Paracoccidioides brasiliensis, Pb18 (chronic PCM; São Paulo, Brazil; P. brasiliensis, Pb18) (Turissini et al., 2017) was used in all experiments. The yeast cells were maintained by sub culturing at 36°C in Fava Netto’s solid medium every 7 days (Fava-Netto, 1955). After this period, the cells were transferred to Fava Netto’s liquid medium for 72 h at 36°C under agitation at 150 rpm and used to perform the experiments.

The alveolar macrophages AMJ2-C11 (Rio de Janeiro Cell Bank – BCRJ/UFRJ, accession number 0039) are originated from Mus musculus. The cells were maintained in DMEM medium (Vitrocell Embriolife, Campinas, São Paulo, Brazil) containing bovine fetal serum 10% (v/v) at 36°C and 5% CO₂. The culture medium was changed after the cells reach complete confluence.

Paracoccidioides brasiliensis, Pb18 infection in AMJ2-C11 alveolar macrophages was performed in triplicates on 12-well polypropylene plates. In each well was plated a total of 10⁶ macrophages cells in DMEM medium containing or not IFN-γ (1 U/mL) (Sigma–Aldrich, St. Louis, MO, United States) and bovine fetal serum 10% (v/v). The plates were incubated for 12 h at 36°C and 5% CO₂ until complete confluence. The medium was removed and a fresh DMEM medium containing or not IFN-γ (1 U/mL) and bovine fetal serum 10% (v/v) plus 5 × 10⁶ Pb18 cells were added to the macrophages. For all infection assays, the yeast cells obtained from a 72 h inoculum in Fava Netto’s liquid medium were passed through 0.70 μm-pore membrane filters, with the aid of syringe and needle. The plates were incubated at 36°C and 5% CO₂ during 3, 6, 9, and 12 h. The wells were gently washed three times with sterile phosphate-buffered saline (PBS) and the macrophages were lysed by the addition of sterile water. The yeast cells were recovered by centrifugation at 8,000 × g for 10 min (Parente-Rocha et al., 2015). The pellet was diluted (1:100) and plated in solid brain heart infusion (BHI) medium, supplemented with bovine fetal serum 4% (v/v) and glucose 4% (v/v). Yeast cells viability was evaluated based on the number of colony-forming units (CFU), determined after growth at 36°C for 10 days. The control was obtained by incubating 5 × 10⁶ cells/mL in DMEM medium, added of bovine fetal serum 10% (v/v) for 6 h at 36°C and 5% CO₂. Control fungal cells were washed with sterile water prior to protein extraction. Obtainment of adhered and internalized fungal cells indexes after interaction with primed and non-primed macrophages were performed by counting a total of 600 events (macrophages) in cover lips stained with Giemsa, as previously described (Gori and Scasso, 1994).

The proteomic analysis was carried out after 6 h post interaction of fungal and macrophage cells. Fungal protein extracts were obtained in biological triplicates from the three conditions to be analyzed: Pb18 cells recovered of primed macrophages (Pb18_P); Pb18 cells recovered of non-primed macrophages (Pb18_NP) and control cells (Pb18 cells not interacting with macrophages). The yeast cells were collected by centrifugation at 8,000 × g for 10 min and washed once with RapiGEST SF Surfactant 0.1% (v/v) (Waters Corporation, Billerica, MA, United States), followed by washing with ultrapure water and PBS 1×, in order to remove any contamination of the macrophage cells (Parente-Rocha et al., 2015; Pigosso et al., 2017). The pellet was ressuspended in extraction buffer (20 mM Tris–HCl, pH 8.8, and 2 mM CaCl₂) and distributed in tubes containing glass beads (Sigma–Aldrich, St. Louis, MO, United States) in equal volume of the material. The suspension was processed on ice in BeadBeater equipment (BioSpec, Products Inc., Bartlesville, OK, United States) during three cycles of 30 s. The cell lysate was centrifuged at 10,000 × g during 15 min at 4°C and the protein content in the supernatant was quantified using the Bradford reagent (Sigma–Aldrich, St. Louis, MO, United States) using bovine serum albumin (BSA) as standard.

Proteins were enzymatically digested with tripsin as described previously, with some modifications (Murad et al., 2011). Briefly, a total of 150 μg of protein (previous item) of each sample was added to 10 μL of 50 mM ammonium bicarbonate, pH 8.5, in a microcentrifuge tube. Next, 75 μL of RapiGEST^TM SF Surfactante (0.2% v/v) (Waters Corporation, Billerica, MA, United States) was added and the sample was vortexed and incubated in a dry bath at 80°C for 15 min. In each sample were added 2.5 μL of 100 mM dithiothreitol (GE Healthcare, Piscataway, NJ, United States), at 60°C for 30 min, while cysteines were alkylated by the addition of 2.5 μL of 300 mM iodoacetamide (GE Healthcare, Piscataway, NJ, United States) for 30 min, at room temperature in the dark. The digestion of proteins was performed by the addition of 30 μL of trypsin 0.05 μg/μL (Promega, Madison, WI, United States) at 37°C, in dry bath, for 16 h. Then, 30 μL of trifluoroacetic acid (TFA) solution 5% (v/v) was added to the samples, followed by incubation for 90 min at 37°C, for digestion stop, and precipitation of the RapiGEST reagent. The samples were centrifuged at 18,000 × g for 30 min and the supernatant was transferred to a new tube and dried in a speed vacuum (Eppendorf, Hamburg, Germany) for 2 h. The pellet containing peptides was suspended in 80 μL of a solution containing 20 mM of ammonium formiate and 150 fmol/μL of PHB (Rabbit Phosphorylase B) (Waters Corporation, Billerica, MA, United States) (MassPREP protein), as internal standard. The tryptic peptides were analyzed using a nanoACQUITY UPLC^® M-Class system (Waters Corporation, Billerica, MA, United States) coupled to Synapt G1 MS^TM mass spectrometer (Waters Corporation, Billerica, MA, United States), equipped with a NanoElectronSpray source and two mass analyzers: a quadrupole and a time-of-flight (TOF) operating in the V-mode. Nanoscale LC separation of tryptic peptides was performed with two reverse phase columns. The peptides were separated using a gradient of 11.4, 14.7, 17.4, 20.7, and 50% (v/v) of acetonitrile/0.1% (v/v) formic acid, with a flow rate of 2.000 μL/min. Data were obtained using the instrument in the MS^E mode, which switches the low energy (6 V) and elevated energy (40 V) acquisition modes every 0.4 s. The lock mass was used for calibration of the apparatus, using a constant flow rate of 0.2 μL/min at concentration of 200 fmol protein GFP [Glu]¹-Fibrinopeptide B human (m/z 785.8426) (Sigma–Aldrich, St. Louis, MO, United States). The samples were analyzed in triplicate, from three independent experiments.

Mass spectrometry raw data of peptide fractions were processed using the ProteinLynx Global SERVER (PLGS) platform (Waters Corporation, Billerica, MA, United States). Then, the processed spectra were searched against P. brasiliensis, Pb18 protein sequences together with reverse sequences. The mass error tolerance for peptide identification was under 50 ppm. Protein identification criteria were as following: detection of at least two fragment ions per peptide; five fragment ions per protein; the determination of at least one peptide per protein; carbamidomethylation of cysteine as a fixed modification; phosphorylation of serine, threonine, and tyrosine; and oxidation of methionine were considered as variable modifications; maximum protein mass (600 kDa); one missed cleavage site was allowed for trypsin; maximum false positive ratio (FDR) of 5% was allowed. For the analysis of the level of protein quantification, the observed intensity measures were normalized using a protein that showed lower coefficient of variance between the different conditions analyzed and present in all replicates. Expression^E informatics v.3.0.2 was used for proper quantitative comparisons. The identified proteins were organized by the expression algorithm, into a statistically significant list, corresponding to induced and reduced regulation ratios between the different conditions analyzed. The mathematical model used to calculate the ratios was a part of the Expression algorithm inside the PLGS software (Waters Corporation, Billerica, MA, United States) (Geromanos et al., 2009). The minimum repeat rate for each protein in all replicates (nine in total for each condition) was 6. Proteins that presented 40% of differences in expression values, when compared among the different conditions, were considered regulated. Tables of peptides and proteins generated by the PLGS were merged, and the data of dynamic range, peptide detection type, and mass accuracy were calculated for each sample, as previously described using the software MassPivot v1.0.1 (Murad and Rech, 2012), FBAT (Lange et al., 2004), Spotfire^® v8.0 program (TIBCO software), and Microsoft Excel^® (Microsoft) was used for table manipulations. Uniprot¹ and Pedant on MIPS² database were used for functional classification. NCBI database was employed for annotation of uncharacterized proteins³. The heat maps were performed by using MultiExperiment Viewer tool version 4.9⁴.

To evaluate the cell wall components glucans, glycosylated proteins, and chitin contents, the yeast cells recovered of macrophages were stained with aniline blue (AB) (Sigma Aldrich, Missouri, United States) for 5 min, concanavalin A (ConA) TYPE VI conjugated to FITC 100 μg/mL (Sigma–Aldrich, St. Louis, MO, United States) for 30 min and calcofluor white (CFW) 100 μg/mL (Sigma–Aldrich, St. Louis, MO, United States) for 30 min (de Curcio et al., 2017). These experiments were performed independently for each substance. Stained samples were visualized under a fluorescence microscope (Zeiss Axiocam MRc-Scope A1, Oberkochen, Germany). Images were obtained at bright field, at 340–380 nm for AB, 470–480 nm for Con A, and at 395–440 nm for CFW. A minimum of 100 cells on each microscope slide were used to evaluate fluorescence intensity in triplicates. The AxioVision Software (Carl Zeiss AG, Germany) determined the fluorescence intensity (in pixels) and standard error of each analysis. Statistical comparisons were performed using the Student’s t-test and p ≤ 0.05 was considered statistically significant.

Paracoccidioides brasiliensis, Pb18 yeast cells recovered from macrophages and control cells (as previously described) were centrifuged at 8,000 × g for 10 min. The pellet was diluted in PBS 1× to a concentration of 1 × 10⁶ cells/mL. To label mitochondria, the cells were stained with Mitotracker Green FM (400 nM; Molecular Probes, M7514) for 45 min at 37°C. Then, the cells were washed three times with PBS 1× and labeled with rodhamine (2.4 μM) for 45 min at 37°C, according to the manufacturer’s protocol (Invitrogen, Carlsbad, CA, United States) and washed three times with PBS 1×. Labeled cells were observed under a fluorescence microscope (Zeiss Axiocam MRc-Scope A1, Oberkochen, Germany) and photographed at bright field, and at 450–490 nm for the Mitotracker and 546–512 nm for rosamine dye probes. A minimum of 100 cells on each microscope slide were used to evaluate fluorescence intensity in triplicate. The AxioVision Software (Carl Zeiss AG, Germany) determined the fluorescence intensity (in pixels) and standard error of each analysis. Statistical comparisons were performed using the Student’s t-test and p ≤ 0.05 was considered statistically significant.

Paracoccidioides brasiliensis, Pb18 infection in AMJ2-C11 alveolar macrophages was performed to determine the best time of infection, with higher rate of internalization/adhrence of viable fungi cells. Times of 3, 6, 9, and 12 h of infection were analyzed regarding the number of CFUs in Pb18 yeast cells recovered of primed macrophages (Pb18_P) and non-primed macrophages (Pb18_NP). Figure 1A shows that the number of viable fungal cells increases inside the primed macrophage until 6 h. After this time, the number of viable fungi decreases in primed macrophages, suggesting killing of yeast cells by the macrophages after 9 h of interaction. In this way, the proteomic response of P. brasiliensis, after 6 h of interaction with macrophages, was investigated to analyze the differences in the initial proteomic response of the fungus during interaction with macrophages with different activation patterns. In order to evaluate the index of fungal cells adhered/internalized cover lips stained by Giemsa were analyzed after 6 h of interaction of fungal cells with primed and non-primed macrophages. The results are shown in Figure 1B. After interaction with non-primed macrophages, 42% of fungal cells interacting with macrophages are internalized and 58% are adhered to macrophages. A slight increase in the internalization was observed after interaction of fungal cells with primed macrophages (48%) while the percentage of adhered cells index was 52% after interaction of fungal cells with primed macrophages.

Protein extracts were obtained in biological triplicates. Protein quantification was performed using Nano-UPLCMS^E and the protein and peptide data were generated by the PLGS (Supplementary Figure 1). The false positive rates of proteins obtained from Pb18_P were 3.80%, Pb18_NP 5.88%, and control 2.99%. Those experiments resulted in 53,745; 47,941; and 61,561 identified peptides, from Pb18_P; Pb18_NP; and control, respectively. The values of 50.1, 49.2, and 53.9% were obtained from peptide match type data in the first pass and 7.5, 7.6, and 9.7% from the second pass, for Pb18_P, Pb18_NP, and control, respectively. The percentages of 16.2, 16.5, and 14% of total peptides were identified by a missed trypsin cleavage, whereas an in-source fragmentation was 11, 11.2, and 11.8% for Pb18_P, Pb18_NP, and control, respectively. The peptides identification within the first and second pass (PepFrag 1 and 2) were predominantly higher than 56%, and source fragmentation and missed cleavage values did not exceed 20% in all analyzed conditions (Supplementary Figure 1A; Murad and Rech, 2012). The peptide parts per million errors (ppm) indicated that the majority, 82, 83.9, and 79.3% of identified peptides, were detected with an error of less than 10 ppm for Pb18_P, Pb18_NP, and control, respectively (Supplementary Figure 1B). The obtained results from dynamic range detection indicated that a 3-log range concentration and a good detection distribution of high and low molecular weights were obtained in all analyzed conditions (Supplementary Figure 1C).

Five hundreds and thirty eight proteins were detected, considering the proteomic analysis. The description of all detected proteins is shown in Supplementary Table 1. Supplementary Figure 2 ilustrates this data. Among proteins identified during interaction of fungal cells with primed macrophages, 199 proteins were detected in similar amounts and were considered not regulated proteins. A total of 135 proteins were up-regulated in fungal cells after interaction with primed macrophages. Proteins detected only in yeast cells recovered from primed macrophages, in number of 54 proteins were also considered up-regulated. The down regulated proteins include those detected in higher amounts in the control fungal cells (67 proteins) and proteins detected only in the control condition in number of 34. The data obtained from fungal cells after interaction with non-primed macrophages include: 140 proteins that were detected in similar amounts in control fungal cells and in yeast cells after interaction with non-primed macrophages, 222 proteins up-regulated in the fungal cells after interaction with non-primed macrophages, and 54 proteins detected only after interaction with primed macrophages. Concerning to down regulated proteins, 15 were detected in higher amounts in control fungal cells and 49 proteins were exclusively detected in control fungal cells.

Protein extracts of P. brasiliensis, Pb18 recovered of primed macrophages were compared with control samples by using MS^E technology. A total of 290 regulated proteins was detected in Pb18_P; from those 189 were up-regulated proteins and 101 down-regulated proteins (Supplementary Table 2). Supplementary Table 1 shows that after 6 h of interaction with primed macrophages, P. brasiliensis presents accumulation of proteins and enzymes from the TCA cycle, electron transport chain, and ATP synthesis, proteins related to oxidative stress protection, such as thioredoxin reductase (TrxR) (accession number PADG_01551) and SOD (accession number PADG_07418), and virulence factors such as serine proteinase (PADG_07422). Also, potential transcription factors are accumulated. In counterpart, down-regulation of proteins related with the β-oxidation of fatty acids was described. The regulation of proteins related to the tricarboxylic acid cycle, electron transport chain, ATP synthesis, and oxidation of fatty acids is visualized in the heat map presented in Figure 2.

A total of 340 regulated proteins were detected in Pb18 yeast cells recovered from macrophages not incubated with IFN-γ. From the total, 276 were up-regulated proteins and 64 were down-regulated (Supplementary Table 3). P. brasiliensis, Pb18 cells recovered from macrophages not incubated with IFN-γ showed an increase of proteins related to pentose-phosphate pathway, TCA, methylcitrate cycle, electron transport, and ATP synthesis, as well as transcription factors. Molecules related to cell rescue, defense, and virulence are also particularly accumulated. Enzymes of carbohydrate metabolism, related to the synthesis of cell wall precursors, were also induced. The negative regulation of enzymes from beta-oxidation of fatty acids was also described (Supplementary Table 3).

Yeast cells interacting with non-primed macrophages depicted faster metabolic adaptation, activating alternative routes such as methyl citrate cycle, in addition to amino acid degradation, pentose-phosphate pathway, electron transport chain, and ATP synthesis, as depicted in Figure 3. These data indicate that the intracellular environment of primed macrophages presents barriers that may hinder the adaptation of the fungus to the cell environment. Corroborating this hypothesis, Pb18_NP cells present a higher number of proteins and abundance of enzymes related to response to oxidative stress and virulence factors, as shown in Table 1 and Figure 3. SOD2, for example, had its expression more than twice higher in Pb18_NP compared to Pb18_P.

We observed an increase of enzymes that participate in the synthesis of precursors of cell wall components as chitin, glycan, glycoproteins, and glycosylated compounds, mainly in Pb18_NP (Table 2; Figure 4). Enzymes as neutral alpha-glucosidase AB, mannosil-oligosaccharide glucosidase, and alpha-mannosidase are involved in glycoprotein biosynthesis through the N-glycosylation process (Almeida et al., 2016). UDP-galactopyranose mutase, UTP-glucose-1-phosphate uridylyltransferase, and UDP-N-acetylglucosamine pyrophosphorylase form UDP-sugars, substrates used in hundreds of glycosylation reactions (e.g., for protein and lipid glycosylation, synthesis of sucrose, and cell wall polysaccharides). UDP-N-acetylglucosamine pyrophosphorylase generates UDP-N-acetyl-D-glucosamine, which is used in the synthesis of chitin, for example (Decker et al., 2017). The up-regulation of the proteins involved in carbohydrate synthesis, interconversion, and utilization is shown in Figure 4. These data indicate that P. brasiliensis may be synthesizing components of the cell wall during infection, especially within non-primed macrophages.

Proteomic data were confirmed by fluorescence microscopy using aniline blue, ConA, and CFW as markers for glucans, carbohydrates residues in proteins (predominantly), and chitin, respectively (Figure 5). P. brasiliensis yeast cells recovered from non-primed macrophages presented higher fluorescence intensity of aniline blue, ConA, and CFW, suggesting higher content of glucans, carbohydrates residues from proteins, and chitin compared to fungal cells recovered from primed macrophages (Figure 5).

Increase of proteins related to the electron transport chain and ATP synthesis complex were observed in yeast cells recovered from primed and non-primed macrophages. We used MitoTracker Green FM as a total mitochondrial dye (green) and the mitochondrial activity was evaluated by using rhodamine dye (red), which stains selectively according to mitochondrial membrane potential, as depicted in Figure 6. The fluorescence intensity of rhodamine was significantly increased in Pb18_NP compared to Pb18_P.