The essential oils of the ground fruits and leaves were obtained by stream stripping. The essential oil analysis were performed by GC and GC-MS, and Table 1 shows the products quantified and identified using the NIST and Wiley databases.

The chromatographic profile of this essential oil (Figure 1) is consistent with the ranges defined by standard IRAM (Argentina) [15] for the essential oil of the fruits and leaves of Schinus molle L. (Table 2).

The essential oil was chromatographed with a low-medium pressure pump system, using a silica-gel flash type column prepared in the laboratory. Fractions were collected with an automatic collector. The obtained fractions were analysed by ¹H-, ¹³C-NMR, bidimensional 2D NMR and GC-MS to determine the components. In previous studies [16], several components of Schinus molle L. oil were identified, among which sabinene was the main component, followed by (α, β)-pinene and terpinen-4-ol, accounting for a total of 94.0% of the essential oil. In this work we isolated and identified more polar components of the essential oil, which correspond to terpinenediol derivatives and formyloxy derivatives of sabinene and terpinene. The fractions obtained in this work by column chromatography used for antibacterial testing (and their compositions) are summarized in Table 3.

The most polar fraction F07 has as a major component a product with an M⁺ to 198, which corresponds to the molecular formula C₁₁H₁₈O₃ (compound 1). HMRSNa was measured as 221.1146 (calc. for the formula C₁₁H₁₈O₃ 221.1148). The ¹H-NMR spectrum shows a one H singlet at 8.11 ppm that can be attributed to a formyloxy group (HCOO-); at 4.18 ppm and 4.12 ppm there are two doublets of 1H each, J = 12.5 Hz, which can be attributed to the methylene group attached to the formyloxy group. At 0.96 ppm and 0.87 ppm two doublets of three H each (J = 7.0 Hz) are observed, corresponding to two methyl groups that can be attributed to an isopropyl group. The signals at 0.43 ppm (dd, J = 5.4 and 8.3 Hz) and 0.25 ppm (dd, J = 3.6 and 5.4 Hz) 1H each, can be attributed to a cyclopropane methylene group in a bicyclic system [5.3] like in the skeleton of sabinene [17]. The ¹³C-NMR spectrum confirms the skeleton carbons of sabinene, with the carbon of the methylene group of the cyclopropane ring at 13.0 ppm (Table 4). Thus, compound 1 corresponds to a new derivative of the sabinene skeleton.

The major component of fraction F08 corresponds to a diol that shows in the mass spectrum a molecular ion at m/z 170, corresponding to a molecular formula C₁₀H₁₈O₂. The ¹H-NMR spectrum shows a multiplet integrating for 1H at 5.43 ppm, due to an olefinic hydrogen and at 3.80 ppm a 1H, m, signal of hydrogen atom geminal to an -OH group. Other signals are observed at 1.20 ppm (3H, s), corresponding to a methyl group geminal to a hydroxyl group (Me-7), at 1.03 ppm (3H, d, J = 7.1 Hz), for a methyl group at position 9 and at 1.00 ppm (3H, d, J = 7.1 Hz) corresponding to a methyl group at position 10. Taken together this data suggest a terpinenediol [18]. The ¹³C-NMR data of this new derivative 2 is given in Table 4 [19,20].

In fraction F09 another terpinene derivative was observed. The mass spectrum shows a molecular ion at 170 amu which corresponds to C₁₀H₁₈O₂. Its spectroscopic data was identical to that described in the literature for compound 3 [21].

These formylated hydroxyl or hydroxylated derivatives are minor components that cannot be detected by GC or GC/MS. These products easily dehydrated at the high injector temperature giving rise to the corresponding hydrocarbons. No major influence of these derivatives has been observed in antibacterial activities studied.

Only the crude extract and fraction F01 showed vapor inhibition halos derived from the effect of essential oil of S. molle. The smallest inhibition halo fraction of F01 is due to the fact that this fraction is more diluted (with an adjusted concentration of 90 mg/mL), below the concentration of components in the crude extract. Analysis of the results led us to conclude that the effect of vapor is mainly due to the presence of low molecular weight monoterpenes, namely (α, β)-pinene (Table 5).

After analyzing the oil volatiles, analytical tests by the agar diffusion method in cylindrical cavities were done, whose results are summarized in Table 6. From the results, we conclude that the oil of S. molle has the highest inhibitory effect against S. aureus ATCC 25923 strains, which showed resistance only to the 1:32 dilution. For E. coli ATCC 25922 the inhibitory effect is smaller but still (according to the criteria defined in 3.1) only has a strong inhibitory effect for the crude extract (undiluted). P. aeruginosa ATCC 27853 strains are less sensitive, with a maximum range of moderately sensitive, for this oil. We carried out the test with an intermediate dilution 1:10.5 (90 mg/mL) in order to carry out a comparative analysis of the fractions obtained by column chromatography and described in Table 7, Table 8 and Table 9.

To analyze of each of the fractions obtained by CC, three different concentrations (90 mg/mL, 45 mg/mL and 22.5 mg/mL) were prepared for each strain (Table 7, Table 8 and Table 9). Analyzing these results, we conclude (as for crude extract) that the inhibitory effect of each fraction decreases with increasing dilution, demonstrating direct proportionality between the total concentrations of components and increased inhibition zone. All fractions have a greater inhibitory effect for the highest concentration (90 mg/mL), with the halo of inhibition (29.5 ± 0.3 mm) caused by fraction F01 (19% α-pinene, 2.5% β-pinene and 25% sabinene) against the Gram-positive strains standing out. Comparing this result with the halo of inhibition obtained by the effect of the vapor (14.6 ± 0.7 mm) it follows that there may be a synergy between these components (pinenes and sabinene) when they act against S. aureus ATCC 25923. In future studies would be interesting to separate available fractions with pinenes and sabinene isolated in equal concentration, in order to reliably conclude which of the two components has higher antibacterial activity against S. aureus ATCC 25923 and if indeed there is synergy between them. For the lowest concentration of 22.5 mg/mL the Gram-positive strain did not exceeded a value of 17.7 ± 0.3 mm for the halo of inhibition (moderately susceptible) while the remaining strains, E. coli ATCC 25922 and P. aeruginosa ATCC 27853, were resistant to oil at this concentration.

For S. aureus ATCC 25923 and comparing fractions F03 and F02, it was found that the inhibition zone is slightly higher than for F03, probably due to the increased concentration of sabinene and not the presence of (+)-spathulenol so there appears to be antagonism between these two components. Comparing F04, F05 and F06, it is notable that the fraction containing the highest percentage of (+)-spathulenol showed the lowest activity against the Gram-positive strain, possibly due to an antagonism between (+)-spathulenol and terpinen-4-ol. E. coli ATCC 25922 is more sensitive to terpinen-4-ol. F05 (14% (+)-spathulenol and 25% terpienen-4-ol) is the fraction with the highest inhibitory effect in this strain, followed by F06 (7% (+)-spathulenol and 19% terpinen-4-ol). Comparing the results, we highlight the possible effect of the antagonistic (+)-spathulenol on terpinen-4-ol and sabinene in both Gram-negative strains. Comparing the fractions F01, F02 and the absence of inhibition zone by the effect of vapor in Gram-negative strains, we conclude that this may also be associated with an antagonism between the pinenes and sabinene at this group. For P. aeruginosa ATCC 27853, the fraction that showed highest rate of inhibition was F02 (11% sabinene) followed by F03 (12% sabinene and 16% (+)-spathulenol). On the contrary, E. coli ATCC 25922 showed better inhibition halos for the terpinen-4-ol, possibly due to the greater symmetry of the biphospholipidic outer membrane characteristic of the genus Pseudomonas spp, compared to the family Enterobacteriaceae.

According to the methodology described in 2.9, we obtained the values summarized in Table 8. Given the difficulty in visualizing the turbidity in the tubes due to the oil emulsion, the MIC results were evaluated according to the resazurin test. Predictably, and in accordance with the agar diffusion test, the oil has a stronger inhibitory effect on the Gram-positive strain with a MIC of 2.81 mg/mL corresponding to the dilution 1:32. The actual value for the MIC and MBC will be between 1.41 mg/mL and 2.81 mg/mL. For Gram-negative strains, the MIC and MBC observed were equal (1:8 dilution, 11.25 mg/mL) with the actual concentrations varying between 5.63 mg/mL and 11.25 mg/mL. Although a count of 440 CFU/mL was obtained for P. aeruginosa ATCC 27853, this value is less than 0.1% (1,000 CFU/mL) of the original inoculum (10⁶ CFU/mL), strengthening the lower bacterial effect (due to the proximity of MIC values and MBC) of oil of S. molle on this strain. Comparing the macrodilution and cylindrical cavities agar diffusion methods, there appears to be a good correlation between MIC values and inhibition halos for the lower concentrations, possibly due to several factors, such as the fact of being an oil (lipophilic), in aqueous emulsion it may not occur an efficient contact between the oil and the bacterial cell and poor diffusion in the agar.

Despite the low antibacterial activity, sabinene showed the best inhibitory effect on P. aeruginosa ATCC 27853, indicating that this strain is very insensitive/resistant to this oil, possibly due to low outer membrane permeability of strongly/moderately hydrophobic compounds and the multidrug efflux systems characteristic of this kind of organism. (+)-Spathulenol is the component that has a lower activity, or lack of it, with possible antagonism with the terpinen-4-ol level in all tested strains and the sabinene only in Gram-negative strains, Table 9.

The essential oil of the fruits and leaves of Schinus molle L. studied in this paper came from specimens of aguaribay (“molle”) grown in the “Angel Gallardo” Experimental Operation Center of the School of Chemical Engineering at the Universidad Nacional del Litoral, Ministerio de la Producción de la Provincia de Santa Fé, Santa Fé, Argentina, located at 31° 32' 55'' S and 60° 41' 27'' O. Final identification of Schinus molle L. was made by Juan de Dios Muñoz and the specimen is deposited in the Herbarium ERA 4350 (1996)-Jardín Botánico Oro Verde, Facultad de Ciencias Agrarias, UNER - Ruta provincial 11, km 10,5, Oro Verde, Entre Ríos, Argentina [22].

The oil extraction was performed on the crushed leaves and ripe fruits (industrial extraction) by steam distillation using percolation. In this technique the plant material is immersed in the mass of water, which is heated and generates saturated steam in contact with the mass of vegetable oil-bearing particles, physically removing it, and this mixture is then condensed and decanted, to separate the essential oil layer from the aqueous layer.

The overall analysis was performed by GC/MS on a Shimazdu QP-5000 gas chromatograph. The separation of oil components was performed by column chromatography on flash silica, using as eluent hexane and mixtures of hexane/EtOAc of increasing polarity. The fractions were analyzed by TLC. The compounds were identified by spectroscopic methods such as ¹H-, ¹³C-NMR, 2D NMR, GC/MS and High Resolution Mass Spectrometry.

The strains tested were: Staphylococcus aureus (ATCC 25923), Escherichia coli (ATCC 25922) and Pseudomonas aeruginosa (ATCC 27853) from the American Type Culture Collection (ATCC, LGC Standards S.L.U., Barcelona, Spain). The criterion for selection of these strains was in accordance with the standard CLSI M26-A [14].

To determine the antibacterial activity, we used the following resources and solutions: pre-prepared commercial cards of Muller-Hinton Agar (MH2, Ref 43301, BioMérieux SA, Linda-a-Velha, Lisboa, Portugal) with 4 mm thick broth Muller-Hinton Broth (MHB, Ref 724245, Oxoid Ltd, Hampshire, England), aqueous solvent for the oil emulsion (0.15% agar + 5% DMSO) [16] and 1% resazurin solution in saline (NaCl 0.85%) to colorimetrically detect bacterial growth.

For reconstitution of lyophilized bacterial strains in the swabs provided (Culti-Loops, Oxoid Ltd, Hampshire, England), we proceeded as follows: the swab was allowed to reach room temperature, the solid MH2 medium was inoculated by passing the swab on the agar and incubated 18 h–24 h/37 ± 0.5 °C. A second plate was inoculated by streaking from an isolated colony and incubated at the same temperature for 18 h–24 h. Thus we obtain the primary culture (stock culture). The working cultured is obtained by subculturing the primary culture in the same way and under the same incubation conditions [23].

All inocula were prepared and standardized from the working culture in accordance with the methodology proposed by the CLSI [24]. A sterile loop was removed from 1 to 2 colonies, emulsified in a sterile solution of NaCl 0.85% (Ref 20040, BioMérieux S.A., Linda-a-Velha, Lisboa, Portugal) and read on a DENSIMAT densitometer (BioMérieux S.A., Linda-a-Velha, Lisboa, Portugal) to ensure a microbial load of approximately of 10⁸ CFU/mL (turbidity corresponding to McFarland standard 0.5). Followed by addition of 0.1 mL of standardized inoculum and 9.9 mL of sterile MHB broth, resulting in a microbial load of approximately 10⁶ CFU/mL, the latter being the solution inoculated to perform all tests. Confirmation of the inoculum was made by successive dilutions of the working solution up to 10³ CFU/mL and 0.1 mL of further spread in the MH2 medium. The obtention of approximately 100 CFU/mL confirms the microbial load of the original inoculum.

To study the antibacterial activity due to volatility (vapor effects) of the crude oil and fractions of S. molle, we used the method described by Lysine et al. [10]. The crude oil and corresponding fractions were introduced in the swab inoculum (10⁶ CFU/mL) and pressed into the wall of the tube to remove excess liquid. It then spread evenly with the swab, the inoculum on the agar plate MH2 and left to stand for 15 min so they absorb the culture medium around the inoculum. In order to do a comparative test with the agar diffusion method using the technique of cylindrical cavities, it was added the same amount, 70 µL of the sample in the centre of the lid of the Petri dish, at a distance of approximately 1 cm from the agar inoculated, inverted the plate and incubated during 18 h–24 h/37 ± 0.5 °C. Assays were performed in hexaplicate.

The antimicrobial activity was evaluated according to the agar diffusion method proposed by the Clinical and Laboratory Standard Institute (CLSI, Wayne, Pennsylvania, PA, USA), but replacing the sterile paper discs impregnated with the sample by cylindrical cavities of 4 mm in height and 5 mm in diameter, punched into agar with a thickness of 4 mm [24,25]. As a positive control, we used the antibiotic gentamicin (Sigma-Aldrich, Sintra, Portugal) at a concentration of 10 mg/mL as negative control and the dilution solution (aqueous solution agar 0.15% + 5% DMSO). The cylindrical cavities were filled with 70 mL of oil in different concentration and their controls, after a rest period of 15 min at room temperature, which allows the diffusion of the components, the plates were incubated for 18 h–24 h/37 ± 0.5 °C. After the incubation period, the inhibition halos were measured by visual observation and a stereoscopic microscope. All assays were performed in hexaplicate.

To determine the MIC and MBC the technique of macrodilution in MHB broth was used, according to the methodology of CLSI M07–A8 [13]. This approach was complemented by the colorimetric resazurin test for improved detection of bacterial growth [26]. The resazurin was added only at the end of the incubation period. Previously, the inoculum of the microorganism to be tested was prepared at a concentration of 10⁶ CFU/mL in MHB (standardized to 3.6), the solution of the crude extract of S. molle was emulsified in the dilution solution to obtain a concentration of 90 mg/mL and finally the solution of resazurin was sterilized through a 0.22 µm porosity filter (02 Filtropur S, Sarstedt, Germany). In eight sterile tubes containing 1 mL of each MHB broth in double concentration, 1 mL of essential oil at a concentration of 90 mg/mL was added to the first tube and homogenized. Then 1 mL was withdrawn from the first tube and transferred to the second tube and so on until the last dilution. In the end, 1 mL of inoculum was added to all tubes (final concentration of oil is equivalent to 1/3) except for the negative control. As positive control we used MHB broth (double concentration) with 1 mL of inoculum. As a negative control, we used 1 mL of MHB broth (double concentration) with 1 mL of the crude extract oil. All test tubes, mixed thoroughly, were incubated for 18 h–24 h/37 ± 0.5 °C. After incubation, the tubes were visually examined and tested with resazurin, adding 9 mL of resazurin to 1 mL of sample. For the determination of the MBC, samples were inoculated by spreading 0.1 mL of each tube (including controls) on MH2 agar and incubating during 18 h–24 h/37 ± 0.5 °C and then proceeding to count the colony forming units (CFU). All assays were performed in triplicate. The tube with the lowest concentration of oil discolored from resazurin, was considered as corresponding to the MIC. The MBC corresponds to a dilution that shows inhibition of bacterial growth of at least 99.9% of the original inoculum (106 CFU/mL) [14].

The values of the inhibition halos in relation to the agar diffusion methods were measured in mm and include 5 mm diameter cylindrical cavity in accordance with the standard M100–S19, CLSI [27]. The scale of measurement (cavity diameter included) was as follows: ≥20 mm strong inhibitory effects, <20 to 12 mm moderate inhibitory effects, >12 mm weak inhibitory effects. Assays were performed in hexaplicate and data presented as mean ± standard deviation. Readings were made with a ruler and confirmed with a stereoscopic microscope.

Data analysis was performed using the SPSS 17.0 statistics program, by analysis of variance, using two-way ANOVA and comparison of means by a study comparing a posteriori (post hoc)—Tukey Test—allowed to differentiate minimum between means when the probability value p < 0.05.