Abstract

In the last three decades, invasive fungal infections caused by Candida species have become an important public health problem, because they are associated with high rates of morbidity and mortality in immunocompromised and hospitalized patients. The diagnosis and treatment of candidiasis are difficult and usually inefficient. Accordingly, a diversity of available drugs, currently employed to attack candidiasis, frequently induce resistance in patients promoting toxicity due to long-term treatments. Therefore, development of accurate diagnoses and novel antifungals is of high priority to improve life’s quality and expectancy of individuals infected with this pathogen. Plants are invaluable sources of new biologically active compounds. Among the plants used in Mexico in traditional herbolary medicine which have empirically been demonstrated to have antifungal activity are Pedilanthus tithymaloides, Thymus vulgaris, and Ocimum basilicum. In the present study, we analyzed whether these plants contain metabolites with antifungal activity against five Candida species. The extracts from the different plant organs were obtained by macerating them in ethyl alcohol or hexane and filtering. The obtained extracts were preserved in amber flasks at 4°C until used. The minimum inhibitory concentrations (MICs) of the active compound were determined by a microdilution assay. In addition, the following secondary metabolites were identified: linalool (3,7-dimethylocta-1,6-dien-3-ol), eugenol (4-allyl-2-methoxyphenol), limonene (1-methyl-4-(1-methylethenyl)-cyclohexene), and borneol ([(2R)-1,7,7-trimethyl-2-bicyclo[2.2.1]heptanyl] formate). All these compounds were found in the three plants, traditionally used in everyday life, and proved to be effective against Candida species and therefore a viable alternative to conventional antifungals.

1. Introduction

It has been estimated that there are over 5 million fungal species worldwide, and approximately 300 out of these are known to cause diseases in humans, while 20-25% do it with relative frequency. Candida species are among these pathogens. Most of the fungal infections are nontransmissible among people and routinely do not affect healthy individuals [1]. Several species from the Candida genus that can be classified as part of this group of pathogens are widely distributed in nature and are part of the normal microbiota in the oral cavity, the gastrointestinal tract, and the urogenital system in human host. These organisms do not trigger infections in healthy hosts; however, some of these fungi can behave as opportunistic pathogens when the immune system of the host is compromised, causing infections called superficial candidiasis (cutaneous and mucous infections) and systemic or invasive candidiasis (infection of the bloodstream and invasive candidiasis of organs) [2, 3]. These infections amount, as a whole, to approximately 40 million per year worldwide [4]. Invasive candidiasis (IC) represents one of the most common nosocomial infections due to fungi, particularly, in cancer patients and in individuals under immunosuppression regimes [5, 6]. Candidemia can reach a mortality rate from 30 to 60% in hospitalized immunocompromised patients [7, 8]. In more than half of these mycosis cases, death occurs in the first week after the diagnosis of infection due to Candida [9]. Candida albicans has been identified as the most prevalent and pathogenic species; it is responsible for most oral and systemic candidiasis cases, as well as for community and nosocomial origin candidemia [10]. In the last decades, Candida non-C. albicans species like Candida glabrata, Candida tropicalis, Candida parapsilosis, Candida dubliniensis, and, recently, Candida auris have been increasing [1115]. Invasive candidiasis causes significant morbidity and mortality despite intensive treatments with antifungal agents [16, 17]. The mortality attributed to this disease has not diminished significantly despite the new diagnostic methods, the new antimycotic treatment options, and the better control of the infection [18]. Starting at the second half of the last century, commercialization of amphotericin B, although toxic, helped importantly in the treatment of systemic mycoses [19]. The advent of topic azoles like miconazole and clotrimazole, and later of systemic azoles like ketoconazole, itraconazole, fluconazole, and voriconazole, has simplified the treatment of superficial and systemic mycoses, improving the healing expectancy of these infections. However, with the increase in the HIV and cancer incidences and the indiscriminate use of steroids, surgical procedures, and transplants, failed cases have been reported with the antimycotic therapy using diverse compounds like ketoconazole, fluconazole, and even amphotericin B [20]. Hence, the mortality attributed to fungi is still too high even with the current antifungal agents; thus, a greater emphasis must be placed in improving the time used for the fungicide activity of the new antifungal agents. Currently, the general treatment course with common antimycotics is too long and, hence, presents the potential of a deficient short-term fungicide effect, a diminution in compliancy and/or tolerability by patients, or even the appearance of direct resistance to the antifungal drugs [21]. A complication that arises with the treatment of mycoses is the resistance to the antifungal agents; this is defined as the capacity acquired by an organism to resist the effects of a chemotherapeutic agent to which it is habitually sensitive [22]. Plants are a viable option to obtain a wide variety of pharmaceuticals because they are easily accessible and can be applied to diverse pathologies [23]. In this way, plants constitute an excellent source of substances that can be used in the formulation of new antifungal agents [24]. However, the development of the pharmaceutical industry and the synthesis of molecules with diverse activities in the clinical field displaced the use of medicinal plants in many regions of the world, mostly in large cities [25]. Although the current available drugs are usually efficacious, the therapeutic failures and the toxicity after a long treatment are common; thus, plants are a good alternative to be explored to improve treatments of severe infections [26, 27]. It has been calculated that there are from 200 to 500 thousand species of higher plants worldwide; in Mexico, the diversity is estimated from 23 to 30 thousand species [28]. Knowledge on the chemical diversity of plants is still limited worldwide; it is calculated that the chemical structure of around 100 thousand secondary metabolites is known, and there could be at least one million of them in all the species that have not been studied yet [29]. Likewise, the biological properties of secondary metabolites are unknown; in most cases, research is centered on priority problems, like the search of anticancer agents, and they do not attempt to perform integrated studies on all the biological activities. Regarding candidiasis treatment, even though there are currently a large variety of available antifungal agents, it is increasingly more frequent that patients do not respond to treatment, causing toxicity after a long treatment. For this reason, it is necessary to identify new alternatives for the treatment of candidiasis. Plants are an invaluable source of new biological active compounds. Among the reported secondary metabolites in plants with antifungal activity are flavonoids, phenols, glycosides of phenols, and saponins. Among the plants used in traditional medicine that have empirically shown to possess antifungal activity are Pedilanthus tithymaloides, Thymus vulgaris, and Ocimum basilicum. In the present study, we analyzed the secondary metabolites from these plants as well as their antifungal activity against five Candida species.

2. Materials and Methods

2.1. Plant Material

Pedilanthus tithymaloides was donated and collected from a private orchard in the municipality of Leon (state of Guanajuato, Mexico). Lateral stems of the shrub, detached from the nodes, were collected. Thymus vulgaris and Ocimum basilicum plants were acquired from a nursery in Guanajuato, Mexico.

2.2. Sampling of Plants

The leaves and stems of each plant were separated, placed in individual recipients, and washed with tap water to eliminate any residues; then, they were washed again with sterile deionized water. The leaves and stems were placed separately on a grid that allowed dripping the excess water to dry the plant material. This was carried out at room temperature and away from light until use.

2.3. Extract Preparation

The plant material (leaves or stems) was placed on an analytical balance on top of a polyethylene tray until a mass of 40 g was reached and further sterilized in 2% sodium hypochlorite for 10 min. Keeping sterile conditions, this plant material was placed in a mortar, adding 100 mL of 99.5% ethyl alcohol or hexane (which was previously cooled in an ice bath), and crushed to homogeneity. The obtained macerate was filtered with a vacuum filtration system (Kitasato flask-Büchner funnel-TYPE HVLP, 0.45 μm membrane). The obtained extracts were kept in amber flasks at 4°C until their use.

2.4. Fungal Strains and Culture Media

The strains of C. albicans, C. dubliniensis, C. glabrata, C. krusei, and C. parapsilosis species were obtained from the strain collection of the mycology laboratory, which is part of the Department of Biology of the Natural and Exact Sciences Division, at the Universidad de Guanajuato. Strains were grown in YPD (1% yeast extract, 2% Bacto-Peptone, and 2% dextrose) medium. The solid medium was supplemented with 2% agar [30].

2.5. Susceptibility Assays to the Different Extracts by Means of the Disk Diffusion Method

The minimal inhibitory concentrations (MICs) of the active compounds were determined through a microdilution assay as previously described [31]. The effect of the different extracts on the five Candida species was determined with yeasts in the stationary growth phase. For the latter, the five Candida species were grown during 48 h at 28°C under constant agitation at 120 rpm. Once the incubation period had ended, the optical density was measured at 600 nm (OD600nm) and adjusted to 0.5 and 0.1 in 1 mL of sterile deionized water. The cells adjusted to the different ODs were streaked on YPD plates with the help of a sterile Digralsky loop, adding 75 μL of the suspension to each plate. Afterward, the plate was divided in sections, and with previously sterilized curved point steel tweezers, sterile filter paper disks (Whatman 40) of ca. 6 mm in diameter were placed in each section. Of each extract to be evaluated, 50 μL was placed on top of each disk (ethanol extract of stem or leaf of P. tithymaloides, T. vulgaris, or O. basilicum. Hexane extracts of leaf or stem of P. tithymaloides, T. vulgaris, or O. basilicum were also used. To assess whether the solvents used for the extracts affected or not the cell growth, ethanol and hexane were used as negative controls. Additionally, another solvent, reported as relatively innocuous, that is, dimethyl sulfoxide (DMSO), was used. As positive control, phenol at different proportions (1 : 10, 1 : 100, 1 : 1000, and 1 : 10000) and absolute phenol were used. Plates were incubated at 28°C for 36-48 h. After the incubation period, photographs were taken using the gel documentation system (GeneGenius Bio Imaging System, Artisan Technology Group, Champaign, IL, USA, from Syngene). Experiments were performed in triplicate.

2.6. Analysis of Extracts
2.6.1. Gas Chromatography Coupled to Mass Spectrometry (GC/MS)

The secondary metabolites contained in the hexane extract of leaves and stems of P. tithymaloides, T. vulgaris, and O. basilicum were identified by means of GC [32] coupled to a Clarus SQ8 MS (Perkin Elmer, Inc., Waltham, MA, USA) equipped with an ion deflector that allowed identifying low volatility compounds and those thermally labile. This equipment is provided with a capillary DB-5 column of phenyl methyl silicone (30 m length, 0.25 mm in diameter, and 0.25 μm of phase thickness). From each sample, 1 μL was taken and injected in splitless mode, with a total execution time of 30 min. The chromatogram was interpreted with the aid of the AMDIS software. As a previous step to the GC/MS analysis, a derivatization reaction was performed to increase the volatility and thermal stability of the polar compounds which implies to derivatize one or more polar groups of one compound to a less polar group. In 2 mL Eppendorf tubes, 2 mL of the ethanol extract was deposited and lyophilized in the Speed-vac equipment to evaporate the solvent during 3 to 5 h. The lyophilized extract was resuspended in 350 μL of pyridine, then, 50 μL of the derivatizing agent, bis-(trimethylsilyl) trifluoroacetamide (BSTFA), which contains 1% chlorotrimethylsilane, was added, and the tubes were vortexed. Afterward, the mixture was heated in a thermoblock at a constant temperature of 50°C for 1 h. Once the thermal exposure time was concluded, it was centrifuged at 1200 rpm for 4 min. From the derivatized samples, 200 μL was taken, and the metabolites were identified by means of GC/MS. The analysis was performed in triplicate.

2.6.2. High Performance Liquid Chromatography (HPLC)

To obtain the pure fraction of the ethanol or hexane extracts, these were fractionated by means of HPLC (Model Altus A30, Perkin Elmer, Inc., Waltham, MA, USA) [33] with a quaternary pump and a diode detector arrangement (DDA). To inject the sample, the air sampler of the same equipment was used. Separation was performed with a C18 column in isocratic mode with a mobile 45% acetonitrile phase and 55% water. The flow of the mobile phase was of , performing detection at 275 nm. A 1.5 mL volume was taken from the extracts and filtered through a 0.2 μm pore membrane; the filtered extracts were recovered in 1.5 mL Eppendorf tubes and then lyophilized. The lyophilized extract was resuspended in 1.5 mL of HPLC-grade methanol and sonicated for 30 min. The process was repeated twice. To eliminate the solvent of the collected fractions from HPLC, these were lyophilized. The obtained lyophilized samples were resuspended in 2 mL of the initial solvent of each extract, ethanol or hexane, to perform the susceptibility assays against the five Candida species according to the previously described protocol. Experiments were performed in triplicate.

2.6.3. Prediction of Targets for the Identified Metabolites

To evaluate whether the metabolites identified in P. tithymaloides, T. vulgaris, and O. basilicum with possible antifungal activity did not have targets in human cells, the SwissTargetPrediction (http://www.swisstargetprediction.ch.) free-access server was used. This server predicts accurately the targets of the bioactive molecules based on 2D and 3D similitude measurements with known ligands [34].

2.6.4. Statistical Analysis

In order to determine which of the P. tithymaloides, O. basilicum, and T. vulgaris extracts showed the highest antifungal effect against the five Candida species, growth inhibition halos’ diameters in three independent experiments were measured and reported as a error of the mean (SEM). Statistical differences between means were determined using a one-way ANOVA test followed by Tukey’s posttest for multiple comparisons. Normality and homogeneity of data were evaluated using both graphic evaluation and analytical tests such as Shapiro Wilk and Bartlett’s test. Minitab 21.1 Software (State Collage PA, USA) was utilized to perform the statistical analysis. Statistical significance was considered when .

3. Results

3.1. Susceptibility Assays to the Different Plant Extracts with Possible Antifungal Activity

To rule out effects of the solvents employed to prepare the plant extracts, the susceptibility of the five Candida species to hexane and ethanol treatment was tested. Additionally, DMSO and phenol were evaluated as a negative and positive controls, respectively. Assays revealed that non-Candida species exhibited susceptibility to hexane, ethanol, or DMSO (Figure 1(a)). In contrast, phenol treatment, at the highest dilution (1 : 10) affected the growth of all the Candida strains tested (Figure 1(b)). In conclusion, ethanol and hexane were adequate to obtain the extracts from the studied plants to investigate the presence of metabolites with possible antifungal activity.

We investigated whether P. tithymaloides, T. vulgaris, and O. basilicum possessed antifungal activity against five Candida species; to this end, ethanolic and hexane extracts prepared with stem and leaves from these plants were tested in disk diffusion assays as described in Materials and Methods. To investigate if the cell concentration impacts the antifungal efficiency of the plant extracts, the cultured Candida strains were tested at OD600nm of 0.5 and 1.0, respectively, considering that for Candida species, an OD600nm of 1.0 corresponds to cells. It was decided to work with this amount of cells because susceptibility must be evaluated with a number of cells that will allow observing the antifungal effect of the compounds to be analyzed, which is not possible if working with a large amount of cells, where the antifungal effect is masked by the high cell density.

As shown in Figure 2(a), at both ODs, the ethanol extract of the stems presented a higher inhibition of the five Candida species, as compared to the leaf extracts. In contrast, except for leaf extract of T. vulgaris, the hexane extracts of P. tithymaloides, O. basilicum, and T. vulgaris did not present a significant antifungal activity against the Candida species, except the leaf extract of T. vulgaris (Figure 2(b)). These results indicate that the metabolites with possible antifungal activity are found mainly in ethanol extracts. Furthermore, they indicate that the antifungal effect is independent from the number of cells and may correspond to the intrinsic effect of the extracts (Figures 2(a) and 2(b)).

To determine which of the P. tithymaloides, O. basilicum, and T. vulgaris extracts showed the highest antifungal effect against the five Candida species, we measured the growth inhibition halos’ diameters in three independent experiments at both an OD of 0.5 (Table S1) and an OD of 0.1 (Table S2), and the means of all experiments were taken.

For the ethanol extract of the P. tithymaloides stem, at both ODs, the C. dubliniensis species presented the highest susceptibility, followed by C. parapsilosis, C. glabrata, C. krusei, and C. albicans (Figures 3 and 4). For the leaf extract of P. tithymaloides, a nil antifungal activity was observed. Assessment of the stem extract of T. vulgaris revealed that C. parapsilosis and C. krusei were the most susceptible species at an OD of 0.5 and 0.1, respectively (Figures 3 and 4).

With the ethanol extract of the O. basilicum stem, C. glabrata was more susceptible at OD600nm 0.5 and C. dubliniensis at OD600nm 0.1 (Figures 3 and 4), whereas C. albicans and C. parapsilosis were the species with the highest resistance to this extract (Figures 3 and 4). With the ethanol extracts of T. vulgaris leaves, at an OD600nm of 0.5, C. albicans presented the highest susceptibility (Figure 3), whereas at an OD600nmof 0.1, it was C. dubliniensis (Figure 4); the resistance to this extract by C. parapsilosis, C. glabrata, and C. krusei at both ODs was different (Figures 3 and 4). With the stem extract of O. basilicum, at an OD600nm of 0.5, C. albicans was the species with the highest resistance, followed by C. parapsilosis, C. krusei, C. dubliniensis, and C. glabrata (Figure 3). In contrast, at an OD600nm of 0.1, C. glabrata, C. krusei, and C. parapsilosis were the species with the highest resistance to this extract (Figure 4). The highest susceptibility to the ethanol extract of O. basilicum leaves was shown by C. albicans at both ODs tested (Figures 3 and 4), whereas C. parapsilosis and C. krusei were the species with the highest resistance to this extract (Figures 3 and 4). Interestingly, the T. vulgaris leaf extracts prepared with hexane but not those from stems affected the growth of the five Candida species, when these were tested at both cell’s concentrations (Figures 3 and 4).Overall, our results revealed that the solvents employed to prepare the extracts impacted the antifungal activity exhibited by the leaves and stems of the plants tested and that such effects may obey to the existence of a differential profile of metabolites with various degrees of antifungal activity.

3.2. Analysis of the Extracts Obtained with Ethanol and Hexane

The stems and leaf extracts obtained with hexane were analyzed without any GC-MS treatment, whereas the ethanol extracts were subjected to a derivatization reaction to be analyzed with this technique. The chromatograms obtained of each extract were analyzed with the AMDIS software; this approach allowed us to identify a set of secondary plant metabolites (Fig. S1). Table 1 depicts the compounds identified through GC/MS analysis of the extracts that presented antifungal activity against the five Candida species.

The analysis of the T. vulgaris extracts revealed 15 potential compounds with antifungal activity; the leaf was the plant tissue with the highest number of chemical compounds with 15 compounds, and ethanol was the extraction agent with the most extracted metabolites. Compounds like carvacrol, eugenol, carveol, and p-cymene were identified only in ethanol extracts of both leaves and stems. The thymol and thymol-methyl-ether were identified in all plant extracts. In O. basilicum, 11 compounds were identified in ethanol extracts (Table 1). The compounds were identified in both the leaves and stems, revealing that metabolites are present in both plant organs, but at different concentrations, this would explain why the best antifungal effect against Candida species is observed in stems (Figure 2(a)). Regarding P. tithymaloides stem extracts, eight compounds were identified. In this plant, no metabolites were identified in the leaf extracts; i.e., no antifungal activity was recorded (Figure 2). Eugenol, linalool, limonene, and borneol were found in more than one of the plants (Table 1); however, linalool was the only compound identified in the extracts of the three plants (Table 1). This finding indicates that possibly, this compound is the main one implicated in the antifungal effect against C. albicans, C. dubliniensis, C. glabrata, C. krusei, and C. parapsilosis. Once the metabolites had been identified in the different extracts of P. tithymaloides, T. vulgaris, and O. basilicum, it was decided to purify the different metabolites, to assess the antifungal activity of each identified metabolite and, in this way, know which metabolite is responsible for the antifungal activity. For this, the plant extracts were fractionated at 12-elution times through HPLC, collecting 12 fractions in 60 min. Each fraction was used to perform susceptibility assays against C. albicans, C. dubliniensis, C. glabrata, C. krusei, and C. parapsilosis. Some of these fractions exhibited inhibitory activity against these fungal species, indicating that the fraction must contain some compound with antifungal activity. Figure 5 and Table 2 summarize the results of these assays.

The susceptibility assays with the different fractions obtained through HPLC revealed specific activity against some of the Candida species (Figure 5). To analyze these results, a table was constructed considering all experiments performed in triplicate and independently (Table 2).

For the P. tithymaloides stem extract, fractions 8, 9, and 12 showed antifungal activity against the five Candida species. The T. vulgaris leaf extract presented greater antifungal activity in fractions 1, 3, 5, 6, 9, 10, and 12, whereas only fraction 12 of the stem extracts presented antifungal activity. The O. basilicum leaf extract presented the largest number of antifungal fractions against Candida, these fractions were 1, 5, 6, 7, 8, 9, 10, 11, and 12 (Table 2) during the susceptibility assays. For the stem, fractions 10, 11, and 12 (Table 2) presented susceptibility to the five Candida species. To assess whether the four metabolites identified in the three analyzed plants could be good candidates against Candida, but apparently innocuous to the human cells, the targets of these metabolites were mapped using the SwissTargetPrediction software. The bioinformatics analyses revealed that the probability that any of the proteins of human cells could be an actual target was practically inexistent for the four secondary metabolites (Figure 6).

4. Discussion

In the last years, it has been reported that Candida species represent one of the most important causes of systemic nosocomial infections. C. albicans is the most important opportunistic pathogen; however, the prevalence of other species such as C. glabrata, C. tropicalis, C. parapsilosis, C. krusei, and, recently, C. auris is increasing. The developments in medicine, associated with an increase in invasive procedures, increasingly aggressive immunosuppressing treatments, and the generalized use of wide-spectrum antibiotics, have favored the increase in the selective pressure and the development of antifungal resistance, contributing to the incidence of candidemia and IC. Although a variety of antifungal agents is available and because patients do not respond adequately to treatment, it is necessary to identify new antifungal agents that are safe for the patients, but highly selective against Candida species. The main inconvenience when designing new antifungal compounds is that the only difference between mammalian and fungal cells is the cell wall of Candida and, thus, the target site of excellence, leading to a lesser possibility for the effect of antifungal agents. For this reason, the search and use of new active principles, derived from natural products for alternative therapies, are utterly important. Among these natural products are plants which contain a large variety of biologically active molecules, which can have different target sites and action mechanisms from those of traditional antimicrobials [35, 36]; in fact, the antifungal activity of different crude plant extracts against different microorganisms has been reported.

To identify the metabolites with antifungal activity against C. albicans, C. dubliniensis, C. glabrata, C. krusei, and C. parapsilosis, we chose P. tithymaloides, T. vulgaris, and O. basilicum, which are plants used in traditional medicine to treat different diseases. Crude extracts were prepared with two different solvents: ethanol that is a solvent of polar character and hexane that is a polar. The variety of the solvents allowed obtaining different metabolites from the plants. The ethanol extracts presented higher antifungal activity (Figure 2(a)) against the five studied Candida species, whereas for the hexane extracts, the T. vulgaris leaf extract was the only one with antifungal activity against the Candida species (Figure 2(b)). The anatomical parts, i.e., leaves and stems, were assessed independently to analyze whether there is some difference in the composition and concentration of metabolites. This was done because some of the secondary metabolites in plant exert defense functions against predators and can also inhibit the development of insects [37], fungi [38], and bacteria [39], which determine their preferential location in one or another anatomical site. Results allowed identifying a total of 30 metabolites in P. tithymaloides, T. vulgaris, and O. basilicum in leaves and/or stems. Among these metabolites, monoterpene phenols like thymol, carvacrol, and eugenol were identified in T. vulgaris. However, of the identified metabolites, we focused in those identified in at least two plants. In this way, linalool was identified in three plants, and eugenol, limonene, and borneol were found in T. vulgaris and O. basilicum (Table 1).

Linalool (3,7-dimethylocta-1,6-dien-3-ol) has been reported to have properties as acaricide, bactericide, and fungicide. Its antifungal activity has been tested against diverse microorganisms, such as Fusarium moniliforme at a dose of 1000 ppm, Candida species, and Acinetobacter baumannii [4048]. Our results agree with other reports that have evaluated linalool as an antifungal against Candida species; in strains isolated from individuals with oral candidiasis, the antifungal activity was good against C. tropicalis and moderate against C. albicans [42]. In another study, the fungicide activity of linalool against 39 C. albicans isolates and 9 isolates of Candida non-C. albicans was evaluated and found to be effective against both Candida species [46, 47]. Linalool has also been found in the essential oil of Lavandula angustifolia; in this work, the antifungal activity against C. albicans was shown, as well as its participation in avoiding the morphological transition from yeast to mycelium, because linalool can inhibit the formation of the germ tube and reduce the elongation of hyphae. This finding indicates that linalool, by avoiding dimorphism, can be considered as an effective compound against C. albicans [41]. Besides, by preventing the morphology change in Candida, linalool also avoids the formation of biofilms by C. albicans [43] and C. tropicalis [44]. Biofilms are considered as virulence factors in Candida species, because they induce recurring candidiasis symptoms, leading to fatal outcomes. As a whole, these data show that linalool is a compound with efficient antifungal activity against the studied Candida species, and it is present in different plant species, as shown in this and other studies; hence, it is a good candidate to be used against these fungi. The mechanism by which terpenoids, like linalool, act as antifungal agents has not been elucidated yet; but a model has been described proposing that terpenoids induce membrane fluidization, modulating the functions of the proteins bound to the membrane and involved in signaling and transport [47]. Besides, it has been reported that terpenoids arrest the cell cycle in Candida [47] and other organisms, like Staphylococcus aureus and Escherichia coli [49]. Linalool exhibited an estimated low affinity for distinct families of proteins from Homo sapiens (Figure 6(a)).

Another identified metabolite in T. vulgaris and O. basilicum was eugenol (4-allyl-2-methoxyphenol). This compound is found in a variety of plants, like Syzygium aromaticum (L.) and Myristica fragrans (Houtt); besides, finding eugenol in O. basilicum agrees with other works [50, 51]. Its common name comes from the plant Eugenia caryophyllata (syn. Syzygium aromaticum) [50] where it was found for the first time. This compound, like linalool, has been shown to have antifungal activity against Candida species [47]. The antifungal activity of this compound has been investigated for almost four decades [52]. These authors analyzed the antifungal activity of eugenol in 31 strains of C. albicans and found that it had indeed antifungal activity against this fungus [52]. Several years later, the activity of eugenol against several opportunistic fungi, including C. albicans, was investigated, and authors found that the analyzed strains were inhibited by this metabolite [53]. Aiming at administering eugenol as a treatment against fungi, toxicity studies were performed in mice, but the maximal tolerated dose was of 62.5 mg/kg; therefore, treatment of mycoses with eugenol is not allowed [52]. To be able to use eugenol as an antifungal drug, derivatives of this metabolite have been synthetized, finding that a derivative known as peracetyl glucoside, which is more potent and less cytotoxic than eugenol, inhibited the growth of C. albicans, C. glabrata, and C. tropicalis [54]. In a later work, it was demonstrated that the use of a new derivative of eugenol against these fungi inhibited 90% the growth of C. glabrata [44, 45]. These data together with our findings indicated that the eugenol in the fractions of two of the analyzed plants is possibly responsible for the observed antifungal activity (Table 1). The action mechanism of eugenol in fungi has already been studied, and it is suggested that because eugenol is lipophilic, it can enter the fatty acid chains of the lipid bilayer of the membrane, upsetting its fluidity and permeability [55, 56]. It has also been reported that eugenol inhibits the ATPase activity and that it is capable of producing oxidative stress [57]. The analysis by SwissTargetPrediction revealed a low affinity of eugenol for enzymes from H. sapiens (Figure 6(b)) which would be beneficial for the treatment of candidiasis.

Limonene (1-methyl-4-(1-methylethenyl)-cyclohexene) is another of the compounds identified in T. vulgaris and O. basilicum; this cyclic monoterpene has been identified in a large variety of citrus plants and other plants, i.e., Thapsia villosa, Dyssodia decipiens, Helichrysum italicum [58, 59]. Other reports agree with our results by reporting limonene as a compound of Thymus vulgaris [6062]. Limonene is widely used in the pharmaceutical industry as insecticide and antimicrobial [6367]. Few studies have analyzed the antimicrobial effects of limonene; however, there are some works that evaluated the biological activity of limonene against species of genera like Aspergillus, Trichophyton, and Candida, reporting the antifungal efficiency of this compound [68, 69]. Our results agree with studies that have evaluated limonene as an antifungal against species of Candida, in which it was efficient against clinical isolates of C. albicans strains, with a MIC of 12.5–188.4 μg mL-1 [59]. Another study demonstrated the antifungal activity of limonene on the planktonic growth of 35 clinical isolates and two standard strains of C. albicans, in which the fungicide activity was efficient in all cases at a 20 mM concentration [68]. No specific mechanisms on how limonene induces cellular disturbances have been proposed; however, it is known that limonene exerts an inhibitory effect on the formation of the germ tube in C. albicans, affecting the morphological change from yeast to mycelium [46, 47, 70, 71], which is an important virulence factor for the pathogenicity of this species [72]. The hydrophobicity of limonene apparently facilitates the dissolution of lipids located in the microbial plasmatic membrane, inducing a loss of the membrane’s integrity and, thereby, affecting functions like permeability, signaling, and transport [73, 74]. Besides, it affects respiration and the energetic metabolism by interfering with the ATP synthesis through the inhibition of the respiratory complex and ATPase activities [75, 76]. These data support the capacity of limonene to act as an efficient antifungal against Candida, because it is a component of a large variety of plants, and as shown in this and other studies, it is capable of acting on multiple targets. The aforementioned makes this compound a good candidate to be used against fungi. This compound was identified in this work as one of the metabolites responsible for the antifungal activity of T. vulgaris and O. basilicum. In addition, limonene exhibits a low activity on family A of G protein-coupled receptor, nuclear receptor, and cytochrome P450 (Figure 6(c)).

The fourth metabolite identified in T. vulgaris and O. basilicum was borneol ([(2R)-1,7,7-trimethyl-2-bicyclo[2.2.1]heptanyl] formate). It is an organic compound belonging to the bicyclic monoterpenoid class. Its presence has been reported in more than 260 plants, mainly in those belonging to the Lamiaceae family, including T. vulgaris and O. basilicum species [6062]. These reports agree with our present results (Table 1). Borneol is widely used in the pharmaceutical industry and as an antimicrobial agent [7782]. The fungicide activity reported in other studies on borneol against C. albicans is effective at a MIC of 0.320 mg/mL [83], which places this compound as a potential candidate for the treatment of candidiasis. Like other terpenoid compounds, as linalool and limonene, borneol interrupts the integrity of the microbial plasmatic membrane [46, 47, 73, 74] and affects the yeast to mycelium morphological change in C. albicans [71, 84]. In addition, borneol is found in a large variety of plants, and the analysis in silico indicated a low affinity for proteins such as lyase, phosphatase, nuclear receptor, and family A of G protein-coupled receptor (Figure 6(d)). These observations and results described in this work allowed us to propose thus terpenoid as one of the compounds mainly responsible for the antifungal activity of T. vulgaris and O. basilicum. However, besides borneol, linalool, eugenol, and limonene were identified as additional candidate drugs with anti-Candida activity. These results together with our in silico studies suggest a low percentage of binding of these drugs to human targets, but future studies will be required to fully elucidate their mechanism of action.

5. Conclusions

Linalool, eugenol, limonene, and borneol found in P. tithymaloides, O. basilicum, and T. vulgaris plants, traditionally of everyday use, are effective compounds against Candida species, thus allowing them to be considered as viable alternatives to traditional antifungal agents once their toxicity has been discarded in animal models.

Data Availability

The data used to support the findings of this study are included within this article.

Conflicts of Interest

The authors declare that there were no conflicts of interest with any organization or entity with a financial interest or financial conflict with the material discussed in this work.

Acknowledgments

We thank the technical assistance of QFB Claudia Karina Sánchez-Sánchez from the Departamento de Farmacia, DCNE, Universidad de Guanajuato, for the chromatography analyses. The authors acknowledge Ms. Ingrid Mascher for the English revision of this manuscript. This work was carried out with the financial support granted to Dr. M. Cuéllar-Cruz by Proyecto-Institucional-009/2020 from the Universidad de Guanajuato, Mexico.

Supplementary Materials

Figure S1: representative chromatograms obtained from each extract that were analyzed with the AMDIS software, in which some of the secondary metabolites present in the studied plants could be identified. Table S1: diameters of the inhibition halos of the P. tithymaloides, O. basilicum, and T. vulgaris extracts against the five Candida species at an OD600nm 0.5. Superscript letters indicate significant differences between treatments (Tukey’s test ), . Lowercase letters indicate significant differences on inhibition halos between stem extracts for each Candida species. Uppercase letters indicate significant differences on inhibition halos between leave extracts for each Candida species. EtOH: ethanol; Hex: hexane; NGI: no growth inhibition; SD: standard deviation. Table S2: diameters of the inhibition halos of the P. tithymaloides, O. basilicum, and T. vulgaris extracts against the five Candida species at an OD600nm 0.1. Superscript letters indicate significant differences between treatments (Tukey’s test ), . Lowercase letters indicate significant differences on inhibition halos between stem extracts for each Candida species. Uppercase letters indicate significant differences on inhibition halos between leave extracts for each Candida species. EtOH: ethanol; Hex: hexane; NGI: no growth inhibition; SD: standard deviation. (Supplementary Materials)