Triterpene Esters and Biological Activities from Edible Fruits of Manilkara subsericea (Mart.) Dubard, Sapotaceae

Manilkara subsericea (Mart.) Dubard (Sapotaceae) is popularly known in Brazil as “guracica.” Studies with Manilkara spp indicated the presence of triterpenes, saponins, and flavonoids. Several activities have been attributed to Manilkara spp such as antimicrobial, antiparasitic and antitumoral, which indicates the great biological potential of this genus. In all, 87.19% of the hexanic extract from fruits relative composition were evaluated, in which 72.81% were beta- and alpha-amyrin esters, suggesting that they may be chemical markers for M. subsericea. Hexadecanoic acid, hexadecanoic acid ethyl ester, (E)-9-octadecenoic acid ethyl ester, and octadecanoic acid ethyl ester were also identified. Ethanolic crude extracts from leaves, stems, and hexanic extract from fruits exhibited antimicrobial activity against Staphylococcus aureus ATCC25923. These extracts had high IC50 values against Vero cells, demonstrating weak cytotoxicity. This is the first time, to our knowledge, that beta- and alpha-amyrin caproates and caprylates are described for Manilkara subsericea.


Preparation of Extracts.
Extracts were obtained from fruits, leaves, and stems. e M. subsericea freshly fruits (1.14 kg) were crushed and macerated with ethanol (EtOH) 96% (v/v) at room temperature until exhaustion. is ethanolic extract was concentrated in vacuum using a rotary evaporator to obtain ethanolic crude extract from fruits (170 g). is extract was dissolved into 500 mL EtOH/H 2 O 90% (v/v) mixture and partitioned with hexane (2 × 600 mL) to obtain, aer evaporation of the hexanic portion, 14.0 g of hexanic extract from fruits (FH). e hydroalcoholic portion from this partition was evaporated in vacuum and resuspended in 500 mL distilled water. is aqueous suspension was successively partitioned with ethyl acetate (2 × 600 mL) and butanol (2 × 600 mL), furnishing, aer evaporation, 4.5 g of ethyl acetate extract (FEA), and 6.8 g of butanol extract (FB) from fruits. Leaves (1.93 kg) and stems (0.96 kg) were individually dried at 40 ∘ C for two days, triturated and macerated with ethanol (EtOH) 96% (v/v) at room temperature until exhaustion. Each ethanolic extract was concentrated in vacuum using a rotary evaporator to obtain 530 g of ethanolic crude extract from leaves (LET) and 169.3 g of ethanolic crude extract from stems (SET).

Analysis of FH by Gas
Chromatography-Mass Spectrometry. e hexanic extractfrom fruits (FH) was analyzed by a GCMS-QP5000 (SHIMADZU) gas chromatograph equipped with a mass spectrometer using electron ionization, according to these experimental conditions: injector temperature, 270 ∘ C; detector temperature, 290 ∘ C; carrier gas, Helium; �ow rate 1 mL/min; split injection with split ratio 1 : 50. e oven temperature was programmed from 60 ∘ C (isothermal for 3 min), with an increase of 10 ∘ C/min to 290 ∘ C, ending with a 59 min isothermal at 290 ∘ C. One microliter of the sample, dissolved in CHCl 3 (1 : 100 mg/ L), was injected into a ZB-5MS column (i.d. = 0.25 mm, length 30 m, �lm thickness = 0.25 mm). Mass spectrometry (MS) conditions were ionization voltage, 70 eV and scan rate, 1 scan/s. e identi�cation was performed by comparison of the MS fragmentation pattern of the substances of FH with NIST mass spectra libraries. Quantitative analysis of the chemical constituents was performed by �ame ionization gas chromatography (CG/FID), under same conditions of GC/MS analysis and percentages obtained by FID peak area normalization method.

Diffusion Disk Assay.
Qualitative antimicrobial tests were carried out by disk diffusion method [11]. Brie�y, a suspension of microorganism (10 8 UFC/mL) was spread on the solid media plates of Muller-Hinton agar (Difco). Disks (6 mm in diameter) were impregnated, until saturation, with the ethanolic crude extracts from leaves and stems, hexanic, ethyl acetate, and butanol extracts from fruits. en, disks were placed on the inoculated agar. Vancomycin (30 g) and ampicillin (30 g) were used as positive reference standards of the test. Disks impregnated with solvents used for solubilization of extracts were used as negative control. e inoculated plates were incubated at 37 ∘ C for 24 h. Antimicrobial activity was evaluated by measuring the zone of inhibition against the test organisms. Each experiment made in triplicate.

Minimum Inhibitory Concentration (MIC).
A microdilution technique using 96 well micro-plates, as described by Eloff [12] was used to obtain MIC values of extracts against S. aureus. e method comprised of �lling all the wells of a 96 well microplate with 100 L of Muller-Hinton broth (Vetec). Triplicates (100 L) of the samples (ethanolic crude extracts from leaves and stems, hexanic, ethyl acetate, and butanol extracts from fruits) at starting concentrations of 2 mg/mL in DMSO were introduced into the �rst well. Serial doubling dilutions were then performed, rejecting 100 L from each well and adding a mixture of test microorganism (100 L) having an inoculum size of approximately 1 × 10 6 CFU/mL. e �nal concentrations per well were 500, 250, 125, 64, and 32 g/mL. e microplates were sealed and incubated at 37 ∘ C for 24 h. Aer incubation, 50 L of a 2.5% solution of the biological indicator TTC (Triphenyl Tetrazolium Chloride) solution was added, and the plates were incubated again for 2 h to visualize growth inhibition. e lowest concentration of the sample that inhibited the bacterial growth (colourless) aer incubation was taken as the MIC. Vancomycin and DMSO were used as positive and negative controls, respectively.

Cell Viability by LDH Assay.
To evaluate the toxicity of extracts, Vero cell line was incubated with samples (ethanolic crude extracts from leaves and stems, hexanic, ethyl acetate, and butanol extracts from fruits) for 24 hours and cell viability measured using LDH assay (Doles). In brief, 5 × 10 4 cells/well were seeded in a 96-well microplate and incubated for 24 hours to attach. In the following day, cells were washed with PBS, and fresh media DMEM without serum were replaced containing the plant extract at different concentrations (500-31.25 g/mL). Plates were incubated for further 24 hours and LDH activity measured by colorimetric assay using spectrophotometer (micronal-B582) at 510 nm.
As control for maximum LDH release, cells were treated with 0.1% triton-X100 in DMEM medium for 10 min before running the assay. To determine the normal LDH release, cells were cultured in serum-free medium in presence of DMSO. Cell viability was determined using absorbance of treated cells at DMSO as a reference for 100% viability (absorbance of extract-treated cells × 100/absorbance of DMSO-treated cells).

Statistical Analysis.
For antibacterial assay, statistical analysis was performed by ANOVA (one-way Anova) with 95% con�dence interval, using the GraphPad Prism 5.0 soware package. Differences were considered signi�cant when values were ≤0.05. Vero cell viability (%) was determined by averaging three repeated experiments and IC 50 , representing the concentration at which cell viability was reduced by 50%, was calculated by linear regression using the GraphPad Prism 5.0 soware package. On another study, we described the obtainment and identi�cation of a mixture containing beta-amyrin acetate and alpha-amyrin acetate from edible fruits of this species [13]. us, comparison of the previously fragmentation pattern obtained for these substances con�rmed the ma�or substances, with retention time (min) of 34.63 and 36.15, as beta-amyrin acetate (E) (10.27%) and alpha-amyrin acetate (F) (42.34%), respectively. e substances with retention time (min) of 53.31/56.55 and 71.70/76.99 also showed a typical fragmentation pattern for pairs of triterpenes from the Δ12-oleane/Δ12-ursane series. e characteristic peaks at m/z 218 (base peak), 203 and 189 due to Retro-Dials-Alder fragmentation [14] were observed for these substances. Betaamyrin type triterpenes presented peak at m/z 203 higher than peak at m/z 189, while alpha-amyrin type triterpenes showed an equal abundance (Figure 2). According to Oyo-Ita et al. [15], the amyrin caproates have molecular ion peak (M + ) at m/z 524, followed by loss of CH 3 or the acid moiety to m/z 509 and 408, respectively. us, the substances with retention time (min) of 53.31 and 56.55 could be identi�ed as betaamyrin caproate (G) (5.46%) and alpha-amyrin caproate (H) (7.26%). e mass fragment at m/z 408, due to the loss of 144 (caprylic acid) mass unit from the molecular ion peak (M + ) at m/z 552 was in accordance with literature data [16,17] and suggested substances with retention time (min) of 71.70 and 76.99 as beta-amyrin caprylate (I) (2.44%) and alpha-amyrin caprylate (J) (5.04%), respectively.
e identi�ed substances corresponded to 87.19% of the total relative composition of the hexanic extract from fruits of M. subsericea. e individual amounts of each substance are illustrated in Figure 4. Furthermore, to our knowledge, this is the �rst time that the beta-and alpha-amyrin caproates and caprylates are described for the Manilkara subsericea species.
Antibacterial assay was performed against Staphylococcus aureus ATCC25923 and Escherichia coli ATCC36298. ere were signi�cant differences ( 0 05) in the antibacterial activity of ethanolic crude extract from leaves (7 ± 0 28 mm), ethanolic crude extract from stems (8 ± 0 mm), and hexanic extract from fruits (6 ± 0 mm), which were considered active against S. aureus. (Table 1). Ethyl acetate and butanol extracts from fruits did not inhibit the S. aureus growth. All extracts were considered inactive against E. coli. (Table 1). e extracts that exhibited antimicrobial activity during the disk diffusion method were evaluated for their Minimum Inhibition halo (mm) ± SD Staphylococcus aureus Escherichia coli inhibitory concentration (MIC). All extracts tested inhibited the bacterial growth of the S. aureus strain with MIC of 250 g/mL. Terpenoids are active against bacteria but the mechanism of action of terpenes is not fully understood, although it is speculated to involve membrane disruption by lipophilic compounds [20]. e isomeric mixture of betaamyrin and alpha-amyrin is known by its antimicrobial activity [21], and 72.81% of the relative amount of the hexanic extract from fruits is constituted by esters of these substances. Beta-and alpha-amyrin acetates are also known by their anti-in�ammatory activity and also inhibitory effects on Epstein-Barr virus early antigen (EBV-EA) in Raji cells [22]. Furthermore, according to Hichri et al. [23], the triterpene beta-amyrin acetate was able to inhibit the bacterial growth of the Staphylococcus aureus ATCC25923 reference strain at 90 g/mL. us, our results suggest that the antibacterial activity found in the hexanic extract from fruits may be modulated by the beta-and alpha-amyrin esters identi�ed. All tested extracts demonstrated weak cytotoxic effects on the mammalian Vero cells. e Cell viability on treatment with hexanic and ethyl acetate extracts from fruits was 69.66% and 56.07% in concentration of 250 g/mL, respectively ( Figure 5). Ethanolic crude extract from leaves (1658 g/mL; 1164-2525) had highest IC 50 value, followed by ethanolic crude extract from stems (1112 g/mL; 757-2525), butanol extract from fruits (683.4 g/mL; 451-2200), hexanic extract from fruits (482.6 g/mL; 385-677), and ethyl acetate extract from fruits (307.6 g/mL; 276-346).

Conclusions
Although Manilkara subsericea fruits are used as food, to our knowledge, only one article regarding its phytochemicals and biological activities was published [13]. e present study describes the identi�cation of a high percentage of substances from the hexanic extract from edible fruits of Manilkara subsericea, in which beta-and alpha-amyrin caproates and caprylates are reported for the �rst time for this species. Our results suggest that this hexane extract from fruits and ethanolic crude extract from leaves and stems presented antimicrobial activity against S. aureus ATCC25923. In addition, these extracts had low cytotoxicity on Vero cells, in the same concentration which inhibited S. aureus growth. Several biological studies are carried out for mixtures of beta-and alpha-amyrin type triterpenes [21,27,28], since their separation by conventional chromatographic methods is quite difficult [29].