Anti-Candida albicans action of the glyco-protein complex purified from metabolites of gut bacterium Raoultella ornithinolytica isolated from earthworms Dendrobaena veneta

Authors


Correspondence

Marta J. Fiołka, Department of Immunobiology, Institute of Biology and Biochemistry, Maria Curie-Skłodowska University, Lublin, Poland. E-mail: mfiolka@poczta.onet.pl

Abstract

Aims

The aim of our research was to isolate the compounds from the metabolites of Raoultella ornithinolytica with the activity against Candida albicans and to analyse the action of the compounds on the metabolic activity and morphology of the fungus cells.

Methods and Results

The effect of active protein fractions on the cell morphology, growth, and metabolic activity of C. albicans was analysed under a light microscope with Nomarski contrast and after staining with calcofluor-white. The LIVE/DEAD Yeast Viability Kit F-7030 FUN 1 was used for determination of C. albicans metabolic activity. The biomolecules obtained after isolation by ion exchange chromatography were further fractionated by Sephadex G-50 medium gel filtration. Then, after molecular sieve, the fractions were analysed by FTIR and SERS Spectroscopy. A subfraction was isolated from the antifungal protein fraction above 100 kDa. The active subfraction identified as the glyco-protein complex caused a decrease in the metabolic activity and morphological changes of C. albicans cells.

Conclusions

The glyco-protein complex obtained from metabolites of bacteria Raoultella ornithinolytica possesses antifungal activity against C. albicans and shows minimal toxicity (1%) against fibroblasts.

Significance and Impact of the Study

Studies on the glyco-protein complex obtained from earthworm gut bacteria R. ornithinolytica can lead to their application in biological fungicide and pharmaceutical industry.

Introduction

Medicines are mostly obtained from natural resources such as plants. It is known that animals are the source of many therapeutics. Zootherapy is widespread across most human cultures (El-Kamali 2000; Lev 2003; Machkour-M'Rabet et al. 2011). The medicinal value of earthworms has been widely known since the history of Asian ancient medicine. Earthworms have been widely used in Indonesia, China, Japan and Far East in the treatment of various chronic diseases (Trisina et al. 2011).

Profound exploration has been carried out to reveal the use of earthworms as antimicrobial (Popović et al. 2005), anti-inflammatory (Balamurugan et al. 2009) and anticancer agents (Nagasawa et al. 1991; Engelmann et al. 2004; Chen et al. 2007; Yanqin et al. 2007). These activities may derive from various sources in the earthworm body. Mathur et al. (2010) revealed that extracts of the earthworms Eudrilus eugeniae possessed antibacterial and antifungal activity also against Candida albicans. Powder from E. foetida has antifungal properties, which are demonstrated in inhibition of C. albicans growth (Ansari and Sitaram 2010, 2011). The powder could be effective in treating a fungal infection, Candidiasis.

During the past decades, the frequency of fungal infections has increased because of diverse factors increasing people's vulnerability to fungal diseases. Among these factors are the AIDS epidemic, changes in lifestyle into a more stressful one, ageing of the population, increased frequency of organ transplantation, use of anticancer and immunosuppressive therapies, development of modern invasive medical implants and principally widespread use of broad-spectrum antibiotics (Wingard 2003). The problem is further complicated by the lack of effective antifungal drugs available to treat systemic infections and by acquiring resistance to the currently available antifungal agents by human pathogenic fungi. In numerous medical journals, authors have underlined that C. albicans is the most frequent cause of fungal infections.

Many researchers have noticed that antifungal activity derives from the digestive tract of invertebrates. Pathogens passing through the arthropod host's alimentary tract are exposed to both the resident micro-organisms and the intestinal environment. The midgut of earthworms possesses strong antimicrobial activity (Khomyakov et al. 2007; Fiołka et al. 2012).

We found bacteria Raoultella ornithinolytica in the midgut tissue to be a probable symbiont of earthworm Dendrobaena veneta (Fiołka et al. 2010). This micro-organism produces antibiotic-like substances showing antimicrobial activity, both against bacteria and fungi. We noticed antifungal activity of the metabolites secreted by earthworm bacteria Rornithinolytica directed against C. albicans. The aim of our research was to find the compounds responsible for the activity against C. albicans and to analyse the action of the active substances on the activity and morphology of fungal cells in comparison with the known antifungal antibiotic, amphotericin B.

Materials and methods

Micro-organisms

Raoultella ornithinolytica bacteria were isolated from the midguts of a representative group of D. veneta after intense cleaning of the intestines. The homogenates obtained from the midguts were used for bacterial isolation. The isolated bacteria strains with antimicrobial activity from D. veneta guts were identified using biochemical and molecular methods as described previously (Fiołka et al. 2010). The 16S rRNA gene sequence of the R. ornithinolytica strain determined in this study has been deposited in GenBank under accession number FJ905041. Bacteria R. ornithinolytica were multiplicated in liquid Hawiger medium, pH 6·0 for 24 h at 37°C.

A culture of the wild strain of C. albicans was used in the study of the action of the active fractions obtained on fungal cells. The yeast cells were grown overnight in YPD medium at 28°C with shaking at 200 rpm.

Ammonium sulphate precipitation

The bacterial culture was centrifuged at 1157 g for 1 h and filtered through a Millipore filter (0·22 μm). The cell-free culture supernatant was concentrated to 50 ml by lyophilization, precipitated by ammonium sulphate (Sigma) to 90% saturation at 4°C and then centrifuged at 15 557 g at 4°C for 15 min. The pellet was resuspended into 5 ml of 33 mmol l−1 phosphate buffer, pH 6·0, and dialysed against the same buffer in a cellulose membrane bag with cut-off points of 12–14 kDa (Roth) to remove residual salt. To divide the high molecular fraction obtained, ultrafiltration was performed with centrifugal filter devices Amicon Ultra-4 (Millipore) with cut-off points of 50 and 100 kDa. The filtrate and retentate were tested for antifungal activity. The protein concentration was determined by the Bradford method (Bradford 1976).

Preparation for microscopy techniques

The effect of the action of the antimicrobial active fraction (AAF – a fraction containing proteins above 14 kDa) on C. albicans was examined in cultures of Candida in liquid YPD poor medium (1 g agar, 0·1 g yeast extract, 0·2 glucose, 0·05 g peptone, dissolved in 1 l sodium–phosphate buffer 0·01 mol l−1, pH 7, according Vilcinskas and Matha 1997). The protein of AAF in 25 and 50 μl (at the final concentrations of 1·25 and 2·5 mg ml−1) was added to 100 μl YPD poor medium containing the C. albicans culture (107 CFU) and streptomycin sulphate (Sigma, St Louis, MO) to the final concentration of 0·17 mg ml−1. The samples were filled up with YPD poor medium to the final volume 200 μl and incubated for 3 days at 37°C with gentle shaking. After that time, the antifungal action of AAF was analysed under a light microscope.

Amphotericin B was dissolved in 0·2% dimethyl sulphoxide (DMSO) (Okutomi et al. 1997) and at the concentrations of 0·25 and 0·50 μg ml−1 and incubated with C. albicans in YPD poor medium simultaneously in the same way.

Minimal inhibitory concentration assay

The determination of the lowest concentration of the compounds with anti-C. albicans activity and amphotericin B was accomplished using twofold serial dilutions of the test compounds in YPD poor medium at 37°C for 72 h. The wells were assessed by a 96-well plate reader (Benchmark Plus) to determine the concentration exhibiting in vitro inhibition of C. albicans growth.

Fluorescence and Nomarski microscopy

The cells of the C. albicans wild-type strain were observed at 1000× magnification with Nomarski contrast (Zeiss/LEO LEO 912AB). After incubation with AAF, the liquid cultures of C. albicans after incubation with the AAF of R. ornithinolytica and the control culture were stained with calcofluor white (Fluka) for 10 min in the dark (Monheit et al. 1984). The morphology and cell structures of C. albicans were analysed in the cultures after staining and using two-dimensional scan.

Fluorescent staining for quantification of yeast viability

The LIVE/DEAD Yeast Viability Kit F-7030 FUN 1 was used for determination of the C. albicans metabolic activity (Mazzoni et al. 1993). The C. albicans wild strain suspension was incubated with AAF (at the concentration of 2·5 and 1·25 mg ml−1), AAS (2·5, 5, 10 μg ml−1), amphotericin B (0·25 and 0·50 μg ml−1).

Of 0·2 ml of cell suspension (OD 1·4) was centrifuged for 5 min at 12 857 g at room temperature, and the cells were resuspended in 100 μl GH solution (2% D-glucose containing 10 mmol l−1 Na-HEPES, pH 7·2). 30 μl of the C. albicans suspension in GH solution was added to 30 μl of GH-FUN solution. The samples were incubated for 30 min at 30°C before microscopic observation.

Only metabolically active cells were clearly marked with fluorescent intravacuolar structures. Cells with intact membranes and little or no metabolic activity had diffuse green cytoplasmic fluorescence and lacked fluorescent intravacuolar bodies.

Scanning electron microscopy

After fixation with 4% glutaraldehyde (Verma et al. 1999) in 0·1 mol l−1 phosphate buffer, pH 7·0, the cells were treated with OsO4, dehydrated stepwise in a graded acetone series, dried and sputter coated with gold using a K550X sputter coater (Quorum Technologies). The samples were examined using a Vega 3 scanning electron microscope (Tescan) with ×10 000 magnification.

Isolation of protein subfractions

Antimicrobial active fraction from the R. ornithinolytica culture was fractionated by centrifugation using filter devices Amicon Ultra-4 (Millipore) with cut-off points of 100 kDa. The fraction of MW above 100 kDa was collected and lyophilized. Next, it was solubilized in 5 ml of 25 mmol l−1 Tris–HCl pH 9·0 buffer (about 5 mg) and applied to a column of an anion exchanger, DEAE-Sepharose Fast Flow (2 × 10 cm), which had been equilibrated with 25 mmol l−1 Tris–HCl pH 9·0 buffer. After washing off unbound proteins, the column was eluted at the flow rate 5 ml min−1 with a linear gradient of NaCl (0–0·5 mol l−1). Ten millilitres fractions were collected. All fractions were measured at 280 nm to obtain an elution profile of the column. Fractions with the highest levels of protein concentrations were collected as separate peaks, desalted and lyophilized. The antiyeast activities of the peaks were analysed.

The biomolecules obtained after fractionation on ion-exchange chromatography were further fractionated by Sephadex G-50 medium gel filtration. After molecular sieve, the fractions were analysed by surface enhanced Raman spectroscopy (SERS) and Fourier transform infrared spectroscopy (FTIR).

Polyacrylamide gel electrophoresis

The protein antimicrobial active subfraction (AAS) obtained was analysed using native (without SDS) and SDS/PAGE electrophoresis. Polyacrylamide gel electrophoresis was performed by the method of Laemmli (1970) in 10% polyacrylamide gels for 1·5 h at 120 V. Samples containing 20 μg of protein were heated at 100°C for 10 min in sample buffer and used for electrophoresis. Spectra Multicolor Broad Range Protein Ladder, SM1841 (Fermentas) was used for the determination of the molecular mass of the proteins detected. The protein concentration was estimated by the Bradford method using bovine serum albumin as a standard (Bradford 1976). Twenty microgram of the protein was used in the samples for both types of electrophoresis. The gel was stained with Coomassie Brilliant Blue R-250. Homogeneity and MW of fractions were analysed on the gel.

Cytotoxicity assay

Cultures of human skin fibroblasts were incubated in Falcon plastic dishes in RPMI supplemented with 10% foetal bovine serum (FBS; Sigma) with antibiotics, at 37°C, 5% CO2 and 90% humidity. 1000 cells from this culture were collected for each recess of a 96-well plate and placed in the culture medium. The fraction tested was added to each well at a concentration 10 μg ml−1 and incubated under standard conditions. The effect was determined after 24, 48 and 72 h, using the BrdU assay (Roche). Cells were visualized on a Zeiss Axiovert 40 CFL microscope (Carl Zeiss) with magnification ×200.

Chemical identification of active subfraction using FTIR

Fourier transform infrared spectroscopy, an analytical technique used for the identification of organic materials, was used for identification of the compounds of the active subfraction (Kumar and Prasad 2011). The FTIR Nicolet 8700A spectrometer and FTIR microscope iN10MX (Thermo Scientific) were used. This technique measures the absorption of infrared radiation by the sample material vs wavelength. Infrared absorption bands identify molecular components and structures. To identify the material analysed, the unknown IR absorption spectrum is compared with standard spectra in computer databases or with a spectrum obtained from a known substance.

SERS analyses of the compounds from the active subfraction

In our research, surface enhanced Raman spectroscopy was used for the measurements. This technique involves measurement of radiation of Raman scattering of particles adsorbed on the surface of metal or metal sol particles. This technique results in significant strengthening of the measured signal in relation to the classical Raman spectroscopy. SERS is an attractive approach for the identification of Raman-active compounds and biological materials. Using SERS, the chemical features of the surface of an active substrate can be detected and analysed in an extremely sensitive manner (Liu et al. 2009).

In this study, measurements were carried out by placing the sample on a Renishaw Diagnostics Klarite plate. Klarite plate is a substrate that features a systematically designed submicron scale patterning of a gold-coated silicon surface.

The instrumentation used in this study was a Renishaw InVia Raman Microscope. Excitation was provided using a 785 nm semiconductor laser with 300 mW output and 150 mW at the sample. A microscope objective at magnification ×50 in a 180 backscatter collection configuration was used. The spectra obtained were processed (smoothed and baseline corrected) using the Wire 2·0 Renishaw software (Renishaw, Gloucestershire, UK).

Statistical analysis

The results are expressed as mean SD (standard deviation). Differences between the means were tested with one-way anova and the post hoc Tukey's test as the correction for multiple comparisons. Normal distribution of data was examined using the W. Shapiro–Wilk test, and equality of variance was tested by Levene's test. The P-value of <0·05 was considered statistically significant.

Results

In the experiments, we observed the effect of the active antifungal fraction and subfraction obtained from the culture supernatant of the gut bacterium of the earthworm D. veneta on C. albicans; additionally, the compounds of the active protein subfraction were identified.

Determination of the minimum inhibitory concentration

The Antimicrobial active fraction (AAF), subfraction (AAS) and amphotericin B were used to find the minimum inhibitory concentration (MIC) in vitro of the C. albicans wild strain. The MIC of AAF was 1·25 mg ml−1 and of AAS was 2·5 μg ml−1; and MIC of amphotericin B was 0·25 μg ml−1.

Morphological changes of C. albicans cells after AAF action

The morphology of C. albicans cells in the control culture grown for 72 h is shown in Fig. 1a,b. The untreated control cells were oval, regular in shape, and they occasionally formed buds. The tube-like, filamentous hyphae indicated that some of the cells had elongated in a manner that was similar to mycelial growth, with regular formation of septa that were clearly visible in the calcofluor-stained preparations (Fig. 1a). Calcofluor is a well-known fluorochrome binding preferentially to areas containing chitin (Roncero et al. 1988). The fluorescence of the untreated control cells was weak, and the septa were recognized as bright fluorescent lines in the hyphae and at the junction between the mother cell and the buds or daughter cells. The septa were intensely fluorescent in the calcofluor-stained preparations, suggesting that these were the chitin-rich regions. A majority of the oval cells contained big vacuoles, which are clearly visible in Fig. 1b. The vacuoles usually occupied about half of the whole cell volume. In the same figure, the hyphae are divided into segments with elongated vacuoles.

Figure 1.

Morphological and cell structure changes in Candida albicans after the action of the active fraction (AAF) of Raoultella ornithinolytica observed under CLSM microscopy (×10 000 magnification) after staining with calcofluor white (left column) and using two-dimensional scan in Nomarski contrast (right column); (a, b) Control cells of C. albicans, (c, d) C. albicans after incubation with the AAF at the concentration of 1·25 mg ml−1, (c- short and elongated and enlarged cells are marked with arrows and the chain cell structure is pointed with the arrowhead; (d) cells with enlarged vacuoles are marked with arrows and thicker hyphae are pointed with the arrowhead). (e, f) C. albicans after incubation with the AAF at the concentration of 2·5 mg ml−1, (e – ballooned cells are pointed with arrows and thicker hyphae are pointed with the arrowhead; f – ballooned cells with big vacuoles are pointed with arrows and thicker hyphae are pointed with the arrowhead).

After supplementing the yeast culture with the AAF at the final protein concentration of 1·25 mg ml−1, in most cases, the yeast cells became more differentiated in size. The fungal cells frequently became swollen and formed chains of interconnected cells (Fig. 1c). In comparison with the control culture, the single cells of C. albicans formed numerous short, elongated hyphae. The hyphae were also of different length and sporadically formed chains of three connecting cells. In comparison with the control culture, the cells of C. albicans were filled with much bigger vacuoles (Fig. 1d). The vacuole usually occupied most of the cell volume both in single cells and in the cells germinating into hyphae. Numerous vacuoles (bigger and more numerous than in the control) were present in the elongated, short hyphae.

The morphological changes were more pronounced after adding the AAF at the final concentration of 2·5 mg ml−1. Short hyphae were formed more frequently and were found more often than after using a lower protein concentration (Fig. 1e). More cells were visibly ballooned. The vacuoles in the maternal cells, which grew into the hyphae, were very big, while, in the hyphae themselves, the vacuoles were more numerous and smaller (Fig. 1f).

Comparison of the action of the AAF with amphotericin B

The relative metabolic activity of C. albicans cells after incubation with AAF and amphotericin B was significantly decreased in comparison with the control, but the action of the antifungal antibiotic was more effective. The metabolic activity was decreased by 41% after the AAF treatment at the concentration of 1·25 mg ml−1 and by 46% after 2·5 mg ml−1, in comparison with the control. After incubation of C. albicans with amphotericin B at the concentration of 0·25 μg ml−1, the metabolic activity was lower by about 77%, and at 0·50 μg ml−1, the metabolic activity was inhibited almost completely – by 94%, in comparison with the control culture (Fig. 2a). The differences were significant at P < 0·05 vs the control group.

Figure 2.

The relative metabolic activity of Candida albicans cells after incubation with the active fraction (AAF), amphotericin B (A) and fraction (AAS) (B): A (a) C. albicans control cells, (b) C. albicans after incubation with AAF at the concentration of 1·25 mg ml−1, (c) C. albicans after incubation with AAF at the concentration of 2·5 mg ml−1, (d) C. albicans after incubation with amphotericin B at the concentration of 0·25 μg ml−1, (e) C. albicans after incubation with amphotericin B at the concentration of 0·50 μg ml−1; B (a)- C. albicans control cells, (b) C. albicans after incubation with AAS at the concentration of 2·5 μg ml−1, (c) at the concentration of 5 μg ml−1, (d) at the concentration of 10 μg ml−1. Bars represent ±SD calculated from three independent experiments; *P < 0·05 vs control group. LIVE/DEAD Yeast Viability Kit F-7030 FUN 1 was used for the determination of C. albicans metabolic activity.

Next, more effective doses of AAF and amphotericin B were used for comparison of the morphology of C. albicans cells and the vacuolization effect (Fig. 3a–c). After incubation of C. albicans with AAF (2·5 mg ml−1), the fungal cells formed cell clusters (Fig. 3b), but in the presence of amphotericin B, (0·50 μg ml−1) they were usually visible as separate cells (Fig. 3c). After the action of AAF, C. albicans showed extensive intracellular vacuolization. A majority of the oval cells observed had enlarged vacuoles and, additionally, the long filaments included these visible organelles (Fig. 3b). After incubation with amphotericin B, cells with big, clear vacuoles and chain cell structures were observed (Fig. 3c). However, after the action of the antifungal antibiotic, the vacuoles were not as much ballooned as after the action of AAF.

Figure 3.

Candida albicans cells after incubation with the active fraction from the Raoultella ornithinolytica culture [antimicrobial active fraction (AAF)] and amphotericin B. (a) C. albicans control cells, (b) C. albicans after incubation with antimicrobial active fraction at the concentration of 2·5 mg ml−1, (c) C. Albicans after incubation with amphotericin B at the concentration of 0·50 μg ml−1. Arrows point to the cells with enlarged vacuoles and the arrowhead points to the chain cell structure. Microscopy Zeiss/LEO LEO 912AB with two-dimensional scan (×10 000 magnification) was used.

Morphological changes in C. albicans exposed to protein fractions with different molecular mass

In comparison with the control cells, changes in C. albicans cells were observed after incubation with the protein fractions obtained after ultrafiltration of the precipitated R. ornithinolytica culture supernatant. The observations were conducted using two-dimensional scan with Nomarski contrast. The fractions contained proteins with the molecular mass 14–50 kDa, 50–100 kDa and above 100 kDa. After incubation with the 14–50 kDa and above 100 kDa fractions, most C. albicans cells contained swollen, clear vacuoles occupying almost the entire volume of cells. After the action of the 50–100 kDa protein fractions, few cells contained enlarged vacuoles and the organelles were less visible. Additionally, we observed that after exposure of C. albicans cells to fraction above 100 kDa, some cells were ballooned, and cell clusters were formed. The fungal cells of this fraction only showed strong adhesive properties, and therefore, this protein fraction was used for further analysis.

Morphological analysis of C. albicans after the action of the AAS using fluorescence microscopy

The morphology of the control culture of Calbicans after staining with calcofluor white shown in Fig. 4a was typical for this species. The cells formed buds only occasionally. All the control yeast cells stained with calcofluor white were regularly oval and usually similar in size. Their walls of untreated cells were thin, and fluorescence was evenly bright around each cell and uniformly distributed on the peripheral and septal walls (Fig. 4a).

Figure 4.

Morphological and cell structure changes in Candida albicans after the action of the active subfraction (AAS) at the concentration of 2·5, 5, 10 μg ml−1 observed under CLSM microscopy (×10 000 magnification) after staining with calcofluor white. (a) C. albicans control cells, (b) C. albicans after incubation with AAS of 2·5 μg ml−1, (c) after incubation with AAS of 5 μg ml−1, (d) after incubation with AAS of 10 μg ml−1.

After incubation of C. albicans in the presence of subfraction AAS at the concentration of 2·5, 5 and 10 μg ml−1, the yeast cells were ballooned and formed chains consisting of three or four cells that differed in size and shape (Fig. 4b,c,d). After the addition of AAS at the concentration of 5 and 10 μg ml−1, it was revealed that the cell walls surrounding single cells were irregular in thickness (Fig. 4b,c). The clearly thickened walls were at the junction of two maternal and daughter cells (Fig. 4c,d). The biggest changes in the morphology of yeast cells were observed after administration of AAS at the concentration of 10 μg ml−1 to the culture. Besides the fact that the single cells differed in size, the cells formed branched chains (Fig. 4d). After the treatment of the yeast cells with 10 μg ml−1, the cell walls showed uneven fluorescence. The fluorescence of the walls of ballooned cells was brighter than the fluorescence of unchanged, typical cells.

The size of 100 C. albicans cells from the control culture and from the culture after incubation with AAS at the concentration of 5 μg ml−1 was measured. The length and width of morphologically similar cells were compared. In the control culture, the average length of cells was 5·56 μm, and the average width was 4·65 μm. In the culture after incubation with AAS, the sizes were 6·33 μm and 5·03 μm, respectively. The differences were statistically significant at P < 0·001.

After incubation of C. albicans cells with separate ingredients (polypeptides and polysaccharides separately) of the analysed subfraction (AAS), no morphological changes were observed. The subfraction acted only as a glyco-protein complex.

Morphological analysis of C. albicans after the action of the AAS using scanning electron microscopy

Electron micrographs of the untreated control culture of C. albicans and the culture after incubation with AAS observed by scanning electron microscopy (SEM) are presented in Fig. 5. The untreated cells showed a characteristic ovoid shape with a smooth surface (Fig. 5a1,a2). The average diameter was 3–5 μm. Figure 5b1,b2 show a SEM image of cells treated with AAS at the concentration of 2·5 μg ml−1. Most cells were normal in contour, although swollen and deformed cells were also observed. Fig. 5c1,c2 show deformed, ballooned and elongated cells. The cells were sticking together and formed clumps or agglomerates. The SEM image of the cells after exposure to 10 μg ml−1 AAS showed chains of oval cells with sometimes wrinkled surfaces Fig. 5d1,d2. Probably, the cell wall at the junction of the cells was dissolved.

Figure 5.

Scanning electron microscopy image of a1,2 – untreated control culture of Candida albicans, b1,2- culture after incubation with AAS of 2·5 μg ml−1, swollen and deformed cells are observed, c1,2 – after incubation with AAS of 5 μg ml−1, deformed ballooned and elongated cells formed clumps, d1,2- after incubation with AAS of 10 μg ml−1, chains of oval cells with sometimes wrinkled surfaces; (×10 000 magnification).

Quantification of yeast viability after incubation with the AAS

After incubation of C. albicans cells with the AAS at the concentration of 2·5, 5 and 10 μg ml−1, the relative metabolic activity of fungal cells was significantly decreased (Fig. 2b). After the action of the protein subfraction obtained at the concentration of 2·5 μg ml−1, the metabolic activity was diminished by 15%, at the concentration of 5 μg ml−1 – 30%, and after 10 μg ml−1 – 35%. The differences were significant at P < 0·05 vs the control group. It was observed that the C. albicans cells connected in chains were not metabolically active.

Biochemical analysis of the AAS

A multiprotein band appeared in the analysis of AAS using native electrophoresis (Fig. 6a). The SDS/PAGE electrophoresis revealed several high molecular proteins with the molecular mass of 35, 38, 41, 43, 56, 65 and 68 kDa (Fig. 6b).

Figure 6.

Analysis of the active isolate obtained after ion-exchange chromatography from the fraction containing proteins above 100 kDa, (a) using native electrophoresis; arrow points to the multiprotein band; (b) using SDS/zelectrophoresis; Spectra Multicolor Broad Range Protein Ladder, SM1841 (Fermentas) were used to determine the molecular weight. Twenty micrograms of protein was used in the sample. Analyses were performed in 10% polyacrylamide gels.

Antimicrobial active subfraction has trace amounts of lipids and oligosaccharide groups. It has no proteolytic or lysozyme activity. AAS exhibits inhibitor activity against pepsin-like enzymes and weak activity against trypsin-like enzymes.

Cytotoxicity determination of the AAS

The study subfraction shows a minimal cytotoxic effect on human skin fibroblasts. After 24 and 48 h of incubation with the AAS at the concentration of 10 μg ml−1, there was no cytotoxicity, whereas after 72 h, cytotoxicity was found at 1% (Fig. 7).

Figure 7.

Cytoxicity effect of the AAS on the human skin fibroblasts. (a) control fibroblasts, (b) fibroblasts after incubation with AAS (10 μg ml−1) for 72 h. Cells were visualized on a Zeiss Axiovert 40 CFL microscope (Carl Zeiss) with magnification ×200.

Chemical identification of the AAS using FTIR

Using Fourier transform infrared spectroscopy, the substances of the antifungal subfraction (AAS) were identified as a polysaccharide with 73% similarity to dextran and 83% similarity to protein (Fig. 8).

Figure 8.

Identification of the anti-Candida albicans active substance subfraction using fourier transform infrared spectroscopy (FTIR). (a) Identification of a substance similar to protein: X1 – the substance obtained, X2 – protein from Library-NICODOM IR, Index No 11856 (83% similarity); (b) Identification of a substance similar to dextran: Y1 – the substance obtained, Y2 – dextran from Library- Sadtler ATR of Polymers W, Index no. 1526 (73% similarity).

SERS analyses of the compounds from the AAS

The SERS analysis confirmed that the compounds obtained were protein and carbohydrate (Fig. 9a,b). The main SERS peaks of the protein obtained corresponded to band assignment of bacterial proteins (Table 1). In the case of the dextran analysis, seven peaks corresponded to the known peaks of carbohydrates (Table 2). The last three peaks 1394, 1506, 1591 cm−1 probably originate from contamination of the protein obtained, as the peaks were also detected in the spectrum of the protein.

Figure 9.

Surface enhanced Raman spectroscopy (SERS) analyses of the substances from anti-Candida albicans active subfraction. (a) The SERS spectrum of the obtained protein; (b) The SERS spectrum of the obtained dextran.

Table 1. Assignment of the main surface enhanced Raman spectroscopy (SERS) vibrational bands observed in the obtained protein Raman spectra
 Raman shift (cm−1)Assignments
LabelStrongWeakVibrational modesLocation
  1. v, stretching vibration; δ, deformation vibration.

A831 

v(C–N) from tyrosine group in proteins

v(C–C) ring breathing

Protein
B1101 

v(C–C)

v(C–N)

Protein
C 1172Aromatic amino acids δ(C–H) in tyrosineProtein
D1257 Amide IIIProtein
E1329 

δ(CH2)

δ(C–H)

Protein
F1521 V(C=C)Protein
G 1688Amide IProtein
Table 2. Assignment of the main surface enhanced Raman spectroscopy (SERS) vibrational bands observed in the obtained dextran Raman spectra
 Raman shift (cm−1)Assignments
LabelStrongWeakVibrational modesLocation
  1. v, stretching vibration; δ, deformation vibration.

A 861

v(COC)

v(CO)

Carbohydrate
B 907v(CO)Carbohydrate
C1096 δ(O–H)Carbohydrate
D1150 v(C–C)Carbohydrate
E1172 δ(C–H)Carbohydrate
F1218 δ(CH2)Carbohydrate
G 1270δ(CH2)Carbohydrate
H1394 v(COO)Protein
I1506 v(C=C)Protein
J1591 v(C=C)Protein

Discussion

We found antifungal activity directed against C. albicans of the metabolites secreted by the gut bacterium Rornithinolytica. This micro-organism was isolated from the walls of the D. veneta midgut. It was observed that the midgut fluid of earthworms contained a selectively active ingredient against soil micro-organisms (Byzov et al. 2007). Stephens et al. (1993, 1994) Stephens and Davoren (1997) showed that earthworms reduced the severity of plant diseases caused by fungal pathogens Rhizoctonia solani and Gaeumannomyces graminis. A suppressive action of the gut environment of earthworms has been demonstrated for fungal spores (Moody et al. 1996). Khomyakov et al. (2007) revealed significant inhibition of germination of Aspergillus terreus and Paecilomyces lilacinus spores and complete inhibition of the growth of colonies from hyphae of Trichoderma harzianum and Penicillium decumbens after the action of the digestive fluid of earthworms.

Khomyakov et al. (2007) also suggested that microbiocidal agents were formed in the body of the earthworm and not by the soil micro-organisms entering their digestive tract. They revealed that the digestive fluid of earthworms feeding on soil and sterile sand showed the same antimicrobial activity. Sampedro and Whalen (2007) also confirmed the suggestions that the microbial community in the earthworm gut was not a casual combination of micro-organisms already present in the soil. Wang et al. (2007) tested the antifungal activity of the coelomic fluid of earthworm Eisenia foetida and proved that the antimicrobial peptides isolated from the coelomic fluid did not show any activity against C. albicans. This confirmed the assumption that the antifungal activity in earthworms was associated with the digestive tract.

We obtained the AAF and subfraction (AAS) containing high molecular weight proteins from the supernatant of the R. ornithinolytica culture. After incubation of C. albicans cells with the protein fraction, we observed morphological changes in the vacuoles and a decrease in the metabolic activity. After exposure to the protein subfraction, the C. albicans culture also showed decreased metabolic activity, the cells formed chains, and some of the cells were swollen. After the action of AAS, the C. albicans cells were ballooned and elongated and formed chains of oval cells. A similar effect was noticed after the action of lysozyme on C. albicans cells (Nishiyama et al. 2001). These observations suggest that the target of action of the glyco-protein complex obtained is the cell wall of C. albicans.

However, in our experiments, chain formation and an increase in the C. albicans cell size were also observed after treatment with amphotericin B, an antifungal agent with vacuole-targeting activity. Amphotericin B is a very powerful antibiotic for the treatment of fungal infections. It inserts into lipid bilayers, binds to sterols and forms pores that disturb plasma membrane integrity and permit the efflux of cations such as K+. This effect is fungicidal for C. albicans (Cannon et al. 2007). The vacuole-targeting fungicidal activity of amphotericin B against C. albicans was described by Borjihan et al. (2009). Vacuole disruption was observed in parallel to amphotericin B-induced cell death when the antibiotic was used at a lethal concentration. Vacuoles are involved in osmoregulation, ion homoeostasis and the cell volume in fungal cells (Thumn 2000; Wicker 2002). Various hydrolytic enzymes, including proteases and nucleases, accumulate in vacuoles and damage these organelles. This phenomenon is considered a critical step in cell death induction (Obara et al. 2001). Necrotic cells of C. albicans have no defined nuclei and show extensive intracellular vacuolization (Phillips et al. 2003).

The effect that was not observed after amphotericin B but occurred after treatment with AAS and AAF was the tendency for cell aggregation and formation of conglomerates. These observations suggest that the properties of C. albicans cell walls were markedly changed. The cells that closely adhered to each other have lost metabolic activity. The molecular mechanism of AAS action on C. albicans cells will be explored.

The active anti-C. albicans subfraction obtained from the R. ornithinolytica culture was identified using FTIR as a glyco-protein complex that contains protein and polysaccharide. The SERS analyses confirmed the identification of the compounds from the active subfraction.

The anti-Candida action of glycoproteins is widely known. Lactoferrin, an iron-binding glycoprotein naturally presents in glandular secretions (milk, tears, saliva) has a broad-spectrum antimicrobial activity with potent activity against Candida biofilm (Bink et al. 2011). Rydengård et al. (2008) reported that histidine-rich glycoprotein also had antifungal activity. Its protein binds to Candida cells and induces breaks in the cell walls of the micro-organism. In earthworms, glycolipoprotein G-90, a protein complex isolated from the whole tissue with a unique biological activity, was found (Grdiša et al. 2001; Popović et al. 2005; Grdiša and Hrženjak 2007; Yanqin et al. 2007).

One of the components of the active glyco-protein complex was polysaccharide similar to dextran. Wang et al. (2007) found that earthworm polysaccharides (Eisenia fetida) performed an antibacterial function on plant–pathogen microbes in vitro. The polysaccharides have broad-spectrum antibacterial activities against pathogenic bacteria and fungi.

Because of the fact that the glyco-protein complex obtained showed minimal toxicity against human skin fibroblasts, the studies on it will be continued as they can lead to application of the complex in biological fungicide and pharmaceutical industry.

Acknowledgements

The research was carried out with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-06-024/09 Center of Functional Nanomaterials).

Ancillary