This study involves using of serum samples obtained from patients as part of routine care. The Ethics Committee on Research at Hospital das Clínicas da FMRP-USP (protocol #13982/2005) did not require the study to undergo review or approval from an ethics committee because the study used sera from patients who had already undergone laboratory testing.

The virulent strain of P. brasiliensis was obtained from Gustavo H. Goldman at the University of São Paulo, Brazil and was used for all experiments in this study. Cultures were maintained on solid, modified BHI (Brain Heart Infusion) medium (Kasvi, São José dos Pinhais, Paraná, Brazil) at 37°C for the yeast phase. For the hyphal phase, cultures were grown in liquid BHI medium at 25°C and 37°C for yeast phase. Fungi were cultivated for 7 days in liquid BHI medium with shaking.

Total RNA was extracted from yeast and hyphae cells using Trizol (TRIzol RNA isolation reagents, Invitrogen, Thermofisher, Massachusetts, USA) following the manufacturer’s instructions. RNA was resuspended in DEPC-treated water, quantified by spectrophotometry, and stored at -80°C.

Total RNA was treated with DNase using the protocol provided by the manufacturer (Biolabs Inc., New England, USA). First-strand cDNA synthesis was carried out using the Improm-II reverse transcriptase (Promega Corporation, Madison, Wisconsin, USA) following the manufacturer’s instructions. Quantitative real-time PCR was performed using SYBR Green PCR master mix (Applied Biosystems, Foster City, CA, USA). The reaction conditions were as follows: 50°C for 2 min, 95°C for 5 min, and 40 cycles of 95°C for 15 sec and 60°C for 1 min. The primer concentration used in each reaction ranged from 2 to 10 pmol. The amount of cDNA used in the qRT-PCR reaction was determined empirically by diluting the samples from 1:5 to 1:100. The expression levels of the predicted genes were calculated using the ΔCt (cycle threshold) method (Kubista et al., 2006), and normalized to the expression of the L34 gene. The specificity of the reaction was confirmed by dissociation curve analysis (95°C for 15 min, 60°C for 20 sec, and 95°C for 15 sec). The results were presented as the means of duplicate measurements in three independent experiments. Negative controls were included in each round of reactions to ensure the absence of RNA or genomic DNA contamination.

To identify and express GPI-anchored proteins in Saccharomyces cerevisiae, we selected four ORFs based on their expression levels by western blot analysis, molecular weight similarities to predicted conservation levels in other fungi, and enhanced expression in yeast. We used the pEGH vector, which was modified to be compatible with the Gateway System to create the pEGH-A vector, and expressed the proteins as fusions with a tail of six histidine (6His) and GST (Glutathione-S-Transferase) at the N-terminus (Zhu et al., 2001). The pEGH-A vector contains a galactose-inducible promoter, an ampicillin resistance marker, and a URA3 auxotrophic marker. Transformants were plated on SC without URA solid medium (2% glucose) and incubated at 30°C for 48 h, followed by inoculation in 25 ml of SC without URA liquid medium (2% raffinose) and incubation in a shaker at 30°C and 170 rpm for 48 h. Upon reaching optimal optical density, ORF expression was induced by the addition of 4% galactose filtered for 18 h at 30°C. After induction, cells were centrifuged at 3,800 x g for 20 min, and the insoluble fraction was prepared for GPI-anchored protein extraction.

The purification was carried out as follows: cells expressing the target proteins were resuspended in 800 µL of lysis buffer per 289.3 mL of culture, along with 800 µL of 0.5 mm zirconium beads, and homogenized by vortexing for 10 cycles of 1 min with 20 sec intervals at 4°C. After lysis, the samples were washed with 800 µL of lysis buffer, and the supernatant was collected by centrifugation at 13,000 x g for 10 min at 4°C. The supernatant was further clarified by centrifugation, and the soluble extract containing the GST-fused target proteins was incubated overnight at 4°C with glutathione sepharose resin equilibrated with Buffer A. The flowthrough was collected, and the resin was washed with 12 mL of Wash Buffer, and the eluate was collected. The resin was incubated with 10 mL of elution buffer (Buffer A with 30 mg of reduced glutathione) for 3 h at 4°C under vigorous stirring, and the elution was collected. The samples were then concentrated by means of the Amicon system with three washes of 15 mL of 1 M Tris HCl pH 7.4 at 1800 x g for 40 min. The lysis buffer contained 100 mM Tris HCl pH 7.4, 1 mM NaCl, 1 mM EDTA pH 8.0, 0.1% β-mercaptoethanol, 0.5 mM PMSF, 0.1% Triton x-100, and protease inhibitor cocktail. The resin was first treated with the column buffer and then equilibrated with Buffer A. The column buffer contained 100 mM Tris HCl pH 7.4, 150 mM NaCl, 2 mM DTT, 1 mM EDTA, and 0.2 mM sodium azide.

The immunoblot analysis was performed to assess the reactivity of the recombinant GPI-anchored proteins with sera from patients with various infections or healthy individuals. The polyacrylamide gel containing the proteins was transferred to a nitrocellulose membrane using a transfer buffer. Non-specific interactions were blocked by incubating the membrane with TBS-T containing skim milk powder. The membrane was then incubated with sera from patients and anti-GST antibody, and the reactivity of the recombinant GPI-anchored proteins was compared to that of the protein GP43 used in the diagnosis of paracoccidioidomycosis (PCM). After the incubation period, the membrane was washed with TBS-T, and human anti-IgG secondary antibody conjugated with peroxidase was added. The membrane was then revealed using a substrate for peroxidase and exposed to X-ray film. The sera used in the analysis were obtained from Prof. Dr. Roberto Martinez-HCFMRP-USP.

Polystyrene microplates (Nunc®, Denmark) were coated with recombinant proteins from P. brasiliensis (1 g/well), which were diluted in 0.2 M sodium carbonate buffer (pH 9.6) and incubated for 16 h at 4°C. The microplate wells were then washed three times with TBS wash buffer containing 0.05% Tween 20 (v/v) and blocked with TBS containing 0.05% Tween 20 and 3% gelatin (Difco, Merck, Sigma-Aldrich®, Darmstadt, Germany) (200 µL/well), and incubated for 2 h at 37°C. After blocking, the plates were incubated for 2 h at 37°C with patient sera. The wells were then washed and incubated for 1 h at 37°C with human IgG conjugated with peroxidase. The wells were washed again, and the reaction was developed with a substrate solution for peroxidase, which contained 5 mg ortho-phenylenediamine (OPD) (Merck, Sigma-Aldrich®, Darmstadt, Germany), 12.5 mL of 0.1 M citric acid (Merck), Na2HPO4 0.2 M (Synth, Sigma-Aldrich®), and 19 µL of 30% H2O2 (Synth, Sigma-Aldrich®). The reaction was stopped by adding 50 µL/well of 2 M H2SO4 (Merck, Sigma-Aldrich®, Darmstadt, Germany), and the absorbance was measured at 450 nm using an ELISA reader (PowerWave X Biotek Instruments). All trials were monitored without antigen controls, without sera of patients, and with only peroxidase conjugated to human IgG.

The data were analyzed using the statistical program “one-way ANOVA” (analysis of variance), followed by a Tukey test to compare the means of the groups. Prism 9.3.1 software (GraphPad, CA, USA) was used for all analyses. The difference between the data was considered significant when p < 0.05.

A set of 22 GPI-proteins were found to be differentially expressed (>2-fold) in hyphae as compared to yeast (H/Y) ( Figure 1 ), including hydrophobin 2 (PADG_07472), which exhibited a remarkable 355-fold increase in mycelia relative to yeast and an 8.07-fold increase compared to L34 (H/L34) ( Figure 1 ). Additionally, a group of GPI-proteins of unknown function, including PADG_03914, PADG_02867, PADG_04289, PADG_05482, PADG_07620, PADG_04649, PADG_08385, PADG_07354, PADG_06677, PADG_02955, PADG_06557, and PADG_00497, showed elevated expression in mycelia, indicating their potential roles in regulating the mycelial phase of P. brasiliensis ( Figure 1 ).

Two predicted proteins with the CFEM domain, including the homolog of ECM33 (PADG_04499), a glucanosyltransferase (GEL2/PADG_05076), and a CRH-GH16 glycosylase (PADG_04416), exhibited differential expression in hyphae. Notably, the homolog of ECM33 was highly expressed in mycelium (5.48-fold increase as compared to L34) and exhibited a 44-fold increase in expression in hyphae compared to yeast ( Figure 1 ). The expression of the glucanosyltransferase and CRH-GH16 glycosylase were also significantly higher in the mycelial phase (6-fold and 5.06-fold, respectively, compared to L34) and exhibited higher expression in hyphae than in yeast ( Figure 1 ). We found a glycoside hydrolase (DCW1), namely PADG_1494, which showed the similar expression in mycelium and yeast cells (data not shown). These differentially expressed GPI-proteins may play important roles in the regulation of mycelial growth and morphogenesis in P. brasiliensis.

We identified a group of 10 GPI-anchored proteins that showed significantly higher expression levels in yeast cells (Y/L34), with over 2-fold induction compared to hyphal growth (Y/H) ( Figure 2 ). This group includes several proteins of unknown function (UK), PADG_067040, PADG_04373, PADG_08006, PADG_03262, PADG_06841, and PADG_02891, as well as three proteins with known functions (PADG_00795, PADG_02842, and PADG_08524). Of particular interest, a member of the Gas family was found to be highly expressed in yeast cells, suggesting a potential role in the processing of glucans during this morphological phase ( Figure 2 ). Further studies are needed to elucidate the specific functions of these differentially expressed GPI-anchored proteins in yeast cells.

We cloned 46 predicted GPI-anchored proteins in a yeast expression vector and purified them as N-terminal GST-tagged proteins in a high throughput 96-well format (data not shown) (Zhu et al., 2001). We selected four GPI-anchored proteins to evaluate their immunoreactivity to sera obtained from 7 patients with paracoccidioidomycosis (PCM) and to the healthy controls, using the Western blot and ELISA analyses. Two of these proteins (i.e., PADG_04617/cutinase and PADG_01494/glycosyl hydrolase) had similar RNA expression during yeast and hyphal growth, one protein (PADG_06557, unknown function) had higher expression during hyphal growth, and one protein (PADG_02842, SOD3) had higher expression during yeast growth.

Therefore, we decided to examine the reactivity of these proteins using the Western blot analysis. We showed that rPADG_02842 was recognized by sera from six patients with PCM (PB1-PB7) and showed no reactivity to serum sample of those patients diagnosed with histoplasmosis, aspergillosis, or candidiasis ( Figures 3A–D ). Since one patient serum (C6) showed non-specific reactivity, we investigated the patient’s history and found that this patient is presented fungal pneumonia, possibly caused by Paracoccidioides spp. ( Figure 3D ). PADG_02842 has homology to superoxide dismutase (SOD3) in other organisms and has counterparts in 12 other fungi close to P. brasiliensis and P. lutzii, including Blastomyces dermatitidis, Histoplasma capsulatum, Coccidioides immitis, Microsporum canis, Trichophyton rubrum, Aspergillus nidulans, Penicillium marneffei, Neurospora crassa, Fusarium oxysporum, Magnaporthe grisea, and Candida albicans (Gonçales et al., 2021). Recent studies by Tamayo and colleagues (2016) have shown that PADG_02842 is a virulence factor in P. brasiliensis in knockdown studies (Tamayo et al., 2016). The present study found that PADG_02842 has 15.9% serine and threonine and 9 predicted domains for N-glycosylation and 5 for O-glycosylation. Despite the presence of homologs in other fungi and a small number of glycosylation sites, this protein showed specificity for patients with PCM.

Our analysis showed that rPADG_01494 was recognized by five sera from patients with PCM (PB1, PB2, PB4, PB5, and PB6) ( Figure 4A ). In some of these samples, a fragment predicted for this protein was also detected. rPADG_01494 showed no cross-reactivity with sera from patients with histoplasmosis (H1 and H2), aspergillosis (A2), candidiasis (CA1 and CA2) ( Figures 4B, C ), or unrelated diseases (C1, C4, and C6) ( Figure 4D ). In these samples, fragments of different molecular weights to the predicted rPADG_01494 were also detected. As noted in other recombinant proteins, serum of a patient diagnosed with fungal pneumonia (C6) showed cross-reactivity ( Figure 4D ). PADG_01494 is predicted in other organisms as a glycoside hydrolase (DCW1) and is conserved in 15 other fungi, including B. dermatitidis, H. capsulatum, C. immitis, M. canis, T. rubrum, A. nidulans, P. marneffei, U. reesi, N. crassa, F. oxysporum, M. grisea, C. albicans, S. cerevisiae, S. pombe, and C. neoformans (Gonçales et al., 2021). The ORF shows 15.3% serine and threonine, with 5 possible domains for N-glycosylation and 10 for O-glycosylation.

The Western blot analysis showed that rPADG_06557 was recognized by sera from six patients with PCM (data not shown). In addition, rPADG_06557 was recognized by a serum of a histoplasmosis patient (H2) and a candidiasis patient (CA2) and showed no reactivity to sera of patients with aspergillosis (data not shown). However, sera from two patients with unrelated diseases (C4 and C6) showed nonspecific reactivity (data not shown). PADG_06557 is of unknown function and is preserved only in three other fungi: A. nidulans, M. canis, and C. immitis (Gonçales et al., 2021). The ORF shows 24.1% serine and threonine, with 6 possible domains for N-glycosylation and 71 for O-glycosylation.

The rPADG_04617 protein was recognized by six sera from patients with PCM in our Western blot analysis. The protein showed cross-reactivity with sera from patients with histoplasmosis (H2), aspergillosis (A1) or unrelated diseases (C1, C2, and C4) (data not shown). Similarly, two proteins, namely rPADG_02842 and rPADG_04617, showed non-specific reactivity against serum from a patient diagnosed with pneumonia (C6). PADG_04617 is predicted in other organisms as a possible protein belonging to the family of cutinases. This protein is conserved in 10 other fungi, including B. dermatitidis, H. capsulatum, C. immitis, M. canis, T. rubrum, P. marneffei, U. reesi, N. crassa, F. oxysporum, and M. grisea (Gonçales et al., 2021). The percentage of serine and threonine is 13.3%, with 3 possible domains for N-glycosylation and O-glycosylation.

To firmly establish a sensitive diagnostic test for Paracoccidioides spp., it is important to identify the epitopes produced by the pathogen in the blood. To this end, we analyzed the reactivity of recombinant GPI-anchored proteins (rGPI-proteins) using an enzyme-linked immunoenzymatic assay. Specifically, we examined the reactivity of four rGPI-proteins, including rPADG_02842, rPADG_04617, rPADG_01494, and rPADG_06557.

Our results showed that the rPADG_02842 protein ( Figure 5A ) demonstrated significant differences in reactivity compared to sera from healthy subjects, as well as sera from patients with other fungal diseases, such as candidiasis and histoplasmosis, and unrelated diseases. The other three proteins tested in this study, including rPADG_01494, rPADG_04617 and rPADG_06557 ( Figures 5B–D ), were found to exhibit significant differences in reactivity for sera from patients with PCM compared to those with candidiasis and histoplasmosis. In Figure 5E , the reactivity of GST is compared to the reactivity of rPADG_02842 and gp43. The rPADG_02842 and GP43 proteins exhibited statistically significant differences in reactivity compared to the GST-tag ( Figure 5E ). These proteins did not show any difference in reactivity in healthy patients’ sera, or sera from patients with aspergillosis or unrelated diseases.