In Vitro Anti-Leishmanial Activity of Essential Oils Extracted from Vietnamese Plants

Leishmania mexicana is one of the pathogens causing cutaneous leishmaniasis which is associated with patient morbidity. In our researches for new safe and effective treatments, thirty-seven essential oils (EOs) extracted from Vietnamese plants were screened in vitro for the first time on Leishmania mexicana mexicana (Lmm) promastigotes at the maximum concentration of 50 nL/mL. Active EOs were also analyzed for cytotoxicity on mammalian cell lines (WI38, J774) and their selectivity indices (SI) were calculated. Their composition was determined by GC-MS and GC-FID. Our results indicated that EOs extracted from Cinnamomum cassia, Zingiber zerumbet, Elsholtzia ciliata and Amomum aromaticum, possessed a moderate anti-leishmanial activity, with IC50 values of 2.92 ± 0.08, 3.34 ± 0.34, 8.49 ± 0.32 and 9.25 ± 0.64 nL/mL respectively. However, they also showed cytotoxicity with SI < 10. The most promising EO was extracted from Ocimum gratissimum, displaying an IC50 of 4.85 ± 1.65 nL/mL and SI > 10. It contained 86.5% eugenol, which was demonstrated to be effective on Lmm with IC50 of 2.57 ± 0.57 nL/mL and not toxic on mammalian cells, explaining the observed activity.


Introduction
Cutaneous leishmaniasis (CL), the most common form of leishmaniasis, largely affects poor and developing countries in Africa, the Mediterranean Basin, the Middle East, central Asia (the Old World) and South America (the New World). The estimated number of new CL cases worldwide ranges from 0.7 million to 1.3 million annually [1].
CL is caused by protozoan parasites of the Leishmania (L.) genus. Infected female phlebotomine sand flies inject these parasites from their proboscis to human when they take their blood meals. Skin lesions, the clinical signs of CL, develop within several weeks or months after exposure. These lesions are sometimes self-healing without treatment, or become chronic. In cases of ulcers healing, they leave permanent deep scars, which often cause a serious social prejudice. Conversely, if the disease becomes chronic, there is a high risk of severe bacterial infection. CL in the New World, mainly caused by L. mexicana, L. amazonensis, L. venezuelensis, L. braziliensis, L. guyanensis, L. panamensis, and L. peruviana, These selected EOs were further analyzed for dose-response activity to calculate IC 50 value and for cytotoxicity on mammalian cells. The results are summarized in Table 2. All selected EOs revealed high potential against promastigote form of Lmm with IC 50 values lower than 10 nL/mL. However, four of them also showed toxicity on mammalian cells, as indicated by their SI < 10. Our data demonstrated that the most active and selective EO was extracted from Ocimum gratissimum with an IC 50 value on Lmm promastigotes of 4.85 ± 1.65 nL/mL and no toxicity towards mammalian cells at the highest tested concentration (50 nL/mL), as evidenced by more than 80% viable cells after 72 h of incubation. To define their composition and further determine active compounds, the five selected EOs were analyzed by GC-MS and GC-FID (Table 3). Identified compounds accounted for more than 90% for each EO. Eucalyptol (55.2%), trans-cinnamaldehyde (83.6%), citral (neral and geranial) (40.2%), eugenol (86.5%), and zerumbone (60.3%) were characterized as the main components of EOs extracted from A. aromaticum, C. cassia, E. ciliata, O. gratissimum, and Z. zerumbet, respectively.
Neryl acetate 1218 Geraniol Cubenol t: trace (peak area less than 0.05%); RI: the retention index was calculated using a homologous series of fatty acid methyl esters C 5 -C 27 ; MS: mass spectra (matching coefficient >700 compared with NIST database); Co-GC: co-injection with pure compound.
It is important to point out that eugenol is the major compound (86.5%) of the EO extracted from O. gratissimum, which was shown to be the most selective active sample. In order to understand the activity of this EO, in vitro anti-leishmanial activity and cytotoxicity of eugenol were studied. Interestingly, the IC 50 value against Lmm promastigotes of eugenol was 2.57 ± 0.57 nL/mL (=2.72 µg/mL, 15.67 µM) and more than 80% of mammalian cells were living after 72 h when incubated at the maximal concentration used (50 nL/mL).

Discussion
This study analyzed the in vitro activity of 37 EOs extracted from Vietnamese plants against Leishmania mexicana mexicana. According to the classification of anti-parasitic activity from literature, extracts that showed effect against parasites with IC 50 value ≤2 µg/mL (or 2 µM for pure compounds) are cited as having good activity, those with IC 50 between 2 and 20 µg/mL (or micro molar for pure compounds) are considered as having a moderate activity, while those with higher IC 50 value were considered as less interesting [11]. In this study, the primary screening was performed against Lmm promastigotes at concentrations of 25 and 50 nL/mL in order to quickly identify interesting EOs (1 nL is a little less than 1 µg depending on the density of the EO) [11]. Our data showed that percentages of viable parasites incubated with the lower concentration of five EOs, extracted from A. aromaticum, C. cassia, E. ciliata, O. gratissimum, and Z. zerumbet, were lower than 1%, meaning that these EOs revealed a potential activity. Indeed, the IC 50 values of these selected EOs were lower than 10 nL/mL in the second analysis.
To have a general look at anti-leishmanial activity and cytotoxicity of the five more active EOs and to understand their observed effects, we compiled literature data of EOs extracted from similar plant species and also of their major components in Table 4.  As shown in this table, among the five selected samples, EOs obtained from A. aromaticum, C. cassia, and E. ciliata were analyzed here for the first time for their leishmanicidal effect. In our experiment, they showed IC 50 values of 9.25 ± 0.64, 2.29 ± 0.08, and 8.49 ± 0.32 nL/mL, respectively against Lmm promastigotes. Unfortunately, this activity was not very selective on Leishmania as SI values compared to non-cancer mammalian cells (WI38) and cancer cells (J774) were approximately 5 and 2, respectively. Two EOs extracted from O. gratissimum and Z. zerumbet were already investigated for anti-leishmanial activity [13][14][15][16] however on other L. species. The EO extracted from fresh leaves of O. gratissimum collected in Brazil did not reveal interesting effects against both L. amazonensis and L. chagasi. On the contrary, in our study, the EO extracted from O. gratissimum collected in Vietnam demonstrated notable activity against Lmm promastigotes, with an IC 50 value of 4.85 ± 1.65 nL/mL. These results indicated a specific activity of this EO related to parasite species (Leishmania mexicana mexicana) and/or collection place and composition. Another important feature of this EO is its absence of cytotoxicity as the percentages of viable WI38 and J774 cells at the maximal tested concentration (50 nL/mL) were higher than 80%. We therefore selected the EO extracted from O. gratissimum as the most active and selective sample. Regarding the EO extracted from fresh rhizomes of Z. zerumbet, its activity against L. donovani was in the same range than what we found against Lmm promastigotes (IC 50 = 3.34 ± 0.34 nL/mL). Unfortunately, we also observed cytotoxicity of this EO on WI38 and J774 cells as shown by SI values of 1.10 and 0.72, respectively.
Citral was characterized as the main component of the EO extracted from E. ciliata (40.2%). It showed a moderate activity against L. donovani, L. amazonensis and a weak effect against L. infantum, L. tropica and L. major [20][21][22]. However, it is important to point out that this compound is not selective on Leishmania as shown by the toxicity on kidney epithelial and J774 cells [20,21]. Our results indicated activity on parasites and mammalian cells of the EO extracted from E. ciliata in the same range than citral with IC 50 value of 8.49 ± 0.32 nL/mL and SI of 5.58 (compared to WI38) and 1.56 (compared to J774). Nevertheless, citral being present at only 40.2% in the E. ciliata EO, other compounds should also have anti-leishmanial activity as β-(E)-ocimene (14.0%), linalool (8.3%), 1-octen-3-ol (7.1%) and β-(E)-farnesene (6.2%). Dutra et al. have already examined leishmanicidal effect of linalool on L. infantum chagasi. However when axenic amastigotes were treated with linalool, IC 50 was 550 µg/mL [24].
EO extracted from C. cassia contained 83.6% of trans-cinnamaldehyde. The anti-leishmanial activity of this compound is unknown but its toxicity on two human cell lines was determined [19]. In our sample, the high percentage of cinnamaldehyde may explain the toxicity of this EO on both mammalian cell lines, WI38 and J774, with IC 50 of 14.19 ± 0.54 and 6.26 ± 0.80 nL/mL respectively.
Given the known cytotoxicity of this compound and its high concentration in the C. cassia EO, we did not find interesting to further analyze its anti-parasitic activity.
As mentioned previously, eugenol was characterized as the major compound (86.5%) of the most interesting EO extracted from our Vietnamese sample of O. gratissimum. Literature data indicated a moderate effect on L. amazonensis and a less interesting activity against L. infantum chagasi of eugenol [23][24][25]. On L. donovani, along with a moderate in vitro activity, in vivo effect of eugenol (emulsified) was determined. The intra-peritoneal administration of an eugenol emulsion at the dose of 75 mg/kg b.w. for 10 consecutive days decreased by 87.01 ± 5.85 and 86.68 ± 5.42% parasitic load in spleen and liver, respectively, in 8-weeks infected BALB/c mice. Moreover, a significant reduction in spleen size, and spleen and liver weights were also found at this dose [26]. Our results further support the anti-leishmanial activity of this compound with IC 50 value of 2.57 ± 0.57 nL/mL (=2.72 µg/mL or 15.67 µM) against Lmm promastigotes. Interestingly, eugenol was not toxic on both non-cancer and cancer mammalian cells in our models at the highest analyzed concentration (50 nL/mL). From these results, and its high percentage in the O. gratissimum EO, it can be concluded that activity of eugenol can explain the anti-leishmanial activity of this EO.
Zerumbone, accounting for 60.3% of the EO extracted from fresh rhizomes of Z. zerumbet, was reported to be active against L. donovani, however it revealed toxicity on human leukemia cells HL-60 [27,28]. These data can explain the anti-leishmanial activity but also cytotoxicity of the Z. zerumbet EO found in our study.
Mechanisms of anti-leishmanial activity of these EOs and their major compounds are not well known. Usually, effects were analyzed on the morphology of treated parasites or as a consequence of immunostimulatory activities.
Most of the studies used transmission electron microscopy (TEM) and scanning electron microscopy (SEM) to analyze the morphology of EOs treated parasites. The EO extracted from O. gratissimum caused considerable mitochondrial swelling in L. amazonensis promastigotes at 135 µg/mL and amastigotes at 100 µg/mL [13], swelling of cell body, flagellar pocket and mitochondria in L. chagasi promastigotes at 50 µg/mL [15]. Ultrastructure alterations were also observed in L. amazonenesis and L. infantum promastigotes treated with citral at concentrations of 8.0 and 42 µg/mL respectively [21,22]. Mukherjee et al. detected morphological alterations by SEM in L. donovani promastigotes treated with 9.36 µM of zerumbone associated with induced ROS-mediated apoptosis [27].
Using other experiments, citral at the concentration of 42 µg/mL was shown to trigger programmed cell death of L. infantum promastigotes, as indicated by the externalization of phosphatidylserine, loss of mitochondrial membrane potential and cell-cycle arrest at the G(0)/G(1) phase [22].
Regarding immunostimulatory activity, an increase of nitric oxide produced by infected macrophages has been suggested to be responsible for the activity of the O. gratissimum EO against L. amazonensis at 100 and 150 µg/mL [13]. Islamuddin et al. explored the synergic effect between eugenol emulsion and the immune system. In BALB/mice infected by L. donovani, treatment with 75 mg/kg b.w. of eugenol emulsion enhanced IFN-γ and IL-2 serum levels, as well as increased CD4+ and CD8+ T cell population and expanded IFN-γ producing CD4+ and CD8+ splenic T lymphocytes [26].
Concerning the mechanism of eugenol activity, although no data are available on Lmm promastigotes, some experiments were carried out on bacteria and fungi. Eugenol at concentrations of 5.3 and 10.6 mg/mL was reported to significantly damage both the cell wall and membrane of the treated Gram-negative and Gram-positive bacteria [31]. Khan et al. observed damaging effects of eugenol at 200 µg/mL on cell wall, cell membrane, cytoplasmic contents and other membranous structures of treated Candida albicans [32]. Indeed, because of the lipophilicity of eugenol, it could easily diffuse between the fatty acyl chains of lipid bilayers modifying the fluidity, integrity and permeability of cell membranes [31]. We now intend to analyze possible effects of this compound on leishmanial membranes.

Plants Collection
Thirty-seven plants were collected from different areas of Vietnam in November 2014 and from May to August 2015. They were identified by matching with literature and herbarium specimen at the Botanical Department, Hanoi University of Pharmacy, Vietnam. The information of sample name, genus, species, family, and collector name is given in the supplementary material.

Essential Oils Extraction
Fresh samples were extracted by hydro-distillation using a modified Clavenger apparatus for two hours. Each essential oil obtained from hydro-distillation was dried using sodium sulfate, filtered and kept refrigerated prior to analysis. Stock solutions at the concentration of 20 µL/mL were prepared in DMSO and diluted further in culture medium to achieve a maximum final DMSO concentration of 0.25%.

Anti-leishmanial Assay
The Lmm promastigote density was counted on a haemocytometer and adjusted to 10 5 parasites/mL. The assays were performed in 96-well plates. For primary screening (repeated two times at concentrations of 50 and 25 nL/mL in triplicate) essential oils were diluted in culture medium from the stock solutions (20 µL/mL). Each well was filled with 50 µL of the diluted essential oil and 50 µL of the Lmm promastigote culture (total volume 100 µL). After 72 h of incubation, 10 µL Alamar blue (Thermo Fisher Scientific, diluted with PBS at the ratio 1:1) was added to each well and the plates were further incubated for 4 h. Fluorescence was measured on a spectrophotometer (SpectraMax-Molecular Devices, Berkshire, UK) at 530 nm excitation and 590 nm emission wavelengths. Pentamidine was tested as standard drug. The essential oils that inhibited more than 50% the growth of Lmm promastigotes at the concentration of 25 nL/mL in the primary screening were submitted to a second analysis for accurate IC 50 determination. Selected essential oils were screened at least three times at concentrations ranging from 50 to 0.02 nL/mL in duplicate. IC 50 values were calculated from dose response growth inhibition curves by Microsoft excel files.

Cytotoxicity Assay
The essential oils which were analyzed for IC 50 value on anti-leishmanial assay were also analyzed for cytotoxicity against non-cancer (WI38) and cancer (J774) mammalian cells. The culture of cells was diluted with medium to the adequate density of 5 × 10 3 cells/mL and then 180 µL was added into each well of 96-well plates. After 24 h of incubation, 20 µL of essential oils diluted in culture medium was added to each well (to obtain concentrations ranging from 50−0.02 nL/mL) for further 72 h of incubation. After removing the medium, 100 µL of MTT was added to each well and the plates were further incubated for 45 minutes. 100 µL of DMSO was used to dissolve formed formazan crystals after MTT removing. Absorbance was measured on a spectrophotometer (SpectraMax-Molecular Devices, Berkshire, UK) at 570 nm with a reference wavelength at 620 nm. All experiments were made at least two times in triplicate. The computation of the IC 50 values was performed with GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA, USA). The selectivity index (SI) values were calculated using the formula below: SI = IC 50 for mammalian cell/IC 50 for protozoan parasite

Essential Oil Analysis
The GC-MS analyses were carried out on a TRACE GC 2000 series (Thermo-Quest, Rodano, Italy), with a DB-WAX capillary column (30 m × 0.25 mm × 0.25 µm) using the following operating conditions: injection volume: 1 µL (TBME solution); injection mode: splitless; injector temperature: 230 • C; oven temperature: increased from 45 • C (held on 5 min) to 250 • C (held on 5 min) at 3 • C/min; helium was used as a carrier gas at a constant flow of 1.3 mL/min; detector temperatures: 260 • C; ion source: 70 eV. The oil components were identified using linear retention indices in relation to a series of fatty acid methyl esters (C 5 -C 27 ), pure compounds and NIST mass spectral library (matching coefficient > 700). GC-FID was done using the same column and conditions on a FOCUS GC (Thermo Finnigan, Milan, Italy) with modifications of injection mode (split at ratio 1:50) and detector temperature (250 • C). Percentage of compounds was calculated by the normalization procedure.

Conclusions
The present study analyzes for the first time the anti-leishmanial activity of 37 essential oils extracted from Vietnamese plants. Those extracted from A. aromaticum, C. cassia and E. ciliata are shown here for the first time to be effective on a Leishmania species, while the effects of O. gratissimum and Z. zerumbet EOs are reported for the first time on L. mexicana species. More than 90% of their contents was characterized. Eugenol was identified as the major compound of the O. gratissimum EO, the most active and selective one. This compound, showing a moderate anti-leishmanial activity and low cytotoxicity in the tested models, can explain this EO activity. However, further results are necessary before developing it in the treatment of cutaneous leishmaniasis such as in vivo assessment in Lmm infected animals and determination of its mode of action.