Introduction

Plant defensins are small, basic peptides of 45–54 amino acids comprised in a three-dimensional structure formed by three anti-parallel β-strands and one α-helix. This structure is stabilized by four disulfide bounds which form a cysteine-stabilized α-helix β-strands motif common to these peptides [4]. Plant defensins, such as insect and mammal defensins, possess antimicrobial activity. The discovery of the antimicrobial activity of plant defensins was reported in the early 1990s by Terras et al. [20]. The antimicrobial activity of plant defensins is principally observed against fungi. However, some bacteria, especially Gram-positive species, are also inhibited, though the activity is less pronounced in comparison with the activity against fungi. Growth of several fungal species, including several filamentous fungi and yeast cells, was inhibited when incubated with these peptides [3, 5, 7, 11–14, 19, 21, 22, 24, 28–32]. More recently, it was shown that Rs-AFP2 (Raphanus sativus antifungal peptide 2) induces reactive oxygen species (ROS) production in Candida albicans in a dose-dependent manner but not at all in an Rs-AFP2-resistant Δgcs C. albicans mutant that lacks the Rs-AFP2-binding site in its membranes. These findings indicate that upstream binding of Rs-AFP2 to GlcCer is needed for ROS production, which leads to yeast cell death. Moreover, the antioxidant ascorbic acid blocks Rs-AFP2-induced ROS generation and Rs-AFP2 antifungal activity [1].

PvD1 defensin was purified from Phaseolus vulgaris (cv. Perola) seeds. PvD1 inhibited the growth of the different yeast cells and filamentous fungi [8]. In this study, the antimicrobial activity of PvD1 was investigated by studying its permeabilization of fungal membranes, inhibition of glucose-stimulated acidification of the medium, and also ultrastructural analysis, and production of ROS and nitric oxide (NO) in the different filamentous fungi cells and yeast cells.

Materials and Methods

Plant Material

Phaseolus vulgaris L. (cv. Pérola) seeds were provided by the Laboratório de Melhoramento Genético Vegetal, Centro de Ciências e Tecnologias Agropecuárias, Universidade Estadual do Norte Fluminense, Campos dos Goytacazes, Rio de Janeiro, Brazil.

Fungi

The yeasts Candida parapsilosis (CE002), Candida guilliermondii (CE013), Candida tropicalis (CE017), Candida albicans (CE022), Kluyveromyces marxiannus (CE025), Pichia membranifaciens (CE015), and Saccharomyces cerevisiae (1038) were obtained from the Departamento de Biologia, Universidade Federal do Ceará, Fortaleza, Ceará, Brazil. The phytopathogenic fungi Fusarium oxysporum, Fusarium solani, and Fusarium laterithium were obtained from the Laboratório de Fisiologia e Bioquímica de Microrganismos, Universidade Estadual do Norte Fluminense, Campos dos Goytacazes, Rio de Janeiro, Brazil. Yeasts and fungi were maintained on Sabouraud agar (1% peptone, 2% glucose and 1.7% agar-agar).

Purification of the Phaseolus vulgaris Defensin PvD1

Purification of the defensin from P. vulgaris (cv. Pérola) seeds was performed largely as described by [8].

Gel Electrophoresis

PvD1 defensin purification was monitored by SDS-Tricine-gel electrophoresis performed according to the method of Schägger and Von Jagow [18].

Inhibition of Glucose-Stimulated Acidification of the Medium by Yeasts

In order to evaluate the influence of PvD1 interference with fungal metabolism, the authors have analyzed the H+ extrusion by yeast. It is known that this acidification is due to the activity of the plasma membrane H+-ATPase, and it is stimulated by glucose [9]. The effect of PvD1 was determined by incubation of yeasts cells (1 × 107) or filamentous fungi cells (1 × 107) with 0.8 ml of 10 mM Tris–HCl, pH 6.0. PvD1 was added to make up to final concentrations of 100 and/or 200 μg/ml. After the pre-incubation time (1, 2, and 4 h for yeasts and 30 min and 1 h for filamentous fungi), 0.2 ml of 0.5 M glucose solution was added. Measurements of pH were taken at each subsequent minute for the next 30 min. Controls (Tris–HCl buffer was added instead of PvD1) were run to evaluate the influence of peptides on H+ extrusion by yeast cells [22]. The concentration of H+ extrusion was calculated as the difference (ΔpH) between initial (T = 0) and final (T = 30 min) pH, and the obtained values were applied to the equation pH = −log [H+]. The results of inhibition of the glucose-stimulated acidification are shown by average values and standard deviations of triplicate for each experiment. Graphs were drawn based on the assumption that the values obtained for controls corresponded to 100% acidification and the significant test were done according to the Student’s t-test, calculated with the Statistica software, among the pre-incubation time and their respective controls.

SYTOX Green Uptake Assay

Fungal plasma membrane permeabilization was measured by SYTOX Green uptake as described previously by Thevissen et al. [26] with some modifications. SYTOX Green is a dye that only penetrates cells when the plasma membrane is structurally compromised. Once inside the fungal cytoplasm, it binds to nucleic acids, resulting in a fluorescent complex. Therefore, this dye could be used for visualization of the permeabilization of the fungal plasma membrane. The different fungal species (1 × 104 cells/ml) were incubated in the presence of PvD1 at the concentration of 100 μg/ml for 36 and 60 h growth for yeast and filamentous fungi, respectively. One hundred-microliter aliquots of the fungal cell suspension were incubated with 0.2 μM SYTOX Green in 1.5 ml microcentrifuge tubes for 2 h at 25°C with periodic agitation. The cells were observed with a DIC microscope (Axiophoto Zeiss) equipped with a fluorescence filter set for fluorescein detection (excitation wavelengths, 450–490 nm; emission wavelength, 500 nm). The results of membrane permeabilization are representative of one triplicate experiment.

Transmission Electron Microscopy

In order to evaluate the influence of PvD1 on fungal cells morphology, the authors performed analyses of transmission electron microscopy. C. albicans and C. guillermondii cells, grown for 36 h in Sabouraud broth in the presence (100 μg/ml) or absence of PvD1, were fixed for 30 min at room temperature in a solution containing 2.5% glutaraldehyde (vol/vol) and 4% paraformaldehyde (vol/vol) in 50 mM cacodylate buffer (pH 7.2). After fixation, the materials were washed, post-fixed in 1% (wt/vol) osmium tetroxide in the same buffer for 1 h at room temperature. The samples were dehydrated in a graded acetone series [30, 50, 70, 90, and 100% (vol/vol)] and embedded in Epon resin (Polybeded). Ultrathin sections (0.1 μm) were laid on copper grids, stained with uranyl acetate for 10 min followed by lead citrate for 5 min and were then observed with a ZEISS 900 transmission electron microscope (TEM) (Zeiss company, Germany) operating at 80 kV. The results of electron microscopy are representative of one triplicate experiment.

ROS and NO Induction Assay

It has been demonstrated that some plant defensins could induce oxidative stress in fungal cell exposed to them [1]. In order to evaluate whether the action mechanism of PvD1 involve the induction of oxidative stress the authors have used two dyes that indicates the presence of reactive species. Induction of endogenous production of ROS in C. albicans and F. oxysporum fungi and NO in C. albicans yeast cells, treated with 100 μg/ml of the defensin PvD1 after growth inhibition assay, was evaluated using the fluorescent dye 2′,7′ dichlorofluorescein diacetate (Calbiochem - EMD) and 3-amino, 4-aminomethyl-2′,7′-difluorofluorescein diacetate (Calbiochem - EMD), respectively, using methods described by [1] with some modifications. Incubations were performed as described in the SYTOX Green uptake assay section. The incubation time for this analysis was 24 h of growth in the presence or absence of PvD1. An aliquot was incubated with constant agitation for 2 h with fluorescent dye to a final concentration of 20 μM, according to instructions provided by the manufacturers. After this period, these cells were transferred to slides, covered with coverslips, and analyzed with a fluorescence microscope (Axiophoto Zeiss) equipped with a fluorescence filter set for fluorescein detection (excitation wavelengths, 450–490 nm; emission wavelength, 500 nm). The results of oxidative stress are representative of one triplicate experiment.

Results and Discussion

The ability of PvD1 to permeabilize the plasma membrane of different fungal cells, especially pathogenic yeasts, was examined. SYTOX Green membrane permeabilization was assessed after 60 and 36 h of growth for filamentous fungi and yeasts, respectively, in the presence of PvD1 and 2 h after the addition of SYTOX Green. When observed with a fluorescence microscope, all filamentous fungal cells showed strong SYTOX Green fluorescence in the presence of PvD1 (Fig. 1), as compared with controls, in which fungi were grown in the absence of PvD1 (Fig. 1). All these yeast cells also showed strong SYTOX Green fluorescence in the presence of PvD1 (Fig. 2), especially P. membranifaciens and C. tropicalis, as compared with controls, in which cells of C. tropicalis and S. cerevisiae were grown in the absence of PvD1. The controls for the other yeasts showed the same results (data not show). Different defensins have also been found to permeabilize membranes and to modulate ion flux across membranes, and they have since become popular models for understanding how ion channel proteins function [1]. Thevissen et al. [25], for example, demonstrated that when the fungi Neurospora crassa and F. culmorum were treated with the plant defensins Rs-AFP2 and Dm-AMP1 (Dhalia merckii antimicrobial peptide 1), ion flux across the fungal plasma membrane was observed. The fungal membrane permeabilization was also shown in response to other antimicrobial peptides, such as lipid transfer proteins and 2S albumins [2, 6, 17], and membrane permeabilization is related to the inhibition of fungal growth [26].

Fig. 1
figure 1

Membrane permeabilization assay of different filamentous fungal cells treated with PvD1 (100 μg/ml). 200×

Full size image
Fig. 2
figure 2

Membrane permeabilization assay of different yeast cells treated with PvD1 (100 μg/ml). 400×

Full size image

Interference with the function of H+-ATPase in fungi by antagonists commonly leads to cell death. In this study, the authors investigated whether PvD1 could interfere with the fungal H+-ATPase. For this, the authors monitored glucose-stimulated acidification of the incubation medium by S. cerevisiae and C. albicans cells in the presence of various concentrations of PvD1 defensin, a phenomenon that is dependent on the activity of the H+-ATPase. As shown in Fig. 3a, PvD1, at concentrations of 100 and 200 μg/ml, was able to decrease this acidification in S. cerevisiae by 10 and 20%, respectively, when used for a pre-incubation time of 1 h, been the higher concentration used statistically significant. In Fig. 3b, PvD1, at the same concentrations, was able to lower this acidification by 65 and 75%, respectively, when used for a pre-incubation time of 2 h. Finally, when held for a pre-incubation time of 4 h in the presence of PvD1 at the concentration of 100 μg/ml, the decrease in acidification observed was 80% (Fig. 3c). These results suggest that for S. cerevisiae cells, the inhibition of acidification is dose and pre-incubation time-dependent. For C. albicans, as shown also in Fig. 3d, PvD1, at the concentrations of 100 and 200 μg/ml, strongly inhibited acidification of the medium; the inhibitions observed were 97 and 99%, respectively, when a pre-incubation time of 1 h was used. As shown in Fig. 3e, PvD1 was able to inhibit this acidification by 68% and 90%, respectively, when a pre-incubation time of 2 h was used. Interestingly, when a pre-incubation time of 4 h was used, the inhibition of acidification observed was 72% at the concentration 100 μg/ml, suggesting that for C. albicans cells (Fig. 3f), the inhibition of acidification is not dose or pre-incubation-time dependent.

Fig. 3
figure 3

The effect of PvD1 on the glucose-dependent acidification of the medium by S. cerevisiae and C. albicans cells. PvD1 was pre-incubated for 1, 2, and 4 h at 100 and/or 200 μg/ml before addition of glucose. Significantly different from the controls according to Student’s t-test (P < 0.05)

Full size image

For filamentous fungi (F. oxysporum and F. solani), as shown in Fig. 4, PvD1, at the concentrations of 200 μg/ml, strongly inhibited acidification of the medium. The inhibition observed was 99%, for both fungi, when a pre-incubation time of 30 min was used; with pre-incubation time of 1 h, were obtained the same results (data not shown). However, the filamentous fungi are not good models for testing acidification, since even in the presence of glucose, they acidify slightly. The authors believe that this effect is due to the sum of the toxicity of the peptide and the low acidifying capacity of filamentous fungi. Different peptides have been studied, and the analysis has been performed to test the ability of some of these peptides to act on the plasma membrane [25–27] and consequently interfering with the function of H+-ATPase, causing alterations in the glucose-stimulated acidification of the medium by the cells.

Fig. 4
figure 4

The effect of PvD1 on the glucose-dependent acidification of the medium by F. oxysporum and F. solani conidia. PvD1 was pre-incubated for 30 min at 200 μg/ml before addition of glucose. Significantly different from the controls according to Student’s t-test (P < 0.05)

Full size image

TEM observations of yeast cells showed normal ultrastructure development in control cells (Fig. 5a, d); however, cells treated with PvD1 (Fig. 5b, c, e, f) exhibited plasma membrane blebbing, disappearing and shrinking cytosol, and disorganization of the nucleus and other organelles. Not all of these features were observed in every cell, but they occurred quite frequently. The percentage of cells observed with abnormal features was approximately 80% as compared with controls.

Fig. 5
figure 5

Transmission electron microscopy of C. albicans (a, b, d, e, f) and C. guilliermondii (c). a, d controls; b, c, e, f cells treated with PvD1 (100 μg/ml). A star indicates condensation and shrinkage of the cytosol with loss of cytosol structure and contents; arrows indicate cell and plasma membrane blebbing; bars: a, 0.4 μm; b, c, e, f, 0.25 μm

Full size image

Another objective of this study was to investigate the ability of PvD1 to induce the production of ROS and NO in yeasts cells. It was recently demonstrated that PsD1 (Pisum sativum defensin 1) is internalized into the cytoplasm and targeted to the nucleus [10]. NaD1 (Nicotiana alata defensin 1) has also been shown to enter the cytoplasm of F. oxysporum f. sp. vasinfectum, but in contrast to PsD1, it was not observed in the nucleus, indicating that defensins may act by different mechanisms [27]. These results provide a direct link between ROS generation and the antifungal effect of the peptide. Using a microscopic assay, the authors were able to demonstrate ROS induction in C. albicans cells incubated with 100 μg/ml PvD1 for 24 h (Fig. 6) compared with control cells (Fig. 6) that were not treated with PvD1. The authors were able also to demonstrate ROS induction in F. oxysporum cells incubated with 100 μg/ml PvD1 for 24 h (Fig. 6) as compared with control cells (Fig. 6) that were not treated with PvD1. The ROS induction capacity of various antifungals has been previously reported. Azole antifungals, as well as the polyenes and other antifungals, which interact with ergosterol in the fungal membrane cause membrane permeabilization and induce ROS production in susceptible fungi [15, 23, 25, 27]. Induction of NO production in C. albicans cells was also demonstrated after incubation of the cells with 100 μg/ml of PvD1 for 24 h (Fig. 6). Little or no fluorescence was observed in the case of control cells (Fig. 6).

Fig. 6
figure 6

Oxidative stress assay for ROS and NO of C. albicans and F. oxysporum cells previously incubated with 100 μg/ml of PvD1 for 24 h. Cells were treated with 2′,7′ dichlorofluorescein diacetate for ROS and/or 3-amino, 4-aminomethyl-2′,7′-difluorofluorescein diacetate for NO detection. 400× for yeasts and 200× for filamentous fungi

Full size image

The authors found in this study that the defensin isolated from P. vulgaris, PvD1, was able to permeabilize the membranes of filamentous fungi and yeasts. Ultrastructural analysis of yeast cells treated with this defensin revealed disorganization of both cytoplasmic content and the plasma membrane. Incubation with the PvD1 caused fungal inhibition of the glucose-stimulated acidification for the filamentous fungi and yeasts and was also able to cause induction of ROS in C. albicans and F. oxysporum and NO in C. albicans. The antimicrobial activity of plant defensins are initially felt at the level of the fungal plasma membrane where receptors were identified. They are for Dm-AMP1 the sphingolipid manosyldiinositolphosphrylceramide [22] and for Rs-AFP2 and MsDEF1 the sphingolipid glucosylceramide [16, 22]. After this initial binding on components of the fungal membranes, secondary effects induced internally in the cell have been demonstrated. Lobo et al. [10] have demonstrated that PsD1 interacts with cyclin F and consequently interferes with the cell cycle [1, 27]; NaD1 and Rs-AFP2 induce ROS production; and that these toxic substances are involved in fungal growth arrest. These results reinforce the idea that different plant defensins must act by different mechanisms and demonstrate a complicated and sophisticated mechanism underlying fungal growth arrest.