Introduction

Serine proteinase inhibitor molecules (PIs) are found throughout the plant kingdom and have been described in many plant species [9, 16, 23, 28]. In plants, serine PIs can be found as constituent components present in the reserve tissues (in the tubers and seeds) or expressed in response to insects and pathogens by inhibiting the action of the digestive peptidases present in mammals and especially insects, as well as the enzymes present in bacteria and fungi [5, 7, 9, 10], and its expression levels can also increase in response to abiotic stresses [3]. Plant Kunitz inhibitors are mainly concentrated in the leguminous seeds of subfamilies Mimosoideae, Papilionoideae, and Caesalpinioideae. The majority of these inhibitors are molecules between 3 and 25 kDa and can inhibit either trypsin or chymotrypsin, with some capable of inhibiting both enzymes [20, 26].

More recently, especially among plant isolates, PIs concomitantly active toward specific proteinase with antimicrobial activity have attracted the attention of many, and various researchers have shown these functions. A protein called PSC-AFP, with molecular mass of 18 kDa, was isolated from Psoralea corylifolia, and by partial sequencing of the N-terminal region, this protein was found to be homologous with plant trypsin inhibitors. The antimicrobial activity of this protein was also tested against different fungi, including Alternaria niger, Aspergillus brassicae, Fusarium oxysporum, and Rhizoctonia cerealis, and it was observed that, in the presence of 10 µM of PSC-AFP, all of the fungi tested were inhibited [30]. Studies related to the antimicrobial roles of serine proteinase inhibitors were also performed by Lopes et al. [15]. Isoform inhibitors of Kunitz-type serine proteinase, with a molecular mass of 20 kDa each, were isolated from Acacia seeds, purified, and named ApTIA, ApTIB, and ApTIC. After characterization tests were performed of their structural properties, as well as tests for biological activity in the presence of different phytopathogenic fungi, it was observed that all of the inhibitors were able to inhibit the growth of Aspergillus niger, Thielaviopsis paradoxa, and Colletotrichum sp. More recently Ribeiro et al. [25] showed that a 6000 Da peptide, called CaTI, isolated from Capsicum annuum L. seeds showed potent inhibitory enzymatic activity against trypsin and chymotrypsin. It was also observed that CaTI inhibited the growth of Saccharomyces cerevisiae, Kluyveromyces marxiannus, and Candida albicans. Chilosi et al. [4] showed that STI, a family proteinase inhibitor of Bowman–Birk, showed strong antimicrobial activity in vitro against the growth of filamentous fungi. This inhibitor was able to inhibit 50 % growth of Botrytis cinerea, Colletotrichum acutatum, Didymella bryoniae, Fusarium culmorum, Fusarium graminearum, and Septoria tritici.

In 2011, antibacterial activity from corms of 15 species of the Xanthosoma genus was also shown. A Kunitz peptidase inhibitor from X. blandum called A Xb-KTI, with a molecular mass of approximately 24 kDa, inhibited the growth of the pathogenic bacteria Staphylococcus aureus, Salmonella typhimurium, and Escherichia coli at a concentration of 0.1 mg mL−1 [13]. However, small Kunitz inhibitors are the most described, e.g., PT-1 and potide-G were isolated from different cultivated varieties of potato tubers (Solanum tuberosum), showing molecular masses of 5600.0 and 5578.9 Da, respectively [11, 12]. PT-1 was active against Rhizoctonia solani and Clavibacter michiganensis and inactive against S. aureus [11], while potide-G inhibited S. aureus, Listeria monocytogenes, C. michiganensis, and E. coli [12].

Inga laurina (SW.) Willd. is a tree belonging to the subfamily Mimosoideae of the Leguminoseae. It is a tropical plant with a widespread distribution in Central and South America. A protein inhibitor of trypsin (ILTI) was isolated from its seeds, yielding a single band with an MM of approximately 20 kDa [16]. ILTI previously showed insecticidal activity against the lepidoptera Diatraea saccharalis and Heliothis virescens [22] and the coleoptera Homalinotus coriaceus [17]. In this work, we aimed to analyze the antimicrobial role of ILTI against pathogenic and non-pathogenic bacteria and fungi. Here, we demonstrated the strong potential of ILTI as a new fungi growth inhibitor, especially against Candida yeasts.

Materials and Methods

Biological Materials

The yeasts Candida parapsilosis (CE002), Candida tropicalis (CE017), Candida albicans (CE022), and Candida buinensis (URM4674) were maintained in the Laboratório de Fisiologia e Bioquímica de Microrganismos, Universidade Estadual do Norte Fluminense Darcy Ribeiro, Campos dos Goytacazes, Rio de Janeiro, Brazil. The yeasts were maintained on Sabouraud agar (1 % peptone, 2 % glucose, and 1.7 % agar–agar).

The bacteria E. coli ATCC 8739, Klebsiella pneumoniae ATCC 13883, and S. aureus ATCC 25923, used as the bacterial model strains, were maintained in the Laboratório de Purificação de Proteínas e suas Funções Biológicas, Universidade Federal de Mato Grosso do Sul. A single colony of each was isolated by the streak method in Mueller–Hinton (MH) solid medium and was propagated in MH broth at 37 °C, and the cell suspension was stored in sterile 10 % (v/v) glycerol at −80 °C. All of the subsequent experiments were performed from this original stock.

Extraction and Purification of ILTI

ILTI was purified from seeds of I. laurina (SW.) Willd., as previously described by Macedo et al. [16].

Preparation of Yeast Cells and Determination of the Effect of ILTI on Yeast Growth

The yeast growth assay was performed following the protocol developed by Broekaert et al. [2] with some modifications described below. To assay the effects of ILTI fraction on yeast growth, cells of C. albicans, C. tropicalis, C. parapsilosis, and C. buinensis (103 cells/mL of Sabouraud broth) were incubated at 30 °C in 100 μL in 96-well microplates in the presence of different concentrations of ILTI fraction (15, 30, 62.5, 125, and 250 μg mL−1) for C. tropicalis and C. buinensis and in the presence of 250 μg mL−1 for C. albicans and C. parapsilosis. The growth of each strain was evaluated by turbidity readings at a wavelength of 670 nm every 6 h for a period of 24 h. Yeast growth without the addition of ILTI fraction was also determined.

SYTOX Green Uptake Assay

Plasma membrane permeabilization was measured by SYTOX Green (Invitrogen, Grand Island, NY, USA) uptake, as described by Thevissen et al. [29]. Cells of C. tropicalis and C. buinensis were grown in the presence of ILTI (30 μg mL−1). Aliquots (100 µL) of yeast cell suspension were incubated with 0.2 μM SYTOX Green in 96-well microplates for 30 min at room temperature with periodic agitation, followed by observation using an optical microscope (Axio Imager A2 Zeiss) equipped with a fluorescence filter set for fluorescein detection (excitation wavelengths, 450–490 nm; emission wavelength, 500 nm). Negative controls without ILTI were also run to evaluate membrane permeabilization.

Visualization of Reactive Oxygen Species (ROS) Using Fluorescence

To determine whether the ILTI mechanism involves the induction of oxidative stress, it was used a fluorescent probe 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA) that indicates the presence of reactive oxygen species. Induction of endogenous production of ROS in yeast cells grown in the absence or presence of 30 μg mL−1 ILTI was evaluated after growth inhibition assay, using the fluorescent probe, according to the methods described by Mello et al. [18] with some modifications. Aliquots (50 µL) of the yeast cell suspension were incubated with 20 μM of the fluorescent probe for 2 h at room temperature with periodic agitation. After this period, the cells were transferred to slides, covered with cover slips and analyzed by fluorescence microscopy (Axiophoto Zeiss), with a fluorescence filter set for fluorescein detection (excitation wavelengths 450–490 nm and emission wavelength 500 nm). The results are representative of one triplicate experiment.

Localization of ILTI Conjugated with FITC for Optical Microscopy

ILTI (100 μg) was coupled to fluorescein isothiocyanate (FITC, 50 μg mL−1 solubilized in DMSO) following the manufacturer’s instructions (Sigma). After coupling, 50 μg mL−1 ILTI-FITC was incubated with cells (103 cells mL−1 of Sabouraud broth) of C. tropicalis and C. buinensis for 24 h in 96-well microplates. Subsequently, an aliquot of each cell suspension was removed and incubated with 50 μg mL−1 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) for 10 min for nuclear staining. Cells were analyzed with a DIC optical microscope (Axio Imager.A2, Zeiss) equipped with a fluorescent filter set for detection of the fluorescein (excitation wavelength, 450–490 nm, emission wavelength 500 nm).

Antibacterial Assay

The growth rate of bacterial cells from the original culture was established by measuring optical density (OD) at 595 nm and monitored at 30-min intervals in MH broth at a constant temperature (37 °C) and with shaking (240 rpm). The direct relationship of OD to colony forming units (CFUs) was established based on the drop surface plate method [21], and a standard growth curve was constructed and calibrated. The evaluation of the antibacterial activity of ILTI was performed at a final concentration of 200 μg mL−1 following the microdilution assay, according to the CLSI document M07-A8 [6]. Briefly, the bacterial culture was adjusted in MH broth supplemented with ILTI to 5 × 105 UFC mL−1 and incubated (37 °C) for 12 h, and the suspension turbidity was monitored at 595 nm every 30 min. Chloramphenicol (40 µg mL−1) was used as the positive control for bacterial growth inhibition, and ultrapure water was the bacteria growth control.

Results

Antimicrobial Activity of ILTI Fraction

In this work, we tested the effects of ILTI as an inhibitor of the growth of pathogenic and non-pathogenic yeast strains and of human pathogenic bacteria. We noted complete growth inhibition of C. tropicalis in the presence of 250 and 125 μg mL−1 ILTI and partial inhibition of 83, 31, and 6 % to 62.5, 30, and 15 μg mL−1 ILTI, respectively (Fig. 1). Regarding C. buinensis, complete growth inhibition was observed only at 250 μg mL−1 ILTI concentration, while for 125, 62.5, 30, and 15 μg mL−1 ILTI, inhibition of 89, 45, 13, and 0 % was noted, respectively (Fig. 1). The growth inhibition curve for C. albicans and C. parapsilosis, determined only at the concentration of 250 μg mL−1, ILTI inhibited 9 and 12 %, respectively (data not shown). Nevertheless, the antibacterial assay determined the inability of ILTI (200 μg mL−1) to inhibit the growth of the human pathogenic bacteria E. coli, K. pneumoniae, and S. aureus (data not shown).

Fig. 1
figure 1

Antimicrobial activity of ILTI against Candida tropicalis and Candida buinensis

Full size image

SYTOX Green Uptake Assay

The ability of ILTI to permeabilize the plasma membrane of C. tropicalis and C. buinensis cells was examined in this study. Membrane permeabilization was assessed after 24 h of growth in the presence of ILTI 30 μg mL−1 and 30 min after the addition of SYTOX Green. When observed with a fluorescence microscope, the cells of both yeasts showed SYTOX Green fluorescence in the presence of ILTI, compared to controls in which cells were grown in the absence of ILTI (Fig. 2).

Fig. 2
figure 2

Membrane permeabilization assay performed with fluorescence microscopy of Candida tropicalis and Candida buinensis yeast cells treated with SYTOX Green for a period of 30 min after being treated with ILTI (30 μg mL−1) for 24 h. Bars 20 µm (Color figure online)

Full size image

Visualization of Reactive Oxygen Species (ROS) Using Fluorescence

Using a fluorescence microscope, we demonstrated ROS induction in C. tropicalis and C. buinensis cells after pre-incubation with 30 μg mL−1 ILTI for 24 h, compared to control cells (Fig. 3). ILTI could induce higher levels of ROS, which might be involved in different metabolic mechanisms of yeast growth arrest or death.

Fig. 3
figure 3

Oxidative stress assay performed with fluorescence microscopy of Candida tropicalis and Candida buinensis yeast cells for ROS detection. Yeasts cells were incubated with ILTI (30 μg mL−1) for 24 h and after this period with the fluorescent probe 2′,7′-dichlorodihydrofluorescein diacetate (H2DCFDA) for 2 h. Bars 20 µm

Full size image

Localization of ILTI Conjugated with FITC for Optical Microscopy

To determine the localization of ILTI in C. tropicalis and C. buinensis, ILTI was coupled to FITC and used for antifungal activity assay. Subsequently, after 24 h of growth, the cells were subjected to DAPI labeling. In this assay, we could observe the internalization of ILTI (Fig. 4). An overlay of the images labeled with FITC and DAPI (Fig. 4) suggests that ILTI could target the intercellular organelles of this organism, such as the nuclei.

Fig. 4
figure 4

Fluorescence microscopy analysis of Candida tropicalis and Candida buinensis cells, incubated for 24 h with 50 μg mL−1 FITC-tagged ILTI (green fluorescence). After the incubation period, the nuclei were stained with DAPI (blue fluorescence). Bars 20 µm (Color figure online)

Full size image

Discussion

The effects of ILTI on the growth of different species of yeasts were observed (Fig. 1). These effects are interesting because this activity might be associated with the presence of important proteinases responsible for the synthesis of fungal components. Other authors have shown the antimicrobial effects in vitro of some proteinase inhibitors of plants and have shown that some proteinases secreted by fungi could be important factor in determining pathogenicity and that its inhibition could significantly reduce the infections caused by these fungi [12, 27]. Despite its antifungal potential, ILTI was ineffective against both of the bacterial strains evaluated. However, ILTI could be selective against other bacteria not tested. However, further experiments should be performed with an extensive list of bacterial strains before disregarding any antibacterial proprieties of ILTI. Selectivity is recurrently observed in peptidase inhibitors [11, 13] and was also seen in ILTI antifungal activity. Many classes of endopeptidases and exopeptidases secreted by pathogenic fungal and bacteria are referred as virulence factors [8, 19]. Furthermore, the variety of secreted peptidases on the surface of fungal and bacterial, as well those in extracellular environment, also varies among fungal [19] and bacterial species [8], what may explain the ILTI selectivity.

In this study, we observed the capacity of ILTI to damage the plasma membranes of the yeast C. tropicalis and C. buinensis (Fig. 2), allowing for the permeation of same through the use of SYTOX Green dye. This ability to induce cellular damage has been viewed in different families of proteins/peptides and plants, and it is capable of causing changes to the cell membrane and regulating its flow of ions, making these molecules ideal models for understanding the functioning of vital proteins that act as ion channels.

Fluorescence assays demonstrated that ILTI increased ROS levels in yeast cells (Fig. 3). However, most ROS can be produced in basal levels in cells when grown under normal conditions free of stress, which might explain the low or absent fluorescence in the control cells. Other authors have reported that the plant defensin Rs-AFP2, isolated from Raphanus sativus, is involved in the induction of oxidative stress in cells of the yeast Candida albicans [1]. From the results obtained in this study, it is reasonable to suggest that ILTI could have some mitochondrial target, causing an increase in the production of ROS. Ribeiro et al. [24] also showed that CaTI, a proteinase inhibitor, induced the generation of nitric oxide and interfered in a dose-dependent manner with glucose-stimulated acidification of the medium, mediated by H+-ATPase from S. cerevisiae cells.

The next experiments were designed to analyze whether ILTI was able to enter C. tropicalis and C. buinensis actively. To that end, FITC-tagged ILTI was monitored by fluorescence microscopy. Because ILTI entered yeasts cells, we suggest that a possible intracellular target for this proteinase inhibitor might be part of a complex mechanism responsible for the death of Candida species. FITC-tagged ILTI overlapped with DAPI staining, suggesting that one of its targets is nuclear (Fig. 4). Important results were shown by Lobo et al. [14], who found that the defensin isolated from Pisum sativum (PsD1) could in fact have a nuclear target. This study and other related studies have suggested that the antifungal activities of plant defensins are not restricted to the plasma membranes of fungi because the defensins can enter cells and target different intracellular compartments. More recently, Zottich et al. [31] purified a new protein from coffee, called Cc-GRP, which is involved in the plant defense system against pathogens by acting through a membrane permeabilization mechanism, as well as being localized in the nuclei of fungal cells. Our work opens new perspectives regarding the antimicrobial mechanisms of plant-derived proteinase inhibitors suggesting that the toxicity of these proteins might have different mechanisms of action.