Mechanisms of action of antimicrobial peptides ToAP2 and NDBP-5.7 against Candida albicans planktonic and biofilm cells

Candida albicans is a major cause of human infections, ranging from relatively simple to treat skin and mucosal diseases to systemic life-threatening invasive candidiasis. Fungal infections treatment faces three major challenges: the limited number of therapeutic options, the toxicity of the available drugs, and the rise of antifungal resistance. In this study, we demonstrate the antifungal activity and mechanism of action of peptides ToAP2 and NDBP-5.7 against planktonic cells and biofilms of C. albicans. Both peptides were active against C. albicans cells; however, ToAP2 was more active and produced more pronounced effects on fungal cells. Both peptides affected C. albicans membrane permeability and produced changes in fungal cell morphology, such as deformations in the cell wall and disruption of ultracellular organization. Both peptides showed synergism with amphotericin B, while ToAP2 also presents a synergic effect with fluconazole. Besides, ToAP2 (6.25 µM.) was able to inhibit filamentation after 24 h of treatment and was active against both the early phase and mature biofilms of C. albicans. Finally, ToAP2 was protective in a Galleria mellonella model of infection. Altogether these results point to the therapeutic potential of ToAP2 and other antimicrobial peptides in the development of new therapies for C. albicans infections.


Minimal inhibitory concentrations (MIC) for C. albicans. We have already evaluated the MIC of
ToAP2 and NDBP-5.7 against C. albicans SC-5314 in our previously work using an inoculum of 2 × 10 3 cells/ mL. However, some assays described in this work, such as flow cytometry and Electron Transmission Microscopy (TEM), required a higher cell density or a non-filamenting strain. To solve the filamentation problem for the flow cytometry analysis, we used the non-filamenting strain SSY50-B 40 , which showed the same MIC values to both AMPs presented by the filamenting strain SC-5314 (12.5 µM for ToAP2 and 25 µM for NDBP-5.7) 38 . In addition to that, we evaluated ToAP2 and NDBP-5.7 MIC for both C. albicans strains at a cell density of 1 × 10 6 cells/mL. The obtained MIC was also the same for both strains (25 µM for ToAP2 and 100 µM for NDBP-5.7).
ToAP2 and NDBP-5.7 permeabilize C. albicans cell membrane. Plasma  ToAP2 alters the ultrastructure of C. albicans. The ultrastructure of C. albicans cells treated with AMPs ToaP2 and NDBP-5.7 was evaluated by transmission electron microscopy. Control samples revealed cells with preserved morphology and cell shape surrounded by a regular cell wall and continuous cell membrane. These cells also presented a homogeneous cytoplasm with nucleus, mitochondria and other membrane-enclosed organelles ( Fig. 2A,B). On the other hand, ToAP2 treated cells show several ultrastructure alterations, including an irregular thickness and shape of the cell wall, regions of discontinuity of the cell wall and cell membrane and an amorphous and diffuse cytoplasm lacking a clear organization of organelles (Fig. 2D,E). This could suggest that in addition to the cell membrane, this peptide could be also acting on intracellular membranes. In contrast, cells treated with NDBP-5.7 displayed a more regular morphology, especially at the cell wall organization (Fig. 2G,H), although we could also observe some cells showing an irregular cytoplasm filled with a large number of tiny vacuoles (Fig. 2H). Interestingly, we could also observe a very distinguishable cell wall fibrillar outer layer in NDBP-5.7 treated cells that was not clear in the control cells.
ToAP2 and NDBP-5.7 induce morphological changes in C. albicans cells at subinhibitory concentrations. We used atomic force microscopy to visualize possible alterations in C. albicans cell surface or morphology induced by a 24 h treatment with both peptides. Control cells not exposed to any of the AMPs showed the typical oval and well-defined morphology and a smooth surface. Treatment with sub-inhibitory concentration of ToAP2 (6.25 μM) and NDBP-5.7 (25 μM) resulted in C. albicans cells with altered cell morphologies in comparison to control cells (Fig. 2C). In overall, the most noticeable changes in ToAP2-treated were deformations in their area including flattened regions in their surface instead of the round morphology observed in the control cells. On the other hand, NDBP-5.7-treated cells had a wrinkled appearance with significant increase in cell-surface deformations (Fig. 2F,I). Results from both microscopy approaches combined with the results of the membrane permeability assay suggest that the cell surface is probably a target for both AMPs.
ToAP2 inhibits C. albicans filamentation. Sub-inhibitory concentrations of ToAP2 (6.25 and 3.12 μM) and NDBP-5.7 (12.5 and 25 μM) were used to evaluate their effect on C. albicans SC 5314 filamentation during 4 h or 24 h treatments. There was no difference in the percentage of fungal cells with filaments or in the germ-tube www.nature.com/scientificreports www.nature.com/scientificreports/ lengths in comparison to the control at the early time points (Fig. 3A,B). In contrast, there was a decrease in fungal filamentation after 24 h of treatment with ToAP2 (Fig. 3C). Considering that filamenting cells normally grow in clumps, impairing the evaluation of the size of individual cells or the number of individuals cells in a clump at later time points, we complement the microscopic analysis presented in Fig. 3C by time-lapse microscopy of cells submitted to the different treatments along 24 h (Supplementary Movie S1). At MIC concentrations we observed very little cell filamentation in response to both peptides, probably because the cells are dying. Interestingly, the treatment with 6.25 μM of ToAP2 not only delayed C. albicans filamentation, but also decreased de overall length of the filaments in comparison to the control, apparently without any major effect on cell viability (Supplementary Movie S1). Although NDBP-5.7 apparently have not affected the filamentation on the tested conditions, at the second half of the Supplementary Movie S1, we can still see several yeast cells lacking germ-tubes.
Combined action of antifungal drugs with ToAP2 and NDBP-5.7. The combined activity of antifungal agents and AMPs was evaluated using a Checkerboard assay 41 , based on the MIC concentrations of each AMP 38 .
ToAP2 is effective in different stages of biofilm formation. We evaluated the effects of ToAP2 and NDBP-5.7 at early-phase (after 4 h of attachment) and mature biofilms (after 24 h of attachment). Treatment with amphotericin B concentrations equal or higher than 0.27 μM produced an inhibition of at least 75% in the viability of early-phase biofilms. Although the effects were less pronounced, the three highest concentrations of amphotericin also produced a 40-60% inhibition of mature biofilms. (Fig. 4A,B). In contrast, fluconazole had no activity at any of the tested concentrations at both phases of biofilm formation (Fig. 4C,D).
ToAP2 displayed significant antimicrobial activity against both early-phase (p = <0.0001) and mature biofilm phases (p = <0.0001) (Fig. 4E,F). Starting with 25 μM of ToAP2 we observed a dose dependent inhibition of early www.nature.com/scientificreports www.nature.com/scientificreports/ phase biofilms ranging from 60.02 to 99.85%. ToAP2 also produced a 97.02% decrease in the viability of mature biofilms, but only at 200 μM. On the other hand, NDBP-5.7 peptide showed no activity in either phases of C. albicans biofilms (Fig. 4G,H). The viability of biofilms after any of the treatments was also evaluated by fluorescence microscopy using the viability dye Phloxine B (Supplementary Figs. S2 and S3).
Peptides ToAP2 and NDBP-5.7 permeabilize the membrane of biofilm cells. Membrane integrity of C. albicans biofilm cells treated with ToAP2 or NDBP-5.7 was evaluated using an adaptation of the fluorophore Rhodamine 6G efflux assay as described before 42 . Both peptides produced a dose-dependent increase in permeabilization of C. albicans biofilm cells at both phases of biofilm formation. There was an increase in the permea- Effect of ToAP2 on the survival of Galleria mellonella infected with C. albicans. The ToAP2 toxicity and antimicrobial activity in vivo was evaluated using the G. mellonella model. For the toxicity evaluation different groups of larvae were injected with 10, 20 and 40 mg ToAP2/kg or with 2 mg/kg of Amphotericin B. None of the AMP doses were toxic to G. mellonella (data not shown). In vivo antimicrobial activity of ToAP2 was evaluated after infecting different groups of larvae with C. albicans and treating the groups with different doses of www.nature.com/scientificreports www.nature.com/scientificreports/ this peptide. All doses of ToAP2 resulted in an increased survival of the larvae when compared to control group (treated with PBS) (Fig. 6A). Doses of 10, 20 and 40 mg/kg resulted in an increased survival rate of 18.7, 31.2 and 25% respectively. ToAP2 doses of 20 and 40 mg/kg resulted in survival rates not statistically different when compared to 2 mg/kg of Amphotericin B, which increased larvae survival to 50%. In the larvae group treated with PBS (control group), all organisms died in the third day after C. albicans infection.
www.nature.com/scientificreports www.nature.com/scientificreports/ ToAP2 treatment decreases fungal burden of G. mellonella larvae infected with C. albicans. To evaluate the in vivo effects of ToAP2 we performed histopathology analysis of G. mellonella larvae infected with C. albicans and treated with this peptide, amphotericin B or PBS (untreated control). Non-infected larvae were used as control (Fig. 6A). Infected and untreated larvae (  . After that, cell viability was evaluated using the alamarBlue reduction assay. Results were expressed as a mean of three independent assays. P-value indicates statistical difference using the Tukey post-test. The results represent mean ± SEM.

Discussion
The structural and physiological similarity between fungal and mammalian cells makes the development of new antifungals a complex challenge 43 . Given this, the AMPs are small conserved molecules part of the innate immune  www.nature.com/scientificreports www.nature.com/scientificreports/ response of several organisms and a promising group of new antimicrobial agents 44 . In the present work, we further evaluated the antifungal activity of two AMPs previously described and partially characterized by our group, ToAP2 and NDBP-5.7. We analyzed the activity of both AMPs on membrane integrity, cell morphology, yeast-to-hypha transition, and biofilm formation of C. albicans.
Flow cytometry analysis of propidium iodine labeled cells revealed that treatment with both peptides produced a dose-dependent increase in membrane permeability and cell death. A similar permeabilization of C. albicans cells was previously described after treatment with Cathelicidin PMAP-36 analogues 45 . AFM and TEM results also suggested the induction of membrane damage and loss of cell surface integrity by both peptides, particularly by ToAP2. AFM images of fungal cells treated with ToAP2 and NDBP-5.7 showed morphological alterations including the loss of the typical oval shape and an increase in surface roughness, when compared to untreated control cells. Those cell deformations might be linked to leakage of cellular contents as a result of cell membrane damage. Comparable effects were observed in other AFM studies describing alterations in C. albicans cell morphology and membrane integrity when in the presence of antimicrobial peptides Psd1 46 and LL-37 47 . TEM of C. albicans treated with ToAP2 revealed several cells showing a loss of cell wall integrity and presenting an internal amorphous cytoplasm with mostly no distinguishable organelles, suggesting that this AMP can induce damage not only on the plasma membrane but also in the membrane of cellular organelles inducing loss of integrity of those structures. These alterations bear similarities to the ones produced by the treatment of C. albicans cells with the phosphorylated peptide K40H derived from the constant region of human IgMs 48 or with the antifungal peptide AMP MCh-AMP1 derived from chamomille (Matricaria chamomilla) 49 .
These observations suggest that membrane damage and permeabilization is one of the mechanisms of action of those molecules as previously reported for many antimicrobial peptides 50,51 . Our findings are also in agreement with previously published circular dichroism analyses showing ToAP2 and NDBP-5.7 tendency to form α-helix structures 38 , a common feature in other antimicrobial peptides targeting microbial cell membranes [52][53][54] . However, we cannot exclude other mechanisms of action for these molecules, as AMPs have been shown to have multiple antifungal actions such as DNA damage, inhibition of protein synthesis or metabolic pathways and induction of apoptosis or oxidative stress in addition to membrane permeabilization 50,51 .
An important factor associated to C. albicans pathogenesis is its ability to perform yeast-to-hypha transition 8 . An alternative therapeutic approach for infectious diseases is targeting microbial virulence attributes, what would increase microbial susceptibility to antimicrobials or to the immune system components with lower chances of inducing microbial resistance 55 . As a result, many research groups have dedicated efforts in the search of molecules that could inhibit C. albicans filamentation [56][57][58] .
As shown in Fig. 3, subinhibitory concentrations of both AMPs presented no effect on the percentage of C. albicans cells presenting germ tubes formation during the first four hours of treatment. However, a prolonged incubation with 6.25 µM of ToAP2 significantly delayed filamentation and reduced the number of cells presenting germ tubes without killing the cells (Fig. 3C). It is important to notice that filamentation is a complex process dependent of many different factors, such as availability of nutrients and the presence of serum in the medium, and all of them can interfere with the peptide inhibitory effects 59 .
ToAP2 showed the most promising antimicrobial effects in most of our assays. However, as other AMPs 60,61 , this peptide presents some toxicity to mammalian cells in concentrations close to its MIC for C. albicans and other tested fungi 38 . One approach to overcome this problem is the optimization of this peptide sequence aiming to decrease its toxicity to mammalian cells/enhance its antimicrobial activity. Another approach could be the combined use of AMPs with conventional antifungals. The use of drug combinations is a strategy to improve drug's efficacy reducing their concentrations to non-toxic levels and at the same time a strategy to decrease resistance development 62 .
Thus, we evaluated the combined actions of ToAP2 and NDBP-5.7 with each other and with amphotericin B or fluconazole against planktonic cells of C. albicans. The combination of ToAP2 and NDBP-5.7 AMPs showed an additive effect, while both peptides presented a synergistic effect with amphotericin B and ToAP2 also with fluconazole (Table 1). In all tested combinations, the active concentrations used were reduced in comparison with the MIC of each of the compounds alone 38 . For instance, the combination of AMPs with amphotericin B produced a reduction of up to 4X in their MIC. Our results were similar to those found by Lum et al. 63 , who found that the combination of antimicrobial peptides with amphotericin B resulted in a synergistic effect, while the combination between peptides was additive or indifferent in C. albicans infection models. The peptides MUC7 12-mer, hepcidin 20 and defensin HsAFP1 have also shown synergism with antifungal drugs [64][65][66][67] . Our results might also suggest that ToAP2 and NDBP-5.7 act on different targets than the antifungal agents, thus their complementary effect as previously suggested for other AMPs 68 . According to Gray and colleagues 69 , the main mechanism of action of Amphotericin B in yeast cells (Saccharomyces cerevisiae and C. albicans) is the binding and sequestration of ergosterol, while membrane permeabilization is a secondary mechanism not necessarily required for its fungicidal effects. In addition to its activities at the membrane level, another important microbicidal activity of amphotericin B is the induction of reactive oxygen species (ROS) accumulation observed in many fungal species. It is believed that this mechanism could be related to the low induction of fungal resistance to this compound 70 . On the other hand, fluconazole, a fungistatic azole, interacts with a key enzyme in fungal sterol biosynthesis inhibiting the synthesis of ergosterol and compromising membrane permeability and fluidity 71 . Microbial cell membrane is also regarded as the primary target of AMPs, being membrane permeabilization probably by pore formation one of the main effects of those molecules. Our flow cytometry and microscopy results point to the same happening in C. albicans cells treated with NDBP-5.7 and ToAP2. However, as we mentioned before, we cannot exclude other mechanisms of actions for those molecules and further studies are needed to elucidate the mechanisms behind the synergistic effects we have observed.
Compared to planktonic cells, biofilms can be up to 1000 times more resistant to antimicrobials, due to certain factors such as the presence of a dense extracellular matrix, changes in cell metabolism and the positive regulation www.nature.com/scientificreports www.nature.com/scientificreports/ of efflux pumps among others [72][73][74] . Consequently, few antimicrobials capable of treating biofilm-related infections are available today 75 . Our results indicate that ToAP2 presented inhibitory action on both phases of biofilm formation. Amphotericin B but not fluconazole was also active in our tests, as previously reported 63,[76][77][78][79] .
Biofilm resistance to fluconazole could be partially explained by the presence of the extracellular matrix, composed by proteins, polysaccharides, nucleic acids and lipids, which serves as a barrier for antifungals 80,81 . It has been reported that β-1,3 glucan, present in C. albicans biofilm matrix, can hijack triazoles preventing these drugs from reaching cell membrane 82,83 . Both ToAP2 and NDBP-5.7 were able to permeabilize C. albicans biofilm cell membrane in both phases of biofilm formation (Fig. 5A,B). Interestingly, NDBP-5.7 affected membrane permeabilization without decreasing the viability of the biofilm (Fig. 4G,H).
We have shown that ToAP2 was not lethal to G. mellonella larvae when used at concentrations up to 40 mg/kg, similar to what was observed with AMPs hMUC7-12, hLF(1-11), DsS3(1-16) by Maccalum et al. 64 . ToAP2 doses of 10, 20 and 40 mg/kg were also able to significantly increase larva survival after C. albicans infections in our assays. At the two highest concentrations of this AMP the larva survival profile was similar to what we observed with the amphotericin B treatment. Although amphotericin B effects were more pronounced, histopathological analysis of larvae infected with C. albicans and treated with any of the two compounds resulted in a reduction in the number of foci of infection and a more limited infection in contrast with the disseminated form observed in the control (Fig. 6). In addition, in the control groups, we observed the presence not only of yeast cells as in the treated-groups, but also filament cells possibly indicating C. albicans tissue invasion as described before 84 . These results are also in accordance to previously described works showing the efficacy of antifungals in the treatment of G. mellonella infected with C. albicans [85][86][87] . Finally, the results observed with ToAP2 indicated a possible systemic action of this peptide in this in vivo infection model.
In conclusion, we have shown that the peptides ToAP2 and NDBP-5.7 are active against planktonic and biofilms of C. albicans. Treatment with these peptides increased cell permeability and generated important morphological alterations in C. albicans cells. ToAP2 showed the most promising antimicrobial effects in most of our assays an can be considered a putative candidate for new antifungal formulations, especially for topical treatments or to be used in combined therapeutic strategies. The observed synergy of this peptide with fluconazole and amphotericin B is particularly encouraging because we could observe an increase in the efficacy of both molecules decreasing their active concentrations and probably also their toxicity to mammalian cells.
Invasive candidiasis affects around 750 thousand people yearly. This disease has a close association with health-care environment where is a major cause of morbidity and mortality specially among elderly and premature babies 88,89 . As for other invasive fungal infections the available therapeutic options are extremely limited and associated with high toxicity, costs and the increase in fungal strains resistant to the few antifungal drugs currently available, such echinocandins and azoles [90][91][92][93] .
The results described in this work are promising in view of the urgent need for alternative antifungal molecules to treat systemic mycoses and to prevent the development of fungal resistance. Currently, we are working on the design and optimization of the antimicrobial features of ToAP2, aiming to circumvent its cytotoxicity against mammalian cells and to increase the biological properties reported here.

Material and methods
Synthesis of antimicrobial peptides. ToAP2 (FFGTLFKLGSKLIPGVMKLFSKKKER) and NDBP-5.7 (ILSAIWSGIKSLF-NH2) sequences are derived from cDNA libraries made from the venom glands of Tityus obscurus and Opisthacanthus cayaporum scorpions 38 . The AMPs were chemically synthesized by Biomatik using an Fmoc / t-butyl on solid support strategy. Peptide purification, characterization and purity evaluation were performed as previously described by Guilhelmelli 38 . Fungal culture conditions. C. albicans strain SC 5314 was used in most assays, except in the experiments of flow cytometry and Atomic Force Microscopy. In these assays we used the non-filamenting strain C. albicans SSY50-B, kindly provided by Stephen P. Saville and José L. Lopez-Ribot 40 . Fungal cells from frozen stocks were grown in Sabouraud dextrose broth for 18 h at 30 °C and 200 rpm. After that, cells were harvested by centrifugation (1000 g at 25 °C), washed three times in sterile phosphate buffer (PBS), counted in a hemocytometer and diluted to the appropriate cell densities.
Minimum inhibitory concentration (MIC) evaluation. As previously described by our group 38 , all antifungal assays were carried out according to guidelines of the broth microdilution susceptibility test by the Clinical and Laboratory Standards Institute (CLSI) M27-A3 with some modifications 94 . Briefly, cells suspensions of C. albicans cells at 2 × 10 3 cells/mL were tested using concentrations of 100 to 0.78 µM of each AMP. The criterium to define the minimum inhibitory concentration (MIC) was the lowest AMP concentration that completely inhibited visible fungal growth at the end of the incubation period. All assays were done in biological triplicates performed on separate dates.
Before performing assays requiring a higher number of cells, such as the evaluation of cell membrane integrity or microscopic analyses, we determined the MIC for C. albicans suspensions at the cell density used in those experiments -1 × 10 6 cells/mL. Therefore, the higher MIC values for all antifungals in those assays.

Evaluation of cell membrane integrity.
An inoculum of 500 µl of 1 × 10 6 cells/mL of C. albicans SSY50-B strain was added to RPMI medium in the presence or absence of different concentrations of ToAP2 or NDBP 5.7 and incubated in a shaker at 37 °C and 200 rpm for 90 minutes 45 . Afterwards, the cultures were incubated with 1 μg/mL propidium iodide (PI) for an additional 30 minutes, in the dark, at 37 °C. The cells were then analyzed on a BD FACSVerse flow cytometer (minimum of 20,000 events per sample). The resulting data was analyzed using the FlowJo X software. The gating strategy consisted of: a) gating on fungal cells on a forward scatter (area) www.nature.com/scientificreports www.nature.com/scientificreports/ x side scatter (area) plot and b) doublet exclusion by gating on a forward scatter (width) x forward scatter (height) plot. Live untreated cells, and 70% ethanol-treated (dead) cells were also incubated with PI and used as controls. The cells were observed in an inverted microscope (Axio Observer Z1 Carl Zeiss Microscopy, USA) (excitation: 546/12 nm; emission: 607/80 nm).

Evaluation of ultrastructure by Transmission electronic microscopy.
An inoculum of 1 × 10 6 cells of C. albicans SSY50-B strain was added to RPMI with concentrations of 25 and 50 µM of ToAP2 and NDBP-5.7 respectively following standard protocol with modifications 45 . After 24 h, the cells were washed with PBS and cell suspensions were fixed by immersion in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) solution for 2 days. After washing, the cells were post-fixating with 2% osmium tetroxide in 0.1 M sodium cacodylate buffer and incubated overnight with 1% uranyl acetate after a new wash. Then the cells were dehydrated in graded series of ethanol and soaked in Epon (EMS). The blocks were cut at 50 nm on a RMC Ultramicrotome (PowerTome, USA) and recovered into 200 mesh copper grids, followed by double contrast method with 2% uranyl acetate. Visualization was performed at 80 kV in a JEOL JEM 1400 microscope (Japan) and digital images were acquired using a CCD digital camera Orious 1100 W (Tokyo, Japan). The combinatory effects between both AMPs and also between each of them and fluconazole or amphotericin B were tested using the checkerboard assay with some modifications 41 . Cells of C. albicans SC 5314 (2 × 10 3 cells/ mL in 100 µL) were inoculated in RPMI 1640 in the presence of different concentrations of combined AMPs or AMPs plus antifungals. The plates were incubated at 37 °C for 24 h. After that, cell viability was evaluated using alamarBlue (Thermo Fisher Scientific) at 9% in a fluorescence plate reader with excitation at 550 nm and emission at 585 nm (SpectraMax M plate reader). This assay was performed in biological triplicates for each of the combinations tested. The synergistic action was evaluated by the Fractional Inhibitory Concentration (FIC). Antifungal action of AMPs against C. albicans biofilms. The antifungal activity of the AMPs against C.

Morphological analysis by atomic force microscopy (AFM).
albicans SC 5314 biofilms was assessed using a microplate alamarBlue susceptibility assay as previously described by Repp et al. 98 . Briefly, a 100 µL inoculum containing 1 × 10 6 cells/mL of C. albicans in RPMI 1640 medium was added to the wells of a 96-well polystyrene microplate and incubated at 37 °C for 4 h (early phase) or 24 h (mature biofilm). After that, biofilms were washed three times with PBS to remove non-adherent cells and each well received fresh medium with different concentrations of ToAP2 (100 to 3.12 μM for early phase and 200 to 6.25 μM for mature biofilms); NDBP-5.7 (200 to 6.25 μM for early phase and 400 to 12.5 μM for mature biofilms), amphotericin B (1.08 to 0.03 μM for both early-phase and mature biofilms) or fluconazole (104 to 3.25 μM for both early-phase and mature biofilms) and the plates were incubated at 37 °C for 24 h. Cell viability was evaluated using the alamarBlue reagent, in a concentration of 9%, and fluorescence was detected on a SpectraMax M plate reader (excitation at 550 nm and emission at 585 nm). Cell viability was also analyzed using the fluorescence dye www.nature.com/scientificreports www.nature.com/scientificreports/ Phloxine B (0.01%) and observing the cells in an inverted microscope (Axio Observer Z1 Carl Zeiss Microscopy, USA) (excitation: 546/12 nm; emission: 607/80 nm). All assays were done in biological triplicates.

Evaluation of cell membrane integrity of biofilm cells. Membrane permeabilization of C. albicans
biofilm cells was evaluated using an adaptation of the rhodamine 6 G efflux assay as described by Szczepaniak 42 One hundred microliters of a cell suspension of C. albicans SC 5314 containing 1 × 10 6 cells/mL was incubated at 37 °C in 96-well microplates for 4 or 24 h, in RPMI medium supplemented with MOPS. After the incubation times, the biofilms were washed three times with PBS and incubated for 2 h in PBS. Afterwards, all the wells received a solution of Rhodamine 6 G in PBS (10 µM) and were incubated for an additional 30 minutes. After this incubation, the biofilms were washed three times with PBS to remove the non-internalized dye, received different concentrations of ToAP2 or NDBP-5.7 diluted in PBS and were incubated for an additional 24 h. Untreated biofilms and biofilms treated with Amphotericin B (1.08 µM in PBS) and Fluconazole (52 µM in PBS) were used as controls. After 24 h, the supernatant was removed, and fluorescence measured in a fluorescence plate reader with excitation at 550 nm and emission at 585 nm (SpectraMax M plate reader). The test was carried out in biological triplicates.

Galleria mellonella infection.
In vivo toxicity and efficacy against C. albicans infection of ToAP2 were tested using G. mellonella larvae weighting 250-300 mg as described before 99 . For the toxicity assay, groups of 16 randomly selected larvae were inoculated with different doses of ToAP2 (0, 40, 20 and 10 mg/kg) diluted in PBS in the last proleg using a Hamilton syringe and the larvae were incubated at 37 °C. Larva survival was evaluated daily for 7 days, and larva death was defined by the absence of movement in response to touch 100 . For protection assays, each larva received an inoculum of 5 × 10 5 C. albicans cells and was incubated for 1 h at 37 °C. After that, different doses of ToAP2 peptides diluted in PBS were injected into the infected larvae and the animals were incubated at 37 °C for 7 days. Amphotericin B (2 mg/kg) and PBS alone were used as controls. The assay was performed twice.
Histopathological analysis of G. mellonella. Histopathological evaluation was performed according to a standard protocol 85 . Three larvae from each group (non-infected, infected, infected + ToAP2 40 mg/Kg and infected + Amphotericin B 2 mg/Kg) were used after 24 h of incubation at 37 °C. The larvae were fixed with 10% buffered formalin for 24 h and then transverse and longitudinal incisions were made. The samples were dehydrated with increasing concentrations of ethanol, followed by xylol and incorporated with paraffin. Tissue sections (6 micron) were made with a microtome and placed on glass slides with 5% albumin. The slides were then stained with periodic acid Schiff (PAS) solution and observed in inverted microscope (Axio Observer Z1 Carl Zeiss Microscopy, USA).
Statistical analyzes. Except for the G. mellonella infection assay, which was performed as two biological replicates in different days with similar results, all other analyses were done in biological triplicates performed on different days. ANOVA and Tukey tests were used as post-tests, being considered statistically significant results when p ≤ 0.05. Long-rank was used in the analysis of Mantel-Cox survival curves.