Combining UFLC-QTOF-MS analysis with biological evaluation of Centrosema coriaceum (Fabaceae) leaves

: Centrosema coriaceum Benth belongs to Fabaceae family and have few studies of biological activity and chemical composition. Thus, the aims of this work were to determine chemical profi le of the ethanolic extract of C. coriaceum leaves (CCE) by UFLC-QTOF-MS and to evaluate its in vitro biological potential. CCE showed MIC value of 1000 µg/mL against Candida glabrata (fungistatic effect) and high affi nity in cell envelope by increasing cell permeability in nucleotide leakage, sorbitol and ergosterol assays. CCE showed antioxidant activity in all assays performed. For the anti-infl ammatory and cytotoxicity activities, CCE, at all tested concentrations, signifi cantly inhibited the production of nitric oxide and did not decrease J774A.1 cell viability below 70%. Finally, rutin, kaempferol-3 O -rutinoside, caffeic acid, and sucrose were identifi ed in CCE by UFLC-QTOF-MS . These results suggest, for the fi rst time, that C. coriaceum has interesting antifungal


INTRODUCTION
Fungal diseases are responsible for much of morbidity and mortality in public health systems (Parente-Rocha et al. 2017), causing at least 1.4 million human deaths per year worldwide (Sanglard 2016).The invasive ability of fungi difficults the disease diagnosis and limits the treatment with the available drugs (Souza & Amaral 2017).Moreover, the presence of pathogenic Candida spp.fungi in the body may lead to candidiasis conditions that are accompanied by inflammatory processes (Roselletti et al. 2017).This process releases chemical mediators in the affected tissue, including excitatory amino acids, oxygenreactive species (ROS) and other free radicals, peptides, lipids and cytokines (Chen et al. 2016).In addition, the presence of resistance genes (Candida Drug Resistance -CDR and Multidrug Resistance -MDR) in yeasts of the genus Candida (Khosravi et al. 2016) associated with side effects, toxicity, and ineffi ciency from current candidiasis treatment, have become challenging antifungal therapy through available drugs (Bhattacharjee 2016, Perlin et al. 2017).
The medicinal plants used to treat many diseases are associated with folk medicine from different parts of the world (Moreira et al. 2014).Different cultures know and use the therapeutic potential of plants in the treatment of disease, and it is a practice that has evolved over centuries (Sofowora et al. 2013).Fabaceae is an important plant family used in traditional medicine, which has a wide variety of properties already described, as antifungal, anti-inflammatory, antimicrobial, anticancer, antileishmanial and antibacterial properties (Ahmad et al. 2016, Rahman & Parvin 2014).Nevertheless, some species of this family have few studies of biological activity and chemical composition, such as Centrosema coriaceum Benth, a sub-shrub native to South America present in Cerrado, forest and rupestrian field of Southeast Brazil (Mendonça et al. 1998).
Based on these considerations, this study was undertaken to evaluate the in vitro antifungal, antioxidant and anti-inflammatory activities of ethanolic extract from C. coriaceum leaves and to identify some of its compounds by UFLC-QTOF-MS analysis.

Plant material
The leaves of Centrosema coriaceum Benth were collected in the district of Toledos, Juiz de Fora, State of Minas Gerais, Brazil, on December 4, 2014, at 13:00h.A voucher was deposited in the Herbarium Leopoldo Krieger, Federal University of Juiz de Fora under the number CESJ 49986 and according to the license number A032F41-SISGEN/BRAZIL.

Preparation of the C. coriaceum ethanolic extract (CCE)
The leaves of C. coriaceum were dried in the shade at room temperature for 15 days, and then pulverized to obtain the dry and homogeneous plant material.The dried material was extracted by static maceration for 24h with the ethanol solvent (5 x 500 mL) and concentrated under reduced pressure using the rotatory evaporator (Heidolph -Laborota 4000), obtaining a yield of 28.7% in relation to the dried plant material, and kept under refrigeration until the time of the chemical and biological tests.

Determination of minimum inhibitory concentration (MIC)
The broth microdilution susceptibility assay was performed using the method described by CLSI (2014) for determination of MIC.All tests were performed using Sabouraud dextrose broth (SDB) and Sabouraud dextrose agar (SDA).Strains of five Candida species (Candida albicans ATCC ® 24433, Candida albicans ATCC ® 10231, Candida glabrata ATCC ® 2001, Candida krusei ATCC ® 6258 and Candida tropicalis ATCC ® 750) were grown at 35°C for 24h up to 48h on Sabouraud Dextrose Agar.Successive dilutions of 4000 to 31.3 µg/mL of CCE were prepared in 96-well microplates.For this, 10000 µg/mL stock solutions in 1% dimethylsulfoxide (DMSO) were used.80 µL of this solution was transferred to the microplates, which already contained 100 µL of SDB.To complete the final volume of 200 µL, 20 µL of inoculum (10 4 colony forming units (CFU/mL) according to McFarland's standard turbidimetric scale reaching 2 x 10 3 CFU/mL per well) was added.The plates were incubated at 35 ºC for 24h to visualize fungal growth, except for Candida tropicalis, which was evaluated after 48h.The same tests were performed simultaneously for growth control (SDB + fungus + CCE vehicle) and sterility control (SDB + CCE vehicle).The positive control used was fluconazole at concentrations of 1000 to 8 µg/mL.The MIC was calculated as the lowest dilution showing complete inhibition of the tested strain.The experiment was performed in duplicate.

Minimum fungicidal concentration (MFC)
To determine the MFC, 20 µL from each well that showed no visible fungus growth in the MIC assay was plated on freshly prepared SDA plates and later incubated at 35 °C for 24h up to 48h, according to Spencer & Spencer (2004).Minimum fungicidal concentrations (MFC) were taken as the concentrations that showed no growth of fungal colonies on agar plates.The assay was performed in duplicate.

Effect on cell envelope
Cell envelope assays consisted of assessing the ability of CCE to disrupt cell envelope integrity (nucleotide leakage), cause fungal cell wall damage (sorbitol protection assay) and membrane steroid binding capacity (exogenous ergosterol binding assay) against the species C. glabrata.For the assays on membranes, nystatin was used as a positive control.For this, the MIC value of nystatin for C. glabrata (concentrations ranging from 25 to 0.19 μg/mL) was identified under the same conditions of the MIC assay.

Nucleotide leakage
The experiment was performed according to Tang et al. (2008) with some modifications.The strains (C.glabrata) were incubated in SDB at 35°C for 24h.The culture was then centrifuged at 10,000g for 10 min, washed and resuspended in 10 mM phosphate buffered saline (PBS) (pH 7.4), reaching the final density of about 10 6 cells/ mL.The fungal cells were incubated with CCE and nystatin (MIC values) at different times (0, 1, 2, 3, 4 and 5h); and also incubated only with 10 mM PBS (pH 7.4) (negative control).After incubation, cell suspensions were centrifuged at 10,000g for 15 min and the optical density (OD) at 260 nm was recorded in a spectrophotometer (Multiskan Go, Thermo Scientific, Waltham, MA, United States) at room temperature (2 °C).The experiment was performed in triplicate.

Sorbitol protection assay
After determining MIC value for CCE against C. glabrata, the wall assay was performed using the osmotic protector sorbitol, according to the methodology of Frost et al. (1995).A serial microdilution was performed, in a 96-well sterile microplate, using SDB enriched with 0.8 M Sorbitol.The CCE stock solution was diluted in concentrations ranging from 4000 to 500 µg/mL.For this assay, fluconazole was used as antifungal positive control at concentrations of 250 to 15.6 µg/mL.New MIC values were determined after incubation at 35°C for 24h.The experiment was performed in duplicate.

Exogenous ergosterol binding assay
For the ergosterol-binding assay, the change in MIC value from CCE to C. glabrata was determined according to the methodology of Leite et al. (2014).A serial microdilution was performed in a 96-well microplate using SDB enriched with ergosterol (400 μg/mL).The CCE stock solution was diluted in concentrations ranging from 4000 to 500 µg/mL.New MIC values were determined after 24h of incubation at 35°C.Nystatin was used as positive control.The experiment was performed in duplicate.

2,2-Diphenyl-1-(2,4,6-trinitrophenyl) hydrazyl (DPPH•) radical-scavenging method
The free radical-scavenging activity of DPPH• was determined by the method described by Brand-Williams et al. (1995).Successive dilutions in methanol of CCE were made in 96well plates and the 20 µg/mL of 2,2-diphenyl-1-(2,4,6-trinitrophenyl) hydrazyl (DPPH) solution in methanol was added, thus obtaining CCE concentrations from 250 to 0.49 µg/mL.The plate was incubated in the dark and after 30 min the absorbance was measured at 517 nm.The percentage (%) of CCE oxidation inhibition was calculated as follows: % Inhibition = [(Abs DPPH -Abs sample )/ Abs DPPH ] ×100.Where: Abs DPPH : measured absorbance for DPPH radical in methanol; Abs sample : measured absorbance of the samples.The positive control used was rutin, whose concentration ranged from 250 to 0.0005 µg/mL.The results were expressed in the form of 50% inhibitory concentration (IC 50 ), which was defined as the concentration (in µg/ mL) of sample required to inhibit the formation of DPPH radicals by 50%.The experiment was performed in triplicate.

Total antioxidant capacity by phosphomolybdenum assay
The total antioxidant capacity (TAC) was determined by phosphomolybdenum assay based on the method proposed by Prieto et al. (1999).Test tubes containing 300 μL of the sample/ positives control (quercetin and rutin) in ethanolic solution (500 µg/mL) and 2 mL reagent solution (1.12 mL sodium phosphate solution, 0.48 mL ammonium molybdate and 0.40 mL H 2 SO 4 ) were heated at 95°C for 90 min.The reading was performed by UV spectrophotometer at 695 nm.Results were expressed as a percentage of relative antioxidant activity (RAA%) in relation to the positive controls.The equation for calculating RAA% is shown below (Negri et al. 2009, Zocoler et al. 2009).RAA% = [(Abs sample -Abs blank of sample )/ Abs positive control -Abs blank of positive control ] × 100.Where: Abs sample : absorbance measured for the samples; Abs positive control : absorbance measured for the positive controls; Abs blank of sample : absorbance measured for the sample blank; Abs blank of positive control : absorbance measured for the positive control blank.The experiment was performed in triplicate.

β-carotene/linoleic acid system
Lipid peroxidation inhibition activity was evaluated by the β-carotene/linoleic acid method described by Melo & Mancini Filho (1989) with some modifications.An emulsion containing β-carotene and linoleic acid was prepared and adjusted for a reading between 0.6 and 0.7 of absorbance at 470 nm.250 µL of emulsion and 10 µL of CCE and rutin (positive control) solutions (Final concentration of 38.46 µg/mL) were added.Negative control was performed with 250 µL of the emulsion and 10 µL of methanol (100% oxidation).The plate was incubated at 45°C to accelerate oxidation reactions and initiate β-carotene discoloration.The absorbance of the samples was measured at 470 nm.Readings were taken at zero time and intervals of 15 min until completing 120 min of reaction.
The decrease in sample absorbance (Abs sample (time 0 min to time 120 min)) was related to the decrease in absorbance of the negative control (Abs control (time 0 min to time 120 min)), obtaining the percentage of lipid peroxidation inhibition (% I) by the formula: % I = (Abs control -Abs sample ) x 100/ Abs control .The oxidation curve between the negative control and CCE/rutin was evaluated to find the values of F1 (blocking capacity of peroxide formation) between 15 and 45 min after the beginning of the reaction and F2 (capacity of inhibiting other reactions during the oxidative process, which produces radical species) between 75 and 90 min after the beginning of the reaction by the following formulas: F1 = tg Abs sample / tg Abs control ; F2 = tg Abs sample / tg Abs control (Duarte-Almeida et al. 2006).Where: Abs control : absorbance measured for negative control; Abs sample : absorbance measured for the sample; %I: percentage of lipoperoxidation inhibition; tg: tangent.The experiment was performed in triplicate.

Malondialdehyde quantification
The lipid peroxidation assay determines the presence of malonaldehyde (MDA), from a colored complex which is measured by spectrophotometric method, after its extraction in acid-aqueous extract (Osawa et al. 2005).The concentration of thiobarbituric acid (TBA) -Malondialdehyde (MDA) complex was established from the MDA standard curve using butanol as a blank.Rutin was used as positive control.The preparation of ground beef homogenates containing CCE and rutin in their different concentrations (10, 20 and 40 µg/ mL) were performed.Homogenates were kept in amber flasks and refrigerated at 5°C for the 5 days of the assay.0.5 g of homogenate, 50 µL of BHT 4% in ethanol, 2.5 mL of 1% phosphoric acid and 1.25 mL of TBA 1% were added using test tubes.The solutions were heated at 95°C for 15 min in a water bath and after that process they were cooled in ice bath for 10 min.After cooling, 3 mL of butanol was added to the tubes and centrifuged at 4000 rpm for 5 min.Readings of the supernatant were taken at 535 nm on days 0, 2 and 4. Samples vehicle was used as negative control (100% oxidation).The experiment was performed in triplicate.

Cell line
J774A.1 cells (murine macrophages) were grown in culture bottles with Roswell Park Memorial Institute (RPMI)-1640 medium supplemented with 2mM L-glutamine, 100 µg/mL antibiotic (streptomycin and penicillin), 5% fetal bovine serum (FBS) and kept in an incubator at 5% CO 2 at 37°C until the day of the test.

NO dosage
Macrophage cells of the J774A.1 cells were transferred to 96-well microplates (2x10 4 cell well -1 ) and treated with CCE at concentrations ranging from 60 to 3.75 µg/mL.Subsequently, Escherichia coli LPS (lipopolysaccharide) at 1 µg/mL and IFN-γ at 0.9 ng/mL were added and the volume made up to 200 µL with RPMI-1640 medium.Negative controls were stimulated with LPS and IFN-γ and cells treated only with 0.06% DMSO (vehicle).The cells were incubated for 48h in a 5% CO 2 at 37°C.Subsequently, nitric oxide (NO) production was evaluated by the Griess method, an indirect NO measurement by nitrite dosing, in recovered culture supernatant after 48h stimuli and treatment as described by Sun et al. (2003).Briefly, 50 μL of Griess reagent in phosphoric acid 2.5% (v/v) were added to 50 μL of culture supernatant.The quantity of NO (μM) was determined by comparison with a standard sodium nitrite curve.The spectrophotometry reading was performed at a 540 nm wavelength.The experiment was performed in triplicate.

MTT assay
J774A.1 cells at 2x10 4 cell/well were transferred to 96-well microplates, and 24h later, treated with CCE at concentrations of 60 to 3.75 µg/ mL.For the negative control, cells treated only with 0.06% DMSO (vehicle) were used.The cells were incubated for 48h in a 5% CO 2 and 37ºC.Cytotoxicity was assessed by cell viability by the 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay (Mosmann 1983).The assay for viability assessment is based on the metabolic reduction of MTT to formazan (violet coloration).After 48h of the treatment, MTT solution (500 μg/mL) was added to each well, and the plate was incubated for additional 2.30h at 37°C, 5% CO 2 .Absorbance (Abs) was read at 570 nm and cell viability (%) of macrophages after treatment obtained by the following formula: % viable cells = [(Abs sample )/ Abs 0.06% DMSO ] × 100.Where: Abs sample : absorbance measured after treatment of cells with the sample; Abs 0.06% DMSO : absorbance measured after treatment of cells with 0.06% DMSO (vehicle).The experiment was performed in triplicate.

Propidium iodide using flow cytometry
The evaluation of CCE influence on J774A.1 murine macrophages viability was performed by flow cytometry following the methodology of Crowley et al. ( 2016) with some modifications.2x10 4 cells per well were transferred to 96-well plates and incubated for 48h at 37°C and 5% CO 2 with CCE (60 and 30 μg/mL).At the end of the incubation time, the supernatants were discarded and 200 µL of flow cytometry staining buffer (FACS) buffer solution (PBS with 1% SFB) was added.The cells were placed on ice and treated with trypsin, stained with propidium iodide (10 µg/ mL) 15 min before reading and collected into FACS (CORNING) tubes.Samples reading were performed in a FACS canto II flow cytometer, where 3000 events were acquired.Analyses were performed using FlowJo® software where the percentage of viable cells (without propidium iodide staining) was determined.Cells treated with 0.06% DMSO were used as negative control.The experiment was performed in triplicate.

Chromatographic conditions
CCE was analyzed by ultra-fast liquid c h ro m a to g ra p h y co u p l e d w i t h m ass spectrometry (UFLC-QTOF-MS) in positive [M + H] + mode using a Shimadzu UFLC (Nexera model) and a Bruker mass spectrometer (QTOF Compact model) with electrospray ionization source.The mobile phase used was acidified water with formic acid, with pH = 3 (phase A) and methanol (phase B), the injection flow was 0.4 mL min -1 and the running time was 12 min.The column used was the Kinetex 2.6 µm -C18 -100A, length 100 mm X 3.0 mm.The chromatographic running began with 40% of phase B in 0.01 min time, reaching up to 70% of B in 8.20 min time and 95% of B from 8.20 to 9.70 min, subsequently the mobile phase returned to 40% B in the time of 10.20 min, following up to 11.50 min, the running ended in 12 min.The ionization conditions were set as follows: ion source electrospray voltage of 40 V, a capillary voltage of 4500 V, and a capillary temperature of 220 °C.The full scan mass acquisition was performed by scanning from 100 up to 1000 m/z range.The experiment was performed in duplicate.

Statistical analysis
Statistical analysis was performed by ANOVA test followed by Bonferroni using the software GraphPrism 5.0.In some cases, the Student's t test was applied.Values of p<0.05 were considered significant.Results were expressed as mean ± standard deviation.

Evaluation of antifungal activity
The antifungal activity of CCE was evaluated by the MIC on different species of Candida spp.The results are shown in Table I.At the MIC value, a growth inhibition was observed in both yeasts (fungistatic effect).The MFC was not able to be determined at the concentrations tested.
C. glabrata was more susceptible to CCE and, therefore, the mechanism of action for this yeast was investigated.The action assay over plasma membrane showed a significant increase (p<0.05) by nucleotide extravasation was performed in absorbance at 260 nm after the cells treatment with CCE and nystatin at MIC values when compared to the growth control (Figure 1).The action on the cell wall and the ability to bind to fungal membrane steroids using sorbitol and exogenous ergosterol were investigated, respectively.The results found in both assays demonstrated an increase in MIC value for CCE (greater than 4000 µg/mL), evidencing its action on C. glabrata cell envelope (Table II).

Antioxidant activity
Antioxidant activity of CCE was determined using DPPH radical scavenging, phosphomolybdenum and β-carotene/linoleic acid system assays and malondialdehyde quantification.The results are shown in Table III and Figure 2. Rutin and quercetin were used as positive controls.

Anti-inflammatory activity
CCE anti-inflammatory activity was assessed by NO dosage (µM) in the J774A.1 macrophage supernatant stimulated with LPS and with INF-γ (Figure 3).CCE reduced NO production in all tested concentrations (p<0.05).It is noteworthy that CCE, at 60 and 30 µg/mL, inhibited 81 and 62% of NO production, respectively.
Compound 1 was identified as sucrose and showed retention time at 1. Compound 2 was identified as rutin and showed retention time of 3.8 min and UV with λ max at 204, 256 and 355 nm, mass fragments in

DISCUSSION
Brazilian flora is one of the most importants and diverses in the world.Despite plant richness, only 15% of species have been investigated for chemical and pharmacological properties (Wurtzel & Kutchan 2016).In the other hand, the high morbidity and mortality rates caused by fungal diseases and the current antifungal arsenal reduced to high toxicity drugs make it important to discover new therapies that may have fewer adverse effects (Scorzoni et al. 2017).Thus, our research group conducted this work, with C. coriaceum ethanolic extract looking for its biological activities and studying its chemical composition, hoping that it could be a new source of bioactive molecules in the treatment of fungal diseases, oxidative stresses and inflammatory process.
According to Simões et al. (2009), extracts from plants with the minimum inhibitory concentration (MIC) ≤ 1000 μg/mL are considered promising antimicrobial agents.Therefore, our present results demonstrate that CCE possesses a promising inhibitory activity for C. glabrata (MIC values =1000 µg/mL), with fungistatic effect.This fungi species has shown increased resistance in recent years (Whaley & Rogers 2016).The possible action mechanism of CCE in C. glabrata was investigated by cell envelope assays (nucleotide leakage, sorbitol and binding to exogenous ergosterol).Nucleotides are molecules that have a peak in the absorbance of light at the wavelength of 260 nm.So, an increase in absorbance at this wavelength considering the extracellular medium is due to the presence of these molecules (Khan et al. 2013).Since the nucleotides are located in intracellular compartments, their extracellular presence is a sign of cell damage due to membrane rupture (Campos et al. 2018).The result found for CCE nucleotide leakage assay was similar to nystatin result (Figure 1).Both led to an increase of nucleotides in the extracellular medium, measured by the increased absorbance.The nystatin mechanism of action is already well known by membrane sterols binding, which leads to a cell membrane permeability alteration and consequent leakage of the cytoplasmic content.This information allows us to infer that CCE acts in the C. glabrata membrane.
Deepening studies on this mechanism of action was carried out using the tests with sorbitol and ergosterol.Sorbitol has an osmoprotective function and is essential for fungal growth (Frost et al. 1995).Fungal cells with cell wall damage cannot grow in the absence of sorbitol due to loss of their osmotic regulation.However, their growth is still possible if sorbitol is supplemented to the culture medium.Therefore, fungal cell wall inhibitors can be identified when MIC values obtained in the presence of sorbitol are higher than those in its absence, since the osmolality of the cells is conserved by sorbitol and this permits their growth even in presence of a cell wall inhibitor drug (Frost et al. 1995).The MIC values of CCE for C. glabrata increased when the assay was performed in the presence of 0.8M sorbitol in the culture medium (Table II).This result indicates a possible action of the CCE on the fungal cell wall inhibition.
Ergosterol is an anchored lipid in fungal membrane whose absence/lesion causes changes in membrane permeability (Iwaki et al. 2008).If the antifungal agent has ergosterol binding activity, the presence of exogenous ergosterol will prevent binding to membrane ergosterol, increasing the MIC of the substance (Leite et al. 2014).Our test to evaluate this ergosterol action resulted also in an increase in MIC value on the ergosterol presence (Table II), suggesting the ability of constituents present in CCE to bind to fungal ergosterol.
Once antifungal activity was detected and the mode of action was reached, others biological properties were researched.The infection process is always linked with the inflammatory process.So, considering that the production of free radicals, including reactive oxygen species, is crucial inflammation mediators that can cause cell damage (Conner & Grisham 1996), different antioxidant tests were performed.CCE was evaluated by different antioxidant tests and showed promising results in all assays (Table III).The free radicals are atoms or molecules containing one or more unpaired electrons in their valence shell (Phaniendra et al. 2015) which are produced by different biological and environmental sources (Arulselvan et al. 2016).When in presence of a hydrogen donor substance, the DPPH free radical (violet coloration) is reduced (yellow coloration) and its absorption intensity decreases (Duarte-Almeida et al. 2006, Kedare & Singh 2011).In this study, it was observed that CCE shows hopeful findings by inhibiting concentration of 50% DPPH radical, and this activity is related to the presence of phenolic compounds in CCE, such as rutin, used as positive control, a flavonoid widely known for its antioxidant property (Koval'skii et al. 2014).The data obtained by phosphomolybdenum assay corroborate the DPPH assay.The principle to assess the antioxidant capacity by phosphomolybdenum assay includes a green phosphate/molybdenum complex formed due to the reduction of molybdate VI to molybdate V by antioxidant substances (Mendes et al. 2017).CCE presented a relevant activity for this assay, demonstrating that it possesses reducing substances in its composition (Table III).
The presence of free radicals due to pathological conditions increases the rates of lipid peroxidation tissues and leads to the formation of the peroxyl lipid radical (LOO⋅), which may posteriorly give rise to peroxides (LOOH), a second lipid radical.Peroxides are decomposable and give rise to aldehydes, reactive substances that are also responsible for causing cellular damage (Yadav & Ramana 2013).The results found by the inhibition of linoleic acid peroxidation are promising, being CCE considered significantly better than the rutin (Figure 2).
The β-carotene/linoleic acid test still allows us to calculate F1 and F2 values that can evaluate the antioxidant capacity of CCE to act on the blockage of peroxide radicals formed in the propagation stage or to act on other radical species in the terminal stage of lipid peroxidation (peroxide decomposition products) (Duarte-Almeida et al. 2006).From the values F1 and F2, we can suggest that CCE possesses a greater ability to block the formation of peroxide radicals than other radical species formed in the terminal phase of lipid peroxidation.
The lipoperoxidation process generates, as one of the main aldehydes, malondialdehyde (Ayala et al. 2014).The blockage of malondialdehyde formation was verified by complexing this substance with thiobarbituric acid.As can be observed in Figure 3, CCE was able to block malondialdehyde formation during the oxidation process from the second day of this assay.
Since the free radicals are mediators of the inflammatory process, from the positive results in the tests of antioxidant activity, tests were made to evaluate the anti-inflammatory activity.Regarding the anti-inflammatory activity, a statistical difference was observed between cells treated with CCE at the concentration of 60 µg/ mL and cells without stimulation.These results are encouraging since NO is a signaling molecule that plays a key role in the pathogenesis of inflammation (Sharma et al. 2007).Thus, the discovery of new medicinal plants that can be used in anti-inflammatory therapies is of great relevance, since some of the commercially available anti-inflammatory drugs, such as nonsteroidal anti-inflammatory drugs (NSAIDs), may increase the risk of gastrointestinal and cardiovascular complications (Sostres et al. 2010).
Ideal pharmacotherapies are those that are effective and have low toxicity to human body cells (Piccolo et al. 2015).Therefore, in order to evaluate the cytotoxicity of CCE, a cell viability assay was performed, and the viable percentage determined according to ISO 10993-5 (2009) (cell viability > 70 %).Thus, it was possible to observe that CCE does not reduce the viability of normal cells for all concentrations tested by MTT and propidium iodide assays.
Considering that CCE presents promising biological activities, it is important to know its chemical constituents.LC-MS is one of the major analytical techniques to determine global metabolite profiles.This tool allows you to analyze and differentiate several compounds from a plant (e.g., alkaloids, glycosides, phenyl propanoids, flavonoids, isoprenes, glucosinolates, terpenes, benzoids) with medium and high polarity and/ or higher molecular weight (Kumar et al. 2016).Therefore, we use UFLC-QTOF-MS to identify the compounds present in CCE.The results suggested the presence of two flavonoids, rutin and kaempferol-3O-rutinoside, one phenolic acid, caffeic acid, and one disaccharide, sucrose.Thereby, rutin and kaempferol-3O-rutinoside may be related to the results obtained in this work.Rutin is a flavonoid well known in the literature for its ability to neutralize free radicals and anti-inflammatory activity due to inhibition of nitric oxide synthase enzymes, phospholipase and suppression of inflammatory genes (Koval'skii et al. 2014), besides presenting antifungal properties (Han 2009).Kaempferol-3O-rutinoside is another flavonoid that also has antioxidant and antifungal activities already reported (Nascimento et al. 2018).The caffeic acid has also been identified in other species of the Fabaceae (Hanganu et al. 2016) family and is known for its antifungal (Teodoro et al. 2015), 2 min, UV with λ max at 195 nm and mass fragments at m/z 365.1051 [M+Na] + , m/z 203.0525 [M+Na-162] + and m/z 138.0551 [m/z 203-65] + by ESI+ .

Figure 2 .
Figure 2. Antioxidant activity of the ethanolic extract of Centrosema coriaceum leaves (CCE).a) Oxidation curve of β-carotene absorbance evaluated at 470 nm, demonstrating the preservation of its coloration during the 120 min of reaction when using CCE and rutin.b) Inhibition of malondialdehyde production after treatment with CCE and rutin.a Statistical difference considering the negative control.Results expressed as mean ± standard deviation of three independent experiments.CONTROL -Negative control.RUT -rutin.

Figure 3 .
Figure 3. Anti-inflammatory activity of different concentrations of the ethanolic extract of Centrosema coriaceum leaves (CCE) by the method of nitric oxide (NO) reduction.a Statistical difference when compared to cells stimulated with INF-γ and LPS and cultured with RPMI + 0.06% DMSO (STI).b Statistical difference when compared to cells treated with CCE (30; 15; 7.5 and 3.75 µg/mL).c No statistical difference between cells without stimulation (NSTI) and cells treated with CCE (60 µg/ mL).d No statistical difference between the concentrations of 30; 15; 7.5 and 3.75 µg/mL of CCE.Results expressed as mean ± standard deviation of two independent experiments.(ANOVA followed by Bonferroni, p<0.05).

Figure 4 .
Figure 4. Evaluation of cell viability in J774A.1 macrophages cell line treated with ethanolic extract of Centrosema coriaceum leaves (CCE).a) MTT method.b) Flow cytometry using propidium iodide (PI) as stain in the concentrations of 30 and 60 μg/mL.Cells treated with 0.06% DMSO was used as negative control.a Statistical difference when compared to negative control.b Statistical difference when compared to CCE (30 µg/mL).Results expressed as mean ± standard deviation of two independent experiments (ANOVA followed by Bonferroni, p<0.05).

Table I .
Minimal Inhibitory Concentration (MIC) of ethanolic extract of Centrosema coriaceum leaves (CCE) against Candida genus.
aValues expressed in µg/mL of sample.
aValues expressed in µg/mL of sample.

Table III .
Antioxidant activity of ethanolic extract of Centrosema coriaceum leaves (CCE) by DPPH, total antioxidant capacity by phosphomolybdenum (TAC) and inhibition of the β-carotene/linoleic acid system assays.