Introduction

Candida species are mild commensals in humans, but opportunistic pathogens that can cause superficial and systemic infections in severely ill or immunocompromised patients [22]. Although more than 100 species of Candida have been described, most of all Candida-associated nosocomial bloodstream infections are associated with Candida albicans, Candida glabrata, Candida parapsilosis, Candida tropicalis, and Candida krusei [22,23], and C. albicans is known as a primary fungal pathogen [14].

Regardless of the extensive efforts to develop new antifungal agents, currently available drugs are very restricted for the treatment of Candida infections. There are four molecular groups targeting three different metabolic pathways that are available for the treatment of systemic fungal infections, including fluoropyrimidine analogs, polyenes, azoles, and echinocandins [29]. Polyene drugs such as amphotericin B target ergosterol. Ergosterol is the analog of cholesterol found in fungal cell membrane, which plays a role in membrane integrity, membrane fluidity, and the function of many membrane-bound enzymes [10,32]. Azole drugs are involved in the ergosterol biosynthetic pathway by inhibiting lanosterol 14α-demethylase [3], and echinocandins target the biosynthesis of (1-3)-β-glucan, which is a major component of the cell wall in fungi [15]. Along with the emergence of resistant yeasts, approach to Candida infections is problematic due to the eukaryotic characteristics of fungi and the close evolutionary relationship between fungal cells and their human hosts, reducing the number of drug targets that can be applied to the pathogen selectively [26]. Thus, it is essential to develop new antifungal agents with good therapeutic effect as well as low toxicity. Plant products have been used traditionally in ethnomedicine as effective antifungals, and are regarded as a part of the defense mechanism in higher plants [6]. Salvia miltiorrhiza Bunge, belonging to Lamiaceae, is a perennial plant that grows in China and East Asia. Roots of the plant have been used in ethnomedicine, and hold lipophilic tanshinone I, tanshinone IIA, cryptotanshinone, and dihydrotanshinone as well as the hydrophilic danshensu and salvianolic acid B as the active constituents [1,16]. S. miltiorrhiza has been used in traditional medicines owing to its pharmacological properties, containing anti-hypertension, anti-platelet aggregation, anti-atherosclerosis, antitumor, anti-oxidative, antimicrobial, and antifungal activities [11,12,17,28,33]

This study was aimed to determine the minimal inhibitory concentration (MIC) of the S. miltiorrhiza ethanol extract against medically important Candida species, including C. albicans, C. krusei, C. glabrata, and C. tropicalis, and evaluate the possible mechanism of action against C. albicans, which is a major fungal pathogen, based on the results of propidium iodide (PI) uptake, release of A260-absorbing materials and protein, changes in DPH binding, increased MIC in the presence of the osmoprotectant, and decreased (1,3)-β-D-glucan synthase activity. The results offer valuable insight into the pharmacological utility of S. miltiorrhiza as a promising antifungal agent against C. albicans infections.

Materials and Methods

Preparation of Ethanol Extract

Dried roots of S. miltiorrhiza were obtained from jchanbang.com, Korea. Dried S. miltiorrhiza roots (30 g) were infused in 300 ml of 70% ethanol for 1 h and boiled for 2 h. The supernatant obtained by centrifugation of the suspension at 2,000 ×g for 20 min was concentrated using a vacuum evaporator (Eyela, Japan) and lyophilized to yield an ethanol extract. The ethanol extract of S. miltiorrhiza (5.1 g) was dissolved in dimethyl sulfoxide (DMSO) to 100 mg/ml, and kept at -20℃ until used (hereafter referred to as S. miltiorrhiza). Positive-control drug amphotericin B was dissolved in distilled water at the concentration of 16 mg/ml, filter-sterilized, and further diluted in broth just before the assay.

Candida Strains

C. albicans (ATCC 18804, KCCM 50235), C. krusei (ATCC 32196, KCCM 11426), C. glabrata (ATCC 2001, KCCM 50044), and C. tropicalis (ATCC 750, KCCM 50075) were purchased from the Korean Culture Center of Microorganisms (KCCM).

Antifungal Susceptibility Test

The in vitro minimum inhibitory concentrations of S. miltiorrhiza against Candida spp. were determined by the modified CLSI M27-A3 protocol of the colorimetric broth microdilution method in the presence of resazurin as a cell growth indicator, according to Liu et al. [18]. Growth and sterility controls in the presence of DMSO used in sample preparation were also included, and no inhibitory effects were observed in the presence of the solvent control at the highest concentration used (1% (v/v)). All assays were repeated at least three times.

Effect on Cytoplasmic Membrane

C. albicans ATCC 18804 cells at the exponential phase (1 × 106 cells/ml) were incubated with S. miltiorrhiza, at a final concentration equivalent to its MIC, at 35℃ with shaking at 200 rpm for 3 h. The suspension was washed with phosphate-buffered saline (PBS, pH 7.4), and PI (10 μM) was added. After the incubation of the cell suspension for 30 min in the dark at room temperature, the effect of S. miltiorrhiza on the cytoplasmic membrane of C. albicans was evaluated using a confocal laser microscope (Olympus).

Loss of 260-nm-Absorbing Materials and Proteins

In order to investigate the antifungal effect of S. miltiorrhiza on the integrity of the C. albicans cell membrane, the release of 260-nm-absorbing materials and proteins was determined spectrophotometrically using the method of Kahn et al. [13] with slight modifications. Briefly, C. albicans cells (1 × 109 cells/ml) at the exponential phase were washed twice and dissolved in 0.85% NaCl. The suspension was treated with 4× MIC S. miltiorrhiza or the same volume of DMSO (control) as S. miltiorrhiza. At each time point (0, 30, 60, and 90 min), 0.5 ml of cell suspension was taken and filtered through a spin column containing a 0.22 μm filter (Costar, USA), by centrifugation at 12,000 ×g for 1min. To measure the leakage of cellular materials that absorb at 260 nm, the absorbance of filtrate was read at 260 nm using a UV-microplate reader (Bio-Tek) and the absorbance of DMSO control was subtracted. To evaluate the protein leakage, filtrate was mixed with Bradford reagent according to the method provided by the manufacturer (Bio-Rad) and absorbance at 595 nm was read using a microplate reader. The data represent the absorbance at 595 nm of DMSO control subtracted from the absorbance at 595 nm of sample at each time point. The data are representative of three independent experiments, carried out in quadruplicates.

Effect of S. miltiorrhiza on DPH Binding to C. albicans Cells

C. albicans cells (5 × 107 cells/ml) were incubated with 2× MIC or 4× MIC S. miltiorrhiza at 35℃ on a shaking incubator at 200 rpm for 120 and 150 min, respectively. The control cells were incubated with the same volume of DMSO as that of S. miltiorrhiza. The cells were separated from the growth medium and the plant extract by centrifugation at 4,000 ×g for 5 min, washed, and resuspended in PBS. Each suspension was adjusted so that the optical density at 595 nm would be 1 × 108 CFU/ml. To evaluate the effect of S. miltiorrhiza on 1,6-diphenyl-1,3,5-hexatriene (DPH) binding to the C. albicans membrane, C. albicans ATCC 18804 cells were incubated with the fluorescence probe DPH at a final concentration of 2 μM at room temperature for 30 min in the dark. The samples were washed with PBS (pH 7.4), and fluorescence was measured in a black 96-well microplate using a spectrofluorometer (Bio-Tek) at 360 nm excitation and 460 nm emission wavelengths. The results represent the average of quadruplicate measurements from three independent assays.

Ergosterol Binding Assay

In order to assess whether S. miltiorrhiza binds ergosterol in Candida membranes, ergosterol binding assay was performed by the modified CLSI M27-A3 protocol [5]. Briefly, duplicate plates were prepared containing S. miltiorrhiza or amphotericin B (positive control). One plate included 2-fold dilutions of S. miltiorrhiza and the other plate included S. miltiorrhiza and 200 μg/ml ergosterol. Each well inoculated with 100 μl of cell suspension (1~5 × 103 cells/ml) was incubated at 35℃. MIC end-points were determined at 2 and 7 days.

Sorbitol Protection Assay

Sorbitol protection assay was carried out by the modified CLSI M27-A3 protocol as described above. Briefly, duplicate plates were prepared containing S. miltiorrhiza. One plate included 2-fold dilutions of S. miltiorrhiza and the other plate included S. miltiorrhiza and 0.8 M sorbitol as an osmotic protectant [8]. All the wells inoculated with cell suspension were incubated at 35℃.

Preparation and Quantification of (1,3)-β-D-Glucan Synthase

Preparation and analysis of (1,3)-β-glucan synthase from C. albicans cells were performed according to the methods of Shedletzky et al. [27] with little modifications. Briefly, C. albicans cells cultivated in 1 L of YM broth at 37℃ for 16 h were harvested and washed with ice-cold 15 ml breakage buffer (50 mM Tris.Cl (pH 7.4), 1 mM EGTA, 20 μM phenylmethylsulfonyl fluoride, 4 mM dithiothreitol, and 5 μm leupeptin), and broken by 12 cycles of 1 min in a Bead Beater (BioSpec Products, USA) with 0.5 mm acid-washed glass beads. The homogenate was then centrifuged at 3,000 ×g for 10 min at 4℃, and the supernatant containing microsomes and membranes were gathered by centrifugation at 100,000 ×g for 1 h at 4℃. The pellet was resuspended in 15 ml of breakage buffer with 33% glycerol and stored at −27℃. Protein concentration was measured using the Bio-Rad protein assay kit. For (1,3)-β-glucan synthase assay [8], 100 μg of C. albicans membrane protein was mixed with 50 mM Tris-Cl (pH 7.4), 20 μM GTP, 4 mM EDTA, 0.5% Brij 35, 6.6% glycerol, and 2mM UDP-glucose. All the reaction samples were incubated for 30 min at 25℃ and stopped by adding 10 μl of 6 N NaOH. Glucans formed were solubilized by incubating the reaction samples at 80℃ for 30 min followed by adding 210 μl of aniline blue mix (40: 21: 59 of 0.1% aniline blue, 1 N HCl, and 1 M glycine/NaOH (pH 9.5)). Reaction samples were further incubated at 50℃ for 30 min and at room temperature for 30 min. Fluorescence was measured in a black 96-well microplate using a spectrofluorometer (Bio-Tek) at 400 nm excitation and 460 nm emission wavelengths. The data represent the mean of quadruplicate measurements from three independent assays.

Flow Cytometry Analysis on Fungal Cell Cycle

C. albicans ATCC 18804 cells (1 × 107cells/ml) at the exponential growth phase were incubated with 1× MIC S. miltiorrhiza at 35℃ with shaking at 200 rpm for 3 h. The cells were collected and washed with PBS (pH 7.4) and then fixed with 70% cold ethanol overnight to permeabilize the cell membrane. The cells mixed with 200 μg/ml RNase A (Sigma, USA) was left to react for 1 h at 50℃. To stain DNA, PI in PBS was added to a final concentration of 10 μM, and the mixture was incubated for 30 min at room temperature in the dark. Cell cycle analysis was carried out utilizing the Muse Cell Analyzer from Millipore (Merck Millipore, USA) according to the manufacturer’s instructions. Data obtained from one run out of two independent experiments, which were conducted in duplicate, are presented.

Statistical Analysis

The differences in relative fluorescence or (1,3)-β-D-glucan synthase activity (mean ± standard error) of each group treated with S. miltiorrhiza and control were calculated using Student's t-test, and were regarded as statistically significant when a p-value was less than 0.05. All experiments were performed at least three times in quadruplicates.

Results

Antifungal Susceptibility Assay

The antifungal activity of S. miltiorrhiza was judged against standard strains of C. albicans ATCC 18804, C. krusei ATCC 750, C. glabrata ATCC 2001 and C. tropicalis ATCC 32196 using a colorimetric broth microdilution assay (Table 1). The MICs of S. miltiorrhiza against the test strains were in close proximity to one another, ranging from 39 μg/ml for C. albicans ATCC 18804 and C. krusei ATCC 750 to 78 μg/ml for C. glabrata ATCC 2001 and C. tropicalis ATCC 32196. MICs of S. miltiorrhiza and amphotericin B against the tested strain of C. albicans ATCC 18804 were 39 μg/ml and 1 μg/ml, respectively.

Table 1.The in vitro MICs of S. miltiorrhiza against Candida spp. were determined by the modified CLSI M27-A3 protocol of the colorimetric broth microdilution method containing resazurin. All assays were repeated three times.

Effect of S. miltiorrhiza on C. albicans Cell Membrane

Cells with severe membrane lesion will internalize PI. Once entering the cell, PI binds to nucleic acids. As a result, cells with damaged membranes present a red fluorescence signal, whereas those with intact membranes do not. Our data showed that 1× MIC S. miltiorrhiza was effective to damage the cell membrane of C. albicans ATCC 18804 cells with 3 h of incubation (Fig. 1).

Fig. 1.Effect of S. miltiorrhiza on the C. albicans ATCC 18804 cell membrane. (A) Untreated cells; (B) cells treated with 1× MIC S. miltiorrhiza for 3 h were stained with PI and fluorescent signals were examined using confocal laser microscopy. Superimposed fluorescent signals with Nomarski differential interference contrast images (left panes) and fluorescent images of PI (right panels) are shown .

Leakage of Cellular Materials Absorbing at 260 nm and Proteins

Disruption of the physical integrity of the fungal cell membrane has been associated with antifungal agents that exhibit rapid fungicidal activity. To assess if S. miltiorrhiza is involved in disrupting the integrity of C. albicans ATCC 18804 cell membranes, the leakage of intracellular contents was analyzed after treatment with S. miltiorrhiza. Suspensions of C. albicans ATCC 18804 cells in 0.85% NaCl were treated with 4× MIC S. miltiorrhiza, or DMSO as a control, at different time intervals. S. miltiorrhiza induced steady leakage of cellular materials in a time-dependent manner (Fig. 2), and treated cells showed significant release of intracellular nucleotide and proteins compared with DMSO-treated control (p < 0.05). This data are consistent with the loss of the plasma membrane integrity evidenced by PI uptake.

Fig. 2.Cell leakage analysis of S. miltiorrhiza against C. albicans ATCC 18804 cells. S. miltiorrhiza (4× MIC) or DMSO (control) was added to C. albicans ATCC 18804 cells, and then samples taken at the indicated time points were filtered through a 0.22 μm filter. (A) Nucleotide leakage: absorbance at 260 nm of filtrate was subtracted from the background absorbance of control. (B) Protein leakage: absorbance at 595 nm of filtrate mixed with Bradford reagent was subtracted from the background absorbance of DMSO control. Each value represents the average ± SE of quadruplicates.

Measurement of DPH Binding to Cell Membrane

Binding of DPH and its derivatives into intramembranes is coupled with strong increment of their fluorescence [31]. Therefore, the influence of S. miltiorrhiza on the C. albicans plasma membrane was examined using fluorescence probe DPH (Table 2). After exposure of C. albicans ATCC 18804 cells to 2× MIC and 4× MIC S. miltiorrhiza for 120 or 150 min, respectively, the C. albicans cells were then incubated with DPH. As Table 2 presents, the fluorescence intensity of DPH exhibited a significant reduction in S. miltiorrhiza-treated C. albicans cells as compared with DMSO control (p < 0.01). In the presence of 2× MIC of S. miltiorrhiza, the fluorescence was reduced to 73% for 120 min and 43% for 150 min, respectively, in comparison with the DMSO control (100%). After 4× MIC S. miltiorrhiza treatment, the fluorescence intensity was decreased by about 65% and 34%, respectively. This result suggests that S. miltiorrhiza is involved in a manner that alters the membrane composition of C. albicans cells to lead to a change of membrane permeability and disruption of the physical integrity of the C. albicans cell membrane.

Table 2.Measurements were performed on whole cells of C. albicans ATCC 18804 with S. miltiorrhiza. Relative fluorescence was expressed as a percent to determine the change in DPH binding caused by each treatment. Each value represents the average ± SE of quadruplicates. AU: arbitrary unit. **p < 0.01.

Ergosterol Binding Assay

To check whether S. miltiorrhiza binds membrane ergosterol, an essential lipid for many aspects of yeast cell physiology, in C. albicans cells, the MIC of S. miltiorrhiza against C. albicans ATCC 18804 was evaluated in the absence and presence of ergosterol in the broth. As shown in Table 3, the MIC of S. miltiorrhiza against C. albicans ATCC 18804 was not varied in the presence of ergosterol, but the MIC for the positive control, amphotericin B, increased 2-fold in the presence of ergosterol, revealing the binding of amphotericin B to the main sterol of yeast membranes. The data confirm that S. miltiorrhiza does not act in a manner that binds ergosterol directly in the C. albicans membrane to change membrane permeability.

Table 3.Effect of exogenous ergosterol on the MIC of S. miltiorrhiza against C. albicans ATCC 18804.

Sorbitol Protection Assay

To analyze if S. miltiorrhiza affects the fungal cell wall, the MICs of S. miltiorrhiza and amphoterin B (negative control) were evaluated against C. albicans ATCC 18804 cells in the absence and presence of sorbitol. Damaging effects of an antimicrobial compound on cell walls can be recovered in the presence of osmoprotectants such as sorbitol. As shown in Table 4, the MIC of S. miltiorrhiza against C. albicans cells increased more than 30-fold at 7 days. In contrast, the MIC of the negative control, amphotericin B, was not changed.

Table 4.Effect of sorbitol on the MIC of S. miltiorrhiza against C. albicans ATCC 18804.

Effect of S. miltiorrhiza on (1,3)-β-D-Glucan Synthase

The increased MIC values observed in the sorbitol protection assay (Table 4) implies that one of the possible targets of S. miltiorrhiza is the cell wall in C. albicans cells. Therefore, it was examined if S. miltiorrhiza inhibits the (1,3)-β-glucan synthase in C. albicans ATCC 18804 cells. The quantity of (1,3)-β-glucan synthase activity after treatment with S. miltiorrhiza was expressed as a percentage of the DMSO control. The content of (1,3)-β-glucan in the cell wall decreased approximately 12% and 16% in 4× and 8× MIC S. miltiorrhiza-treated C. albicans cells, respectively, as compared with DMSO control cells (Table 5), and the changes by 8× MIC S. miltiorrhiza were statistically significant (p < 0.01).

Table 5.Relative (1,3)-β-D-glucan synthase activity was expressed as a percent to determine the changes in (1,3)-β-D-glucan synthase activity caused by each treatment. Each value represents the average ± SE of quadruplicates. AU: arbitrary unit. **p < 0.01.

Effect of S. miltiorrhiza on C. albicans Cell Cycle

Although S. miltiorrhiza seems to impair the function of both cell membrane and cell wall of C. albicans cells, the antifungal effect was not fungicidal in the preliminary test. Therefore, the effect of S. miltiorrhiza on the cell cycle progress of C. albicans was explored, as shown in Fig. 3. In our study, growth control showed 20% cells in the G0/G1 phase in 3 h, whereas S. miltiorrhiza-treated C. albicans cells showed a major peak (42% cells) in the G0/G1 phase, representing a 2.1-fold increase in 3 h. The data indicate that S. miltiorrhiza elicits prolonged G0/G1 arrest in C. albicans cells.

Fig. 3.Effect of S. miltiorrhiza on the cell cycle progress of C. albicans ATCC 18804. Suspensions of C. albicans ATCC 18804 (1 × 108 cells/ml) were incubated without (black bars) or with 2× MIC S. miltiorrhiza (gray bars) for 3 h, and then washed with PBS. The cells were fixed in 70% ethanol overnight, and then stained with 10 μM PI. Cell cycle analysis was performed with the Muse Cell Analyzer.

Discussion

Plant extracts used in traditional medicine are thought to have the advantages of their several active components and their synergistic activity over conventional single component drugs: S. miltiorrhiza seems to contain a few active antifungal components against C. albicans, which affect the structure and function of both the cell membrane and cell wall in C. albicans.

Many cellular processes involved in growth and cell function are associated with changes in the membrane characteristics [2]. Although PI cannot penetrate intact cells since it carries two positive charges [25], it can go through the damaged cell membrane to bind nucleic acids. The results of treatment of C. albicans cells with S. miltiorrhiza in the presence of PI showed an enhancement of red fluorescence caused by membrane damage (Fig. 1). Leakage of cytoplasmic material is considered a serious destruction of the cytoplasmic membrane. The data that 260-nm-absorbing materials and protein of filtrates exposed to S. miltiorrhiza were significantly different from those of control confirm that S. miltiorrhiza acts in a manner that disrupts the physical integrity of the C. albicans cell membrane. Damage of the cell membrane could change the membrane permeability and disturb its ability of osmotic regulation or excluding toxic materials [4]. Fungal cell membrane is a typical lipid bilayer composed of 30-40% sterols, 10-20% sphingomyelin, low levels of glycosphingolipids, and some membrane proteins [20]. Ergosterol is a sterol in the fungal cell membrane, and changes in ergosterol synthesis affect membrane rigidity, fluidity, and permeability [21]. Amphotericin B, which targets ergosterol, is recognized to be involved in changes in membrane permeability of many organisms, bringing a leakage of cellular materials to cell lysis and death [7]. To know the reason why membrane permeability was changed, the ability of S. miltiorrhiza to bind membrane ergosterol was examined in C. albicans cells. The result of ergosterol binding assay suggests that S. miltiorrhiza does not bind to the ergosterol of fungal membranes. Alterations in lipid composition, such as the length, the degree of saturation, and branching of fatty acyl chains, lead to modification of the biophysical characteristics of the cell membrane [2]. Higher levels of polyunsaturated fatty acids were found in mycelial lipids than in C. albicans lipids [9]. Since DPH is soluble in the hydrocarbon tail region, and supposed to be oriented parallel to the axis of the lipid acyl chain, it is often used in measurements of fluorescence anisotropy and provides insight into membrane fluidity and lipid ordering [24]. To determine whether S. miltiorrhiza influences the lipid composition of the C. albicans cell membrane, the ability of DPH binding to the C. albicans cell membrane was assessed. Both 2× and 4× MIC S. miltiorrhiza caused a significant reduction in the DPH fluorescence intensity as compared with control cells. Our data suggest that the lipid composition or fluid properties of the C. albicans membrane have been altered by S. miltiorrhiza. The fluidity of biomembranes is involved in a variety of functions of cells, such as growth of cell, membrane transport, signal transduction, and membrane-related enzymatic activities [19]. Therefore, alterations in membrane permeability in S. miltiorrhiza-treated C. albicans were possibly attributable to the changes in the composition of membrane lipid, such as fatty acids, but not ergosterol. The fungal cell wall surrounds the cell membrane and confers cell rigidity and strength, providing osmotic protection from protoplast turgor. The cell wall consists of many macromolecules such as β-glucans, chitin mannoproteins, and other proteins, and many of these macromolecules are essential to the fungi, and enzymes that synthesize these constituents could be important antifungal targets [30].

Damage to the essential cell wall components by antifungal agents will lyse cells in the absence of an osmoprotectant, but cells will continue to grow if a suitable stabilizer is present in the medium [8]. Increase of MIC in the presence of sorbitol demonstrates that S. miltiorrhiza is involved in cell wall synthesis. This conjecture was confirmed via aniline blue assay: the reduced activity of (1,3)-β-D glucan synthase proves that S. miltiorrhiza contains the component preventing the synthesis of cell wall.

Regardless of the synergistic activity of S. miltiorrhiza targeting the C. albicans cell wall and cell membrane, its antifungal effect on C. albicans cells seems to be modest, since the antifungal activity was significantly efficacious when S. miltiorrhiza was used at slightly high concentrations. Therefore, it was investigated whether S. miltiorrhiza has an influence on cell cycle progression in C. albicans. As shown in Fig. 3, S. miltiorrhiza arrested C. albicans cells at the G0/G1 phase, leading to growth inhibition.

In conclusion, it is hypothesized that the damage in cell membrane and cell wall, by modified cell membrane composition and obstruction of cell wall synthesis, in S. miltiorrhiza-treated C. albicans cells imposes severe stress on C. albicans cells, leading to the arrest of cell cycle progression by controlling signaling pathways and loss of health. Multiple targets of S. miltiorrhiza for the cell membrane and cell wall in C. albicans could be worthwhile to use as a potential drug for future therapeutic option against Candida-associated infections.