Dermatomycoses are common dermatological conditions of the skin, hair and nails caused by superficial fungal infections, affecting more than 20-25% of the people in the world, mainly in tropical and subtropical regions. These diseases alter the individuals’ quality of life and are considered a public health issue (Havlickova et al., 2008; Simonnet et al., 2011). The etiologic agents of dermatomycoses comprise dermatophytes, yeasts and non-dermatophytic filamentous fungi. Epidemiological studies have verified that most patients with superficial fungal infections had dermatophytes or yeast depending on the geographic location (Silva et al., 2014; Kemna and Elewski, 1996). The treatment available for dermatomycoses includes both topical and systemic antifungal drugs, such as the allylamines (terbinafine), triazoles (fluconazole, itraconazole and voriconazole), imidazoles (ketoconazole) and griseofulvin (Meis and Verweij, 2001). The resistance to the available therapeutic agents, for example, in the case of terbinafine, has led to the need for new drugs (Mukherjee et al., 2003). Plant crude extracts have demonstrated antifungal activity in many studies (Kuete et al., 2009; Mbaveng et al., 2008).

Antifungal effects have not been reported for A. othonianum. Therefore, herein, the authors described the antifungal activities of the crude ethanolic extract (EE); the n-hexane (HF), EtOAc (EF), n-BuOH (BF), and hydromethanolic (HMF) fractions; and isolated compounds of interest against the following microorganisms: Candida albicans and Trichophyton rubrun. Additionally, the cytotoxicities of these compounds were also evaluated.

General

¹H- and ¹³C-NMR spectra were recorded in CD₃OD on Bruker 400 and 500 NMR spectrometers using tetramethylsilane (TMS) as an internal standard. A positive-ion mode HRESIMS spectrum was obtained on a Bruker Daltonics HRMS ultrOTOF-Q-ESI-TOF instrument using electrospray ionization. Analytical and preparative HPLC analyses were performed on Shimadzu LC-20AD and LC-6AD systems, with a degasser DGU-20A5, an SPD-M20A series PDA detector or an SPD-20A series UV-VIS detector, a CBM-20A communication bus module, and a Reodyne manual injector or SIL-20A autosampler. A SHIMADZU Shim-pack ODS (particle diameter = 5 μm, 250 × 4.60 mm, and 250 × 20 mm) column equipped with a pre-column were used in the HPLC separations. The MeOH employed was of HPLC grade (J. T. Baker), and ultrapure water was obtained from Millipore Direct-Q UV3 system. Sample preparation was accomplished using silica gel 90 reverse-phase ODS (Fluka, 230-400 mesh).

Plant

The leaves of A. othonianum Rizz. were obtained from the Brazilian Cerrado in the Goiás State in the city of Montes Claros de Goiás (17°48′15.9ʹʹ S and 50°54′19.5ʹʹ W), in September 2012. A voucher specimen (HJ3793) was deposited in the Herbarium Jataiense Germano Guarim Neto of the Universidade Federal de Goiás, Brazil (Herbarium HJ).

Extraction and isolation

The powdered air-dried leaves of A. othonianum (1.17 kg) were extracted with ethanol, and the solvent was eliminated under reduced pressure producing 30.8 g of the crude extract. The crude extract (EE, 24 g) was then resuspended in MeOH:H₂O (2:8, v/v) and partitioned with n-hexane, EtOAc and n-BuOH. The solvent of the obtained phases was removed using a rotary evaporator, yielded 0.2, 1.9, 12.4, and 1.2 g, respectively. The EtOAc fraction (1.9 g) was purified by solid phase extraction with a silica gel 90 reverse-phase ODS chromatographic column using MeOH:H₂O (30:70 v/v, 50:50 v/v) and 100% MeOH as eluents, yielded three fractions. Fraction 3 (1.2 g) yielded compound 1. Fraction 1 was dissolved in MeOH:H₂O (35:65, v/v) and subjected to preparative RP-HPLC purification using MeOH:H₂O:Acetic acid (58:41.9:0.1, v/v/v), UV detection at 254 nm, and a 9.0 mL/min flow rate, which yielded seven fractions. Fractions 3, 4 and 6 gave compounds 2 (140 mg, retention time (t_R)= 3.54 min), 3 (260 mg, t_R= 7.75 min) and 4 (200 mg, t_R= 14.45 min), respectively.

Amentoflavone (1)

¹H-NMR (400 MHz, CD₃OD) δ: 7.83 (d, J=2.3, 1H, H-2ʹ), 7.79 (dd, J=2.3 and 8.6, 1H, H-6ʹ), 7.41 (d, J=8.8, 2H, H-2''' and H-6'''), 7.01 (d, J=8.6, 1H, H-5ʹ), 6.61 (d, J= 8.8, 2H, H-3''' and H-5'''), 6.50 (s, 1H, H-3''), 6.48 (s, 1H, H-3), 6.29 (d, J=2.0,1H, H-8), 6,26 (s, 1H, H-6ʹʹ), and 6.07 (d, J=2.0, 1H, H-6).¹³C-NMR (100 MHz, CD₃OD) δ: 184.2 (C-4), 183.8 (C-4''), 165.9 (C-2), 165.0 (C-7, C-2'' and C-7''), 162.0 (C-5 and C-5''), 161.0 (C-4'''), 159.4 (C-4'), 159.0 (C-9), 156.5 (C-9''), 132.8 (C-2'), 129.3 (C-2''' and C-6'''), 128.9 (C-6'), 123.0 (C-1'), 123.1 (C-1'''), 121.0 (C-3'), 117.4 (C-5'), 116.8 (C-3''' and C-5'''), 105.0 (C-10, C-8'' and C-10''), 104.0 (C-3''), 103.4 (C-3), 100.1 (C-6''), 100.0 (C-6), and 95.1 (C-8). HR-ES-MS m/z 539.0936 (calculated for C₃₀H₁₈O₁₀ [M+H]⁺, 539.0978).

Gallic acid (2)

¹H-NMR (500 MHz, CD₃OD) δ: 6.95 (s, H-2 e H-6, 2H). ¹³C-NMR (100 MHz, CD₃OD) δ: 170.4 (C-7), 146.4 (C-3 and C-5), 139.6 (C-4), 121.9 (C-1) and 110.3 (C-2 and C-6).

Protocatechuic acid (3)

¹H-NMR (500 MHz, CD₃OD) δ: 7.45 (d, J= 2.0, 1H, H-2), 7.44 (dd, J = 8.0 and 2.0 Hz, 1H, H-6) and 6.81 (d, J= 8.0 Hz, 1H, H-5). ¹³C-NMR (100 MHz, CD₃OD) δ: 170.5 (C-7), 151.6 (C-4), 146.1 (C-3), 123.9 (C-6), 123.4 (C-1), 117.8 (C-2), 115.8 (C-5).

Ethyl 3,4,5-trimethoxybenzoate (4)

¹H-NMR (500 MHz, CD₃OD) δ: 6.93 (2H, s, H-2 and H-6), 4.16 (q, J= 7.1 Hz, 2H), 3.38 and 3.10 (OCH₃), 1.24 (t, J= 7.1 Hz, 3H). ¹³C-NMR (100 MHz, CD₃OD) δ: 168.6 (C-7), 146.5 (C-3 and C-5), 139.7 (C-4), 121.7 (C-1), 110.0 (C-2 and C-6), 61.7 (C-1ʹ), 50.0, 50.0 and 49.9 (OCH₃), 14.6 (C-2ʹ).

Chromatographic studies

Chromatographic separations of the samples were carried out on a Phenomenex Gemini C18 (Octadecylsilane) (5 μm, 250 x 4.60 mm) column with a pre-column. For the experiments, the elution conditions were as follows: methanol/water (0.1% acetic acid) gradient from 5 to 100% methanol for 30 min, then elution with 100% methanol for 10 min, oven at 40°C, and flow = 1.0 mL/min. The analysis also includes 3 min to regress to the initial conditions and 15 min of equilibration. The chromatogram wavelength was set at 254 nm, and UV data were recorded between 190 and 500 nm. The 1-mg samples (EE, EF, BF and 1-4) were weighed and dissolved in 1 mL of methanol. The sample solutions were filtered and then transferred to a vial.

Antifungal assay

The following microorganisms were employed for evaluation of the antifungal activity: Candida albicans (ATCC 64548) and Trichophyton rubrum (Tr1, earlier identified by Profa. Dra. Ana Marisa Fusco Almeida). The T. rubrum strain was preserved and cultured in Petri dishes having Sabouraud agar. It was incubated and subcultured for 7 days at 28°C. The strain was suspended in 0.85% saline solution, counted in a Neubauer chamber and final concentration of 5.0 x 10³ CFU/mL was achieved by dilution in RPMI medium (Ghannoum et al., 2006). The strain C. albicans was cultivated in RPMI-1640. The inoculum was adjusted to 0.5 McFarland scale to yield a cell concentration of 1×10⁶-5×10⁶ yeast/mL (CLSI, 2008a). The extracts, fractions and compounds 1-4 were resuspended in dimethyl sulfoxide (DMSO) to concentrations of 4 mg/mL and subsequently diluted in RPMI medium with L-glutamine (pH 7.2) with 0.165 mol/L morpholine propane sulfonic acid (MOPS) complemented with 2% glucose. Amphotericin B (Sigma Chemical Co. Saint-Louis, USA) was prepared in DMSO and tested in the concentrations of 0.00625 to 32 μg/mL. Minimum inhibitory concentration (MIC) values for each sample were obtained in triplicate using document M-38 A2 adapted to 96-well microplates for T. rubrum (CLSI, 2008b). Each well contained 100 μL of a twofold serially diluted sample and 100 μL of RPMI medium, with 100 μL being transferred to the next well, sequentially. Then, 100 μL of the T. rubrum inoculum was added to obtain a final sample volume of 200 μL with concentrations ranging from 1250 to 2.44 μg/mL. For amphotericin B, 100 μL of the twofold sample dilution was added to 100 μL of the inoculum. The microplates were incubated in an orbital shaker at 120 rpm and 28°C for 7 days. The MIC was determined as the minimum concentration of the tested sample for which no growth was visualized, and then 30 μL of a 0.01% resazurin aqueous solution (Sigma-Aldrich) was added to determine the microorganism viability. The minimum inhibitory concentration (MIC) values against C. albicans for each sample were determined in triplicate using the reference method M-27 A3 (CLSI, 2008a) with modifications. Each well contained 100 μL of a twofold serially diluted sample and 100 μL of RPMI medium, with 100 μL being transferred to the next well, consecutively. Then, 100 μL of the C. albicans inoculum was added to obtain a final sample volume of 200 μL with concentrations ranging from 1250 to 2.44 μg/mL. For amphotericin B, 100 μL of the twofold drug dilution was added with 100 μL of the microorganism. The orbital shaker at 120 rpm and 35°C for 2 days was used to incubate the microplates. Then, the viability of C. albicans was determined by the addition of a 2.0 % 2,3,5-triphenyltetrazolium chloride aqueous solution (Sigma-Aldrich). The MIC was characterized as the smallest concentration of the tested sample for which no development was observed.

Cytotoxicity assay

The cytotoxicity was measured using the in vitro Toxicology Colorimetric Assay Kit (XTT; Roche Diagnostics) according to manufacturer´s instructions, using normal human lung fibroblasts (GM07492A), and as previously described (Alvarenga et al., 2015).

Statistical analysis

Inhibitory concentration at 50% cell growth inhibition (IC₅₀) was obtained with the GraphPad Prism 5 software. One-way ANOVA was used for the comparison of means (p Ë‚ 0.05).

Regarding the antifungal activities of the compounds, only 1 displayed significant inhibitory effects against C. albicans, with a MIC value of 19.53 μg/mL. Compounds 2 and 3 presented significant inhibitory effects against T. rubrum TR1 (MIC values of 9.76 and 39.06 μg/mL, respectively). Again, no cytotoxicity was observed for the isolated compounds (IC₅₀ ˃ 400 μM). However, the isolated compounds displayed weaker antifungal activities when compared with amphotericin B, which was used as a positive control (Table 1). Additionally, compounds 1-4 have previously been shown to possess antifungal activities (Mbaveng et al., 2008; Leal et al., 2009; Kuete et al., 2009; Soares et al., 2014). Flavonoids have been found to be active against a wide range of microorganisms; their mechanism of action is possibly due to their capacity to complex with proteins and with bacterial cell walls (Cowan, 1999).

The structures of gallic acid (2) and ethyl 3,4,5-trimethoxybenzoate (4) were quite related, differing in the occurrence of three free hydroxyl substituents and an acidic moiety in 4 versus 3 methoxy groups and an ethyl acetate group in 4. The presence of three methoxyl groups in the latter compound might cause a decrease in the activity, suggesting that the three free OH groups could play a role in the antifungal activity. This finding is in accordance with the data obtained by Leal et al. (2009). Additionally, the mechanism of action of phenolic compounds comprises enzyme inhibition (Cowan, 1999).

Protocatechuic acid (3) and gallic acid (2) also showed related chemical structures, diverging only in the occurrence of two hydroxyl moieties in 3, which decreased the activity. Once more, these results suggested that the presence of the three free OH groups appears to be necessary for antifungal activity.

Regarding the chemical composition of the bioactive extract (EE) and fractions (EF and BF) (Figure 2a-c), an HPLC-DAD analysis was performed using a C18 column. The chromatograms were recorded using a methanol/water (0.1% acetic acid) gradient from 5 to 100% methanol over 30 min, then elution with 100% methanol for 10 min. These conditions were similarly employed to evaluate the presence of compounds 1-4 in the samples, based on an evaluation of retention times (λ = 254 nm) and UV spectra from a DAD detector of the compounds previously isolated from EF (Figure 2d).

Compounds 1-4 were recognized in different proportions (Table 2) in the ethanol extract from the leaves of A. othonianum (EE) and the EtOAc fraction (EF) (Figure 2a and b). Additionally, compound 4 was not detected in the n-BuOH fraction (BF) (Figure 2c). Conversely, the amentoflavone (1) was always present at a higher concentration relative to 2, 3 and 4. Based on the data obtained from the chromatogram and MIC values of EF, BF and compounds 1-3, when compounds 1-3 were combined in the fractions, the antifungal activity was improved, at least in the case of C. albicans.

Thus, the data obtained suggest that the antifungal activity of A. othonianum may be mainly attributed to the effects of amentoflavone (1), gallic acid (2) and protocatechuic acid (3). Frequently plant extracts showed better activity than the isolated compounds, this situation can be explained by synergy effects (Wagner and Ulrich-Merzenich, 2009). Although, in this study the isolated compounds displayed increased antifungal activity, when compared with the crude extract (EE). 

To the best of the authors’ knowledge, the antifungal activity of A. othonianum and the occurrence of compounds 1-4 are reported for the first time in this work. Though, the antifungal activities of compounds 1-4 have been documented previously. In conclusion, the antimicrobial activity of the crude extract of the leaves of A. othonianum may be due to the occurrence of compounds 1-3. The evaluated extract, together with fractions EtOAc (EF) and n-BuOH (BF) and compounds 1-3, could be beneficial for the research of new antifungal medications. Nevertheless, the toxicological and pharmacological studies of the analyzed samples will check this proposition.

The authors have not declared any conflict of interests.

The authors are grateful to the São Paulo Research Foundation (FAPESP) [grant numbers 2011/00631-5; 2012/18060-7; 2013/09280-6]; Coordenadoria de Aperfeiçoamento de Pessoal do Ensino Superior (CAPES); and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for fellowships.