Eight yeast strains were used, six belonging to Candida spp., one Cryptococcus neoformans strain (CBS 132) and one Saccharomyces cerevisiae strain (W303). Candida strains used were C. albicans var. albicans (CBS 562), C. dubliniensis, C. glabrata, C. lusitaneae, C. parapsilosis (PYCC 2597) and C. tropicalis. [CBS – Centraalbureau voor Schimmelcultures; PYCC – Portuguese Yeast Culture Collection; The other strains were a kind gift of Institute of Microbiology, Faculty of Medicine of the University of Coimbra (Paulo et al., 2009)]. All yeasts were grown at 35°C for 24 h, except for C. neoformans that was grown for 72 h, in Glucose Yeast Peptone (GYP) medium [0.5% (w/v) peptone, 0.5% (w/v) yeast extract, 2% (w/v) glucose, 1.5% (w/v) agar]. For the different experiments performed, three media were used; RPMI 1640 (Applichem), pH 7.0, supplemented with 2.08% (w/v) glucose and 6.9% (w/v) MOPS [3-(N-morpholino)propanesulfonic acid]; YNB (Difco), pH 7.0, supplemented with 2% (w/v) glucose and 0.1% (w/v) MOPS; and PDB (DIFCO), buffered at pH 7.5.

Dry seeds of Lupinus albus were germinated and grown in growth chambers with a photoperiod of 16 h light/8 h dark at 18°C, for periods up to 10 days. The seed coats were removed and the intact cotyledons were dissected from the axes and stored frozen at -80°C until needed. BCO is a breakdown product of β-conglutin catabolism, and it was extracted and isolated from the cotyledons of 8-days old seedlings as described by (Monteiro et al., 2015). The protein corresponding to β-conglutin was purified by AKTA anion exchange chromatography followed by AKTA gel filtration chromatography as follows: the total globulin fraction was loaded on the Q-Sepharose column (Ø = 1 cm; h = 8 cm; flow rate = 1.5 mL/min) previously equilibrated in 20 mM Tris-HCl buffer, pH 7.5, and eluted with a linear gradient of NaCl (0 to 1 M). The fraction containing the BCO, eluted between 0.25 and 0.35 M NaCl, was subsequently subjected to gel filtration on an AKTA Superose 12 HR 10/30 column (GE Healthcare Life Sciences), equilibrated in 0.1 M Tris-HCl buffer (pH 7.5). This last purification step does not affect the polypeptide pattern of the protein, but removes unidentified low molecular mass compounds, resulting in a high pure BCO sample.

The BCO was extracted and purified as described above and stored lyophilized at room temperature (RT). AMB was obtained from their respective manufacturers and stock solutions were prepared and stored frozen at -20°C until used.

A sample of BCO was lyophilized and ressuspended in Freund adjuvant. New Zeland female rabbits and rats were immunized with the purified BCO sample. To obtain a high titer, three boosters injections of 100 μg/mL of antigen each were given every 2 weeks in complete Freund’s diluted 1:10 with incomplete adjuvant. Total blood was taken from the heart 12 days after the third booster injection. Blood samples were allowed to clot, and the serum was collected, centrifuged, and stored at -70°C. To purify the IgG present in the serum, a chromatography in a Protein G sepharose column was conducted.

Susceptibility tests were made according to the CLSI – Clinical and Laboratory Standards Institute (former NCCLS – National Committee for Clinical Laboratory Standards) guideline M27-A2 (NCCLS, 2002) with some adjustments, using the broth microdilution method. Yeast cells were grown on GYP medium and the inoculum suspension was prepared by picking fresh colonies and resuspending them in 5 mL of sterile 0.9% (w/v) saline (NaCl). The resulting suspension was vortexed for 15 s and the cell density was adjusted with a spectrophotometer to give an inoculum concentration of 10⁶ cells per mL. The final inoculum suspension was prepared by a 1:50 dilution followed by a 1:20 dilution with double-strength broth medium, which resulted in a final concentration of 10³ cells per mL. One other final inoculum concentration was tested in C. albicans, 10⁵ cells/mL, achieved by a 1:10 dilution with double-strength broth medium, for allowing a sufficient number of cells to be visualized under the microscope and to enable the determination of a MFC based on a 99.9% killing (see Minimum Fungicidal Concentration section below). The inoculum size was verified by enumeration of CFUs obtained by subculturing on GYP plates. The solution of the BCO was prepared in ultrapure sterile water and 200 μL were added to the first line of the microplate. A serial two-fold dilution was made, twelve times, using ultrapure sterile water, in the 96-well microplates. The final concentration of the BCO, after addition of the inocula, ranged from 0.002 to 4.762 μM when using PDB medium and from 0.012 to 23.81 μM with RPMI medium. The serial twofold dilutions of AMB ranged from 0.03 to 17.31 μM. The yeast inoculum (100 μL) was added to each well of the microplate, containing 100 μL of the drug solution (twofold concentrated). The final volume in each well was 200 μL. The microplate was incubated at 35°C and examined after 72 h. Minimum inhibitory concentrations (MICs) are the lowest drug concentration showing absence of growth, as recorded visually. To evaluate the effect of sorbitol on the fungal susceptibility to the BCO, the growth medium (twofold concentrated) was supplemented with 2.4 M sorbitol (final concentration 1.2 M). All these tests were performed with three different batches of the BCO (triplicates).

Minimum fungicidal concentrations were determined according to (Espinel-Ingroff, 1998). After MIC determination, as previously described, 30 μL aliquots were subcultured from each well that showed no visual growth onto GYP plates. This procedure was performed to minimize drug carryover effects (Espinel-Ingroff, 1998). The plates were incubated at 35°C for 24 h. All these tests were performed with three different batches of the BCO (triplicates).

The non-existence of a standard method for determining MFCs in yeasts as led to an indiscriminate use of a wide range of methodologies for determining this value (Vazquez et al., 1997; Johnson et al., 1998; Espinel-Ingroff, 2001). To maintain the standardized methodology for determining MIC, an initial inoculum of 10³ CFU/mL was used, although it does not allow the detection of 99.9% killing. To minimize this restriction, the MFC considered in this study was the lowest drug concentration where no growth was observed after plating 30 μL on GYP plates (0 CFUs). For achieving a 99.9% killing an initial inoculum size of 10⁵ CFU/mL was tested for C. albicans only.

The effect of the BCO on the growth of human fungal pathogens was evaluated by using C. albicans as a model organism and by comparing the results observed with those of AMB, in PDB pH 7.5. The assays were conducted in the presence of different BCO and AMB concentrations. A cell suspension was grown overnight in 20 mL of PDB pH 7.5, at 35°C, 150 rpm and refreshed in 20 mL of PDB pH 7.5, approximately 5 h before addition to the culture medium. To obtain an initial concentration of approximately 10⁵ CFU/mL, the OD_{640 nm} was adjusted to 0.1 and then 10-fold diluted with PDB pH 7.5, to a final volume of 100 mL, in 500 mL Erlenmeyer flaks. The cultures were incubated at 35°C without shaking. At regular intervals, samples were collected for absorbance measurements, viable cell counts and morphological evaluation. For viable cell counts, 30 μL aliquots of the culture were taken, diluted if needed, and plated on GYP agar plates. Each time-kill curve was performed in triplicate and a representative curve is shown in the results.

The effect of the BCO on the yeast cell volume was evaluated using S. cerevisiae (W303) as a yeast model. Two yeast cultures were grown as described in the Time-kill curves section, one in the presence of the BCO and the other kept as control. At regular intervals, samples were collected and an estimate of the cell volume was made, by measuring the diameter of 100 cells. The cell volume was calculated considering the shape of the cells as a sphere. At each sampling time the average volume of the 100 cells was calculated as well as the corresponding standard deviation.

The haemolytic activity of the BCO was analyzed according to (Ling et al., 2015). Briefly, fresh red blood cells from rabbit were collected and washed with PBS until the upper phase was clear after centrifugation. The pellet was ressuspended in PBS to an OD_{600 nm} = 24 and added to a 96-well microplate. The solution of the BCO was prepared in ultrapure sterile water and a serial twofold dilution was made in water and added to the wells. The final concentration of the BCO ranged from 0.04 to 4.76 μM. After 1 h incubation at 37°C, cells were centrifuged at 1000 g and the supernatant was diluted and measured at 450 nm in a BioTek’s Take3^TM Multi-Volume Plate spectrophotometer.

The LIVE/DEAD^® Yeast Viability Kit (Molecular Probes) was used to evaluate fungal viability. A FUN1 100 μM working solution was prepared in 10 mM MOPS buffer, pH 7.2, with 2% (w/v) glucose. A 50 μM calcofluor white working solution was prepared in distilled water. Forty microliter of fungal culture and 5 μL of FUN1 working solution were mixed thoroughly and incubated at 30°C in the dark. After 30 min, 5 μL of calcofluor white working solution were added to the culture and mixed thoroughly. Five microliter of the cell culture were trapped between a microscope slide and a coverslip for visualization on a fluorescence microscope.

After 24 h incubation with the BCO, under the same conditions as described in the Time-kill curves section, cells were incubated with propidium iodide at a final concentration of 7.5 μM, for 10 min at 4°C. Five microliter of the cell culture were trapped between a microscope slide and a coverslip, for visualization on a fluorescence microscope.

Morphological changes in C. albicans cells in the presence of the BCO were assessed by Transmission Electron Microscopy (TEM). The culture was prepared and kept under the same set of conditions as previously described in the Time-kill curves section. At regular intervals, samples were collected, washed twice with saline [9.5% (w/v) NaCl] and concentrated by centrifugation (3500 g for 10 min) to a final concentration of 1 to 5 × 10⁶ CFU/mL. Cell concentration was confirmed at all points by plating in GYP agar. The samples were collected and fixed in 2% (v/v) glutaraldehyde and 2.5% (v/v) paraformaldehyde in 0.1 M sodium cacodylate buffer pH 7.4. Then, they were post fixed in 2% (w/v) OsO₄, dehydrated and embedded in epon. The ultrathin-sections (60 nm) were counterstained with aqueous uranyl acetate solution and lead citrate. Controls were prepared under the same conditions, but in the absence of the BCO.

Immunolocalization of the BCO was accomplished by both indirect immunofluorescence and immunogold methods. In both cases the culture was prepared and kept under the same conditions as previously described in the Time-kill curves section.

Immunofluorescency studies were accomplished according to (Lawrence et al., 2004) with some modifications. After 24 h incubation with a lethal concentration of the BCO, the culture was concentrated by centrifugation (3500 g for 10 min) to a final concentration of 1 to 5 × 10⁷ CFU/mL followed by treatment with lyticase (0.4 mg/mL in 500 mM Tris-HCl, 1 M sorbitol, 0.8 M KCl, 10 mM MgS0₄, pH 7.5) for 2 h at 30°C, to digest the cell wall, allowing the subsequent cell membrane permeabilization. The culture was washed with PBS (137 mM NaCl, 1.5 mM KH₂PO₄, 8.1 mM Na₂HPO₄ and 2.7 mM KCl) followed by fixation in 4% (v/v) formaldehyde, for 30 min, at 30°C. After two washes with PBS and PBS containing 0.1% (v/v) Triton X-100 for cell membrane permeabilization, cells were blocked with bovine serum albumine (BSA) 5% (w/v) in PBS containing 0.1% (v/v) Triton X-100 for 30 min. Cells were washed with PBS and incubated with the first antibody (anti-BCO), produced in rabbit and diluted 1:500 in PBS containing 0.1% (v/v) Triton X-100 and 0.1% (w/v) BSA, for 16 h at 4°C. The cells were then washed in PBS and incubated with a second, anti-rabbit antibody, produced in goat, conjugated with FITC and diluted 1:80 in PBS with 1% (w/v) BSA, for 1 h at 37°C. After washing twice with PBS for 15 min, 50 μM calcofluor white were added and 5 μL of the cell culture were trapped between a microscope slide and a coverslip, for visualization on a confocal microscope.

For immunogold analysis, the samples were collected after 6, 12, and 24 h of incubation with the BCO and subsequently fixed in 0.1% (v/v) glutaraldehyde in 0.1 M sodium cacodylate buffer pH 7.4 for 1 h. Then, they were dehydrated and embedded in LRWhite. Thin sections on TBS were immunolabeled after incubation on etching process: grids were incubated in a humid chamber in large drops of a saturated aqueous solution of sodium metaperiodate, for 1 h at RT. After washing the grids, they were first incubated for 20 min in 2% (w/v) gelatin in TBS, and a second 5 min incubation was done with 15 mM glycine. The grids were blocked with a solution containing 2% (w/v) immunoglobulin-free BSA. Sections were then incubated overnight (16–18 h) with the first antibody (anti-BCO) diluted 1:250 in TBS containing 2% (w/v) BSA and 1% (v/v) Tween-20/3% (w/v) NaCl. The grids were then washed by floating them on drops of 0.1% (w/v) BSA /TBS (four changes, 2 min each) followed by 20 min incubation on TBS with 1% (w/v) BSA. Bound antibodies were visualized by incubating the sections for 1 h with the second antibody-gold conjugate (10 nm diameter particles) diluted 1:25 in PBS with 1% (w/v) BSA, 1% (v/v) TBS. Finally, grids were washed on drops of water (six changes, 10 min each). The immune-complexes formed were visible as little black dots when observed by TEM.

(I) Fluorescence microscopy. Samples were observed under a fluorescence microscope (Axioscope A1 with phase contrast and epi-fluorescence, Zeiss) equipped with a camera (AxioCam ICm1, Zeiss), using three different filters: Filter Set 49 DAPI (Excitation G 365, Emission BP 420/470); Filter Set 10 FITC/GFP (Excitation BP 450-490, Emission BP 515-565) and Filter Set 15 Rodhamine (Excitation BP 540-552, Emission LP 590). (II) Confocal microscopy. The images were acquired with a Leica TCS SP5 II confocal microscope equipped with an objective HCX PL APO CS 63x/1.3 Glycerol (Leica Microsystems, Germany). Samples were excited by 488 nm laser line and emitted signal detected in the range of 496-564 nm with a HyDet detector (fluorescence channel) and a PMT detector (transmission channel). Images were acquired with 512 × 512 pixels and a pixel size of 68 nm. (III) Transmission electron microscopy. The sections were examined under a JEOL JEM 1400 TEM 120kV (Tokyo, Japan). Images were digitally recorded using a CCD digital camera Orious 1100W Tokyo, Japan.

The antifungal activity of three different batches of the BCO was evaluated in six strains belonging to six Candida species and in one strain of C. neoformans, with each strain cultured in two different growth media. C. albicans CBS 562 was also tested for resistance to AMB, one commonly used antifungal drug. The MICs and MFCs of the BCO are shown in Table 1. MIC values in RPMI medium for the entire group varied from 0.19 to 2.98 μM. Regarding the MFCs of the BCO, the values obtained were higher than the highest concentration tested. The exception was the strain of C. albicans, for which the MFC of the BCO was 23.81 μM (Table 1).

When the antifungal activity of the BCO was assessed in PDB medium, the MIC and MFC values obtained were more consistent among the different species, which is in accordance with previous studies that demonstrated that PDB is a more suitable medium for testing the bioactivity of this oligomer (Monteiro et al., 2015). All strains tested presented a similar susceptibility to the BCO, with MIC values ranging from 0.08 to 0.31 μM (Table 1). Regarding the MFCs of the BCO in PDB medium, the values varied from 0.15 to 2.38 μM. For the strains of C. albicans, C. dubliniensis and C. neoformans, the MFCs were always only twice of the respective MICs and for those of C. glabrata and C. lusitaneae were always four times higher. The MFC of BCO for C. parapsilosis was too much higher than the other ones (2.38 μM) (Table 1). Overall the results obtained for the BCO are indicative of a potent antifungal activity for these fungal species.

The antifungal activity of the BCO was then compared to that exhibited by AMB, in PDB medium, using C. albicans as a control model. This specie has long been used as a model in several fungal research studies (Fu et al., 2008; Kabir et al., 2012), and also showed a strong susceptibility to the BCO, as demonstrated by the corresponding MIC and MFC values (Table 1). Since the optimum inoculum density for the subsequent microscopy tests was found to be 10⁵ CFU/mL, and since this concentration allows the detection of a 99.9% killing based MFC, both minimum inhibitory and fungicidal concentrations were determined with this inoculum size. The results are presented in Table 2. As expected, there was an increase in MIC and MFC values, due to a higher number of cells in the initial inoculum. However, these results are still consistent with those obtained before (Table 1). The concentration of the BCO needed to induce death of 99.9% of the microorganisms was only twice the MIC (0.60–1.19 and 1.19–2.38 μM, respectively). The same effect was observed for AMB (MIC and MFC values of 1.1 and 2.2 μM, respectively), which is also consistent with the values reported in the literature (Manavathu et al., 1998; Spreghini et al., 2012).

Given the demonstrated affinity of the BCO to chitin (Monteiro et al., 2015), two studies were conducted for assessing its possible effect in cell wall biosynthesis. S. cerevisiae was used as the model organism for these studies. The cells were exposed to BCO at the MIC value and an estimate of the cell volume at three-time sampling points was made. The results obtained are described in Figure 1 and show that the presence of the BCO leads to a progressive increase in the average volume of the cells. The volume variation between the first and the last sample analyzed is higher than 10 fold (147 μm³ after 1 h of incubation and 1621 μm³ after 48 h).

Considering the results above, if the BCO mode of action is at least in part based on cell wall damage, it is expected for the cell to suffer an increase of volume under hypotonic conditions, until eventually bursts. If this is the case, under isotonic conditions, or even in a slightly hypertonic medium, this effect should not occur, since there is no water entering into the cells. With the purpose of testing if the osmotic stabilizer sorbitol counteracts the toxicity of the BCO, 1.2 M sorbitol was added to the culture medium (creating a slightly hypertonic condition). Tests were carried out in the absence and in the presence of 1.2 M sorbitol in S. cerevisiae. The results demonstrated that the presence of 1.2 M sorbitol in the culture medium did not reduce the antifungal effect of the BCO, since the MIC values were the same in both cases (0.15 μM). This result suggests that ultimately, the toxicity of the BCO to fungi is not dependent on dramatic changes in cell wall integrity despite its ability to bind very tightly to the chitin polymer (Monteiro et al., 2015).

Determination of the “killing” of an isolate over time by one or more antimicrobial agents under controlled conditions is known as the time-kill method (Pfaller et al., 2004). It is a broth based method where the rate of killing of a fixed inoculum is determined by sampling control (organism, no drug) and antimicrobial agent-containing tubes or flasks, at certain time intervals, and determining the survivor colony count (CFU/mL) by spreading each sample onto an agar plate. In order to study the effect of the BCO on the growth of C. albicans, time-kill curves were performed in PDB medium. Several samples were taken during these experiments in order to assess the evolution of the number of viable cells (OD_{640 nm} and CFU counts). Two concentrations of BCO were used, 1.19 and 2.38 μM, corresponding to the minimum inhibitory and fungicidal concentrations, respectively, as determined previously (Table 2). A fraction of the culture grown under the same conditions but without BCO was tested for control purposes. The results are shown in Figure 2 for the BCO and in Figure 3 for AMB.

Figure 2 shows that the addition of BCO to the culture medium had a strong effect in the growth of C. albicans. It is possible to observe that the culture grown in the absence of BCO presents a normal growth curve, being in the exponential phase of growth for approximately 12 h, before entering the stationary phase. This was observed by following both OD_{640 nm} readings and CFU/mL counts. Cells grown in the presence of the MIC of the BCO, showed a decrease in the growth rate when compared to the control, which resulted in a lower final optical density (Figure 2A) and a lower final CFU/mL count (Figure 2B). This result indicates that the concentration of the BCO tested had, indeed, the ability to inhibit the growth of this microorganism. Cells grown in the presence of the minimum fungicidal concentration became non-viable after 16 h of growth. This was observed by both stabilization of OD_{640 nm}, just after 4 h of incubation (Figure 2A), after a slight initial growth, and absence of CFU counts (Figure 2B).

The same assay was performed using AMB as the antifungal agent. Two different concentrations were also used, 1.1 and 2.2 μM, corresponding to the minimum inhibitory and fungicidal concentration, respectively. The results obtained were very similar to those obtained for the BCO and are shown in Figure 3. The control fraction stayed in exponential phase for 12 h, and the fraction exposed to the MIC showed a total absence of growth. When using the minimum fungicidal concentration of AMB cells became non-viable in the first 4 h of exposure, which was observed by the absence of CFU counts (Figure 3B). The prompt reduction in C. albicans viability caused by AMB is in accordance with the data published in the literature (Cantón et al., 2004; Leite et al., 2014). The major difference between the BCO and AMB in terms of killing kinetics is that both fungistatic and fungicidal activities of AMB, under these conditions, act more rapidly than those of the oligomer.

The effect of the BCO on the viability and cellular integrity of yeasts was evaluated using C. albicans as model and was assessed using samples collected along the growth curve, in PDB medium, under three different conditions: without drug (control), with the inhibitory (MIC) concentration (1.19 μM) and with the lethal (MFC) concentration (2.38 μM). Each sample was stained with FUN-1 and calcofluor white and visualized in a fluorescence microscope. FUN-1 binds to nucleic acids producing a yellowish green fluorescence in death cells with a damaged membrane. Cells without metabolic activity but with an intact plasma membrane also present a diffuse green coloration in the cytoplasm. On the other hand, in metabolically active cells, formation of orange cylindrical structures designated CIVS (Cylindrical IntraVacuolar Structures) is observed inside vacuoles. CIVS formation only occurs in metabolically active cells with an intact plasma membrane, meaning they are not observed in dead cells (Chew et al., 2015). Calcofluor white is a compound with high affinity to chitin and is normally used as a marker of the fungal cell wall. The results are shown in Figures 4 and 5.

Figure 4 suggests that during the first 4 h of incubation with the BCO there are no changes regarding the viability and integrity of the cells, for all conditions tested, since the presence of CIVS indicates metabolic activity and the fluorescence with calcofluor white is normal and equal to the control fraction, indicating cell wall integrity. These results are consistent with the ones obtained in the growth curves (Figure 2).

After 12 h of incubation with the BCO, the control fraction continued to exhibit CIVS in the majority of the cells, confirming that they were still metabolically active. The fraction incubated with an inhibitory concentration of the BCO showed a slight lower number of cells with CIVS, meaning that some cells were metabolically active and, therefore viable and culturable, thus explaining the small increase in the OD_{640 nm} and CFU counts observed in the curve of Figure 2. At 12 h, cells incubated with the lethal concentration of the BCO presented very few CIVS, corresponding to a decrease in the metabolic activity (data not shown). This explains the decrease also observed in the CFU counts observed in Figure 2.

Microscopical observations performed at 16 h of incubation revealed to be a turning point in cell viability, which is in accordance to what is also observed with other antifungal drugs (Kim et al., 2011). Although both the control and MIC fractions of the BCO showed no changes as compared to the previous time point studied, the culture incubated with a lethal concentration of the BCO no longer presented visible CIVS. Only a diffuse green coloration in the cytoplasm was visible, corresponding to the absence of metabolic activity (Figure 5). However, when cultivated in a free BCO medium some cells were still able to grow (Figure 2B). At 24 h of incubation, the last time point studied, the control fraction presented some cells without CIVS, typical of an old culture and the MIC fraction presented even fewer metabolically active cells than in the previous time point studied (data not shown). This is in accordance with the stabilization of OD_{640 nm} and with the smaller number of culturable cells observed in the growth curves (Figure 2). The results obtained with the lethal concentration of the BCO were similar to those obtained after 16 h (cells without any metabolic activity), but at this point there were also no records of culturable cells (Figure 2B). This means that approximately between 16 to 24 h of incubation with a lethal concentration of the BCO, C. albicans lost the ability to grow in a free BCO medium, and that the number of culturable cells was beneath the detection limit of the method (<10 CFU/mL). These results suggest that after 16 h of incubation with a lethal concentration of the BCO, cells are metabolically inactive (Figure 5), non-viable and non-culturable (Figure 2). During these periods, the integrity of the cell wall remained unchanged regardless of the concentration of the BCO tested, as showed by calcofluor white staining (Figures 4 and 5).

Cell membrane integrity was evaluated with propidium iodide. This compound binds to DNA and RNA producing a red fluorescence when intercalated with nucleic acids. However, due to its positive charge, it cannot cross an intact cell membrane and, therefore, only dead cells or cells with a damaged membrane are stained. Propidium iodide staining was evaluated at 24 h of incubation of C. albicans with the inhibitory (MIC) and with the lethal (MFC) concentration of the BCO, as determined previously (1.19 and 2.38 μM, respectively). The results, shown in Figure 6, clearly indicate that the BCO somehow destabilizes the plasma membrane, enabling the entrance of the fluorescent dye into the cell.

Immunofluorescence is a technique that allows the visualization of antigen-antibody interactions in cell suspensions. To this end, the BCO was used as the antigen since a first anti-BCO antibody produced in rabbit is then added, followed by a second anti-rabbit antibody produced in goat, conjugated with FITC. In this particular case, C. albicans was incubated with a lethal concentration of the BCO (2.38 μM) in PDB pH 7.5 medium for 24 h and observed by confocal microscopy. Calcofluor white was also added to assess the efficiency of the cell wall digestion. The results are presented in Figure 7 and show a clear green fluorescence inside the cell. Control without the BCO was also tested and no green fluorescence was observed (data not shown). This result suggests that after 24 h of incubation, the BCO was able to cross the cell envelope and is clearly inside the cells. The calcofluor white staining indicates that not all the cell wall was efficiently digested (Figure 7C). However, it was sufficient to allow the cell membrane permeabilization and the subsequent entrance of the antibodies.

To confirm these results we next proceeded with immunogold labeling. As before, cells were exposed to a lethal concentration of the BCO for 24 h and samples were collected after 2, 6, 12, and 24 h of incubation. Each sample was analyzed by immunogold using anti-BCO produced in rat as the first antibody and a second anti-rat antibody coupled to gold particles. The immune-complexes formed were visible as little black dots when observed by TEM (Figure 8). At each sampling point, including time zero, the controls without the BCO were also tested and no black dots were found in none of the controls tested (data not shown). The analysis of Figure 8 shows that the BCO progressively moves from the cell wall into the interior of the cell. Shorter incubation periods show the protein preferentially agglomerating near the cell wall (Figures 8A,B), while longer incubation periods show a migration of the BCO to the cytosol and even to the interior of vacuoles (Figures 8C,D).

The morphological changes undergone by C. albicans after exposure to the BCO, were evaluated by TEM. Cells were exposed to a lethal concentration of the BCO for 48 h and samples were collected at 0, 24, and 48 h. The results are presented in Figure 9. Figure 9A shows a section of a well preserved C. albicans cell, presenting a homogeneous cytoplasm, with an external fibrillar layer (f), a compact cell wall and a normal plasma membrane (cm). After 24 h incubation with the lethal concentration of the BCO, several morphological changes are visible (Figure 9B), mainly focused between the cell wall and the cell membrane. These changes became more clear after 48 h of incubation with a lethal concentration of the BCO, where it is visible an increased thickness at the cell wall level (Figure 9C), appearance of small vesicles in the periplasmatic region (Figure 9D), abnormal density and shape of the cell wall (Figure 9D), an accumulation of high density vacuoles in the cytoplasm (Figure 9E) and cell wall disruption, only visible in some buds (Figure 9F). Although the population of cells showed some heterogeneity, in general, the cells that presented more structural changes also displayed an increased size. The controls performed in the absence of the BCO showed none of these characteristics at the corresponding harvesting times (data not shown).

The interaction of the BCO with mammalian red blood cells was studied by haemolysis experiments. Erythrocytes were incubated with different concentrations of the BCO, ranging from 0.04 to 4.76 μM for 1 h at 37°C. BCO showed no haemolytic effects up to 4.76 μM (data not shown), which is indicative of no detectable interference of the red blood cells.