Hydroalcoholic Extract and Ethyl Acetate Fraction of Bixa orellana Leaves Decrease the Inflammatory Response to Mycobacterium abscessus Subsp. massiliense

The incidence of infections caused by rapidly growing mycobacteria (RGM), especially Mycobacterium abscessus subsp. massiliense (Mabs), is increasing worldwide. Severe infections are associated with abscess formation and strong inflammatory response. This study evaluated the antimicrobial and anti-inflammatory activities of a hydroalcoholic extract (BoHE) and ethyl acetate fraction (BoEA) of Bixa orellana leaves. Antimicrobial activity was evaluated by broth microdilution to determine the minimum inhibitory (MIC) and the minimum bactericidal (MBC) concentrations. Cytotoxicity was evaluated using erythrocytes and RAW 264.7 cells. Nitric oxide (NO) was assayed in stimulated RAW 264.7 cells, and inflammatory cell migration and acute toxicity were evaluated in a Mabs-induced peritonitis mouse model. The compounds present in BoEA were identified by high performance liquid chromatography and mass spectrometry (HPLC-MS). The MIC and MBC values were 2.34 mg/mL and 37.5 mg/mL for BoHE and 0.39 mg/mL and 6.25 mg/mL for BoEA. The extracts did not induce significant toxicity in erythrocytes and RAW 264.7 cells. High levels of NO induced by Mabs were decreased by treatment with both extracts. The anti-inflammatory activity was confirmed in vivo by significant reduction of the cell migration to the peritoneum following BoHE and BoEA pretreatment. Animals treated with BoHE or BoEA did not show signs of acute toxicity in stomach, liver, and kidney. The chemical characterization of BoEA (the most active extract) revealed that kaempferol-3-O-coumaroyl glucose is its major component. The extract of B. orellana may be effective for treating infections caused by Mabs.

Previous studies have described Bixa orellana (also known as urucum) as a source of antimicrobial agents. The seeds of this plant are currently used in folk medicine to treat heart disease, gastrointestinal problems, respiratory infections, burns [14], diabetes, skin infections, fever, measles, gonorrhea, diarrhea, and asthma [15]. The antimicrobial action [16] and antioxidant activity [17] of the methanolic extract of B. orellana seeds and leaves against Gram-positive bacteria (e.g., S. mutans and S. sanguinis) have previously been described. In addition, the antimalarial activity of essential oils and root extracts [18] and the anti-inflammatory effects of aqueous extracts of the leaves [19,20] have been also reported.
This study evaluated the antimicrobial and antiinflammatory activities of BoHE and BoEA leaf extracts of B. orellana, which is a plant widely used in Brazilian folk medicine to treat Mycobacterium infections and other inflammatory disorders [14]. The antimicrobial activity was tested against M. abscessus subsp. massiliense (Mabs), an emerging human pathogen implicated in hospital outbreaks with natural resistance to several last generation antimicrobials. The cytotoxic effects of BoHE and BoEA were assayed using human erythrocytes and murine RAW 264.7 cells. The acute toxicity and the in vivo anti-inflammatory activities of the extracts were evaluated in a murine peritonitis model induced by inactivated Mabs.

Collection of B. orellana Leaves and Preparation of
Extract and Fractions. Fresh B. orellana leaves were collected (June-July 2016) and identified at theÁtico Seabra Herbarium of Federal University of Maranhão (São Luís, Brazil; voucher specimen 1147). The cleaned leaves were oven dried at 40 ∘ C for 3 days, ground in a mill, and extracted for 24 h in 70% ethanol by maceration with agitation. The sample was filtered, and the resulting filtrate was concentrated in a rotary evaporator under low pressure at 50 ∘ C. The concentrate was lyophilized (named BoHE) and stored at -20 ∘ C until use. BoHE was submitted to liquid-liquid fractionation using hexane, chloroform, and ethyl acetate with a series of increasing polarity to produce hexane (BoHex), chloroform (BoCl), and ethyl acetate (BoEA) fractions. BoHE and BoEA were resuspended in a saline solution (0.9% NaCl).

Determination of Total Phenol and Total Flavonoid
Contents. The content of total phenols was estimated using the Folin-Ciocalteau method for total phenols (Singleton et al., 1965), with some modifications. Samples of 20 L of each extract (1 mg/mL) and 100 L of the Folin-Ciocalteu reagent were mixed; and after 3 minutes of incubation at room temperature, 80 L of a solution of sodium bicarbonate (0.7 M) was added. The reaction was kept in the dark for 2 h at room temperature and the absorbance was measured at 735 nm using a microplate reader (BioTek UQuant MQX200). PBS (vehicle) was used as negative controls. Gallic acid was used as standard, and the results were calculated based on the calibration curve of gallic acid (10-100 mg/L) and expressed as mg equivalent of gallic acid per gram of extract (GAE /g extract).
The flavonoid content was determined according to the colorimetric method of aluminum chloride (Woisky and Salatino, 1998). The extracts were tested at the concentration of 1 mg/mL, and the quercetin was used as standard compound (10-100 mg/L). The sample (100 L) was mixed with 100 L of the reagent [2% aluminum chloride (AlCl 3 ) in methanol]. After 1 h on incubation in the dark at room temperature, the absorbance was read at 420 nm. PBS was used as negative control. The results were expressed as mg equivalent of quercetin per gram of extract (mg QE/g).

RGM
Strain. An isolate of M. abscessus massiliense (Go01) obtained from patients with hospital outbreaks of postoperative infections [21] was used in the study. The sample was generously provided by the Laboratory of Immunopathology of Respiratory Infections from the Institute of Tropical Pathology and Public Health, the Federal University of Goiás. The use of the clinical isolate was authorized by freely given written informed consent using a form approved by the institutional ethics committee (N ∘ : 21357413.4.0000.5084).

Antimicrobial Susceptibility. Minimum inhibitory (MIC)
and minimal bactericidal (MBC) concentrations were determined as described in the ATS/IDSA guidelines [5]. In brief, the extracts were diluted (from 200 mg/mL to 0.04 mg/mL) in microplates and microbial suspensions (at 2x10 3 CFU/mL) were added. After 72 h of incubation at 37 ∘ C, a resazurin suspension (0.01% w/v) was added followed by a further incubation (16 h incubation at 37 ∘ C). A pink color indicated the presence of living bacteria and a purple color indicated dead bacteria. Amikacin was used as a control. The extract concentrations without visible growth were inoculated on MH agar plates. The MBC was determined after 5-7 days incubation at 37 ∘ C. The MBC was the lowest extract concentration that prevented 99.9% bacterial growth expressed in CFU/mL [22,23].

Cytotoxicity Assays
2.5.1. Hemolytic Activity. Heparinized blood was obtained from three healthy volunteers (types A, B or O serotypes) that were nonsmokers and had not taken any medication for at least 15 days. They were 18-35 years of age and had given written informed consent. The protocol (N ∘ 1.570.437) was approved by the Research Ethics Committee of Ceuma University. Erythrocytes were obtained by centrifugation at 1500 rpm for 10 min immediately after collection, removal of the plasma, and washing three times in pH 7.4 phosphate Evidence-Based Complementary and Alternative Medicine 3 buffered saline. After washing, a 1% erythrocytes suspension was prepared in PBS. Samples (0.4 mL) of BoHE (0.58 mg/mL to 9.36 mg/mL) or BoEA (0.09 mg/mL to 1.56 mg/mL) were added to 1.1 mL erythrocyte suspension. PBS and 0.05% Triton X-100 solution were used as negative and positive controls, respectively. After 60 min incubation at room temperature, the cell suspensions were centrifuged, and the concentration of hemoglobin in the supernatant was measured at 540 nm [24]. The result was reported as the mean of three independent assays. Hemolytic activity was expressed in relation to Triton X-100 and calculated using the following formula: % hemolysis = [(As − Ab) × 100] / (Ac − Ab), where Ab is absorbance of control (blank without extract), As is absorbance in the presence of the extract, and Ac is absorbance in the presence of Triton X-100.

Cytotoxicity towards Murine RAW 264.7 Cells.
Murine RAW 264.7 cells were maintained in Dulbecco's high glucose modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Sigma-Aldrich), streptomycin 100 g/mL, and penicillin 100 UI/mL. After growth was established, cells were plated in 96-well plates at 2 x 10 5 cells/well and treated with BoHE (1.17 mg/mL to 9.36 mg/mL) or BoEA (0.19 mg/mL to 1.56 mg/mL). DMEM and 0.05% Triton X were positive and negative controls, respectively. The plated were incubated at 37 ∘ C and 5% CO 2 . After 24 h, the medium of each well was removed and DMEN medium containing 5 mg/mL MTT (3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide; Sigma-Aldrich) was added to each well. The plates were incubated for 4 h at 37 ∘ C and 5% CO 2 in the dark. The supernatants were discarded, and the formazan crystals were dissolved in 100 L dimethyl sulfoxide (DMSO). Absorbance was read at 540 nm using a microplate reader (Thermo Plate), and the results were expressed as a percentage of the maximal value of the positive control and reported as means of three independent assays ± the standard deviation [25,26].
2.6. In Vitro Anti-inflammatory Assay. The effects of BoHE and BoEA on nitric oxide (NO) production by murine RAW 264.7 cells induced by Mabs were evaluated using Griessbased assay. For this, RAW 264.7 cells were plated at a density of 2 x 10 5 cells/well in 96-well plates and infected with Mabs at 1 x 10 8 CFU/mL. After 1.5 h, the cells were exposed to BoHE (1.17 mg/mL and 9.36 mg/mL) or BoEA (0.19 mg/mL to 1.56 mg/mL). Mabs-infected cells without any extract treatment were considered as positive control. Cells treated and incubated with 1 g/mL LPS (E. coli 0111 lipopolysaccharide; Sigma-Aldrich, Saint Louis, MO, USA) and 100 g/mL IFN-(Interferon-gamma; BD Pharmingen) were also used as positive control. After 24 h incubation at 37 ∘ C in 5% CO 2 , the supernatants were collected (50 L) and incubated with 50 L Griess reagent (1% sulfanilamide, 0.1% naphthyl ethylenediamine, and 2.5% phosphoric acid) for 10 min at room temperature in the dark, and the absorbance at 540 nm was compared with that of a standard curve derived from 0−300 M NO [26]. Results were expressed as M.

Animal Experimentation
2.7.1. Animals. Female C57BL/6 mice 6 to 8 weeks of age, weighing from 20 g to 25 g, were housed in plastic cages at room temperature (23 ± 1 ∘ C) and a 12 h light-dark cycle and given balanced laboratory food and water ad libitum. All experimental procedures were conducted following the laboratory animal care standards of the Ceuma University Animal Experimentation and Use Committee (approval N ∘ 107/14) in accordance with the UK Animals (Directive 2010/63/EU).

Clinical Isolates and Preparation of Inocula.
Mabs bacterial suspensions were adjusted to 1 × 10 8 CFU/mL in pH 7.4 phosphate buffered saline with vigorous stirring to disperse cell clumps. The bacterial cells were heat killed at 90 ∘ C for 1 h.

Experimental Design and Induction of Peritonitis by
Mabs. Animals were randomly allocated into seven groups (n= 5-6 animals) that received a single oral dose of 0.9% saline solution at 1 mL/kg (groups I and II); BoHE at 50 mg/kg (group III) or 150 mg/kg (group IV); BoEA at 50 mg/kg (group V) or 150 mg/kg (group VI); and 5 mg/kg dexamethasone (group VII). After 1 h of treatment, the animals from groups II to VII received intraperitoneal injections of 0.1 mL 1×10 7 CFU heat killed Mabs.

Determination of Cell Migration to Peritoneal Cavity.
Peritoneal liquid was obtained 4 h and 24 h after induction of peritonitis. For this, each animal received 80 mg/kg ketamine hydrochloride and 10 mg/kg xylazine hydrochloride. Lavage was performed with the introduction of 3 mL of 1 mM EDTA in PBS into the abdominal cavity with a sterile, disposable syringe, and needle. The aspirated contents were transferred to a tube. The leukocytes present in the peritoneal liquid were counted in a Neubauer chamber after a 1:2 v/v dilution in Turk's solution. The differential leukocyte count was performed in a 100 m hanging drop of sample obtained by cytocentrifugation at 600 rpm for 10 min. The slides were Giemsa stained, and the 100 cells were counted by optical microscopy at 1000 × using an oil immersion objective. 2.7.5. Acute Toxicity Assay. Stomach, liver, and kidneys were in fixed 10% paraformaldehyde, embedded in paraffin, sectioned at 5 m, and stained with hematoxylin and eosin (H&E). The slides were evaluated by optical microscopy (Axio Imager Z2; Carl Zeiss, Oberkochen, Germany), at 40× to 400× increments, and ten fields per slide were evaluated by an experienced pathologist for the presence or absence, distribution, and severity of histological changes.

HPLC-DAD-ESI-IT/MS
Analysis. The chemical constituents of BoEA were analyzed in a high-performance liquid chromatography (HPLC) using a LC-10AD system (Shimadzu, Japan) equipped with a photodiode array detector (DAD) and coupled to an Esquire 3000 Plus ion-trap mass spectrometer (Bruker Daltonics, Bremen, Germany), using electrospray ionization (ESI) using argon as collision gas and 80 eV as collision energy. Separation was performed using The injection volume consisted of 25 L of reconstituted sample at 5 mg/mL and a flow rate of 0.6 mL/min. Detection was by a diode array detector (DAD) at 200-500 nm and a direct mass spectrometry/mass spectrometry method in negative electrospray (−ESI) mode with the detector voltage maintained at 4.0 kV, an ion source of 40 V, and capillary temperature of 320 ∘ C. The nebulizing gas was nitrogen (N 2 ) flowing at 7 mL/min and sheath gas provided at a pressure of 27 psi, and helium was used as the four collision gas. The analyses were performed using full-scan mass spectra and data-dependent MS2 scans from 100 to 2000 m/z. The compounds were identified on the basis of their molecular ion mass fragmentation. The mass spectra were compared with those previously reported [27].

Statistical
Analysis. Data were presented as means ± standard variation (SD) or percentages. The normality of distributions was determined by the Shapiro-Wilk test, and the differences between groups were evaluated by analysis of variance (ANOVA) followed by Tukey's multiple comparison test using the Graph Prism 6.0 software. P-values < 0.05 were considered significant.

B. orellana Extracts
Have Antimicrobial Activity against M. abscessus massiliense. M. abscessus massiliense (Go01) was susceptible to the B. orellana leaf extracts. The BoHE showed a MIC of 2.34 mg/mL and the MBC was 37.5 mg/mL, while MIC found in BoEA was 0.39 mg/mL and the MBC was 6.25 mg/mL. For both samples, the MBC/MIC ratios were > 4, which indicated probable bacteriostatic activities [28]. The best results observed for BoEA may be related to its higher phenolic and flavonoid content. BoEA showed a total phenolic content of 1554.75 ± 56.14 mg GAE/g and total flavonoid content of 146.55 ± 2.38 mg QE/g; these values were higher than those found for BoHE (1025.5 ± 66.33 mg GAE/g; 64.82 ± 2.05 mg QE/g).

Cytotoxic Evaluation of B. orellana Extracts.
The cytotoxic effects of BoHE and BoEA towards erythrocyte and RAW 264.7 cells are shown in Figure 1. The extracts did not induce significant erythrocyte toxicity; the results were similar to those observed for erythrocytes treated with PBS (it was not possible to calculate IC50). Regarding the results with RAW 264.7 cells, the IC50 values were 7.44 mg/mL for BoHE and 7.716 mg/mL for BoEA. The selective index (SI, i.e., the ratio between IC50/MIC values) was 19.8 for BoEA and 3.2 for BoHE. The higher SI value for BoEA indicated a greater antimicrobial efficiency. Figure 2 shows the effects of B. orellana extracts on NO production by RAW 264.7 macrophages infected by Mabs. As expected, the cells infected by Mabs exhibited high levels of NO in their supernatant, which were similar to the values found for the treatment with LPS. This effect was significantly inhibited by the treatment with all concentrations of BoEA (Figure 2(b)) and the lower concentrations of BoHE (0.19 mg/mL and 0.39 mg/mL) (Figure 2(a)).  (Figure 3(a)).
Finally, the effects of BoEA and BoHE on mononuclear cells (MNs) migration induced by Mabs inoculation were also determined (Figures 3(e) and 3(f)). At 4 h after the  (Figures 4 and 5). The maintenance of esophageal and gastric mucosa confirmed the satisfactory performance of the therapeutic gavage protocol. The architecture of the evaluated organ tissues was preserved, with normal histological characteristics. Figure 6 illustrates the HPLC-DAD chromatogram of the phenolic compounds from ethyl acetate phase from hydroalcoholic extract from leaves of B. orellana. A total of nine peaks corresponding to compounds 1-9 were tentatively identified on the basis of their retention times and MS pattern with also taking into account data in the related literature.  1 6 . 5 g r a n a t i n B  Table 1.

Discussion
This study evaluated the antimicrobial and anti-inflammatory activities of B. orellana, a species widely used in folk medicine for the treatment of infections, headache, dysentery, fever, indigestion, and skin diseases [29,30]. The tested B. orellana extracts had antimicrobial activity against M. abscessus. Although B. orellana is widely used to treat respiratory infections, this is the first report of its kind. The study results are consistent with a previous study reporting MIC values of 0.3, 0.5, and 0.2 mg/mL for B. orellana leaf, seed, and root extracts against M. tuberculosis [31]. Only few studies have described the antimycobacterial activity of plant-derived products against M. abscessus, for example, the effects of chloroform and ethanolic extracts from seeds of Persea americana (where the activity of the extracts was associated with lignans) [32], Valencia orange terpeneless oil [33], and Pelargonium reniforme and P. sidoides root extracts (containing linear chain fatty acids such as palmitic, oleic, and linoleic) [34]. Inflammatory responses play important roles in host defense and also contribute to immunopathogenesis during mycobacterial infection [35]. M. abscessus is a respiratory pathogen causing chronic lung diseases and infections associated with cystic fibrosis [36]. A strain of M. abscessus forms colonies with rough morphology and is known to induce inflammatory responses associated with invasive disease [37]. The anti-inflammatory activity of BoHE and BoEA was evaluated in a C57BL/6 mouse model of peritonitis induced by heat killed M. abscessus. M. abscessus is an etiological agent of infections associated with laparoscopic gastric bands [38] and causes acute peritonitis that can evolve with ascites and formation of granulomas [39].
In the study model, oral pretreatment with BoHE or BoEA (both at 50 mg/kg or 150 mg/kg) significantly reduced  the migration of leukocytes to the peritoneal cavity. The observation that the effects were similar to those observed with the anti-inflammatory dexamethasone suggests that the extracts had secondary metabolites with anti-inflammatory activity. The anti-inflammatory activity of B. orellana-derived products has been previously reported [19,20,40]. In this study, BoHE and BoEA reduced NO production in Mabs-infected RAW 264.7 murine macrophages. It has been shown that peritoneal macrophages of C57BL/6 mice produce high levels of NO after infection with M. abscessus [41]. NO produced by activated cells has cytotoxic and microbicidal action that promotes the destruction of invading microorganisms [42]. BoHE and BoEA might reduce NO production by inhibiting enzymes responsible for the production of induced nitric oxide synthase (iNOS) [43].
The migration of leukocytes into inflamed tissue from the microcirculation is characteristic of acute inflammation [44]. The anti-inflammatory activity of BoHE and BoEA extracts  was evaluated in a model of peritonitis induced by heat killed M. abscessus in which treatment with the extracts reduced the peritoneal migration of leukocytes. The BoHE and BoEA extracts were both protective against the development of acute inflammation in the mouse model used in this study. The B. orellana active extract contained several flavonoids, such as kaempferol, ellagic acid, and HHDP glucose, which are known immunomodulators [45][46][47][48][49].
Determining the toxicity of natural products with therapeutic potential is fundamental to the investigation of their bioactive potential. Effective and safe therapeutics are more active against pathological agents than the cells and tissues of the organism being treated. Histological analysis of esophagus, stomach, liver, and kidney tissue confirmed that the BoHE and BoEA treatment regimens, the concentration, the route of administration, and the exposure time were neither hepatotoxic nor nephrotoxic. These are essential criteria when evaluating new therapeutic compounds. Oral administration, as used in this study, did not disturb the normal histology of the esophageal or gastric mucosa. The absence of mucosal damage allows for drug absorption and satisfactory biotransformation, both of which are essential [50].

Conclusions
The extracts of B. orellana leaves (BoHE and BoEA) had antimicrobial activity against M. abscessus subsp. massiliense and low toxicity against murine RAW 264.7 cells. The antimycobacterial activity was accompanied by significant anti-inflammatory activity in a peritonitis model induced by M. abscessus subsp. massiliense in mice. Inhibition of the migration of leukocytes to the site of inflammation was associated with secondary metabolites. The extracts did not cause acute toxicity in the evaluated organs. Taken together, the results of the study demonstrated the antimicrobial and anti-inflammatory potential of the BoHE and BoEA extracts of B. orellana, which makes these natural compounds targets for drug development. Further in vivo studies of the antimicrobial effect of BoHE and BoEA are warranted.

Data Availability
The values used to build graphs to support the findings of this study are available from the corresponding author upon request.