All reagents, pulegone, thymol, isomenthone, amphotericin B, itraconazole and fluconazole were purchased from Sigma-Aldrich Co. (St. Louis, MO, United States), GIBCO BRL Life Technologies (Grand Island, NY, United States), and Santa Cruz Biotechnology (Santa Cruz, CA, United States).

C. chilense was collected from Viña del Mar, Valparaiso Region, Central Chile (S: 33°02′38″, W: 71°30′04″) during the spring in October 2020. Mr. Patricio Novoa confirmed species authenticity, and a voucher specimen (Cch-1020) was deposited at the Natural Products and Organic Synthesis Laboratory of Universidad de Playa Ancha, Valparaíso, Chile.

The EO was extracted from the aerial parts (leaves and branches) of C. chilense (500 g) by hydrodistillation carried out using a Clevenger-type apparatus for 4 h. The EO was collected and dried over anhydrous sodium sulfate and stored in sealed brown vials at 4°C. Subsequently, C. chilense EO was diluted with dichloromethane, and 1 μL of the sample was analyzed using a GC-MS/MS (Hewlett-Packard GC/MS 6890 coupled to a Hewlett-Packard 5973 mass-selective detector) operating in EI mode at 70 eV, equipped with a splitless injector (250°C). The transfer line temperature was 200°C. Helium was used as a carrier gas at a flow rate of 1.2 mL/min, and the capillary column used was a HP-5 ms (60 m × 0.25 mm i.d., film thickness 0.25 μm). The temperature program was 40°C (5 min) to 300°C (5 min) at a rate of 5°C/min. Volatile compounds were identified by comparing their mass spectra, obtained using the Thermo Xcalibur 3.1.66.10 program, with the NIST 2020 library database, and by comparison of their retention index with those reported in the literature. The retention indices were determined under the same operating conditions in relation to a homologous n-alkanes series (C₈–C₃₆).

The data were reported as the mean values ±standard deviation (SD). Due to non-parametric data, a Kruskal–Wallis ANOVA was used with a confidence level of 95% with the STATISTICA 7.0 program.

The EO of C. chilense was obtained with a yield of 1.9% (v/w). The EO of C. chilense fresh leaves is composed mainly by oxygenated monoterpenes (44.6%), followed by oxygenated sesquiterpenes (20.3%), monoterpene esters (11.7%), phenols (11.0%) and hydrocarbon sesquiterpenes (8.6%) (Table 1).

Twenty-four compounds were identified in the EO of C. chilense, which corresponded to 96.2% of the total oil analyzed, and the main components were pulegone (18.5%), thymol (11.0%), and isomenthone (10.0%) (Figure 1).

This is the first study on C. chilense EO, a plant often confused with Clinopodium gilliesii due to shared scientific nomenclature and similar uses, especially in culinary applications. However, a crucial difference lies in the concentration of pulegone, the major compound in both. While C. gilliesii exhibits a pulegone content of 93.8% (Carvajal et al., 2017), C. chilense contains only 18.5%, suggesting a lower risk associated its consumption of the latter. This difference is critical because C. chilense is often confused with C. gilliesii due to similar nomenclature and uses, especially in culinary applications. The lower pulegone concentration, coupled with the presence of thymol and isomenthone (monoterpenes also found in edible plants like thyme and peppermint), suggests a more favorable safety profile for C. chilense (de Sousa Barros et al., 2015; Salehi et al., 2018). In contrast, the essential oil of Clinopodium nubigenum from Ecuador, traditionally used to treat colds and stomach aches, exhibited a high pulegone content (72.8%), with 3,7-dimethyl-1,6-octadien-3-ol (7.0%) and cis-isopulegone (4.7%) as secondary components, and demonstrated antimicrobial activity against C. albicans (Gilardoni et al., 2011). On the other hand, analysis of four wild Clinopodium species, traditionally used in folk medicine from countries of the former Yugoslavia, revealed a predominance of oxygenated monoterpenes: C. thymifolium (35.2% pulegone), C. serpyllifolium (13.0% pulegone, 38.7% piperitenone oxide), C. dalmaticum (27.3% pulegone, 41.7% piperitenone), and C. pulegium (22.9% pulegone, 34.1% piperitenone oxide) (Dunkić et al., 2017). It is important to note that not all Clinopodium species contain pulegone as the primary component. The essential oil of Clinopodium gracile collected in China lacks pulegone in its composition, instead exhibiting germacrene D (20.6%), nootkatone (8.2%), and morillol (7.7%) as major compounds, yet it still demonstrated larvicidal activity against Aedes albopictus (Chen et al., 2013). These variations emphasize the necessity of carefully characterizing each species and its geographical origin to determine their potential uses and safety.

The antifungal activity (MIC and MFC) of EO and their main compounds is presented in Table 2.

Table 2 summarizes the activity observed at 48 h for C. chilense EO and its major compounds, specifically those found in abundances exceeding 10%. In terms of results, C. lusitaniae proved to be the most sensitive strain to the different treatments, with Candida tropicalis following closely behind. In contrast, Candida glabrata proved to be the least sensitive species to the various treatments. This observation is consistent with previous research, including a study by Deravi et al. (2021) which also identified C. glabrata as a less susceptible species in clinical isolates of vulvovaginal candidiasis.

Analyzing the relationship between the MFC (Minimum Fungicide Concentration) and MIC (Minimum Inhibitory Concentration) values is essential for a thorough understanding of the mode of action of each treatment. The ratio MFC/MIC was calculated in order to determine if the essential oil or its major compounds had a fungistatic (MFC/MIC ≥4) or fungicidal (MFC/MIC ≤4) activity (Boubrik et al., 2025). In this regard, the C. chilense EO demonstrated broad antifungal activity against all tested Candida spp., with MIC and MFC values ranging from 16 to 128 μg/mL. Notably, the oil exhibited fungicidal activity (MFC/MIC = 1) against C. guilliermondii, C. lusitaniae, and C. tropicalis, with particularly potent activity observed against C. tropicalis at the lowest tested concentration of 16 μg/mL. All compounds exhibited antifungal activity, but the C. chilense EO demonstrated a remarkable ability to inhibit the growth of all strains at significantly lower concentrations compared to most individual compounds. Studies indicate that pulegone, one of the components of the EO, can inhibit the growth of various Candida strains, including C. albicans. Some research also demonstrates that pulegone can affect biofilm formation, a mechanism that helps Candida spp. resist antifungal drugs (Alam and Ahmad Khan, 2021). These results suggest an antifungal potential of pulegone against the analyzed yeast species. Notably, it exhibited a favorable MFC/MIC ratio of 1 against C. lusitaniae at a concentration of 64 μg/mL. However, for the remaining species, the MFC/MIC ratio was 2, requiring higher concentrations to achieve fungicidal effects. Concerning thymol, while it exhibited lower MIC values than pulegone, it demonstrated a superior fungicidal potential with an MFC/MIC ratio of 1 against four strains (C. albicans (ATCC 7978), C. lusitaniae, C. albicans (10,935), and C. glabrata). Notably, this group included C. albicans ATCC and a clinical isolate, with its potency being greater against the clinical isolate at concentrations two times lower than those required for the ATCC strain. For the remaining strains tested, the MFC/MIC ratio was 2. This phenol is well-known for its ability to disrupt the cell membrane and combat biofilm formation (Wang et al., 2017). However, at high concentrations or with prolonged exposure, thymol can be irritating to the skin and mucous membranes (Gavliakova et al., 2013). On the other hand, isomenthone generally inhibited all tested strains at a concentration of 128 μg/mL, exhibiting a fungicidal MFC/MIC ratio of 1, indicating its fungicidal potency at that concentration. The exception was C. lusitaniae, which, similarly to pulegone, was inhibited at a lower concentration of 64 μg/mL. This terpenketone along with other compounds such as menthone, has been shown to inhibit the growth of various fungi. It has been specifically reported that this type of monoterpene exhibits activity against Aspergillus niger (Moghtader, 2013). These findings are consistent with the known mechanism of action of terpenoids, the main components of the essential oil, which destabilize cell membranes, increase their permeability, and consequently disrupt and destroy microorganisms (Gonçalves et al., 2007). Further highlighting the potential of this EO, while amphotericin B proved more effective against ATCC strains, the EO outperformed it against all clinical isolates. While the EO exhibited higher MIC values than the evaluated azoles, it demonstrated superior fungicidal potential, with an MFC/MIC ratio of 1, compared to the value of 2 observed for both azoles across all strains. This MFC/MIC value aligns with that reported by Costa et al. (2017) for C. albicans and C. parapsilopsis. Although fluconazole is prototypically considered a fungistatic agent, our study demonstrated fungicidal activity against diverse Candida spp., with an observed MFC/MIC ratio of 2. This finding, far from being anomalous, is consistent with the growing body of evidence that azole activity is conditional. Our results align with those of Maheronnaghsh et al. (2016), who also reported low MFC/MIC ratios in isolates from cancer patients. Mechanistically, it has been shown that fluconazole, at the high concentrations used to determine the MFC, induces a programmed cell death (apoptosis) cascade in C. albicans (Lee and Lee, 2018), the most prevalent species. Furthermore, the fungicidal activity of this azole is not only dose-dependent but also influenced by the microenvironment. For example, Moosa et al. (2004) demonstrated in vitro that fluconazole becomes fungicidal under the acidic conditions that simulate the vaginal environment. This conditional activity has clear clinical relevance, as it has been validated in vivo by Sobel et al. (2003), who documented the potent efficacy of fluconazole in treating C. albicans vaginitis (Akins, 2005). Therefore, in the face of the conditional efficacy of conventional drugs, these results suggest a promising future for C. chilense EO as a potential alternative against pathogenic yeasts, thus opening new phytopharmacological possibilities.

A molecular docking study assessed the major compounds of C. chilense EO against two key receptors involved in Candida spp. pathogenicity; the results are presented in Table 3.

The interactions detailed in Table 3 were assessed against key receptors critical for Candida spp. virulence. One such receptor, Als9-2, is an adhesin of C. albicans that facilitates fungal attachment to host cells by interacting with fibrinogen. This interaction contributes to biofilm formation and virulence in mucosal, systemic, and device-related infections. Sterol 14-alpha demethylase (CYP51) is a key enzyme in ergosterol biosynthesis, essential for fungal cell membrane integrity. Inhibition of CYP51 disrupts the membrane, enhancing the effectiveness of antifungal treatments. Both represent important therapeutic targets against C. albicans (Lim et al., 2025; Mesileya et al., 2025).

DMSO was used as the solvent control in the molecular docking experiments. As expected, no relevant binding affinities or specific interactions were observed. Although recent evidence indicates that DMSO may influence certain physical or kinetic properties—such as medium viscosity or protein stability—under specific conditions, it does not directly interact with the active site nor act as a competitive ligand (Tunçer and Gurbanov, 2023; Feoli et al., 2024).

The lowest binding energies were obtained for the compounds used as positive controls—amphotericin B, fluconazole, and itraconazole—with values of −8.95, −7.97, and −9.96 kcal/mol, respectively, for protein 2Y7L, and −10.38, −6.90, and −10.51 kcal/mol, respectively, for protein 5TZ1. These results indicate that the major compounds of C. chilense EO have binding affinity for two key proteins associated with the pathogenicity of C. albicans: the Als9-2 adhesin (PDB ID: 2Y7L) and the sterol 14α-demethylase enzyme (CYP51, PDB ID: 5TZ1). As expected, the reference compounds exhibited the best binding energies at both receptors, validating the docking methodology used.

Among the natural compounds evaluated, thymol (compound 2) showed the strongest binding affinities toward both proteins, outperforming pulegone (1) and isomenthone (3). This greater affinity suggests that thymol may significantly contribute to the antifungal activity of the essential oil. Several studies have demonstrated that thymol possesses potent antifungal activity against Candida albicans, including fluconazole-resistant strains. For instance, thymol has been shown to inhibit biofilm formation and reduce fungal cell viability. Moreover, when combined with conventional antifungals such as fluconazole, thymol exhibits synergistic effects that enhance treatment efficacy (Jafri and Ahmad, 2020).

Studies on Cryptococcus neoformans, another human fungal pathogen, have investigated the mode of action of thymol. Results indicate that thymol disrupts ergosterol biosynthesis—an essential component of the fungal cell membrane—by modulating the expression of genes involved in this pathway, such as ERG1, ERG11, and HMG1. Additionally, thymol induces intracellular calcium homeostasis imbalance and interferes with protein glycosylation, processes that are critical for fungal cell viability (Jung et al., 2021).

Pulegone and isomenthone also showed interactions with both molecular targets, but with higher (i.e., less favorable) binding energies, suggesting lower affinity and, likely, a reduced contribution to the overall antifungal activity. A study by Miri et al. (2022) reported that the essential oil of Mentha pulegium, containing 74.81% pulegone, exhibited significant antimicrobial activity against several microorganisms, including C. albicans. These findings suggest that pulegone may be responsible for the observed antifungal activity, although its precise mechanisms of action remain unknown.

The compounds tested exhibited binding energies against 2Y7L ranging from −6.38 kcal/mol (pulegone, compound 1) to −6.98 kcal/mol (thymol, compound 2). For the 5TZ1 protein, binding energies ranged from −6.62 kcal/mol (isomenthone, compound 3) to −7.86 kcal/mol (thymol, compound 2). In both cases, thymol demonstrated the strongest binding affinities. Among the three compounds evaluated against different Candida strains, compound 2 (thymol) showed the highest antifungal activity. Molecular docking studies suggest that this compound binds efficiently to both target proteins (Figure 2).

In the case of the Als9-2 adhesin (PDB ID: 2Y7L), thymol formed a hydrogen bond with residue Glu27 at a distance of 1.87 Å, along with multiple alkyl interactions involving residues Val22, Arg171, Pro29, and Tyr297. These interactions suggest a stable binding to the N-terminal domain of the adhesin, potentially interfering with C. albicans adhesion mechanisms to host cells.

Regarding the sterol 14α-demethylase (CYP51, PDB ID: 5TZ1), thymol also exhibited a favorable interaction pattern, forming two conventional hydrogen bonds with residue Ser378 (at 2.00 and 2.26 Å), a T-shaped π–π interaction with Phe380 (at 5.08 Å), and additional alkyl interactions with residues Leu376, Pro230, and Phe233. These findings indicate that thymol may interfere with ergosterol biosynthesis, a key function targeted by azole antifungal agents.

The analysis of the physicochemical and toxicological parameters of the major compounds found in the essential oil of Clinopodium chilense is presented in Tables 4, 5. These data support the evaluation of their viability as candidates with pharmacological potential.

The three compounds analyzed exhibited similar Log P values, indicating comparable lipophilicity profiles. This property favors permeability across biological membranes without compromising aqueous solubility (Chloe, 2024). Moreover, a high predicted gastrointestinal absorption and the ability to cross the blood–brain barrier was observed, suggesting good oral bioavailability and the potential to exert effects on the central nervous system, depending on the therapeutic application.

None of the compounds showed affinity for P-glycoproteins, thus reducing the risk of active efflux and enhancing their potential for intracellular accumulation. The skin permeability coefficient (Kp) was similar for all three compounds, with values ranging from −4.87 to −5.09 cm/s, suggesting limited dermal absorption—an important consideration for the development of topical formulations.

Regarding toxicological parameters, all compounds showed approximately 30% inhibition of the hERG channel, indicating a moderate but not alarming cardiotoxicity risk. Ames (mutagenicity) and DILI (drug-induced liver injury) tests yielded values below 50%, except for isomenthone, which showed an Ames result of 52.88%. This finding may warrant further evaluation in future studies. Additionally, the lethal dose 50 (LD₅₀) values indicated relatively low acute toxicity, ranging from 1.84 mol/kg for pulegone to 2.01 mol/kg for isomenthone.

Although pulegone has been identified as a potential carcinogen in some toxicological studies involving prolonged high-dose exposure in rodents (Jabba and Jordt, 2019; Voigt, et al., 2024), these findings do not necessarily preclude its use in therapeutic contexts. The carcinogenic effects observed are highly dose-dependent and were associated with sustained oral intake, not short-term or localized administration. In our case, pulegone is considered for antifungal use, which could involve topical or controlled-release formulations that minimize systemic absorption and toxicological risks. Furthermore, many clinically approved drugs carry similar concerns but remain in use when the therapeutic benefits outweigh potential adverse effects. As such, the potential of pulegone as an antifungal agent remains valid, though further preclinical toxicological assessment will be necessary to ensure its safe medical application.

Finally, it is worth noting that all three compounds meet the criteria set by Lipinski’s Rule of Five, supporting their potential as bioactive molecules with pharmacokinetic properties suitable for oral administration. To validate their effectiveness at the molecular level, molecular dynamics simulations were performed, providing information on the dynamic behavior of the complexes, as well as details at the molecular level during the simulation period. The first parameter to analyze is the RMSD, which provides a criterion for stability throughout the simulation period. As can be seen in Figure 3, all the complexes studied stabilized after 2 ns of simulation time. It is important to note that this parameter remained low throughout the 100 ns of the trajectory, with RMSD values below 2 Å, indicating high stability of the ligands in the active sites of both proteins. This suggests that the compounds could be good inhibitors of Als9-2 and CYP51 proteins.

From the graph shown in Figure 3, we can see that the lowest RMSD values were found in the complexes formed by the Als9-2 protein with pulegone and thymol, with values of 1.2663 ± 0.1628 Å and 1.2656 ± 0.1843 Å, respectively. These values indicate no significant differences between these complexes and the others, so it is necessary to analyze other variables obtained from the trajectory.

Hydrogen bonding interactions between the ligand and the protein confer stability to the complex over time. In our case, and as seen in Figure 4, this parameter’s behavior was weak during the 100 ns of simulation time.

The largest number of hydrogen bonding interactions (H-Bond) were found in the complexes formed by Isomenthone-CYP51 and thymol-Asl9-2. It is important to note that the standard deviation of this parameter was much higher than the mean for all simulated systems, indicating that this type of interaction was not stable over time. Based on the results obtained for this parameter, we can conclude that hydrogen bonding interactions do not explain the stability of the complexes studied.

To corroborate the behavior of the hydrogen bond interactions obtained during the simulation, we will analyze the interaction occupancy term. This term calculates the percentage of hydrogen bond interactions that remained below 3 Å during the simulation. This parameter is shown in Table 6.

For a hydrogen bond interaction to be considered stable during the trajectory, the occupancy percentage should be above 50% (Tsujimura et al., 2025). As can be seen in Table 6, in none of the complexes did the occupancy percentage exceed 15%. This corroborates the previously stated finding that these interactions were not stable during the 100 ns of simulation time, which does not explain the stability found in the RMSD parameter. Therefore, it is necessary to analyze other parameters such as radius of gyration (Rg) (Figure 5).

The variable radius of gyration is defined as the root mean square distance of the mass center of atom or collection atoms from a common mass centers and shows us the degree of compaction that a system presents during the trajectory (Lobanov et al., 2008).

As we can see in Figure 5, the lowest values for this parameter were found in the complexes formed by isomenthone and pulegone with CYP51, indicating greater compaction of these systems and, consequently, greater stability. The complex formed by isomenthone-CYP51 had the highest number of hydrogen bonds calculated and the highest occupancy, indicating that this compound could be a good inhibitor of the CYP51 enzyme.

Due to the side effects and resistance that can arise from synthetic drugs used to treat Candida infections, both cutaneous and systemic, the search for alternatives in natural products, such as essential oils, has intensified for effective therapies against candidiasis or candidemia. In this context, it is crucial to know the cytotoxic effect that these products may have on healthy cells, in order to ensure their safe application in therapies for these pathologies. Consequently, in this study, the cytotoxic effect of C. chilensis EO on two non-tumorigenic human cell lines was investigated using an SRB assay (Table 7).

In contrast to the chemotherapeutic agent 5-fluorouracil (5-FU), the C. chilense EO exhibited low cytotoxicity against the non-tumorigenic cell line, as indicated by its IC₅₀ value. This finding suggests a favorable safety profile for the EO, indicating a minimal impact on normal cells. This finding aligns with observations from other essential oils containing similar major components. As an example, a commercial Mentha piperita EO (commonly known as peppermint, a sterile hybrid of watermint and spearmint, widely cultivated for its menthol flavor) containing approximately 29.90% isomenthone, exhibited very low toxicity in human epidermal keratinocyte (HaCaT) cells up to a concentration of 400 μg/mL (Silva et al., 2022). Similarly, Thymus caramanicus Jalas (commonly called Iranian thyme, a thyme species endemic to Iran whose leaves have been used traditionally in Iranian medicine to treat various conditions and whose extracts have demonstrated antibacterial, neuroprotective, antinociceptive, and anticancer properties), its EO with 20.84% thymol, was non-toxic to the normal gingival HGF1-PI1 cell line but highly cytotoxic to KB (oral carcinoma) cells at low concentrations (Fekrazad et al., 2017). Furthermore, a study on Minthostachys verticillata EO (a wild plant from Argentina and Paraguay, commonly known as peperina or peppermint, primarily consumed as an infusion and to flavor mate and tereré, and used for medicinal purposes as a digestive, antispasmodic, and anti-inflammatory), which contains 60.0% pulegone, demonstrated no in vitro cytotoxicity or in vivo cytogenotoxicity at both low and high concentrations (Escobar et al., 2012). Notably, Minthostachys mollis (a wild plant commonly known as muña, peperina, poleo de Quito, tifo or tipo is a species of woody shrub, native to Peru, Bolivia, Colombia, Ecuador, Argentina and Venezuela), whose essential oil possesses the same three major components as the C. chilense EO and a similar pulegone concentration (12.4%, with 1.0% thymol and isomenthone), showed minimal cytotoxic effects on healthy HEK.230T cells at 687.60 ± 33.50 ug/mL (Benites et al., 2018). These observations support the existing literature, which indicates that thymol, pulegone, and isomenthone are effective inhibitors of cancer cells with a relative cytotoxic activity in normal cells (Da Rocha et al., 2012; Jung et al., 2012; Ezatpour et al., 2023). Moreover, numerous studies suggest that minor components can play a significant role in antibacterial activity, potentially through synergistic effects with other constituents (Miladinović et al., 2021). This synergistic effect is likely also relevant to cytotoxic activities in cells. Therefore, further investigation is warranted to evaluate the cytotoxic potential of these minor components, both individually and in various combinations, to fully elucidate their role(s) in toxicity against both tumor and non-tumor cells. Based on the results obtained and the consulted literature, C. chilense EO shows promising potential as a therapeutic agent. While the toxicity of its main components, such as pulegone, is acknowledged, the levels present in the oil would not reach the intake doses considered risky (Voigt et al., 2024), suggesting a negligible or very low health risk. Nevertheless, given that its application is envisioned in a dermatological solution and considering its apparent lack of cytotoxicity in non-tumor cells, the importance of a careful evaluation of the dosage and of further research to identify potential synergistic or antagonistic effects before its therapeutic use is emphasized.