Antioxidative, Antiproliferative and Antimicrobial Activities of Phenolic Compounds from Three Myrcia Species

Myrcia bella Cambess., Myrcia fallax (Rich.) DC. and Myrcia guianensis (Aubl.) DC. (Myrtaceae) are trees found in Brazilian Cerrado. They have been widely used in folk medicine for the treatment of gastrointestinal disorders, hemorrhagic and infectious diseases. Few reports have been found in the literature connecting their phenolic composition and biological activities. In this regard, we have profiled the main phenolic constituents of Myrcia spp. leaves extracts by ESI(−)Q-TOF-MS. The main constituents found were ellagic acid (M. bella), galloyl glucose isomers (M. guianensis) and hexahydroxydiphenic (HHDP) acid derivatives (M. fallax). In addition, quercetin and myricetin derivatives were also found in all Myrcia spp. extracts. The most promising antioxidant activity, measured by 2,2-diphenyl-1-picrylhydrazyl (DPPH) scavenging activity, was found for M. fallax extracts (EC50 8.61 ± 0.22 µg·mL−1), being slightly less active than quercetin and gallic acid (EC50 2.96 ± 0.17 and 2.03 ± 0.02 µg·mL−1, respectively). For in vitro antiproliferative activity, M. guianensis showed good activity against leukemia (K562 TGI = 7.45 µg·mL−1). The best antimicrobial activity was observed for M. bella and M. fallax to Escherichia coli (300 and 250 µg·mL−1, respectively). In conclusion, the activities found are closely related to the phenolic composition of these plants.


Introduction
The Myrtaceae family is one of the most important genera of the wet tropics, including South America, Australia and Tropical Asia. In Brazil, there are 23 genera and approximately 1034 species distributed throughout all regions of the country. Their leaves and berries are commonly used in traditional medicine and foods. The chemical compositions of plants belonging to the Myrtaceae family (such as Eugenia sp., for example) have been previously studied [1]. Myrtaceae fruits have found commercial use, for example, Eugenia uniflora L. ("pitangueira"), Syzygium sp. ("jambo"),

Results and Discussion
Firstly, this research shows the main polyphenolic compounds identified from Myrcia bella Cambess., Myrcia fallax (Rich.) DC. and Myrcia guianensis (Aubl.) DC. Although basically the same set of ions was detected for those plants, it is clear from visual inspections that there are chemical profile differences ( Figure 1). Table 1 shows the main phenolic acids and hydrolysable tannins identified via their molecular formula and dissociation patterns from all of these extracts by ESI(−)-MS/MS. Table 2 shows the main flavonols identified from these extracts.
The compounds presented in Table 1 were identified as phenolic acids and hydrolysable tannins (gallotannins and ellagitannins), indicated by the presence of deprotonated molecular ions which correspond to gallic acid, galloyl-glucose, ellagic acid, hexahydroxydiphenic acid (HHDP) and hexahydroxydiphenoyl-glucose (HHDP-glucose).   [7,8]. This fragmentation pattern is commonly found in hydroxybenzoic acid derivatives [9]. The deprotonated molecule of m/z 191 was identified as quinic acid. This compound is a primary metabolite, which can be found in great amounts in all plants and fruits [10]. The MS/MS experiment shows fragments of m/z 173 and 111, which correspond to the loss of the water molecule and a retro Diels-Alder (RDA) ring opening mechanism [10], respectively. Quinic acid has already been found in other plants of the genus Myrtaceae [11].
Ellagic acid was characterized by the diagnostic mass ion of m/z 301 and its mass fragments of m/z 257 and 229 [12]. Monogalloyl-glucose (m/z 331) presented fragments of m/z 271, m/z 169 and m/z 125 (Scheme 1) [13,14]. The ion m/z 343 was identified as monogalloylquinic acid and MS/MS fragmentation showed loss of the galloyl moiety of m/z 191, a deprotonated gallic acid of m/z 169 and its fragment of m/z 125 [M − H − CO 2 ] − [14].
The HHDP derivatives are found mostly in M. fallax (Table 1). HHDP-glucose deprotonate ion molecule of m/z 481, which was followed by an ion fragment of m/z 301 (ellagic acid) after the loss of one unit of glucose. After the spontaneous lactonization of HHDP, forming ellagic acid, a subsequent decarboxylation yields an ion of m/z 275 [M − H − 44] − [15,16]. The deprotonated ion molecule of m/z 483 was assigned to digalloylglucose, since it provided the typical MS/MS fragment ions of m/z 331 [M − H − 152] − and m/z 169 [M − H − 162] − (Scheme 1) [15]. Ellagic acid, galloyl-glucose and HHDP-glucose isomers are described in other Myrtaceae species, such as Eugenia punicifolia and E. aurata [11]. Scheme 1. The proposed fragmentation pathways of digalloyl-glucose isomer.
The ion of m/z 783 was assigned to bis-HHDP-glucose and the ion of m/z 935 was assigned to bis-HHDP-galloylglucose [16,17]. In the Scheme 2, the proposal for the fragmentation of these two compounds is provided. The compound detected of m/z 1085 with a fragment ion of m/z 783 corresponded to the loss of ellagic acid [M − H − 302] − and m/z 301 was assigned to eucalbanin A or its isomer cornussin B (Scheme 2). Eucalbanin A and cornussisn B have been previously found in leaves and fruits of Eucalyptus species, but have never been reported in Myrcia spp. The flavonoids identified from Myrcia bella, M. fallax and M. guianensis are mostly flavonol-Oglycosides as quercetin, myrcetin and kaempferol. The most common sugar units linked to them are galactose, glucose, xylose and rhamnose [2]. The analyzed flavonols from in this work are presented in Table 2. In Table 2, it can be seen that the ions of m/z 301 and m/z 317 were identified as quercetin and myricetin aglycone, respectively. Quercetin and myricetin derivatives were also described for M. splendens, M. palustris DC., and M. uniflora. Quercetin was also found in M. myrtillifolia [2].
Quercetin showed fragmentation ions of m/z 257 after decarboxylation [M − H − CO 2 ] − and of m/z 179 after cleavage of the B ring by a Retro Diels-Alder mechanism (RDA). It was also observed, after loss of a carbonyl group from fragment of m/z 179 and ion of m/z 151( 1,3 A − fragment) [18,19]. Those fragmentations were also observed in myricetin, with an additional typical fragment ion of m/z 137 of the trihydroxylated B-ring [20].
The deprotonated molecule ion of m/z 433 was assigned to quercetin-3-O-D-xylopyranoside and ions products of m/z 300, 271 and 151. An ion of m/z 300 was due to the neutral loss of pentoside. From m/z 300, a CO loss was also detected to give an m/z 271 [Y o − H − CO − H] − and an ion of m/z 151 [19,21]. The ion of m/z 447 was assigned to deprotonated quercetin-3-O-β-rhamnoside and showed an aglycone fragment ion of m/z 301 [M − H − 146] − due to the loss of a sugar moiety, whereas the fragment ion of m/z 271 was also typical for a flavon-3-O-monoglycoside [21].
The  Tables 1 and 2, affected the results for total antioxidant capacity, measured by the DPPH scavenging method (Table 3). In this work, M. falllax extracts showed the lowest EC 50 value (8.61 ± 0.22 µg/mL), corresponding to the highest scavenging activity of DPPH radicals (Table 3). A possible explanation for the higher M. falllax activity is its own phytochemical composition, in which phenolic acids and hydrolysable tannins were found which were not present in other Myrcia spp. extracts. It is also remarkable that M. guianensis extract showed a similar EC 50 value (16.2 ± 0.80 µg/mL) to M. bella (EC 50 13.8 ± 0.53 µg/mL), as this plant showed a much lower phenol content compared to the other extracts. A good Pearson's correlation was found between the flavonoid content and the EC 50 value for the DPPH antioxidant assay (r 0.774, p = 0.0144). The total phenolic content of Myrcia extracts did not correlate with the EC 50 (r 0.0997, p = 0.799), so these results suggest that not all phenolic compounds are primarily responsible for the antioxidant activity in the DPPH. On the literature, Myrcia bella leave infusion showed a better EC 50 (7.73 µg/mL), but with greather flavonoid content compared to our studies (312.99 ± 8.60 mgQ/g extract ) [24]. One possible explanation for the EC 50 values shown above is the synergism/antagonism of the phenolic compounds present in Myrcia extracts, since the antioxidant activity is usually correlated to their structures. For flavonoids, in general, structural arrangements impart the greatest antioxidant activity, since these change the ability to scavenge reactive oxygen. The activity is usually associated with the number of available hydroxyl groups in the chemical structure and delocalization of the aromatic nucleus [25]. In addition, antioxidant activity also depends on their donor-proton capacity. For example, rutin, a quercetin glycosylated molecule, decreases its own antioxidant activity [25]. Thus, quercetin and myricetin derivatives found in Myrcia spp. extracts satisfied all of the aforementioned determinants for antioxidant activity. The results found for flavonoids content and their correlation with EC 50 (Table 3) were compatible with those described in the literature. From the literature, Myrcia splendens and M. palustris ethanolic extracts showed different flavonoids contents (16.43 ± 0.69 and 59.17 ± 0.53 mgQ/g extract ) and EC 50 (10.14 and 29.81 µg/mL), which shows that our results described above are within the range, and not that different to those of other Myrcia spp. [5].
As for flavonoids, the antioxidant activity of hydroxybenzoic acid and their ester derivatives also depend on the number of hydroxyl groups in the molecule. This activity is also affected by steric hindrance of the carboxylate group, since the closeness of the carboxylate group and the hydroxyl groups on the phenolic ring negatively affects their donor proton ability [2].
Landete described the strong correlation observed between antiradical activity measurements and the high molecular weight of hydrolysable tannins, like ellagitannins and ellagic acid. As for the other polyphenolics cited above, the major contributors to antioxidant capacity can be attributed to the presence of hydroxy functions, which exhibit a greater ability to donate a hydrogen atom and support the unpaired electron, compared to those of low molecular tannin [4]. However, as the antioxidant efficiency of these compounds is directly correlated with their degree of hydroxylation, the presence of a sugar moiety would decrease this effect. Thus, understanding the different activities of these polyphenols is critical before claiming any antioxidant-related health benefit [4].
According to literature, allelopathic effects were described for M. guainensis leaves extracts [27]. M. bella leaves extract was described promoted necrosis into primary tumor gastric cell proliferation on 300 µg·mL [28]. Moreover, a US patent described a new extractive protocol to produce a M. fallax leaves extract with antiproliferative effect against KB cell line (human nasopharynx carcinoma) [2]. As allelopathic activity could suggest an antiproliferative effect potencial [29], our results for M. bella and M. guainensis leaves extracts are on agreement with the literature, with M. fallax leaves extracts with the less activity (Table 4).  Therefore, the antiproliferative effect (Table 4) observed for the three Myrcia species had negative correlation with the DPPH scavenger ability. These findings are on agreement with other results, showing that highly antioxidant flavonoids did not inhibit cell proliferation or polyphenolic enriched extracts [11]. Therefore, even without significant antiproliferative activity, antioxidant phytochemicals seem be useful for controlling oxidative stress in the microenvironment.
In the last decade, many Myrtaceae species have been extensively studied for their antimicrobial properties, because good activity of this genus in folk medicine has already been described. However, most of these studies, including those on M. fallax, have focused on essential oils from the leaves, as they are rich in sesquiterpenes and monoterpenes [2]. These terpenes are described as capable of permeating cell membranes, thanks to their hydrophobicity, which causes the death of parasites or microorganisms by affecting their metabolic pathways or organelles. For this reason, these oils could have antimicrobial activity against Gram-positive and Gram-negative human pathogens [2,30,31].
Flavonoids are also known to be synthesized by plants in response to microbial infection, thus explaining the in vitro antimicrobial activity of those substances against a wide array of microorganisms. Their activity is probably due to their ability to complex with extracellular and soluble proteins, and lipophilic flavonoids may also disrupt the microbial membrane. From another point of view, a wide range of anti-infective actions, such as the stimulation of phagocytic cells and host-mediated tumor activity, have been assigned to tannins [31].
For this study, Minimum Inhibitory Concentration (MIC) values below 100 µg/mL were considered to indicate good antimicrobial activity; from 500 to 100 µg/mL to show moderate antimicrobial activity; from 1000 to 500 µg/mL to show weak antimicrobial activity; and above 1000 µg/mL to indicate inactivity [32]. Interestingly, in this study, M. bella and M. fallax have presented a moderate activity to the same Gram-negative bacteria Escherichia coli (300 and 250 µg/mL, respectively). M. fallax also showed a weak activity to Gram-positive Enterococcus hirae (900 µg/mL) and M. guianensis also presented a weak activity of 1000 µg/mL to Gram-positive bacteria Rodococcus equi. For all of the tests and yeast lineages there were no positive results. In the literature, Sygyium cumini (Myrtaceae) hydroethanolic extract has already showed a moderate activity 125 µg/mL against Candida albicans (ATCC 10231) yeast which was also correlated to its phytochemical content (polyphenols 100.10 mg/g extract; flavonoids 23.04 mg/g extract) [33] In this case, the antimicrobial effect of Myrcia spp. could be reasoned by the presence of compounds like phenolic compounds and hydrolysable tannins than galloyl glucose isomers or flavonoids.
To summarize all of the results above, the direct infusion ESI(−) QTOF-MS analysis was a helpful to provide a quite comprehensive understanding of the phenolic chemical composition of Myrcia spp. since they are quite different with quercetin and myricetin derivatives in common. The main constituents found which differs from one plant to another were ellagic acid (M. bella), galloyl glucose isomers (M. guianensis) and hexahydroxydiphenic (HHDP) acid derivatives (M. fallax). Moreover, the phenolic Myrcia spp. composition also showed different biological activities, which could be correlated to their chemical profiles. The most promising antioxidant activity was found for M. fallax Regarding in vitro antiproliferative activity, M. guianensis showed good activity against leukemia. The best antimicrobial activity was achieved for M. bella and M. fallax to Escherichia coli. In conclusion, the activities found are closely related to the phenolic composition of these plants. From these data, we could get a better insight into the unexplored pharmacological potential of those Myrcia spp., and further biological studies are underway in our laboratory.

Polyphenol and Flavonoid Contents
The total polyphenol content was determined in all tested extracts using Folin-Ciocalteu's assay with some modifications [34]. Thus, to an aliquot (0.5 mL) of ethanolic solution of each extract it was added Folin-Ciocalteau reagent (2.5 mL, 10% v/v) and sodium carbonate (2.0 mL, 4% w/v) solutions, followed by thorough mixing, incubation for 120 min in the dark at room temperature and absorbance measurement at 750 nm. A calibration curve (0.5-40 µg/mL) of gallic acid was used to express the results as mg gallic acid equivalent (GAE)/g dry extract. All measurements were performed in triplicate. The total flavonoid content was determined by the aluminum chloride colorimetric assay with modifications [35]. Thus, to an aliquot (0.5 mL) of ethanolic solution of each extract, absolute ethanol (1.5 mL), aluminum chloride (0.1 mL, 10% w/v), potassium acetate (0.1 mL, 1 M) and distilled water (2.8 mL) were added, followed by thorough mixing, incubation for 30 min in the dark at room temperature and the measurement of absorbance at 425 nm. A calibration curve (5-60 µg/mL) of quercetin was used to express the results as mg quercetin equivalent (GAE)/g dry extract. All measurements were performed in triplicate.

ESI(−)-Q-TOF-MS Analysis
Mass spectrometry analyses were performed on an Agilent 6500 series quadrupole time-of-flight (QTOF) mass spectrometer (Agilent, Santa Clara, CA, USA), operating in the negative ion mode. The VCap voltage was maintained at 4 kV, the gas temperature was set to 290 • C and the drying gas flow rate was of 11 L·min −1 , while the nebulizer pressure was set in 45 psi. The fragmentor was maintained at 175 V and the skimmer voltage was of 320 V. A mixture of ions with m/z 112.9856 and m/z 966.0007 was used as internal standards. Data were recorded in a mass range of m/z 100-1500, with a scan rate of 1 spectra s −1 and a MS/MS scan rate of 2 spectra s −1 with an isolation width of 1.3 and collision energy of 35 eV. MS/MS precursor selection was set to four precursors per cycle, where these were chosen by the abundance only mode.

Determination of DPPH Radical Scavenging
The extracts for 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity was estimated according to Enujiugha et al. with some modifications [35]. Thus, to an aliquot (2.0 mL) of methanolic solution of each extract (1.56-200 µg/mL), DPPH solution (0.5 mL, 0.03% in methanol) was added, followed by vigorous mixing, incubation for 30 min in the dark at room temperature and the measurement of absorbance at 517 nm. Absolute methanol (2 mL) was used as the blank (negative control) and quercetin and gallic acid solutions were used as positive controls. All tests were carried out in triplicate. The DPPH radical scavenging activity of each extract was calculated according to Equation (1): where Abs sample is absorbance of each extract and Abs control is absorbance of methanol in DPPH.

Minimum Inhibitory Concentration
The antimicrobial activity of the hydroethanolic extract and fractionated extracts of Myrcia spp. was identified by microdilution in broth, determining the minimum inhibitory concentration (MIC) and minimum fungicidal concentration (MFC) following the norms established by the Clinical and Laboratory Standards Institute (CLSI) [32]. The following microorganisms were tested: Bacillus subtilis (gram positive, ATCC 6051), Enterococcus faecium (gram positive, ATCC 5079), Staphylococcus aureus (gram positive, ATCC 6538) Salmonella choleraessuis (gram negative, ATCC 10708), Enterococcus hirae (gram positive, ATCC 10541), Micrococcus luteus (gram positive, ATCC 4698), Rodococcus equi (gram positive, ATCC 25729), Streptococcus epidermidis (gram negative, ATCC 12228), Escherichia coli (gram negative, ATCC 11775), Pseudomonas aeruginosa (gram negative, ATCC 13388) and Candida albicans (ATCC10231). The test was carried out in 96-well microplates containing 100 µL/well of the specific culture medium (brain heart infusion (BHI; Difco, Franklin Lakes, NJ, USA) for bacteria and RPMI 1640 (Angus Buffers & Biochemicals, Niagara Falls, NY, USA) for yeast). The substances were diluted in 40% alcohol (8 mg/mL) and transferred to the first well, before serial dilutions were performed to obtain concentrations between 15.62 and 2000 µg/mL. Chlorhexidine 0.12% (Sigma-Aldrich) and nystatin (7-8 mg/mL, Sigma-Aldrich) were used as the positive controls and 40% alcohol was used as the negative control. In addition, 40% Alcohol was tested and showed no antimicrobial activity against the microorganism studied. The bacterial inocula (colony-forming units (CFU)/mL) and fungal inocula (CFU/mL) were added to the wells and the plates were incubated at 37 • C for 24 h. The MIC was defined as the lowest concentration of the extract or fraction that inhibited visible microbial growth, which was confirmed with 0.01% 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT, Sigma-Aldrich) for bacteria and by the change in color of the RPMI 1640 for yeast. All assays were performed in triplicate in three independent experiments.

Statistical Analysis
The results were expressed as mean values ± standard deviation from each Myrcia spp. extract parameter in triplicate. Experimental data were evaluated by variance analysis (one-way ANOVA, Bioestat v 5.3, Marinauá Institute, Tefé, AM, Brazil) followed by Tukey test. A significance level of 5% was adopted. Statistical significance was set at a level of p < 0.05. Correlation and regression analyses were performed using ORIGIN 8.6 ® software (OriginLab Corporation).