity index (SI)
We calculated the sample selectivity index based on the results described above and compared this index to that of     (Table 2).
In the propidium iodide (PI) uptake assay, we measured the fluorescence of this dye when the plasma membrane was damaged, and PI gained access to the nucleus and bound to DNA. This dye is positively loaded and has a size of 636 Daltons. Therefore, PI does not pass spontaneously through the plasma membrane. In this scenario, after plasma membrane rupture, the cell dies from necrosis-a usual scenario of trypanocidal drug action. However, treatment with a concentration of 1 µM increased PI fluorescence after 24 h of continuous treatment ( Figure 5).

In silico results
The furanaphthoquinone series compound molecular docking against T. cruzi was performed for the protein 14-α-lanosterol demethylase (CYP51). The results indicate that residues Ala115, Tyr116, and Leu98 appear essential for the affinity of compounds 9a, 9b, 9c, 11a, and 11b to the molecular target. Furthermore, compounds 9a and 9c present similar conformational poses, showing the planar aromatic molecular moiety interacting with the heme group Fe atom by cation-π interactions. In addition, both compounds seem to form hydrogen bonds with the Tyr116 residue and hydrophobic interactions with residues Leu98 and Ala115 (Figure 7). Compounds 11a and 11b exhibited similar poses. Both molecules show the planar aromatic molecular moiety interacting with the heme group Fe atom by cation-π interactions. Both compounds seem to form hydrogen bonds with the Tyr116 residue and with the heme group. Moreover, the molecular poses of both ligands 11a and 11b indicate hydrophobic interactions with residues Leu98 and Ala115 (Figure 7). However, the molecular docking result of compound 9b indicates a different pose compared to 9a and 9c. For ligand 9b, the molecular pose indicates that the intermolecular interactions are mainly hydrogen bonds and hydrophobic interactions. The hydrogen bond interaction is between the cyclic oxygen atom and Tyr116. The hydrophobic interactions are between the ring moieties of compound 9b and the residues Leu98 and Ala115.
The heat of formation (ΔH f ) was calculated using ligands separated and complexed with protein CYP51 to indicate the affinity between interacting molecules    ( Table 3). The binding enthalpies of the studied ligands are different. The results indicate that 9b, 9c, and 11a have more favorable binding enthalpies with T. cruzi CYP51.

Conclusions
In summary, we have developed an operationally simple tandem synthetic protocol for the synthesis of six furanaphthoquinones by enamine intermediates in good yields. Analogs 9a, 9b, 9c, 11a, and 11b exhibited potent and efficacious trypanocide activity against epimastigote forms. These analogs reduced trypomastigote viability by 75% within 24 h. All of these analogs exhibit trypanocide effects acting on the 14-α-lanosterol demethylase enzyme. Thus, these furanaphthoquinone series compounds represent promising candidates to potently and selectively eliminate T. cruzi with low toxicity.

Supplementary Information
Supplementary information is available free of charge at http://jbcs.sbq.org.br as PDF file.