Two Paracoccidioides isolates were used: Pb18 (P. brasiliensis) and Pl01 (P. lutzii), maintained in vitro in Fava Netto’s medium at 37°C (Fava Netto, 1961). The handle of Paracoccidioides spp. cultures was made in a Class II Biological Safety Cabinet according recommendations of American Type Culture Collection (ATCC) and Brazilian Ministry of Health, Secretariat of Science, Technology and Strategic Inputs (Ministério da Saúde and Estratégicos, 2006).

Phage libraries were produced using the vector fUSE55 and the methodology developed by George Smith (Smith and Scott, 1993), with modifications to improve the efficiency and the number of final clones (Koivunen et al., 1993; Koivunen et al., 1999). The libraries were produced and provided by Dr. Ricardo José Giordano from the Laboratory of Vascular Biology, University of Sao Paulo (USP) and were X6 (a linear library) and CX8C (a cyclic library; X represents any amino acid, and C = cysteine that allows for cyclical peptides) libraries (Michaloski et al., Submitted).

The BRASIL (Biopanning and Rapid Analysis of Selective Interactive Ligands) technique (Giordano et al., 2001) was used for the selection of peptides that bind to Paracoccidioides spp. This technique consists of the selection of phages that bind to live cells and the use of a two-phase system to separate the cells and phages in a single centrifugation step. First, 10⁹ transduction units (TU) of each phage display library were separately incubated with Saccharomyces cerevisiae (S288C) (10⁶ cells) in 100 μl of PBS. After 3 h on ice, the suspensions containing the phages and the cells were transferred to the top of a nonmiscible organic lower phase known as BRASIL oil [dibutyl phthalate:cyclohexene (9:1) v/v] and centrifuged (10,000 × g, 10 min, 25°C). After centrifugation, the supernatants containing phages that did not bind to S. cerevisiae were recovered and transferred to a suspension of 10⁶ Pb18 cells. After 3 h on ice, the cell suspensions were again transferred on to BRASIL oil and centrifuged at 10,000 × g for 10 min. The pellets of Pb18 cells that formed at the bottom of the organic phase were collected, and the phages bound to the cells were recovered via infection with log-phase E. coli K91 in LB medium supplemented with kanamycin and tetracycline (100 μg/mL and 20 μg/mL, respectively) and incubated at 37°C, with rotation at 250 rpm, for 20 h. The amplified phages were recovered from the culture medium via precipitation with polyethylene-glycol/NaCl and were used for subsequent rounds of selection (second and third) to further enrich for phages from each library that bound to Pb18 cells.

Paracoccidioides brasiliensis, P. lutzii, and S. cerevisiae cells were incubated with 10⁹ TU of the phage to be tested, as described above. After 2 h of incubation on ice, they were transferred to the top of BRASIL oil as previously described in the item 2.4 and centrifuged (10,000 × g, 10 min, 25°C). The pellet was placed in a new tube, and the phages were recovered via infection with E. coli K91. Serial dilutions were seeded onto LB/Kan/Tet plates to quantify the number of phage bound to each receiver. As a control P. brasiliensis and P. lutzii cells were incubated with 10⁹ of the insertless phage Fd-tet. Three independent experiments were performed.

Ninety-six-well plates were separately pre-coated with 10 μg/mL of each tested ECM component (laminin, fibronectin, and type I and type IV collagen) in a 0.05 M carbonate buffer (pH 9.6) at room temperature for 1 h. The plates were then washed with PBS/Tween-20, 0.05% (PBS-T) and blocked with 1% bovine serum albumin (BSA) for 2 h at 37°C. The two Paracoccidioides isolates (Pb18 and Pl01, 10⁶ cells/mL) were incubated with 200 μg/mL of each studied peptide, which were synthesized by the Chinese Peptide Company (China), at 37°C, with rotation at 250 rpm, for 1 h. These inocula were incubated with the ECM components for 15 h at 37°C. The plates were washed with PBS-T, and an anti-cell free Paracoccidioides antibody (1:100 in PBS-T and 0.5% BSA) (Blotta and Camargo, 1993) was added for 1 h at 37°C. Another wash cycle was performed, and a secondary antibody, anti-rabbit IgG-HRP (1:2000 in PBS-T and 0.5% BSA), was added for 1 h at 37°C. The colorimetric reaction was observed by adding a solution of 0.2 M sodium phosphate, 0.1 M citric acid, 0.4 mg/mL o-phenylenediamine, and 30% hydrogen peroxide for 10 min, followed by the addition of 3 M hydrochloric acid. Absorbance was measured at 490 nm and converted to percent inhibition. For this conversion, Paracoccidioides spp. cells without treatment were used as a control, and the absorbance of this control was set as 100% adhesion. The difference in the absorbance in the treated Paracoccidioides spp. cells relative to absorbance in the control was considered the percentage of adhesion inhibition. This experiment was performed in triplicate, with three independent experiments.

The continuous culture of A549 cells line took place in F-12 Nutrient Mixture (HAM-F12) medium supplemented with 10% fetal bovine serum (FBS) in sterile plastic bottles maintained at 36.5°C with 5% CO₂. After 3–4 days, the bottles were trypsinized by washing the cell monolayer with 1 mL of a 0.2% trypsin ATV solution (Adolfo Lutz) supplemented with 0.02% EDTA. After 1–2 min, the cells were homogenized with variable volumes of HAM-F12 medium supplemented with 10% of FBS to neutralize the trypsin. The total volume of the cell suspension was determined to achieve a concentration of 10⁶ cells/mL. For the assay, 10⁵ A549 pneumocytes were added to each well of a 96-well plate and then infected over 15 h (37°C with 5% CO₂) with the different Paracoccidioides isolates (Pb18 and Pl01, 10⁶ cells/mL) that were previously incubated with 200 μg/mL of each studied peptide at 37°C for 1 h, with rotation at 250 rpm. After incubation, ELISAs were performed as described in the Section “Inhibition of Paracoccidioides spp. Adhesion to Extracellular Matrix (ECM) Components by the Selected Peptides Using Enzyme Linked Immunosorbent Assay (ELISA).” In this experiment, untreated Paracoccidioides spp. cells were also used as controls. This experiment was performed in triplicate, with three independent experiments.

The minimum inhibitory concentration (MIC) of each peptide was determined following the standardized method in the M27-A3 document CLSI (2008). Amphotericin B (Sigma) was used as a control, and the final concentrations of peptides used ranged from 400 to 0.008 μg/mL after the addition of the inocula. Both peptides and amphotericin B were prepared with RPMI 1640 medium (Gibco) supplemented with 2% glucose (RPMI 1640-2%). The fungal cells were suspended in PBS, and the number of viable cells was estimated by staining them with Trypan blue and counting them with a haemocytometer. A suspension with 10⁶ cells/mL was diluted at 1:50 in PBS, and then was diluted at 1:20 in RPMI 1640-2% Glc. The plates were incubated at 35°C, with rotation at 150 rpm, for 48 h, and the MABA (Microplate AlamarBlue Assay) was employed. Samples were incubated for an additional 24 h, for a total of 72 h for the final determination of MIC values, according de Paula e Silva et al. (2013). Three independent experiments were performed.

The cytotoxicity of the selected peptides was determined using the A549 pneumocyte linage. The peptides were evaluated at concentrations ranging from 400 to 25 μg/mL using the MTT cell viability test according Denizot and Lang (1986) and Ferrari et al. (1990). For this, plates were prepared with 10⁶ cells/well in HAM-F12 medium. Before the experiments, the medium was removed, and 100 μL of each tested dilution of the peptides in HAM-F12 was placed in contact with the cells. The plates were incubated at 37°C under 5% CO₂ in the dark for 24 h. The solutions were then removed, 10 μL of MTT solution (5 mg/L) was added, and the plates were incubated again at 37°C for 4 h. After incubation, the MTT solution was removed, 100 μl of isopropanol was added to dissolve the precipitate, and the plates were read spectrophotometrically at 595 nm DO. The dead control was 10% hydrogen peroxide. The live control was untreated cells, and the reading from this control was considered as 100% living cells. This experiment was performed in triplicate, with three independent experiments, and the percentage of viable cells was calculated using the following formula:

The toxicity of the peptides was evaluated using 2, 4, 8, 50, 100, 200, and 400 μg/larva of each peptide. For this assay, groups of 16 larvae were treated with the various concentrations tested and incubated at 37°C. The survival of the larvae was evaluated every day for 7 days after treatment. As controls, we used a group of larvae treated with PBS and a group of untreated larvae. Three independent experiments were performed.

Larval groups (n = 16) were separately treated with 100 μg/larvae of each peptide for 3 h. The larvae were subsequently infected with 5 × 10⁶ cells/larva of Paracoccidioides spp. isolates (Pb18 or Pb01) according to Scorzoni et al. (2015), and larval survival was then observed daily for 1 week. The peptide treatments and fungal inoculations were performed with 10 μL Hamilton syringes (Hamilton, USA). Larval death was evaluated based on the visual confirmation of a lack of movement after the larvae were touched with tweezers. As controls, groups of larvae treated with PBS and infected with Pb18 or Pb01 and a group of uninfected larvae only treated with PBS were used. Three independent experiments were performed.

Groups of 16 larvae were inoculated with 100 μg/larvae of each peptide and incubated at 37°C for 3 h. The peptide inoculations were performed with 10 μL Hamilton syringes (Hamilton, USA). After incubation, the haemolymph from each larva was collected separately and diluted 1:10 in PBS. The cells were counted in a haemocytometer under a bright field microscope. A group treated with PBS was used as the control. Three independent experiments were performed.

Statistical analyses and graph design were performed using GraphPad Prism version 5.00 (GraphPad Software, San Diego, CA, USA), and the results were considered significant when p < 0.05. Binding assay, MTT test and the haemocyte density test results were analyzed using one-way ANOVAs and Tukey’s post hoc tests. Survival assays were analyzed using Mantel–Cox log-rank tests.

We employed the BRASIL technique to select peptides that bound to P. brasiliensis using linear (X6) and cyclic (CX8C) peptide libraries of different sizes to increase the variety of peptides available for study. A pre-clearing step was added to our selection protocol to favor the selection of peptides that specifically bound to this species. Phage libraries were first incubated with the non-pathogenic yeast S. cerevisiae to remove ubiquitously binding peptides. The remaining unbound phages were transferred to the pathogenic P. brasiliensis strain, and bound phages were recovered for further rounds of selection. After three rounds of selection, we observed an enrichment of phages bound to P. brasiliensis compared to the second round (2.3 times for the X6 library and five times for CX8C library), indicating an increase in phages that could bind to the fungal cells with a high affinity.

At the end of the selection process, 96 individual phage clones from each library were randomly picked from the third round and sequenced, and 84 and 76 sequences were obtained from the X6 and CX8C libraries, respectively. Four peptides were selected because they were displayed by multiple phages (Table 1): three peptides from the X6 library and one from CX8C library. These peptides were selected for further study because they were likely to have higher affinities for P. brasiliensis cells.

We evaluated the binding of each peptide to P. brasiliensis and P. lutzii and to the non-pathogenic yeast S. cerevisiae. As a control, we evaluated the binding of the insertless phage Fd-tet to P. brasiliensis and P. lutzii cells. We observed that all peptides showed greater binding to P. brasiliensis (phage LVGRVV, 9.7-fold; phage LDFVVG, 7.5-fold; phage CSVSALGGAC, 10.5-fold; and phage VVAGSV, 8.1-fold) and P. lutzii (phage LVGRVV, 9.7-fold; phage LDFVVG, 30.8-fold; phage CSVSALGGAC, 4.1-fold; and phage VVAGSV, 5.9-fold) compared with the binding of the Fd-tet control phage (Figure 1). None of the phages tested were found to significantly bind to S. cerevisiae, indicating that the pre-clearing step efficiently removed the peptides that bound to these cells. In summary, our validation assays confirmed that all selected peptides bound to specific cell surface ligands of the two species of the genus Paracoccidioides, P. brasiliensis (isolate Pb18) and P. lutzii (isolate Pl01).

Assays were conducted using four different ECM components (laminin, fibronectin, and collagen types I and IV) (Figure 2) and the A549 pneumocyte line (Figure 3). The results indicate the strong potential of the peptides to inhibit the adhesion of P. brasiliensis and P. lutzii and indicate that these peptides might bind to some components of the fungal cell wall important for the Paracoccidioides-host interaction.

All the peptides tested inhibited the adhesion to all of the ECM components, with inhibition ranging from 10 to 60% depending on the strain and the ECM component (Figure 2). LVGRVV more efficiently inhibited the adhesion of P. brasiliensis to fibronectin (39%) and the adhesion of P. lutzii to laminin (57%). LDFVVG efficiently inhibited the adhesion of both fungi to fibronectin and type I and type IV collagen (from 21 to 34% inhibition). CSVSALGGAC and VVAGSV efficiently inhibited the adhesion of both fungi to laminin, with an inhibition rate ranging from 38 to 51%.

All tested peptides were able to inhibit the adhesion of P. brasiliensis and P. lutzii to A549 pneumocytes in a similar manner. The inhibition of the adhesion of the Pb18 and Pl01 strains to A549 pneumocytes ranged from 40 to 60%, respectively, for all the studied peptides (Figure 3).

The peptides did not kill or inhibit the growth of P. brasiliensis and P. lutzii at any tested concentration, which ranged from 400 to 0.008 μg/mL (data not shown).

The peptides did not show cytotoxicity. The percentage of viable cells was over 92% for LVGRVV and LDFVVG and was 100% for CSVSALGGAC and VVAGSV at a concentration of 200 μg/mL, which was the concentration used in the inh-ELISA. All concentrations lower than 100 μg/mL showed no cytotoxicity (data not shown).

No larvae died at any concentration tested, demonstrating that the peptides are nontoxic to the larvae. These results complemented the in vitro tests in which the peptides showed no toxicity to A549 pneumocytes.

All peptides significantly (p < 0.05) protected the larvae against P. brasiliensis infections (Figure 4). Survival rates of 50%, 31%, 64%, and 37% were observed for larvae treated with the LVGRVV, LDFVVG, CSVSALGGAC, and VVAGSV peptides, respectively. The survival time increased when the larvae were treated with each of the peptides. Untreated larvae began to die after 1 day, but the treated larvae began to die after 3 days for LVGRVV, 4 days for LDFVVG and CSVSALGGAC and 2 days for VVAGSV, and in all cases, 7 days was not sufficient to kill all treated larvae, whereas 7 days were sufficient to kill all untreated larvae.

Figure 5 shows the results obtained for peptide-treated larvae infected with P. lutzii. Only the VVAGSV peptide was able to significantly (p < 0.05) protect the larvae against P. lutzii infections, resulting in a survival rate of 60% at the end of the seventh day of infection. However, the LVGRVV and CSVSALGGAC peptides were able to significantly (p < 0.05) increase the survival time. Treatment with LDFVVG did not affect larval survival after P. lutzii infection. The survival time increased when the larvae were treated with each peptide. Untreated larvae began to die after 4 days, whereas larvae treated with the different peptides began to die after 5 days. After treatment with the VVAGSV peptide, 7 days were not sufficient to kill all of the larvae (Figure 5).

Larvae were treated with the selected peptides for 3 h, and we observed that peptide VVAGSV stimulated significantly the haemocyte production (p < 0.05) compared with production in the PBS control (Figure 6).