Soulamarin Isolated from Calophyllum brasiliense (Clusiaceae) Induces Plasma Membrane Permeabilization of Trypanosoma cruzi and Mytochondrial Dysfunction

Chagas disease is caused by the parasitic protozoan Trypanosoma cruzi. It has high mortality as well as morbidity rates and usually affects the poorer sections of the population. The development of new, less harmful and more effective drugs is a promising research target, since current standard treatments are highly toxic and administered for long periods. Fractioning of methanol (MeOH) extract of the stem bark of Calophyllum brasiliense (Clusiaceae) resulted in the isolation of the coumarin soulamarin, which was characterized by one- and two-dimensional 1H- and 13C NMR spectroscopy as well as ESI mass spectrometry. All data obtained were consistent with a structure of 6-hydroxy-4-propyl-5-(3-hydroxy-2-methyl-1-oxobutyl)-6″,6″-dimethylpyrane-[2″,3″:8,7]-benzopyran-2-one for soulamarin. Colorimetric MTT assays showed that soulamarin induces trypanocidal effects, and is also active against trypomastigotes. Hemolytic activity tests showed that soulamarin is unable to induce any observable damage to erythrocytes (cmax. = 1,300 µM). The lethal action of soulamarin against T. cruzi was investigated by using amino(4-(6-(amino(iminio)methyl)-1H-indol-2-yl)phenyl)methaniminium chloride (SYTOX Green and 1H,5H,11H,15H-Xantheno[2,3,4-ij:5,6,7-i′j′]diquinolizin-18-ium, 9-[4-(chloromethyl)phenyl]-2,3,6,7,12,13,16,17-octahydro-chloride (MitoTracker Red) as fluorimetric probes. With the former, soulamarin showed dose-dependent permeability of the plasma membrane, relative to fully permeable Triton X-100-treated parasites. Spectrofluorimetric and fluorescence microscopy with the latter revealed that soulamarin also induced a strong depolarization (ca. 97%) of the mitochondrial membrane potential. These data demonstrate that the lethal action of soulamarin towards T. cruzi involves damages to the plasma membrane of the parasite and mitochondrial dysfunction without the additional generation of reactive oxygen species, which may have also contributed to the death of the parasites. Considering the unique mitochondrion of T. cruzi, secondary metabolites of plants affecting the bioenergetic system as soulamarin may contribute as scaffolds for the design of novel and selective drug candidates for neglected diseases, mainly Chagas disease.


Introduction
The tree Calophyllum brasiliense is known in Brazil as ''Guanandi'' or ''Jacareúba''. It can reach up to 40 meters in high, 1-3 meters in diameter and is usually found in Brazil in the rain forest regions of the Amazon. Its stem bark is used in folk medicine to treat rheumatism, varicose veins, haemorrhoids and ulcers, whereas the leaves have anti-inflammatory properties [1]. Previous chemical studies on C. brasiliense resulted in the isolation of several interesting natural products, e.g. xantones, flavonoids, triterpenoids, and coumarins [2]. Some coumarins isolated from C. brasiliense displayed trypanocidal activity, but unfortunately, no information about the underlying mechanism was available [3].
The parasite Trypanosoma cruzi causes American trypanosomiasis or ''Chagas disease'', which has high mortality and morbidity rates [4]. Chagas disease is common to the Americas, including Mexico and the South of the USA and has become a global public health problem [5]. Due to high levels of migration, the disease has already reached non-endemic countries. An estimated 10 million people are currently infected and 14,000 deaths per year are documented. In Brazil alone, over 6 million people are infected and approximately 6,000 deaths per year are registered. The migration of millions of Latin Americans to more developed countries such as e.g. the USA, accounts for approximately 300,000 chronically infected patients there [6]. More than a dozen infections acquired from blood transfusions or transplantations have been reported in several European countries, the USA, and Canada [7]. Nifurtimox (7-10 mg/kg/day) and benznidazole (5-7 mg/kg/ day) are the two prevalent drugs, currently used in the treatment of Chagas disease. Unfortunately they suffer drawbacks from high levels of toxicity and long treatment periods (ca. 60 days) [8]. Nifurtimox, a nitrofuran, inhibits the ability of T. cruzi to deplete free radicals through the generation of a nitro-anion in the presence of oxygen. Benznidazole, a nitroimidazole, binds to the DNA, lipids and proteins of T. cruzi [9]. The average rate for successful cures among acute and recent cases is 80%, while it is less than 20% for chronic cases [10]. Several studies have identified numerous potential candidates for more effective and less toxic drugs. Amidines [11,12], azoles [13], amiodarones [14], natural naphthoquinone derivatives and megazols [15] as well as calcium channel blockers [16] have been proposed, but clinically effective compounds still remain elusive. Therapeutic drug combinations have also been proposed as treatment strategies, e.g. benznidazole/nifurtimox, orbenznidazole/nifurtimoxin combination with antifungals which inhibit ergosterol in double or triple associations [17].
Natural products isolated from plants are commonly used as drug prototypes or precursors to treat parasitic diseases. Natural coumarins are an important class of plant products with antitrypanosomal activity [3,18]. The coumarins mammea A/ BA, A/BB, A/AA, A/BD and B/BA, isolated from C. brasiliense and Mamea americana, showed activity towards epimastigotes and trypomastigotes of T. cruzi for concentrations between 15 and 90 mg/mL [3]. Other coumarins isolated from the stem bark of Kielmeyera albopunctata showed in vitro activity against the trypomastigotes of T. cruzi, killing 80% of the parasites after 24 hours at 125 mg/mL [19].
Continuing the investigation of bioactive compounds from Brazilian flora, the present study was undertaken in order to determine the antitrypanosomal effects of soulamarin, which is the main compound isolated from the stem bark of C. brasiliense, against T. cruzi. This study moreover investigated the lethal action of soulamarin towards the parasite.

General experimental procedures
NMR spectra were recorded on a Bruker DRX-500 ( 1 H: 500 MHz, 13 C:125 MHz) spectrometer at ambient temperatures. Chemical shifts (d) are reported in ppm and coupling constants (J) in Hz. All resonances were referenced to residual NMR solvent resonance. Low-resolution electrospray ionization mass spectra (LR-ESI-MS) were measured in positive mode on a Platform II-Micromass (quadrupole) mass spectrometer.

Plant material
Samples of the stem bark of C. brasiliense were collected in the Amazonian rain forest of Brazil during September 2011. The authenticity of the plant material was verified by Dr. Eliana Rodrigues from ICAQF-UNIFESP. Sample specimens were deposited at the herbarium of the Instituto de Botânica -SEMA of São Paulo (SP, Brazil).
Extraction and isolation of 6-Hydroxy-4-propyl-5-(3hydroxy-2-methyl-1-oxobutyl)-60,60-dimethylpyrane- [20,30:8,7]-benzopyran-2-one (soulamarin) Dried and powdered stem bark samples of C. brasiliense (72 g) were washed exhaustively with hexane (106500 mL) at room temperature in order to remove any residual fats. Subsequently, the plant material was extracted with MeOH (1061 L) at room temperature. The combined organic fractions afforded, after removal of all solvents under reduced pressure, 4.7 g of crude residue. This crude extract was dissolved in MeOH:H 2 O (1:2) and extracted with EtOAc. The removal of the solvent under reduced pressure resulted in the deposition of a residue (3.0 g), which was subsequently subjected to column chromatography (Sephadex LH-20) with MeOH as the eluent. Nine fractions (I-IX) were separated like this. Fraction III (1.31 g) was further purified by column chromatography over silica gel with a crude solvent gradient of hexane/EtOAc (starting with pure hexane and finishing with pure EtOAc). This second purification step afforded 544 mg of soulamarin (see Figure 1). 1

Animals
Mice (swiss and BALB/c) were supplied by the animal breeding facility of the Adolfo Lutz institute (São Paulo, Brazil). Animals were kept in sterilized cages in a controlled environment, with water and food ad libitum. All experimental procedures were

Author Summary
Chagas disease is a parasitic protozoan that affects the poorest population in the world, causing a high mortality and morbidity. As a result of highly toxic and long-term treatments, the discovery of novel, safe and more efficacious drugs is essential. Natural products isolated from plants are commonly used as drug prototypes or precursors to treat parasitic diseases. As part of our investigation of bioactive compounds from Brazilian flora, the present study was undertaken in order to determine the antitrypanosomal effects of the soulamarin, a coumarin isolated from the stem bark of Callophyllum brasiliense (Clusiaceae), against Trypanossoma cruzi. This study moreover investigated the lethal action of soulamarin towards the parasite. Considering the obtained results, secondary metabolites of plants affecting the bioenergetic system as soulamarin may contribute as scaffolds for the design of novel and selective drug candidates for neglected diseases, mainly Chagas disease.

Parasites and mammalian cells
In all in vitro assays, Y strains of T. cruzi trypomastigotes were used, which were kept at 37uC in LLC-MK2 (ATCC CCL 7) cells using RPMI-1640 medium with calf serum (2%) [20]. To keep the Y strains infective, trypomastigotes were also kept in swiss mice and regularly harvested from the bloodstream by heart puncture of infected animals at the peak of the parasitemia [21]. LLC-MK2 cells were maintained at 37uC in RPMI-1640 medium with fetal calf serum (10%) in an incubator (5% CO 2 atmosphere).

Antitrypanosomal activity
Trypomastigotes were counted in a hemocytometer (Neubauer) and deposited on a microplate (96 wells; 1610 6 cells/well). Subsequently, soulamarin was added to the cells in concentrations up to 386 mM and the cells were allowed to incubate for 24 hours at 37uC (5% CO 2 atmosphere). Benznidazole was used as standard. Trypomastigote activity was based on the conversion of the soluble tetrazolium salt 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyltetrazolium bromide (MTT) into the insoluble formazan by mitochondrial enzymes. The extraction of formazan was carried out for 18 hours at 24uC with sodium dodecylsulfate (10%v/v; 100 mL/well) [22].
In order to determine the IC 50 value for soulamarin against intracellular amastigotes, the method described by De Souza and co-workers [23] was used with minor modifications. Peritoneal macrophages were collected from the peritoneal cavity of BALB/c mice and deposited on a 16-well chamber slide (1610 5 cells/well) before being incubated for 24 hours at 37uC (5% CO 2 atmosphere). Trypomastigotes from LLC-MK2-infected cultures were washed twice in RPMI-1640 medium, counted in a hemocytometer and added to the macrophages (parasite:macrophage ratio = 10:1). After an incubation period of 18 hours at 37uC (5% CO 2 ), residual free parasites were removed with two washings with medium. Soulamarin was subsequently incubated with infected macrophages (60 h, 37uC, 5% CO 2 ) in a non-toxic concentration range between 3.01 and 386 mM. Benznidazole was used as a standard. At the end of the assay, slides were fixed with methanol and stained with Giemsa prior to counting under a light microscope. IC 50 concentrations were obtained by counting 300 macrophages per well (in duplicate) and determining the number of amastigotes per infected macrophage.

Cytotoxicity against mammalian cells
Peritoneal macrophages were collected from female BALB/C mice, seeded at 1610 5 cells/well in 96-well microplates and incubated with soulamarin for 72 h at 37uC in an incubator with 5% CO 2 . The viability of the cells was determined using MTT [16]. The data represent the mean of two independent assays (triplicates).

Hemolytic activity
The hemolytic activity of soulamarin in concentrations up to 1,300 mM was evaluated from the erythrocytes of BALB/c mice [24]. A suspension (5%) of erythrocytesin PBS (phosphate buffered saline) was incubated with soulamarin at 25uC for 1 hour in a Ushaped microplate (96 wells). The absorption of the supernatant at 550 nm was recorded (FilterMax F5 Multi-Mode Microplate Reader-Molecular Devices).

Spectrofluorimetric detection of the permeability of the cell membrane
Trypomastigotes were washed with PBS (phosphate buffered saline), deposited on a microplate (1610 6 cells/well) and incubated with SYTOX Green (1 mM) for 15 minutes at 24uC [25]. Soulamarin was added in three concentrations (IC 100 = 386 mM, IC 50 = 219 mM and IC 25 = 103 mM) and the fluorescence was measured after 20, 40 and 60 minutes. The maximum permeability possible was observed with 0.1% Triton X-100 (positive control). The fluorescence intensity was determined using a fluorimetric microplate reader (FilterMax F5 Multi-Mode Microplate Reader-Molecular Devices) with excitation and emission wavelengths of 485 and 520 nm, respectively. Untreated trypomastigotes and 0.5% (v/v) DMSO-treated parasites were used as negative controls.

Effect of soulamarin on the mitochondrial membrane potential
Trypomastigotes were washed with PBS, deposited on a microplate (2610 6 cells/well) and incubated with soulamarin (IC 50 = 219 mM) for 60 minutes at 37uC. Then MitoTracker Red CM-H 2 XROS (500 nM) was added and the incubation was continued for 40 minutes in the dark. Cells were washed twice with BSS (Hank's buffered salt solution) and the fluorescence was measured using a fluorimetric microplate reader (FilterMax F5 Multi-Mode Microplate Reader-Molecular Devices) with excitation and emission wavelengths of 540 and 595 nm, respectively [26]. Untreated trypomastigotes and DMSO-treated parasites were used as negative controls. Mesoxalonitrile 4-trifluoromethoxyphenylhydrazone (FCCP; 10 mM) was used as a positive control [27]. For the fluorescence microscopy analysis, trypomastigotes were co-stained with 49,6-diamidino-2-phenylindoledihydrochloride (DAPI; 10 mM) and examined at 10006 magnification. Merged images of blue (DAPI) and red (MitoTracker Red) images were obtained using the Nikon NIS -Elements AR software. A Nikon D-FL Epimicroscope equipped with a DS-U3 digital camera was used for the experiment. Analysis of reactive oxygen species (ROS) Trypomastigotes (2610 6 cells/well) were washed in HBSS (Hanks Balanced Salt Solution) medium and incubated with soulamarin (IC 50 = 219 mM) for 60 minutes at 37uC. To these cells 29,79-dichlorodihydrofluorescein diacetate (H 2 DCf-DA) was added (5 mM) and incubation was prolonged for 15 minutes. Then the fluorescence was measured using a fluorimetric microplate reader (FilterMax F5 Multi-Mode Microplate Reader-Molecular Devices) with excitation and emission wavelengths of 485 and 520 nm, respectively. Oligomycin (20 mM) was used as positive control [28]. Untreated trypomastigotes and parasites treated with DMSO were included as negative controls.

Statistical analysis
Results are displayed as mean values 6 standard deviations, which were obtained from at least two independent assays (n$2). IC 50 values were calculated from sigmoidal dose-response curves using the Graph Pad Prism 5.0 software. Confidence intervals of 95% are included in parentheses. The Student's t-test was used for significance testing (p,0.05) for all assays.

Chemical characterization of soulamarin
The structure of soulamarin is shown in Figure 1. The assigned structure is consistent with the results obtained from NMR data and LR-ESI mass spectrum. The individual assignment of proton and carbon atoms was accomplished by 1D ( 1 H, 13 C) and 2D (HMQC, HMBC and NOESY) NMR measurements. The 1 H NMR spectrum of soulamarin in CDCl 3 /CD 3 OD displayed two doublets at d 6.35 and 5.26, with coupling constants of 10.0 Hz, which were assigned to H-9 and H-10. Together with the presence of a singlet at d 1.20 (H-12 and H-13), these peaks suggested the presence of a chromene moiety [29]. The presence of a dihydrocoumarin segment was based on the multiplets at d 2.4-2.5 (2H) and  [31]. The configuration of C-4 was assigned as S*, due to the cross peaks between H-4 ax and H-3 eq as well as between H-4 ax and H-14, observed in the NOESY spectrum. All these results are consistent with a structure of 6hydroxy-4-propyl-5-(3-hydroxy-2-methyl-1-oxobutyl)-60,60-dimethylpyrane- [20,30:8,7]-benzopyran-2-one (see Figure 1). The assigned structure was furthermore supported by comparison of our spectroscopic data with those reported in the literature [32].

Antitrypanosomal, cytotoxicity and hemolytic activity of soulamarin
Soulamarin was incubated with trypomastigotes and the activity of cells was determined after 24 hours via MTT assay. Soulamarin thereby demonstrated activity against parasites, killing all the cells at the highest tested concentration. An IC 50 value of 219.8 mM (95% confidence interval for 186.9-258.5 mM) was established (see Table 1). Benznidazole was used as standard against and resulted in an IC 50 value of 440.7 mM (95% confidence interval for 272.4-478.4 mM). Soulamarin was also effective against intracellular amastigotes (IC 50 = 210.1 mM; 95% confidence interval for 174.5-252.6 mM), while benznidazole showed an IC 50 of 319.7 mM (95% confidence interval for 283.8-360.1 mM). The cytotoxicity of soulamarin was determined with peritoneal macrophages by the MTT assay. Soulamarin showed an IC 99 value of 988.95 mM and IC 50 value of 278.3 mM (95% confidence interval for 229.4-342.8 mM). The hemolytic activity was also examined, but soulamarin did not induce any observable hemolysis up to concentrations of 1,300 mM (Table 1).

Modified permeability of the plasma membrane induced by soulamarin
Three different concentrations of soulamarin were incubated for up to 60 minutes with trypomastigotes and the permeability of the plasma membrane was examined by SYTOX Green assay. Soulamarin induced significant increased (p,0.05) fluorescence for all tested concentrations. Highest fluorescence intensities were observed after 60 minutes of incubation ( Figure 2). Relative to fully permeabilized parasites (Triton X-100, 60 min), soulamarin induced the following percentages of permeability: i) 81% for IC 100 = 386 mM (standard error of the mean SEM 6.2) (p,0.05); ii) 60% for IC 50 = 219 mM (SEM 8.5); iii) 28% for IC 25 = 103 mM (SEM 1.02). DMSO was used as internal control and resulted in lack of alteration.

Soulamarin-induced depolarization of the mitochondrial membrane potential of T. cruzi
Soulamarin was incubated with trypomastigotes (60 min) and the mitochondrial membrane potential was examined using MitoTracker Red. Spectrofluorimetric measurements indicated that soulamarin induced a significant (97%, p,0.05) decrease in fluorescence levels compared to untreated trypomastigotes ( Figure 3A). The control group showed a typical mitochondrial membrane potential. FCCP was used as positive control, which reduced the fluorescence levels by 70% (p,0.05) relative to untreated parasites. Additional fluorescence microscopy experiments corroborated the spectrofluorimetric analysis, demonstrating a substantial reduction of fluorescence levels in soulamarintreated parasites ( Figure 3B), as wells as in FCCP ( Figure 3D). Untreated parasites showed intense fluorescence levels of mitochondria after labeling with MitoTracker Red, which is consistent with a normal mitochondrial membrane potential ( Figure 3C). Panels I represent images with blue fluorescence channel labeled with the fluorescent probe DAPI; panels II represent images with red fluorescence channel labeled with the fluorescent probe MitoTracker Red and panels III, represent the merged images.

Analysis of reactive oxygen species (ROS)
Soulamarin was incubated with trypomastigotes and the up/ down-regulation of ROS was examined using 29,79-dichlorodihydrofluorescein diacetate (H 2 DCf-DA). No changes in the production of ROS could be observed after 60 minutes. Oligomycin was used as positive control (100% ROS up-regulation). Untreated parasites were used as a negative control, showed a normal level of ROS production and were used for normalization (data not shown here).

Discussion
Antitrypanosomal drugs targeting the bioenergetic metabolism as well as the plasma membrane have been considered as potentially chemotherapeutics for Chagas disease [33]. As part of an ongoing search aiming at the isolation of antiparasitic compounds from Brazilian plants [34][35][36], the coumarin 6-hydroxy-4-propyl-5-(3-hydroxy-2-methyl-1-oxobutyl)-60,60-dimethylpyrane- [20,30:8,7]-benzopyran-2-one (soulamarin) was isolated for the first time from the stem bark of C. brasiliense and characterized by 1 H-and 13 C-NMR analysis as well as by mass spectrometry. Ee and co-workers have recently isolated soulamarin from C. soulattri, but no biological activity was described [32]. To the best of our knowledge, this is the first time that an antiparasitic activity against Trypanosoma cruzi is reported for soulamarin. The comparable IC 50 values of soulamarin and benznidazole, which is the currently drug in clinical use, suggests a similar effectiveness against trypomastigotes and intracellular amastigotes. In our assays, benznidazole showed an IC 50 value of 440 mM against trypomastigotes. Different protocols have been described in literature for the   [37]; ii) spectrophotometric evaluation of MTT oxidation by mitochondrial dehydrogenases [38]; iii) culture-derived trypomatigotes and bloodstream trypomastigotes; iv) parasite strain; v) number parasites per well; vi) time of incubation with beznidazole, and vii) the source of the drug, which has been produced by different laboratories. Then, comparisons to other IC 50 values of benznidazole should be carefully analyzed. Natural products and synthetic compounds affecting the biosynthesis [33] or the permeability of the plasma membrane [39] of T. cruzi have been identified as interesting targets for drug discovery studies. By way of using different fluorimetric probes (SYTOX Green, MitoTracker Red), we targeted in this study the mechanistic aspects on how soulamarin kills T. cruzi trypomastigotes. Our data demonstrated that soulamarin rapidly altered the permeability of the plasma membrane, resulting in a dose-and time-dependent influx of the vital dye SYTOX Green into the cell. Fluorescence levels were constant during the observation period (60 min) for all tested concentrations of soulamarin. At 386 mM (IC 100 ), soulamarin induced the highest fluorescence intensity, which was close to the positive control Triton X-100, suggesting long-term effects on the membrane of the parasite. A similar effect was also observed at the lowest tested concentration (IC 25 = 103 mM). Edelfosine, a synthetic lysophosphlipid drug has also been shown to induce alterations in the plasma membrane and mitochondria of T. cruzi, suggesting that these organelles could be potential targets [40]. However, it has also been proposed that a drug-induced change of the permeability of the plasma membrane is not necessarily accompanied by changes of the mitochondrial function. Digitonin for example, a natural glycoside obtained from Digitalis sp., has been shown to increase plasma membrane permeability in T. cruzi, without affecting the mitochondria [41]. The respiratory apparatus of protozoans typically displays a greater diversity in electron pathways compared to their host cells. The mitochondrion of the protozoan can be considered as a valuable drug target, because of its unique structure and function compared to mammalian cells [42]. The predominant physiological function of the mitochondrion is the generation of ATP by oxidative phosphorylation. Additional functions include the generation and detoxification of ROS, the involvement in some forms of apoptosis and the regulation of cytoplasmic and mitochondrial calcium [43]. Furthermore, a proper mitochondrial membrane potential is essential for the survival of cells and changes can result in a variety of consequence, such as the inhibition of the electron transport chain, the inhibition of ATP synthase, the stimulation of uncoupling proteins or the permeability of the inner membrane [44]. In our assays, soulamarin rapidly induced a depolarization of the mitochondrial membrane potential in trypomastigotes, resulting in a reduction of the fluorescence intensity by 97%, relative to the untreated group. This effect may have contributed to deleterious cellular damages associated with bioenergetic system. A similar, potent dosedependent collapse of the mitochondrial membrane potential, resulting in the killing of T. cruzi parasites, has been reported for synthetic naphthofuranquinones [45].
Under physiological conditions, the oxidative phosphorylation involving an electron transport to pump hydrogen ions across the inner membrane, releases ROS, amounting to 3-5% of the total amount of oxygen consumed [46]. Under pathologic conditions, several pathways result in excessive ROS production, which causes -if not efficiently scavenged by the antioxidant system -oxidative stress. Proteins, lipids, and DNA are readily oxidized by ROS, resulting in dysfunction of vital physiological processes, oxidative damage, and cell death [47]. In our assays, despite the substantial depolarization of the mitochondrial membrane potential of T. cruzi, soulamarin induced no up-regulation of ROS compared to untreated trypomastigotes. Within the mitochondria, the primary site of ROS production is the electron transport chain, which involves four protein-associated complexes [48]. Several cellular enzyme systems are potential sources of ROS:NAD(P)H oxidase, xanthine oxidase, uncoupled endothelial nitric oxide synthase (eNOS), arachidonic acid metabolizing enzymes such as cytochrome P-450 enzymes, lipoxygenase and cyclooxygenase, as well as the mitochondrial respiratory chain. Considering that a large number of drugs, which affect mitochondria also contribute to an up-regulation of ROS [49][50], we propose that soulamarin could target T. cruzi mitochondria without affecting the enzymes mentioned above. Conclusion Soulamarin was isolated for the first time from the stem bark of C. brasiliense and showed desirable anti-trypanosomal activity. Our results furthermore indicated that soulamarin-induced death in T. cruzi is associated with mitochondrial dysfunction and a modified permeability of the plasma membrane. Therefore, the natural product soulamarin could serve as a scaffold for the development of selective new drugs against neglected diseases, in particular Chagas disease.