Material and methods

Plant material. The plant material was collected of adult organisms in Oaxaca state, Mexico in September 2015, rainy season and taxonomically determined by Rosa Linda Medina Lemos. Collection data is presented in table 1, voucher specimens were deposited in the herbariums of the Faculty of Sciences, UNAM. The extracts were prepared from air-dried leaves and stems of the plants (150 g each), using ethanol (70 %) in proportion 1:5 and maceration at room temperature for 24 h. The solvent was removed with a rotavapor by reduced pressure distillation.

Flavonoid profiling. The extracts were analysed by high-performance liquid chromatography (HPLC) on a Hypersil ODS (125 × 4.0 mm) Hewlett Packard column eluted with a gradient of (A) H₂O adjusted at pH 2.5 with trifluoroacetic acid and (B) acetonitrile; 0-10 min, in the following mixtures, A:B 85:15; for 20 min, and A:B 65:35 for 25 min. The following parameters were used: flow at 1 mL/min at 30 °C, detection wavelengths, 254, 316 and 365 nm; injection volume, 20 µL and analysis time, 25 min. The analysis of naringin was performed in a Nucleosil 100 A column 125 × 4.0 mm i. d., 5 µm C-18. The flow of the mobile phase remained at 1.3 mL/min and consisted of (A) acetonitrile, (B) water adjusted at pH 3.0 con H₃PO₄ with the following gradient: 0.10 min, 12 % A, 88 % B; 10 min, 18 % A, 82 % B; 15 min, 18 % A, 82 % B; 30 min, 45 % A, 55 % B and 32 min, 100 % A, at a temperature of 30 ^oC. The injection volume of the sample was 20 µL. The equipment was calibrated to a wavelength of 350 nm. The analysis time for this method was 32 minutes. Identification was done by comparison with retention time of standards. Quantification was based on the calibration curves of the reference compounds, obtained under the same chromatographic conditions. The standards used were rutin, phlorizin, myricetin, quercetin, naringenin, naringin, phloretin and galangin.

Antibacterial activity. Bacterial strains used: Bacillus subtilis (Ehrenberg 1835) Cohn (ATCC6633), Escherichia coli (Escherich) (ATCC11229), Staphylococcus epidermidis (Winslow & Winslow) Evans (ATCC12228), Pseudomonas aeruginosa (Schroeter) (ATCC27853), Staphylococcus aureus Rosenbach (ATCC 6538), Salmonella typhimurium (Kauffmann & Edwards) Le Minor & Popoff (ATCC14028), Salmonella typhi (Schroeter) Warren and Scott (ATCC6539) and Staphylococcus aureus ATCC 29213 (donated by the Microbiology Laboratory at the Department of Biology of the Chemistry Faculty, UNAM).

Bacterial strains were grown to exponential phase in Mueller-Hinton broth (Merck) at 37 °C for 18 h and adjusted to a final density of 10⁸ CFU/mL by diluting fresh cultures and comparing with the 0.5 McFarland turbidity standard (^{Kiehlbauch et al. 2000}).

Eloff’s microplate method (1998) was used to determine the minimal inhibitory concentration (MIC) values for plant hydroalcoholic extracts (70 % EtOH) with antibacterial activity. Plant extracts were dissolved at 10 mg/mL with the bacterial culture medium. All extracts were initially tested at 3.84 mg/mL in 96-well microplates and serially diluted two-fold to 1.87 × 10^-3 mg/mL.

The antibiotic ampicillin (Sigma-Aldrich), was included as reference in each assay, and a standard antibiotic solution was made in an appropriate solvent as recommended by CLSI guidelines, dilutions of 0.05 to 5 mg/mL were used. Extract-free solution was used as a blank control. The microplates were incubated at 37 °C for 24 h. As an indicator of bacterial growth, 40 µl of 0.02 mg/mL 2,3,5-triphenyl-2H-tetrazolium chloride (TTC) (Sigma) dissolved in water were added to the wells and incubated at 37 °C for 30 min. MIC values were defined as the lowest extract concentration that prevents visible bacterial growth after 24 h of incubation at 37 °C. The colourless tetrazolium salt acts as an electron acceptor and is reduced to a red-coloured formazan product by biologically active organisms. Where bacterial growth was inhibited, the solution in the well remained clear after incubation with TTC. All the experiments were conducted in triplicate.

Antifungal screening. The fungal strain Colletotrichum musae (Berk. & Curt.) von Arx was obtained from plantain in Monterrey, Nuevo León, Mexico and the strain of Fusarium boothii O´Donnell, T. Aoki, Kistler & Geiser was isolated from wheat. Both were kept for 7 days in PDA (Potato dextrose agar), light/dark 12 h, 22 ºC. The fungal culture was obtained by making a suspension of the conidia of both species (20 × 10⁴ conidia/mL) in sterile distilled water + 0.1% Tween 20. This procedure was performed at the time of inoculation of the 96-well plates.

For screening purposes, the hydroalcoholic extracts of leaf and stem were dissolved with DMSO at a concentration of 1,000 mg/mL. Subsequently, the corresponding dilutions were made in a PDA medium to reach the following final concentrations: 480, 240, 60, 30, 15 μg/mL in 96-well plates. Three repetitions were performed at each concentration. The following controls were used: PDA, PDA + DMSO 0.3% and the broad-spectrum contact fungicide Captan 50 (Bayer) at the same concentrations as the extracts. Finally, 50 μL of the dilutions of the extracts in PDA medium and 50 μL of the conidia suspension were added per well until reaching a final concentration of 10 x 10⁴ conidia/mL, with a final volume per well of 200 μL. Before inoculating the microplate, the optical density was read at 490 nm using a microplate reader (Synergy HT). After the plate was inoculated, this procedure was repeated every 24 h for 72 h. The microplates were incubated under light/dark 12 h at 22 °C (^{Raposo et al. (1995)}.

The difference in optical density is used to calculate the biomass of the fungus at a certain time. From these data, it is possible to calculate growth inhibition as Equation 1:

The results of the inhibition percentages were analysed statistically by means of a two-way ANOVA. All analyses were performed in SPSS Statistics 21.0. The differences between means were calculated with Tukey’s post hoc test (p ≤ 0.05). The IC₅₀ was calculated for each extract (leaf and stem) of each species at 72 h, using a Probit regression. The results were analysed statistically by means of a one-way ANOVA. The differences between the means were calculated with the Ryan-Einot-Gabriel-Welch posthoc test, based on a test of (p ≤ 0.1). To establish the assumptions of each respective ANOVA, the normal distribution was verified via the Shapiro-Wilk test (p ≤ 0.05) and the homogeneity of variances with the Levene test (p ≤ 0.05).

DPPH radical scavenging activity. 2,2-Diphenyl-1-picrylhydrazyl radical (DPPH) is a free radical that is widely used to test the free radical-scavenging ability of molecules. DPPH forms stable-radical intermediate species due to electron delocalization (^{Blois et al. 1958}). Ethanolic solutions of DPPH have a deep violet colour and a strong absorption band at 517 nm. If DPPH reacts with a hydrogen-atom donor, the solution becomes colourless. The test was performed by mixing 50 μL (4 mg/mL) of Bursera extract with 150 μL of 133.33 μM DPPH ethanolic solutions in 96-well microplates. The final concentration of extract in the solution was adjusted to 1, 10, 100 y 1000 μg/mL, and the final concentration of DPPH in the solution was 100 μM. The plates were incubated at 37 °C for 30 min at a constant shaking speed. Subsequently, absorbance was measured at 517 nm using a microplate reader (Synergy HT). The decrease of absorbance was attributed to the amount (in %) of DPPH reduction.

The scavenging activity of each concentration was calculated as a percentage of the reduction of the DPPH concentration as follows (Equation 2):

The radical scavenging activity was expressed as IC₅₀, the concentration in μg/mL of the sample or positive control that reduces the absorbance of DPPH by 50 % compared to the negative control. This value was calculated by a linear model in Microsoft Excel 2010. The results were analysed with Student’s t-test (p ≤ 0.05) to determine the differences between the experimental and control groups. Quercetin and α-tocopherol were used as a positive control.

Anti-inflammatory activity: TPA-induced mouse-ear oedema assay. The ear oedema in mice assay was based on the method described by ^{Merlos et al. in 1991}. A group of six male CD1 mice (25-30 g) were anaesthetised with Sedalphorte^® (63 mg/kg i.p.) and a solution of 12-O-tetradecanoylphorbol-13-acetate (TPA 0.25 mg/mL) dissolved in ethanol was topically applied to both sides (5 μL each) of the right ear. The left ear received only ethanol (10 μL). After 10 minutes of TPA-treatment, the doses of 1 mg of the test extracts, indomethacin (0.358 mg/ear) as a reference drug was applied to both sides (10 μL each) of the right ear. Control animals received only ethanol:acetone (1:1) as a negative control. Four hours later the mice were anaesthetised and killed by cervical dislocation. A 7-mm-diameter plug was removed from each ear. The swelling was assessed as the difference in weight between right and left ear plugs oedema inhibition (EI %) was calculated by the equation: The edema inhibition was measured as the weight difference between the two circular samples.

The % inhibition of oedema was calculated by Equation 3:

The oedematous response was measured as the weight difference between the two circular samples. Data were analysed by ANOVA and Dunnet’s test, to compare with the control.

Results

Flavonoid profile. The quantified flavonoids belong to the structural subgroups flavonol (Quercetin, rutin, myricetin, galangin), flavanone (naringenin, naringin) and chalcone (dihydro) (phloretin, phlorizin). The results are summarised in table 2. Content was generally higher in leaves than stems, with exception of galangin, which showed the highest content in stems of B. aptera. It is to be noted that this species contains all of the flavonoids evaluated. Phlorizin and quercetin were present in all the species, while phloretin was the least represented, found only in B. aptera followed by naringenin in B. aptera and B. fagaroides. The highest quantity detected corresponded to phlorizin in the leaf of B. galeottiana. According to a survey of the literature, this is the first report of the occurrence of phlorizin and its aglycone phloretin in the genus Bursera.

Antibacterial activity. Four Bursera species were active, mostly against the Gram-positive bacteria, but the Gram-negative S. typhimurium, was the most susceptible (Table 3). The leaf and stem extracts of B. aptera and B. fagaroides were active against both bacterial types. Leaf extract of B. aptera was the most effective, with an MIC value of 0.002 mg/mL. However, B. schlechtendalii was not effective at the concentrations tested.

Antifungal screening. The activity of the Bursera extracts on fungal growth is expressed as the half maximal inhibitory concentration (IC₅₀) as shown in Table 4. The most active extract against F. boothii had an IC₅₀ value lower than 50 μg/mL and was obtained from stem of B. morelensis, on the other hand, the leaves of this species were among the least active, along with the stems of B. galeottiana with values above 2000 μg/mL. Contrastingly, these were the two most active against C. musae.

DPPH radical scavenging activity. The most active extracts were from B. fagaroides and B. aptera leaves, followed by B. galeottiana and B. schlechtendalii stems, with values lower than α-tocopherol (IC₅₀ 7.73 ± 0.51, 5.72 ± 1.02, 7.20 ± 0.61, 10.88 ± 0.73, 13.67 ± 0.45 μg/mL respectively). The results indicate that the reducing capacity of DPPH in Bursera depends on the species and the organ of the plant. Well, in the case of B. galeottiana, B. schlechtendalii and B. morelensis the stems are more active, while for B. aptera and B. fagaroides are the leaves (Table 5).

Anti-inflammatory activity in the mouse ear oedema model. The anti-inflammatory activity of Bursera extracts are summarised in table 6. All extracts showed an efficacy lower than reference drugs, indomethacin (COX-1 inhibitor) and celecoxib (COX-2 inhibitor).

Discussion

The Bursera species that were studied (Table 1) are found in environments where stress due to excess UV light and temperature produces reactive oxygen species (ROS). The presence of flavonoids in the five species analysed enriches the knowledge of the chemistry of the genus, because the reported studies have focused mainly on terpenes and lignans. The detection of flavonoids, as well as antioxidant, antibacterial and antifungal activities, becomes important when we consider that flavonoids act as a defense system in plant tissues exposed to different stress types, both biotic and abiotic. The flavonoids are found in the nucleus of the cells of the mesophyll in the chloroplast where they have an important role in diminishing the effect of reactive species, such as hydrogen peroxide, hydroxyl radicals and oxygen singlets. In particular, the phlorizin is contained in other plant species, which belong mainly to the Rosaceae and Ericaceae families. Furthermore, dihydrochalcones are especially abundant in the species M. domestica (Rosaceae), in which the quantity of dihydrochalcones can be variable depending on the tissue but represent the majority (> 70 %) of the total phenolic compounds (^{Gosch et al. 2010}). The other compounds quantified in this work (Table 2) are frequently found in plants with dietary and medicinal uses and have antimicrobial and anti-inflammatory properties (^{Cushnie & Lamb 2005}, ^{Cushnie & Lamb 2006}, ^{Li et al. 2016}, ^{Barreca et al. 2014}).

The biological activities of species of the genus Bursera are broad, regarding to the antibacterial activity detected in the species studied is important not only in the research for compounds useful for medicine, but also for the importance of plants being able to synthesize flavonoids and phenolic compounds in response to microorganisms. Extracts from plants rich in flavonoids are reported to have antibacterial activity. These compounds include: apigenin, galangin glycosides, flavones, flavonols and chalcones which have potent antibacterial properties (^{Cushnie & Lamb 2006}). The antibacterial mechanism of polyphenols may be related to their ability to deactivate enzymes of crucial importance for bacterial survival, towards to form complexes with proteins through hydrogen bonds with consequently formation of quinone, probable targets for quinone bindings on the bacterial cell are proteins of cell envelope and cell wall, surface-exposed adhesins and membrane-bound enzymes (^{Brudzynski & Maldonado-Alvarez 2015}). It has been reported the use of B. simaruba bark for skin infections, as well as antibacterial, antifungal and cytotoxic activities (^{Canales-Martínez et al. 2017}). The results indicate the existence of antibacterial compounds in the hydroalcoholic extracts of species studied (Table 3). However, the flavonoids detected in the extracts have records of antibacterial activity, and furthermore, the most effective species, B. aptera, contained all of those evaluated, these compounds might be only partially responsible for antibacterial activity (^{Cushnie & Lamb 2005}, ^{Cushnie & Lamb 2006}, ^{Barreca et al. 2014}). This suggest that the leaf extract of B. galeottiana, which had the highest amount of phlorizin and quercetin, did not present antibacterial activity. In a recent work, it has found that the activity of phloretin was reduced by the addition of glucose to the chalcone structure (^{Barreca et al. 2014}). Previously, it has been reported that B. aptera also has antifungal and antiprotozoal activities (^{Rodríguez-López 2014}, ^{Nieto-Yañez 2017}). The authors found that podophyllotoxin was the most abundant (13.79 %) compound in methanolic bark extract, suggesting that it is responsible for the leishmanicidal activity, based on the relation between podophyllotoxin and its use in the treatment of skin infections. Furthermore, the flavonoids isolated in Bursera extracts (Table 4) have found to be effective antifungal agents against a wide range of fungus (^{Al Aboody & Mickymaray 2020}).

On the other hand, there are many publications, which associate flavonoids with antioxidant activity (^{Panche et al. 2016}); for example, hisperidine, rutin, quercetin, apigenin, naringenin and kaempferol have been reported for the bark B. simaruba (^{Bah et al. 2014}). Which explains the reduction effect of the DPPH radical of the leaves and stems of the five Bursera species in this study (Table 5). The antioxidant activity of other Bursera species have been associated with lignans (^{Nieto-Yañez 2017}, ^{Marcotullio et al. 2018}) and phenols (^{Martinez-Elizalde et al. 2018})

Flavonoids such as hesperidin, apigenin, luteolin and quercetin, have anti-inflammatory and analgesic effects (^{Ferraz et al. 2020}). In a previous work, anti-inflammatory effects have been reported with an average effective dose (ED₅₀) of B. excelsa (0.26 ± 0.01 mg/ear), B. galeottiana (0.23 ± 0.02 mg/ear) and B. schlechtendalii (0.25 ± 0.02 mg/ear) from bark using chloroform extracts (^{Acevedo et al. 2015}). These reports suggested that the highest concentration of anti-inflammatory metabolites was found in the bark, however, our results indicated that the flavonoids were also found in stems and leaves (Table 6).

Finally, the presence of flavonoids in leaves and stems indicates the richness of these compounds in Bursera genus and, the antifungal and antibacterial, anti-radical and anti-inflammatory activities suggest the potential medicinal use of these species and the role that the flavonoids play in plant physiology.