ABSTRACT
Cryptococcus spp. are common opportunistic fungal pathogens, particularly in HIV patients. The approved drug miltefosine (MFS) has potential as an alternative antifungal against cryptococcosis; however, the mechanism of action of MFS in Cryptococcus is poorly understood. Here, we examined the effects of MFS on C. neoformans and C. gattii yeasts (planktonic and biofilm lifestyles) to clarify its mechanism of action. MFS presented inhibitory and fungicidal effects against planktonic Cryptococcus cells, with similar activities against dispersion biofilm cells, while sessile biofilm cells were less sensitive to MFS. Interestingly, MFS had postantifungal effect on Cryptococcus, with a proliferation delay of up to 8.15 h after a short exposure to fungicidal doses. MFS at fungicidal concentrations increased the plasma membrane permeability, likely due to a direct interaction with ergosterol, as suggested by competition assays with exogenous ergosterol. Moreover, MFS reduced the mitochondrial membrane potential, increased reactive oxygen species (ROS) production, and induced DNA fragmentation and condensation, all of which are hallmarks of apoptosis. Transmission electron microscopy analysis showed that MFS-treated yeasts had a reduced mucopolysaccharide capsule (confirmed by morphometry with light microscopy), plasma membrane irregularities, mitochondrial swelling, and a less conspicuous cell wall. Our results suggest that MFS increases the plasma membrane permeability in Cryptococcus via an interaction with ergosterol and also affects the mitochondrial membrane, eventually leading to apoptosis, in line with its fungicidal activity. These findings confirm the potential of MFS as an antifungal against C. neoformans and C. gattii and warrant further studies to establish clinical protocols for MFS use against cryptococcosis.
INTRODUCTION
Miltefosine (hexadecylphosphocholine [MFS]) is a drug belonging to the alkylphosphocholine class. It was first used in the 1980s for its antitumoral and anti-inflammatory activities (1, 2) and is currently used in the treatment of certain breast cancer types and against leishmaniasis (3). Interestingly, MFS is effective in vitro against other protozoan species, such as Trypanosoma cruzi, Trichomonas vaginalis, Plasmodium falciparum, and the pathogenic amoebae Naegleria fowleri, Balamuthia mandrillaris, and Acanthamoeba (2). In addition, MFS has considerable antifungal potential in vitro against a wide range of pathogenic fungi, including Cryptococcus spp., Candida spp., Aspergillus spp., Fusarium spp., Scedosporium spp., Sporothrix spp., Paracoccidioides spp., Histoplasma capsulatum, Coccidioides posadasii, and dermatophytes (4–14). Therefore, MFS represents a potential alternative to conventional therapy for fungal infections, including cryptococcosis.
Aside from having a broad spectrum of activity against several fungal species, MFS has a fungicidal activity profile; however, its mechanism of action on pathogenic fungi is poorly understood. The fungicidal effect of MFS on Saccharomyces cerevisiae relies on the disruption of the mitochondrial membrane potential, which results in cell death by apoptosis (15, 16). On human fungal pathogens, the only alterations described after MFS treatment were abnormalities in plasma membrane morphology in Sporothrix brasiliensis (5) and Paracoccidioides spp. (10) and increased plasma membrane permeability and a reduction in total ergosterol in C. posadasii and H. capsulatum (7). For Cryptococcus neoformans yeasts, MFS inhibited, in vitro, both the fungal growth and the activity of phospholipase B1, an important virulence factor for Cryptococcus pathogenesis (4). However, other effects of MFS treatment on Cryptococcus spp. have not been examined, and a detailed analysis of MFS physiological effects in Cryptococcus yeasts is essential to elucidate the mechanisms behind the antifungal effect described in vitro.
Here, we evaluated the fungicidal and antibiofilm effects of MFS on C. neoformans and Cryptococcus gattii and examined the effects of MFS treatment on different cell biology parameters to clarify the mechanism of action of MFS in Cryptococcus. In addition, we assessed the postantifungal effect (PAFE) of MFS, representing a suppression of fungal growth that persists after limited exposure to an antifungal (17).
RESULTS
Miltefosine is fungicidal to planktonic Cryptococcus spp. but has reduced antifungal effect against biofilm sessile cells.Although planktonic Cryptococcus species yeasts are susceptible to MFS (4), we tested whether this susceptibility extended to biofilms given the importance of biofilm formation during cryptococcosis. Planktonic cells of Cryptococcus spp. were susceptible to low concentrations of MFS (MICs from 0.5 to 2 μg/ml) and amphotericin B ([AMB] MICs from 0.03 to 0.25 μg/ml), and both antifungals showed fungicidal effects against C. neoformans and C. gattii (Table 1). In contrast, fluconazole (FLC) was the least active drug against C. neoformans and C. gattii and had fungistatic activity (Table 1). The planktonic cells from capsular and acapsular strains of C. neoformans presented similar susceptibilities to the antifungals tested (Table 1).
Susceptibility of Cryptococcus neoformans and Cryptococcus gattii yeasts (planktonic and biofilm) to MFS
Sessile biofilm cells from Cryptococcus spp. were less susceptible to MFS than planktonic cells (MIC90s of sessile cells from biofilms [BSMIC90s] from 8 to >16 μg/ml), although biofilms from the acapsular C. neoformans mutant CAP59 were more susceptible to MFS than capsular C. neoformans (H99) biofilms (Table 1). Interestingly, dispersion cells from Cryptococcus spp. biofilms had similar susceptibility to MFS as their planktonic counterparts (BDMIC ≈ MIC) (Table 1).
C. gattii is less susceptible than C. neoformans to the postantifungal effect of miltefosine.After treatment with MFS for 1 h, we observed PAFEs for all strains tested here, although no PAFE was observed after the treatment of C. gattii with MFS at the MIC (Table 1). The PAFEs varied considerably between the three different Cryptococcus species strains, and it increased with higher MFS concentrations (Table 1). At 4× MIC, MFS treatment for 1 h led to a total loss of cell viability in C. neoformans CAP59; thus, the PAFE could not be determined for this strain at 4× MIC, and it was considerably high (8.15 h) at the MIC (Table 1). Overall, the C. gattii ATCC 56990 strain was less susceptible to postantifungal effects than the C. neoformans strains (Table 1).
Miltefosine treatment reduces the capsule and alters the mitochondrial ultrastructure in Cryptococcus.To assess the effects of miltefosine on Cryptococcus cell ultrastructure, we examined yeasts of C. neoformans H99 treated with 1 μg/ml of MFS for 72 h at 35°C by transmission electron microscopy (TEM). Untreated yeasts were typically round and displayed a clear capsule surrounding a compact and regular cell wall (Fig. 1B). The cytoplasm of untreated cells contained elongated profiles of mitochondria with well-defined mitochondrial ridges (Fig. 1A and B). In contrast, treated yeasts had thinner cell walls and a reduction in the capsule (Fig. 1C and D). Notably, the mitochondria of treated cells appeared swollen and had altered mitochondrial ridges (Fig. 1C and D).
Transmission electron microscopy (TEM) images showing the effect of miltefosine (MFS) on Cryptococcus neoformans yeast. (A, B) Images of untreated C. neoformans H99 yeasts grown in RPMI 1640 medium buffered with 0.16 M MOPS for 72 h at 35°C. The capsule (C) is indicated by the brackets. (C, D) Images of yeasts treated with 1 μg/ml MFS for 72 h, showing a loss of the mucopolysaccharide capsule, alterations in the mitochondrial ridges, and mitochondrial swelling. B and D are higher magnifications of the images in A and C, respectively. m, mitochondrion; cw, cell wall.
To confirm the capsule reduction in a large cohort of cells, we measured capsule thickness by light microscopy in cells resuspended in Chinese ink. The treatment with subinhibitory concentration of MFS (1 μg/ml) for 72 h reduced significantly the capsule thicknesses of C. neoformans H99 and C. gattii ATCC 56990 yeasts (by 72.4% and 33.5%, respectively; P < 0.0001) compared with those of the untreated ones.
Miltefosine increases membrane permeability and reduces yeast viability in Cryptococcus spp.To evaluate the effects of miltefosine on Cryptococcus membrane permeability, we treated yeasts with MIC, 2× MIC, and 4× MIC values for 1, 4, 8, and 24 h (or with AMB, as a reference antifungal) and measured the release of DNA and protein in the supernatant after the treatments. MFS increased the amount of DNA (see Fig. S1 in the supplemental material) and protein (see Fig. S2) detected in the supernatant of Cryptococcus spp. yeast suspensions in time- and dose-dependent manners, suggesting an increase of membrane permeability after treatment, with similar effects for all strains tested. MFS at the MIC value did not increase the membrane permeability relative to that of the untreated control (Fig. S1 and S2); however, 2× MIC and 4× MIC values increased drastically the DNA and protein concentrations in the extracellular milieu, with peaks of macromolecule leak at 4 to 8 h of MFS exposure. The treatment with the standard antifungal AMB resulted in the release of DNA and proteins at all concentrations tested, in a time-dependent response (Fig. S1 and S2).
An analysis of fungal viability by CFU counting (time-kill curves) (Fig. 2) showed that treatment with MFS at 4× MIC reduced the cell viability of C. neoformans (CAP59 and H99) even after only 4 h of treatment; however, MFS at lower concentrations (MIC and 2× MIC) did not decrease yeast viability, even after 24 h of treatment (Fig. 2). In contrast, AMB treatment decreased the viability of C. neoformans strains even at the lower concentrations (MIC and 2× MIC), particularly after 24 h of treatment. For C. gattii, yeast viability was not affected by MFS, even at the highest concentration tested (4× MIC); in contrast, AMB treatment induced cell death in C. gattii as early as 4 h after the start of treatment (Fig. 2).
Time-kill curves of Cryptococcus spp. treated with miltefosine (MFS) (A, C, E) or amphotericin B (AMB) (B, D, F) at the MIC or multiples thereof (2× MIC and 4× MIC). (A and B) C. neoformans acapsular CAP59 strain. (C and D) C. neoformans H99 reference strain. (E and F) C. gattii ATCC 56990 reference strain. The dotted lines indicate the detection limits of the method (50 CFU/ml or 1.5 log).
In general, the treatment with a higher concentration of MFS or AMB (4× MIC) led to the loss of yeast viability, with DNA and protein leakage to the medium. Although MFS did not affect the fungal viability of C. gattii, even at the highest concentrations, the membrane permeability of yeasts from this species was also altered after treatment, as indicated by DNA and protein extravasation (Fig. 2; Fig. S1 and S2). Thus, both MFS and AMB treatments reduced fungal viability in a time-dependent manner, and this effect may be related to the increased membrane permeability of Cryptococcus species yeasts after treatment.
Exogenous ergosterol reduces the antifungal effect of miltefosine on Cryptococcus spp.Previous studies showed that MFS can interact with the membranes of cancer and Leishmania cells (2) and alter the permeability of fungal membranes (7), as we observed for Cryptococcus species yeasts. Ergosterol is an essential lipid component of the fungal cell membrane and is an important target for many drugs (18); therefore, we tested whether the addition of exogenous ergosterol to the culture medium would alter the antifungal effect of MFS against Cryptococcus spp. The presence of exogenous ergosterol in the broth microdilution assay reduced the antifungal activity of MFS, as shown by increases in the MIC values (up to 16-fold) when exogenous ergosterol was added to the culture medium (Table 2). This behavior is similar to that shown previously (19) and confirmed here for AMB (Table 2), an antifungal compound that binds to ergosterol directly. Thus, the ergosterol competition assay results suggest that MFS may interact directly with ergosterol molecules in the fungal plasma membrane.
Effect of exogenous ergosterol on the MICs of MFS and AMB on Cryptococcus neoformans (H99 reference strain and CAP59 acapsular strain) and C. gattii (ATCC 56990 reference strain)
Miltefosine compromises plasma membrane integrity and induces ROS production and DNA damage in Cryptococcus.We evaluated the physiological effects of miltefosine treatment in Cryptococcus by labeling treated yeasts with different fluorescent probes for flow cytometry and fluorescence microscopy. The staining of cells with propidium iodide (PI) indicated that treatment with MFS or AMB (at 4× and 8× MICs) increased the membrane permeability of C. neoformans yeasts, as shown by significant increases in the number of PI+ cells after treatment (Fig. 3A). These results are in agreement with the membrane permeability data showing leakage of DNA and proteins to the supernatant of treated cells (Fig. S1 and S2), suggesting an increase in the cell membrane permeability.
Physiological effects of miltefosine (MFS) treatment on Cryptococcus neoformans yeasts compared to those of amphotericin B (AMB) treatment. Yeasts from the C. neoformans H99 reference strain were kept untreated or were treated with fungicidal concentrations of MFS or AMB (used as a reference antifungal) at the 4× MIC and 8× MIC for 24 h at 35°C and then processed for flow cytometry analysis of the following parameters: membrane permeability, using propidium iodide (PI) (A), mitochondrial membrane potential, using rhodamine123 (B), reactive oxygen species (ROS) levels, using 2′7′-dichlorofluorescin diacetate (DCFH-DA) (C), and DNA fragmentation, using terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) (D). Data represent means ± SEMs from three independent experiments. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 versus untreated (one-way ANOVA with Dunnett's posttest). (E) Light microscopy images of yeast cells labeled with the fluorescent DNA dye Hoechst 33342 (right column), showing chromatin condensation after treatment (white spots, indicated by arrows). The corresponding phase contrast images are shown on the left.
Given that MFS treatment altered the mitochondrial ultrastructure, according to the TEM analysis, we used rhodamine 123 labeling to evaluate the effect of MFS in the mitochondrial membrane potential of Cryptococcus yeasts, because this stain distributes into the mitochondrial matrix in response to the mitochondrial membrane potential (20). The treatment of yeasts with MFS at 4× and 8× MICs reduced significantly the rhodamine 123 fluorescence intensity in C. neoformans yeasts (Fig. 3B), indicating a loss of mitochondrial membrane potential after the treatment. This effect was also apparent after the treatment with AMB (Fig. 3B). We also observed a significant increase in reactive oxygen species (ROS) levels after MFS (only at 8× MIC) and AMB treatments, as assessed by labeling with 2′7′-dichlorofluorescin diacetate (DCFH-DA) (Fig. 3C). Importantly, we observed both DNA fragmentation (by terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling [TUNEL] for flow cytometry) (Fig. 3D) and chromatin condensation (by Hoechst 33342 labeling for fluorescence microscopy) (Fig. 3E) after the treatment of C. neoformans yeasts with MFS or AMB at 4× and 8× MICs, whereas untreated yeasts had a significantly lower or undetectable fluorescence signal.
In summary, we detected an increase in the plasma membrane permeability, as well as a loss of mitochondrial membrane potential, and increases in ROS production and DNA damage in C. neoformans yeasts treated for 24 h with fungicidal concentrations of MFS. These effects were similar to those observed after the treatment with the reference antifungal AMB.
DISCUSSION
Here, we confirmed that MFS has an antifungal effect against Cryptococcus spp. in vitro, with MIC values ranging from 0.5 to 2 μg/ml and fungicidal activity, and these data are in agreement with previous reports (4, 21). Besides the fungicidal effect, the inhibition of biofilm cell proliferation is a favorable pharmacological characteristic of antifungal compounds for cryptococcosis treatment, because biofilm cells are more resistant to host immune mechanisms and drug therapy than planktonic cells (22). We observed that MFS inhibited the proliferation of sessile cells from Cryptococcus species mature biofilms, but only at high concentrations (≥8× MIC for planktonic cells). These results correlate with previous reports suggesting that the biofilm phenotype confers a reduced susceptibility to conventional antifungal therapy (23–25) and also with the reduced effect of MFS on mature biofilms of Candida spp. (14, 26). On the other hand, the dispersion cells from mature biofilms of Cryptococcus spp. were as susceptible to MFS as their planktonic counterparts. We reported similar results for Candida spp., where MFS had comparable antifungal effects against dispersion biofilm cells and planktonic cells (BDMIC and MIC values of 1 to 4 μg/ml) (26). The activity of MFS against dispersion biofilm cells indicates that this drug may be able to reduce infection dissemination by dispersion cells during cryptococcosis treatment.
The PAFE is an important antimicrobial property related to the efficacy of antifungal compounds and represents the persistent suppression of fungal growth after a brief exposure to the drug (27, 28). This interesting effect may be associated with the antifungal dosing regimen during clinical use (27, 28). PAFE values for the treatment of fungi with MFS have not yet been described, and our work demonstrates a relevant PAFE for MFS on Cryptococcus spp. (proliferation delay of up to 8.15 h after MIC pretreatment), suggesting that MFS has prolonged inhibitory activity even when the drug pressure is no longer present. Ernst et al. (17) observed that the fungicidal compounds echinocandins and AMB have prolonged and dose-dependent PAFEs; similar values were observed for MFS, wherein the 8× MIC of AMB led to a PAFE of 7.1 h on C. neoformans (29). In contrast, fungistatic antifungals, such as FLC, do not produce a measurable PAFE against C. albicans or C. neoformans (17). Our data showing that the fungicidal effect of MFS against Cryptococcus spp. is associated with a relevant PAFE supports the notion that fungicidal drugs have a measurable and dose-dependent PAFE.
In Leishmania parasites and human cancer cells, the mechanism of action of MFS involves a direct interaction with membrane lipids, in particular, phospholipids and sterols (30), and the multiplicity of contradictory data and potential mechanisms proposed in different studies indicates that MFS may have more than one molecular target (2). However, the mechanism of action of MFS on fungi remains poorly understood. Our data from competition assays show that exogenous ergosterol decreases the antifungal activity of MFS against Cryptococcus spp., indicating that the direct interaction with ergosterol is part of the mechanism of action of MFS in Cryptococcus, as observed for Leishmania and cancer cells. This in vitro effect was similar to that reported for AMB, whose mechanism of action involves the formation of a complex with plasma membrane ergosterol, resulting in increased membrane permeability and, consequently, fungal death (18, 31). Due to target competition, the concentration of ergosterol in the medium was inversely proportional to the antifungal activities of MFS and AMB (19) (Table 2), which indicates that the efficacy of these drugs is likely to depend on the concentration of ergosterol in the target membranes.
The direct interaction of MFS with ergosterol, suggested by the competition assays, is expected to increase plasma membrane permeability in Cryptococcus. In agreement with this hypothesis, the fungicidal effect against Cryptococcus observed at higher MFS concentrations was associated with an increase in plasma membrane permeability, detected as the leakage of DNA and proteins to the supernatant and by positive staining of yeasts with PI (Fig. 3A; Fig. S1 and S2 in the supplemental material). These data are in agreement with disturbances in the plasma membrane observed after MFS treatment in other fungi, such as S. brasiliensis, Paracoccidioides spp., Coccidioides immitis, and H. capsulatum (5, 7, 10).
In addition to the plasma membrane permeability increase, we observed, by both TEM and light microscopy, that the treatment with a subinhibitory concentration of MFS (1 μg/ml) reduced dramatically the mucopolysaccharide capsule, caused mitochondrial swelling, and decreased the thickness of the cell wall. The mucopolysaccharide capsule is a major virulence factor of Cryptococcus spp. (32), helping the fungi to evade the host's immune system (33, 34). Our group and others have demonstrated that treatments with antifungal compounds, including MFS, AMB, FLC, itraconazole (ITC), and natural products, reduce the capsule and, consequently, the virulence of Cryptococcus (35, 36). In addition, the capsule polysaccharides are the main constituent of the extracellular matrix of Cryptococcus biofilms; thus, the capsule reduction in Cryptococcus spp. observed in this work indicates that the treatment with MFS may reduce the virulence of yeasts both in the planktonic form and in biofilms. Importantly, the acapsular mutant and strains with reduced capsules have lower tolerances to the antifungals (32–34, 36). In agreement with this interpretation, we observed that biofilms formed by the acapsular strain CAP59 were more susceptible to MFS (BSMIC90 = 8 μg/ml) than the C. neoformans H99 (BSMIC90 > 16 μg/ml).
The effects on mitochondria represent an important cellular response to MFS in C. neoformans. In our study, we observed a significant loss of mitochondrial membrane potential in C. neoformans treated with 4× and 8× MICs of MFS, which is in agreement with the loss of mitochondrial ridges and mitochondrial swelling observed by TEM. Mitochondrial damage may explain the loss/reduction in the capsule observed after MFS treatment, since capsule growth is dependent on mitochondrial activity (37).The reduced mitochondrial function after MFS treatment may also explain the increased ROS production in yeasts exposed to a higher concentration of MFS (8× MIC). In contrast, we observed that AMB treatment increased the ROS production significantly (to higher levels than those observed with MFS). ROS accumulation has multiple deleterious effects on the essential structures of the fungi, such as the plasma membrane, proteins, DNA, and mitochondria, resulting in cell death (31). We also observed damage to the genetic material (DNA fragmentation) and chromatin condensation in yeasts treated with fungicidal concentrations of MFS and AMB (4× and 8× MICs).
Mitochondrial membrane potential dissipation, ROS accumulation, and DNA condensation and fragmentation are all hallmarks of apoptosis (38). Thus, our combined physiological effects data indicate that higher concentrations of MFS trigger cell death by apoptosis in Cryptococcus. This interpretation is in agreement with that described in S. cerevisiae, where cell death by apoptosis occurs after MFS incorporation in the inner mitochondrial membrane via an interaction with a subunit of the electron transport chain COX complex (Cox9p) (15), due to the pivotal role played by the mitochondria in apoptosis induction (39), leading to nuclear DNA degradation (15). Importantly, Biswas et al. (16) described that cell death by apoptosis mediated by metacaspase 1 represents the major mechanism responsible for the antifungal effect of MFS against S. cerevisiae.
On the basis of our combined data, we suggest a model for the mechanism of action of MFS on Cryptococcus (see Fig. S3), whereby MFS interacts with cell membranes leading to fungal death by the following important mechanisms: (i) by increasing the plasma membrane permeability via direct interactions with phospholipids and/or ergosterol and (ii) by binding to the mitochondrial membrane, which increases ROS production and culminates in DNA condensation and fragmentation and fungal death by apoptosis. Furthermore, the capsule is reduced as a consequence of mitochondrial damage by MFS, and this effect is expected to decrease the virulence of Cryptococcus species yeasts.
In conclusion, this study demonstrated that MFS has strong potential as a fungicidal agent against cryptococcosis given its low MIC values, clear postantifungal effect, and a mechanism of action that involves fungal cell death by apoptosis. Our results warrant further investigation of the efficacies of different MFS regimens in the treatment of invasive cryptococcosis to support the clinical use of MFS against this lethal fungal infection.
MATERIALS AND METHODS
Microorganisms.Stocks of Cryptococcus neoformans H99 (or ATCC 208821), C. neoformans CAP59 (H99 acapsular mutant), and C. gattii ATCC 56990 strains were kept in brain and heart infusion medium with 20% glycerol at −80°C. For use in experiments, the fungal strains were recovered by growth in Sabouraud dextrose broth for 72 h at 35°C and then kept in Sabouraud dextrose agar (SDA) at 4°C. Prior to each assay, the yeasts were subcultured twice in Sabouraud dextrose medium for 48 to 72 h at 35°C.
Antifungal drugs.Miltefosine ([MFS] Cayman Chemical Co., Ann Arbor, MI) and fluconazole ([FLC] Sigma-Aldrich, USA) were diluted in sterile distilled water and maintained as stock solutions of 1,600 and 2,560 μg/ml, respectively. Amphotericin B ([AMB] Sigma-Aldrich, USA) was maintained as a stock solution of 1,600 μg/ml in dimethyl sulfoxide. All drugs were stored at −20°C.
Antifungal activity on planktonic cells.MIC values were determined using the broth microdilution technique for planktonic Cryptococcus species cells (40). The antifungal drugs were serially diluted (1:2) in RPMI 1640 medium buffered with 0.16 M 3-morpholinopropane-1-sulfonic acid ([MOPS] Sigma-Aldrich, USA), pH 7.0 (referred to as RPMI-MOPS), in flat-bottom 96-well microplates (100 μl/well). Then, 100 μl of the fungal suspension was added to each well, for a final fungal concentration of 0.5 × 103 to 2.5 × 103 CFU/ml and final drug concentrations of 0.03 to 16 μg/ml for MFS and AMB and 0.12 to 64 μg/ml for FLC. Microplates were incubated in the dark at 35°C for 72 h, and the results were read at 492 nm in an Epoch 2 microplate spectrophotometer (BioTek, USA). The lowest drug concentrations that inhibited 50% of the fungal growth (IC50) was used as the MIC for FLC, and the lowest drug concentration that inhibited 90% of fungal growth (IC90) was used as the MIC for AMB and MFS.
To determine minimum fungicidal concentration (MFC) values, aliquots of 10 μl of each well (from MIC assays) where no fungal growth occurred were transferred to drug-free Sabouraud dextrose agar, and the plates were incubated at 35°C for 72 h. The MFC was defined as the lowest concentration of antifungal that killed 99.9% of yeasts (41), and the effect was considered “fungicidal” when the MFC was equal to or 4× higher than the MIC value (41).
Effect of exogenous ergosterol on the MIC.To evaluate the effect of ergosterol on the growth of treated yeasts, broth microdilution assays (for MFS) were performed as described above, but with the addition of 50, 100, or 200 μg/ml of ergosterol (Sigma-Aldrich, USA) (19). The data represent the results from three independent experiments performed in duplicates.
Antifungal activity on biofilm cells.The antibiofilm activity of MFS was evaluated on sessile and dispersion cells from preformed biofilms. For biofilm formation, the yeasts were grown in Sabouraud dextrose broth for 72 h at 35°C, and then an aliquot of 100 μl of fungal suspension at 1 × 107 CFU/ml in RPMI-MOPS was added to the wells of polystyrene flat-bottom 96-well microplates and incubated at 35°C without shaking to enable biofilm formation (24).
After 48 h, the biofilm supernatant, containing the “dispersion cells,” was collected, the number of cells was adjusted to 1 × 106 to 5 × 106 CFU/ml, and antifungal susceptibility assays were performed with dispersion cells using the broth microdilution technique, as described above (see “Antifungal activity on planktonic cells”). Moreover, mature biofilms (i.e., the “sessile cells” that remained attached after the removal of the dispersion cells) were washed twice with phosphate-buffered saline (PBS) and treated with 100 μl of serially diluted MFS (0.03 to 16 μg/ml) for 48 h at 35°C without shaking. Then, the wells were carefully washed with PBS, and the metabolic activity of sessile cells from biofilms was determined by the XTT [2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-car-boxanilide] reduction assay, as described previously (42). The results were used to calculate the MIC90s of sessile cells from biofilms (BSMIC90s).
Postantifungal effect.Yeasts in RPMI-MOPS medium (1 × 104 CFU/ml) were exposed to MFS (at the MIC and at 4× MIC) for 1 h at 35°C. Then, MFS was removed by centrifugation, the yeasts were washed twice with PBS, fresh RPMI-MOPS medium was added, and the cells were incubated at 35°C for 1, 4, 8, 12, or 24 h. Then, the yeasts were washed with PBS, diluted serially (1:10), and plated on SDA. The plates were incubated at 35°C for 72 h, and then yeast growth was quantified by CFU counting. The PAFE was calculated as the difference in time (in hours) required for the treated and untreated yeast populations to grow by 10-fold (1 log) based on the CFU counts (17).
Transmission electron microscopy analysis.For TEM analysis, C. neoformans H99 yeasts were treated with 1 μg/ml of MFS (subinhibitory concentration) for 72 h at 35°C in RPMI-MOPS (or left untreated). Then, the yeasts were fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2) for 2 h and postfixed in 1% osmium tetroxide in the same buffer containing 1.25% potassium ferrocyanide and 5 mM CaCl2 for 2 h at room temperature. The fixed samples were dehydrated in an ethanol series (30%, 50%, 70%, and 100%), transferred to 100% propylene oxide, and then embedded in Spurr's resin. Ultrathin sections were stained with uranyl acetate and lead citrate and observed in a 100CX transmission electron microscope (JEOL, Tokyo, Japan).
Light microscopy analysis of capsule thickness.Cryptococcus spp. were treated with 1 μg/ml of MFS as described above (in “Transmission electron microscopy analysis”). Then, the yeasts were resuspended in Chinese ink and imaged on an EVOS FL light microscopy imaging system (Thermo Fisher Scientific, Waltham, MA, USA). The capsule thickness—defined as the difference between the total cell diameter (as visualized by Chinese ink exclusion) and the cell body diameter (within the cell wall) (36, 43)—was measured from images of 100 yeasts from each group (treated and untreated) using ImageJ 1.49v software (https://imagej.nih.gov/ij/).
Cell membrane permeability and cell viability analyses.Fungal suspensions at 1 × 107 CFU/ml in PBS were exposed to MFS or AMB (at the MIC, 2× MIC, or 4× MIC) and incubated at 35°C for 1, 4, 8, and 24 h (or left untreated). Then, the yeasts were collected by centrifugation at 5,000 × g for 8 min, and the supernatant was analyzed in a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA) for the quantification of free DNA (260 nm) and proteins (280 nm), as described previously (44). Simultaneously, yeast cell viability was evaluated by CFU counting after plating on SDA, as described in “Postantifungal effect,” and the data from three independent experiments (performed in duplicates) were displayed as time-kill curves.
Flow cytometry analyses.Yeasts of C. neoformans H99 at 1 × 107 CFU/ml in PBS were left untreated or were exposed to MFS or AMB (at 4× and 8× the MICs) for 24 h at 35°C. Then, the cells were collected by centrifugation, washed with PBS, diluted to 1 × 106 CFU/ml, and labeled with 5 μg/ml rhodamine 123 (Sigma-Aldrich, USA) (30 min at 35°C) or 40 μg/ml 2′7′-dichlorofluorescin diacetate (DCFH-DA) (Sigma-Aldrich, USA) (30 min at room temperature) for mitochondrial membrane potential or reactive oxygen species (ROS) analyses, respectively. Then, the yeasts were washed with PBS and fixed with 2% formaldehyde in PBS for 30 min at room temperature. Alternatively, treated and untreated yeasts were fixed first as described above and then washed with PBS and stained with 5 μg/ml propidium iodide ([PI] Invitrogen, Eugene, OR, USA) (for 30 min at room temperature) for membrane permeability analysis or, for DNA fragmentation analysis, the cells were fixed, permeabilized with 0.1% Triton X-100 for 30 min (at room temperature), and then subjected to terminal deoxynucleotidyltransferase-mediated dUTP-biotin nick end labeling (TUNEL) using the APO-BrdU TUNEL assay kit (Invitrogen, Eugene, OR, USA), according to the manufacturer's instructions. All incubations in fluorochromes and subsequent steps were performed in the dark. Finally, all samples were analyzed in an Accuri C6 flow cytometer (BD, Franklin Lakes, NJ, USA), using 100,000 events/sample (in triplicates).
Light microscopy analysis of chromatin condensation.Yeasts treated with antifungals as described above (see “Flow cytometry analyses”) were fixed with 2% formaldehyde in PBS for 30 min at room temperature, washed with PBS, and stained with 1 μg/ml Hoechst 33342 (Invitrogen, Eugene, OR, USA) for 10 min in the dark. Then, the cells were mounted onto glass slides using n-propyl gallate and imaged with a DMI6000B/AF6000 fluorescence microscope equipped with a DFC365FX digital camera system (both from Leica Microsystems, Wetzlar, Germany).
Statistical analysis.For the capsule thickness data, statistical analyses was performed by Student's t tests, while flow cytometry results were analyzed by a one-way analysis of variance (ANOVA) with Dunnett's posttest. Statistical analyses were performed using GraphPad Prism 6.0 software (GraphPad Software Inc., La Jolla, CA), with a confidence level of 95% being considered statistically significant.
ACKNOWLEDGMENTS
This work was supported by the Conselho Nacional de Desenvolvimento Cientifico e Tecnológico ([CNPq] Brazil), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior ([CAPES] Brazil), and Fundação de Amparo à Pesquisa do Estado de São Paulo ([FAPESP] Brazil; grant no. 2015/07993-0).
We declare no conflict of interest.
FOOTNOTES
- Received 15 February 2018.
- Returned for modification 8 May 2018.
- Accepted 18 May 2018.
- Accepted manuscript posted online 29 May 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00312-18.
- Copyright © 2018 American Society for Microbiology.
