Recombinant Paracoccin Reproduces the Biological Properties of the Native Protein and Induces Protective Th1 Immunity against Paracoccidioides brasiliensis Infection

Background Paracoccin is a dual-function protein of the yeast Paracoccidioides brasiliensis that has lectin properties and N-acetylglucosaminidase activities. Proteomic analysis of a paracoccin preparation from P. brasiliensis revealed that the sequence matched that of the hypothetical protein encoded by PADG-3347 of isolate Pb-18, with a polypeptide sequence similar to the family 18 endochitinases. These endochitinases are multi-functional proteins, with distinct lectin and enzymatic domains. Methodology/principal findings The multi-exon assembly and the largest exon of the predicted ORF (PADG-3347), was cloned and expressed in Escherichia coli cells, and the features of the recombinant proteins were compared to those of the native paracoccin. The multi-exon protein was also used for protection assays in a mouse model of paracoccidioidomycosis. Conclusions/Significance Our results showed that the recombinant protein reproduced the biological properties described for the native protein—including binding to laminin in a manner that is dependent on carbohydrate recognition—showed N-acetylglucosaminidase activity, and stimulated murine peritoneal macrophages to produce high levels of TNF-α and nitric oxide. Considering the immunomodulatory potential of glycan-binding proteins, we also investigated whether prophylactic administration of recombinant paracoccin affected the course of experimental paracoccidioidomycosis in mice. In comparison to animals injected with vehicle (controls), mice treated with recombinant paracoccin displayed lower pulmonary fungal burdens and reduced pulmonary granulomas. These protective effects were associated with augmented pulmonary levels of IL-12 and IFN-γ. We also observed that injection of paracoccin three days before challenge was the most efficient administration protocol, as the induced Th1 immunity was balanced by high levels of pulmonary IL-10, which may prevent the tissue damage caused by exacerbated inflammation. The results indicated that paracoccin is the protein encoded by PADG-3347, and we propose that this gene and homologous proteins in other P. brasiliensis strains be called paracoccin. We also concluded that recombinant paracoccin confers resistance to murine P. brasiliensis infection by exerting immunomodulatory effects.


Introduction
Paracoccidioides brasiliensis [1,2] is a thermo-dimorphic fungus that grows as a mycelium at 25uC and as a yeast at 37uC. P. brasiliensis infection in humans is known as paracoccidioidomycosis (PCM), a systemic granulomatous disease with high prevalence in Latin America [3]. An estimated 10 million people are infected with P. brasiliensis most people do not show clinical evidence of PCM.
The infection is acquired by inhalation of airborne propagules [4]. Once in the lungs, the fungus is stimulated by body temperature, and then it activates a number of genes that transform the conidia into the pathogenic form [5,6]. Such dimorphism is an important feature of several pathogenic fungi, including P. brasiliensis [7]. The cell wall is involved in the morphogenetic changes that occur during this transition, and it plays key roles in the pathobiology of PCM [8].
Cell-mediated immunity is the main mechanism of defense against PCM [9], whereas a high humoral response correlates with disease dissemination [10]. Therefore, resistance to PCM is linked to high IFN-c production by T helper type 1 (Th1) lymphocytes, and PCM susceptibility is associated with the lack of a cellular response and B cell activation [11,12]. Early secretion of IL-12, followed by sustained secretion of IFN-c, are important for resistance against P. brasiliensis infection [13]. The most physiologically important IL-12 target cells are T lymphocytes, which proliferate and differentiate into cells that produce type-1 cytokines, particularly IFN-c [14].
We previously demonstrated that extracts of P. brasiliensis yeast contain an N-acetylglucosamine-binding lectin called paracoccin [15]. This protein binds to laminin in a sugar recognitiondependent manner, induces macrophages to produce TNF-a and NO, and is related to fungal growth [16] and morphogenesis [17]. In addition, paracoccin exerts N-acetylglucosaminidase activity [18]. Proteomic studies have showed that a paracoccin preparation requires a multi-domain protein, annotated as PADG-3347 (http://www.broadinstitute.org/annotation/genome/paracoccidio ides_brasiliensis/GeneDetails.html?sp = S7000001960922883) that shares sequence similarity with the dual function proteins of chitinase family 18, and has an additional domain that contains the N-acetylglucosamine-binding activity [19]. These features explain the influence of paracoccin on fungal cell growth and morphogenesis, and its ability to interact with the host extracellular matrix and immune cells [20]. In this study, we cloned and expressed the putative recombinant paracoccin to compare the features of the recombinant protein to those of native paracoccin and to investigate the effect of administering the recombinant protein on an experimental model of PCM.

Ethics statement
The study was approved by the Ethical Committee for Ethics in Animal Research (CETEA) of the School of Medicine at Ribeirão Preto of the University of São Paulo. All animal experiments were conducted in accordance with the Ethical Principles in Animal Research adopted by the Brazilian College of Animal Experimentation (COBEA) (Protocol 20/2013-1).

P. brasiliensis strain and accession number
The P. brasiliensis isolate used in this study, Pb18, was kindly provided by Dr. Roberto Martinez (Faculty of Medicine in Ribeirão Preto, University of São Paulo). Yeast cells were cultivated on Fava-Netto semisolid medium, YPD agar, and BHI broth. Virulence was maintained by consecutive intravenous infections in mice. The yeast were recovered from mouse lung tissue and then cultured on Fava-Netto medium at 37uC for 7 days. The viability of the yeast cells was determined by fluorescein diacetate and ethidium bromide staining [21], and it was always greater than 90%.

Cloning of paracoccin
The gene encoding paracoccin is composed of five exons and four introns. Therefore, paracoccin was cloned in two different ways. First, the largest exon was amplified from P. brasiliensis genomic DNA using the oligonucleotide primers F PADG (59-CTGGATCCATGCAAGCACCCGACCAAC-39) and R PADG (59-CGGGAATTCCTACCAACTCGTTATTGATAGAGCGA-TAA-39), and then cloned into pGEX-4T-1 (GE Healthcare, Upsala, Sweden) flanked by BamHI and EcoRI restriction sites. Second, the full-length, predicted ORF encoding paracoccin was synthesized, and cloned into the pUC57 vector (GenScript, Piscataway, NJ, USA). In the first strategy, the paracoccin fragment, referred to as rPCN exon4 , is fused to a GST tag in the pGEX vector, whereas in the second strategy, the 59-UTR region, which drives transcription of paracoccin (rPCN full ) was also synthesized. Competent E. coli BL21 (DE3) cells were transformed with the recombinant vectors, and clones resistant to ampicillin were screened for the presence of the inserts by PCR. Positive clones were also submitted for DNA sequencing (ABI 3100, Applied Biosystems, Foster City, CA, USA). The two cloning strategies for the construction of the expression vectors are shown in figure 1. All clonings were performed using standardized methods [22].
Protein expression and purification rPCN exon4 . E. coli BL21 (DE3) colonies transformed with pGEX-padg3347/exon4, were cultivated in 5.0 mL of LB medium with ampicillin (100 mg/mL). The culture was incubated overnight at 37uC with shaking (180 rpm). This pre-inoculum was diluted 1:100 in 500 mL of the same medium, and incubated at 37uC with shaking (180 rpm) until the OD 600 = 0.6. Then, IPTG was added to a final concentration of 0.1 mM. The culture was induced overnight at 21uC with shaking (80 rpm). Subsequently, the culture was centrifuged at 3,0006 g for 20 min at 4uC, and then the bacterial pellet was resuspended in cold buffer A (Tris-HCl pH 7.5, 100 mM, 150 mM NaCl, 2 mM DTT, 10 mM EDTA, 0.2 mM sodium azide, and protease inhibitor), homogenized in Ovni Mixer 2000 at maximum speed for 4 pulses of 15 s at 60-second intervals on ice. The bacterial suspension was sonicated (5 pulses for 10 s each at 60-second intervals). Triton X-100 was added to a final concentration of 1%, and the lysed cells were centrifuged at 3,0006 g for 30 min. The supernatant was collected and subjected to affinity chromatography on a glutathione-Sepharose 4B column equilibrated with buffer A and

Author Summary
Paracoccin is a constituent of Paracoccidioides brasiliensis, a human pathogen that causes paracoccidioidomycosis, the most prevalent systemic mycosis in Latin America. Paracoccin is a dual function protein with domains for lectin and N-acetylglucosaminidase activities. Proteomic analysis of paracoccin preparation revealed its correspondence to a hypothetical protein from the P. brasiliensis isolate Pb-18, annotated as PADG-3347, which has a polypeptide sequence similar to family 18 endochitinases. These endochitinases are multi-functional proteins with distinct lectin and enzymatic activity domains. The multiexon assembly of the correspondent gene (PADG-3347) was cloned and expressed in E. coli, and the physical and biological features of the recombinant protein was compared to those of the native paracoccin. Moreover, recombinant PADG-3347 was evaluated for its immunomodulatory properties and its ability to confer protection against murine P. brasiliensis infection. The results presented herein showed that mice treated with recombinant paracoccin displayed lower pulmonary fungal burdens and reduced pulmonary granulomas. These protective effects were associated with augmented pulmonary levels of IL-12 and IFN-c.
incubated with slow rotation for 1 h. Washings were performed with buffer B (buffer A plus 250 mM NaCl and 0.25% Tween-20), until the flow through had an OD 280 of less than 0.002. The bound protein was eluted with buffer C (buffer A containing 10 mM reduced glutathione) and dialyzed against distilled water in an Amicon system (Amicon Division, W. R. Grace & Co., Beverly, MA, USA), using dialysis disks with nominal molecular weight limit (NMWL) of 10 kDa (Millipore Co., Billerica, MA, USA). The purified protein was quantified by using the method of Bradford [23]. rPCN full . E. coli BL21 (DE3) cells were transformed with pUC57-padg3347/fullORF, and plated on selective medium containing ampicillin. Recombinant strains were grown at 37uC with shaking until the OD 600 reached 0.6. Recombinant protein expression was induced by the addition of IPTG at a final concentration of 0.5 mM. The cells were cultivated for an additional 5 h, and then harvested by centrifugation. The pellet was resuspended in lysis buffer (5 mM EDTA, 100 mM NaCl, 1 mM PMSF, and 50 mM Tris-HCl, pH 7.4) and the cells were disrupted by sonication. Triton X-100 was added to a final concentration of 1%, and the suspension was centrifuged at 3,0006 g for 30 min. The homogenate, containing soluble and insoluble proteins, was diluted in 20 mL of urea buffer (8 M urea, 20 mM sodium phosphate, and 200 mM sodium chloride) and incubated overnight with agitation at room temperature. The homogenate was dialyzed against refolding buffer that contained 20 mM Tris, 100 mM NaCl, 0.2 mM PMSF, and various concentrations of urea to refold the protein. For refolding, the dialysis buffer was removed and replaced every 24 h, and with each replacement, the concentration of urea was reduced until the   final dialysis contained no urea. After refolding, the protein was purified by affinity chromatography on an N-acetylglucosamine column, and the LPS was removed using an immobilized polymyxin B agarose column (BioRad, Hercules, CA, USA) according to the manufacturer's instructions.

N-acetylglucosaminidase activity assay
The N-acetylglucosaminidase (NAGase) activity of the recombinant paracoccin was evaluated using 100 mL of p-nitro-phenol-N-acetyl-glucosaminidase (pNP-GlcNAc, 5 mM; Sigma), 350 mL of sodium acetate buffer (100 mM, pH 5.0), and 50 mL of each sample. To stop the reaction, 1 mL of 0.5 M sodium carbonate was added to the reaction tube, and then the absorbance was measured at 405 nm. The reaction conditions were the same as those used by Yabuki [24] and adapted by Ulhoa and Peberdy [25]. A unit of NAGase activity was defined as the amount of enzyme required to convert 1.0 mmol of r-nitrophenol per minute at 37uC. The specific activity was determined as the number of enzyme units per mg of total protein.

Biotinylation of recombinant paracoccin
For biotinylation, 2.0 mg of recombinant paracoccin was incubated with 20 mL of a 1% aqueous solution of Ez-link Sulfo-NHS-LC-Biotin (Pierce Chemical Co., Rockford, IL, USA) for 2 h at 4uC in the dark. The biotinylated proteins were dialyzed against PBS, using a centrifugal filter device NMWL of 30 kDa (Millipore Co.), to completely remove the remaining biotin.

Binding to laminin
Each well of a 96-well microplate (Maxi Sorp Fluoro Nunc, Roskilde, Denmark) was coated with 250 ng of laminin derived from mouse Engelbreth-Holm-Swarm sarcoma (EHS) (Sigma-Aldrich, St. Louis, MO, USA) diluted in carbonate buffer (pH 9.6) and incubated overnight at 4uC. Different concentrations of biotinylated protein were added to each well, and then the plate was incubated at room temperature (RT) for 120 min. The plate was then further incubated with neutravidin-peroxidase (Gibco BRL, Gaithersburg, MD, USA) for 90 min at RT. The binding of the biotinylated protein to laminin was detected by adding Ophenylenediamine dihydrochloride (OPD), and reaction was then stopped by the addition of 2 N sulfuric acid. Readings were made at l = 490 nm (PowerWave X, Bio-Tek Instruments, Inc., Winooski, VT, USA). For the inhibition assays, biotinylated protein was pre-incubated with 100 mM N-acetylglucosamine, Dglucose, or D-galactose for 1 h at RT. Data shown are representative of three separate assays (reported in relative luminescence units, RLU).

Production of TNF-a and NO
C57BL/6 mice were injected with 1.0 mL of sterile 3% sodium thioglycollate (Sigma-Aldrich). After 3 days, the animals were killed, and their cells were recovered by washing the peritoneal cavity with 5.0 mL of sterile PBS. The cells were immediately stored on ice and washed with RPMI 1640 medium (Flow Laboratories, Inc., McLean, VA, USA) containing 5% inactivated fetal calf serum (FCS) (Gibco BRL), and placed in 24-well culture plates (10 6 cells/well). After incubation at 37uC 2 h, non-adherent cells were removed, and adherent cells were incubated in 5% RPMI medium containing rPCN full or rPCN exon4 (250 ng/mL). TNFa and NO levels in the supernatants were measured by capture enzyme-linked immunosorbent assay (ELISA) and by nitrate accumulation, respectively, as previously described [15].

Production of IgY antibodies
The procedure for immunizing chickens was performed according to Akita and Nakai [26], with slight modifications. Briefly, immunizations was performed by injecting (intramuscularly) 80 mg of native (PCN prep ) or recombinant (rPCN full ) paracoccin emulsified in Freund's complete adjuvant (Sigma-Aldrich). In weeks 2, 4, and 6 post-inoculation, booster immunizations were administered with antigens emulsified in Freund's incomplete adjuvant (Sigma-Aldrich). The eggs were collected and kept at 4uC. Eggs were also collected during the pre-immunization period to obtain antibodies used as control IgY (IgY irrelevant). The content of the yolk was diluted in ultra-pure water (1:10), and after 6 h of gentle agitation at 4uC, high density lipids were precipitated by centrifugation (10,0006 g for 25 min at 4uC) to collect the supernatant, which was rich in IgY. The IgY antibodies were purified by precipitation with 40% ammonium sulfate as previously described [27,28]. The soluble fraction was collected and incubated at 4uC under constant agitation during the addition of ammonium sulfate. Then, the solution was centrifuged at 10,0006 g for 10 min, and the first supernatant was discarded. The pellet was resuspended in PBS and 40% ammonium sulfate precipitation was repeated twice as described above. The last pellet was resuspended in PBS-azide and dialyzed against PBS using a membrane with NMWL of 10 kDa (Amicon Division, W.R. Grace Co.) to remove the ammonium sulfate.

Confocal microscopy
The yeast cells were cultivated for 3 days at 37uC with shaking. The cells (10 6 viable cells/mL) were fixed with 3.7% paraformaldehyde for 15 min, centrifuged at 3,0006 g for 15 min, and resuspended again in 3.7% paraformaldehyde. After centrifugation, the cells were resuspended in PBS buffer with 0.05% Tween and 1% BSA (PBS-TB), containing the IgY polyclonal antibodies against anti-rPCN full or anti-PCN prep for 1 h at RT. Cells were then washed three times with PBS, and incubated for 1 h at RT with neutravidin AlexaFluor_488-conjugate (Life Technologies, Carlsbad, CA, USA) in PBS-TB for 1 h at RT in the dark. After 40 min, 10 mg/mL of calcofluor (Sigma-Aldrich) was added for the remaining 20 min of incubation. After washing with PBS, coverslips were placed onto the glass slides with Permount (Fisher Scientific, Pittsburgh, PA, USA) and fluorescence was analyzed using a fluorescence microscope. The slices were analyzed using a laser scanning confocal microscope (LSM-510 NLO; Carl Zeiss, Jena, Germany) using 406 NA 1.3 and 636 NA 1.4 plan apochromatic objectives. For phase-contrast microscopy studies, the samples were observed with an immunofluorescence microscope (DMI 4000B; Leica, Culver City, CA, USA). Z-series were processed with ImageJ (National Institutes of Health, Bethesda, MD, USA).

Protection assay with recombinant paracoccin
Animals. The efficacy of protection with rPCN full was assessed in four groups of five 6-8-week-old male BALB/c mice. They were bred and maintained under standard animal housing conditions in the animal house at the Ribeirão Preto School of Medicine, São Paulo University, Ribeirão Preto, SP, Brazil. All assays were performed in triplicate.
rPCN full therapy and experimental infection. Each animal was injected subcutaneously (s.c.) with 0.5 mg of rPCN full in 100 mL of sterile PBS. The treatment varied in terms of the number of rPCN full injections (1-3), and the time intervals between injections ( Table 1). The control animals were injected with vehicle only (PBS). The animals were then challenged by intravenous (i.v.) injection with 100 mL of a suspension containing  CFU counting. Mice were killed by cervical dislocation on day 30 post-infection, and fungal burden in the lung was examined as previously described [29]. One lung per animal was aseptically removed, weighed, and the tissue was disrupted in 1.0 mL of sterile PBS using a tissue homogenizer (Ultra-Turrax T25 Basic; IKA Works, Inc., Wilmington, DE, USA). The number of viable cells was determined by plating crude and serially diluted homogenates onto brain heart infusion agar (BHI; HiMedia, Mumbai, Índia) supplemented with 4% (v/v) heat-inactivated fetal calf serum (FCS; Life Technologies, Camarillo, CA, USA) and with 96 mg/mL gentamicin (Gibco BRL). Plates were incubated at 35uC for 7-14 days. The number of P. brasiliensis colonies per gram of organ is reported as the mean and standard deviation (SD) of triplicate samples. Lung homogenate supernatants were also separated from the cell debris by centrifugation at 2,0006 g for 15 min, and stored at 220uC for ELISA-based cytokine and nitric oxide (NO) quantification.
Histopathology. The lungs, obtained on day 30 post infection, were fixed in 10% formaldehyde for 24 h and embedded in paraffin. Tissue sections (5 mm) were stained with hematoxylin and eosin (H&E) and Gomori's methenamine silver stain and examined by light microscopy with an Axiophot photomicroscope (Carl Zeiss GmbH, Jena, Germany), coupled with a JVC TK-1270 camera (Victor Company of Japan Ltd., Tokyo, Japan).
Cytokine ELISAs. To determine the levels of IL-12p40, IL-12p70, TNF-a, IFN-c, IL-4, and IL-10 in the supernatants of the lung homogenates, the levels of these cytokines were measured by capture ELISA with antibody pairs purchased from Pharmingen (San Diego, CA, USA). The cytokines were measured according to the manufacturer's protocol. The concentration of each cytokine was determined with reference to a standard curve generated by serial two-fold dilutions of recombinant murine cytokines.
NO production in the lungs. The concentration of NO 2 was quantified by using the standard Griess reaction [30], by incubating 50 mL of supernatant from the lung homogenates with an equal volume of Griess reagent (1% sulfanilamide, 0.1% naphthylene diamine dihydrochloride, and 2.5% H 3 PO 4 ), at RT for 10 min. NO 2 was quantified by measuring the absorbance at 550 nm (Power Wave-X). Results are expressed as mM NO 2 and the values were compared to a standard curve generated using known concentrations of NaNO 2.

Statistical analysis of results
Statistical analyses of the differences between the means of experimental groups were performed using analysis of variance (ANOVA) followed by Tukey's test. Differences with p values less than 0.05 were considered statistically significant. All experiments were performed three different times, with a total of 5 animals in each group.

Refolding and purification of the recombinant proteins
The predicted mRNA transcript sequence of the PADG_03347 gene was used to design the oligonucleotides for the cloning procedures. Recombinant E. coli strains were evaluated according to the expression of the largest exon of the paracoccin gene (rPCN exon4 ) and according to its predicted multiple exon assembly (rPCN full ). After inducing the cells containing rPCN exon4 ( Fig. 2A), the harvested bacterial cells were prepared and the resulting lysate was subjected to affinity chromatography on a column of glutathione-sepharose 4B. The bound protein was eluted with reduced glutathione, and was analyzed by SDS-PAGE, which showed a single, 48-kDa band. The predicted sizes of rPCN exon4 and the GST fusion tag are 22 kDa and 26 kDa, respectively. The band containing rPCN exon4 was recognized by anti-GST and antinative paracoccin antibodies (Fig. 2B). In contrast, the induction procedure for rPCN full generated highly enriched inclusion bodies, which were solubilized with urea and refolded. The refolded material was applied to a GlcNAc affinity column and, after elution with a specific sugar, the bound protein was analyzed by SDS-PAGE, which showed a single 28-kDa band (Fig. 2C) corresponding to rPCN full . The purified rPCN full was also analyzed by western blotting. The 28-kDa band was recognized by anti-native paracoccin IgY (Fig. 2D). These results indicate that there is immunological identity between the native and recombinant forms of paracoccin, and that the GlcNAc-binding property was preserved after the refolding and purification processes. rPCN exon4 and rPCN full reproduce the biological properties of native paracoccin We previously reported that native paracoccin (PCN prep ) induces the production of TNF-a and nitric oxide by murine macrophages and interacts with laminin in a sugar recognitiondependent manner [15]. PCN prep also has N-acetyl-b-D-glucosaminidase activity, which probably accounts for the function of paracoccin during fungal growth and morphogenesis [18]. Comparison of the predicted polypeptide sequence of PADG_03347 with a database of functional domains using a Hidden Markov Model (HMM), HHPred, indicated that the enzymatic and lectin activities could be contained in its largest exon (exon 4). As shown in Fig. 3, rPCN exon4 , rPCN full , and PCN prep can induce murine peritoneal macrophages to produce TNF-a (Fig. 3, panel A) and nitric oxide (Fig. 3, panel B) at levels close to that of the positive control (LPS+IFN-c). The NAGase activity of PCN prep , rPCN exon4 , and rPCN full was also evaluated (Fig. 3, panel C). The rPCN full preparation showed an enzymatic activity of 0.054 U/mL and a specific activity of 2.65 U/mg. PCN prep showed an 0.041 U/mL enzymatic activity and a specific activity of 2.05 U/mg. In contrast, the rPCN exon4 preparation had only residual NAGase activity, which was four times lower than that of PCN prep (panel C).
We also assessed the ability of both rPCNs (at concentrations of 25-200 ng/well) to bind laminin. Our results show that rPCNs were able to bind laminin in a dose-dependent manner (Fig. 3,  panel D). Moreover, this interaction was selectively inhibited by the presence of N-acetylglucosamine (Fig. 3, panel E), as was previously reported for PCN prep [15].
These results have given us some clues about the structure and function of paracoccin, and showed that the lectin domain is preserved in both recombinant proteins. However, rPCN exon4 did not have enzymatic activity. This impairment of the enzymatic activity of rPCN exon4 led us to further characterize only rPCN full .

Detection of paracoccin in yeast cells using IgY against the recombinant protein
To verify whether IgY anti-rPCN full was able to recognize PCN prep in its natural context, we performed an immunofluorescence assay with non-permeabilized P. brasiliensis yeast cells. As shown in figure 4, fluorescence labeling was mostly detected on the cell surface, particularly on daughter cells. A pseudohyphal form, seen in the preparation, was prominently labeled at its tip. This labeling was verified both in the yeast and hyphal forms of P. brasiliensis (data not shown), and reproduces previous results obtained with murine IgG raised against PCN prep .

Effect of rPCN full administration on pulmonary fungal load and histopathology
We evaluated whether prophylactic administration of rPCN full affected the course of PCM using a murine model. To determine the most effective administration regimen, groups of BALB/c mice were administered a variable number of injections spaced at different time intervals (see Table 1, Materials and Methods), and then challenged i.v. with P. brasiliensis yeast. Thirty days after infection, pulmonary fungal burden lowered by approximately 80% in all rPCN full -treated groups compared to the untreated control group (Fig. 5B). Light microscopy examination of lung sections from all rPCN full -treated mice, regardless of the administration regimen, showed that the organ architecture was mostly preserved, with scant compact and well-organized granulomas and a few silver-stained yeasts (Fig. 5A). In contrast, coalescent granulomas were frequently observed in the lung sections of mice in the control group. Notably, a large amount of yeast were observed both inside and outside the granulomas. Morphometric analysis revealed significantly (p,0.05) lower granuloma density in the lungs of mice from group 3 (0.92 granulomas/mm 2 ) and 4 (0.94 granulomas/mm 2 ), which were injected with rPCN full on day 10 and day 3 before infection, respectively, compared to the other treated groups (Fig. 5C). These results suggest that administration of rPCN full could diminish the pulmonary damage in P. brasiliensis yeast-challenged mice.

Pulmonary levels of cytokines and NO in challenged mice
To determine if the advantageous effect of rPCN full administration was associated with induction of a protective immune response to fungal infection, the levels of cytokines and nitric oxide in lung homogenates were measured on day 30 after the challenge. The levels of IL-12p40 were three times higher in groups 3 and 4 than in the control group (Fig. 6A). Consistently, the levels of IL-12p70 were four and three times higher in groups 3 and 4, respectively (Fig. 6B). The production of TNF-a (Fig. 6C) and IFN-c (Fig. 6D) was five and two times higher in mice of group 3, respectively, and approximately two and three times higher in group 4, respectively, than that of the control group. Therefore, mice that received rPCN full following the protocols for groups 3 and 4, presented significantly higher levels of Th1 cytokines than the control mice. This inflammatory profile was strengthened by the detection of higher levels of pulmonary NO in animals that received rPCN full as prophylactic therapy (Fig. 6G).
Regarding Th2 cytokines, IL-4 pulmonary levels in the treated and control groups of mice were not significantly different (Fig. 6E). In contrast, IL-10 levels were significantly higher in groups 2 and 3 (Fig. 6F). These results indicate that prophylactic administration of rPCN full may confer protection against P. brasiliensis infection through induction of T-helper 1 immunity in the host. rPCN full injection three days before the challenge appears to be the most favorable administration protocol, since the Th1 response might be balanced by high levels of pulmonary IL-10, which may prevent tissue damage caused by inflammation.

Discussion
We demonstrated that the recombinant form of paracoccin (rPCN full ) reproduced the biological activities of the native protein (PCN prep ) and conferred remarkable protection against fungal infection, probably due to skewed Th1 immunomodulation.
Native paracoccin, referred to as PCN prep , was originally reported as a constituent of the GlcNAc-binding preparation (GNBP), an enriched extract of P. brasiliensis yeast, was shown to bind laminin [15], exert NAGase activity [18], and induce macrophages to produce TNF-alpha and nitric oxide [15,18]. These activities were putatively attributed to a major 70-kDa protein in the GNBP fraction. This assumption is now tested frequently by newer proteomic and molecular studies. The lectin containing feature and enzymatic activities of PCN prep could be attributed to the 28-kDa conserved hypothetical protein of P. brasiliensis, a double domain lectin/chitinase, whose gene is annotated in the Broad Institute website as PADG_03347.1.
Two products of the PADG_03347.1 gene, the proteins encoded by the largest exon (rPCN exon4 ) and by the multi-exon assembly (rPCN full ), were both able to bind laminin and induce the production of inflammatory mediators by macrophages. However, the NAGase activity exerted by PCN prep and rPCN full was not reproduced by rPCN exon4 . This enzymatic activity plays important roles in the biology of the fungus [17]. However, it remains to be determined if this enzymatic activity is also relevant to the fungal interaction with host immune cells. For this reason, rPCN full was used in the subsequent in vivo experiments because it more accurately corresponded to the native protein.
Motivated by previous studies showing that paracoccin has interesting effects on immune cells [15], and that lectins are potential immunomodulatory agents [31,32], we investigated the effect of recombinant paracoccin (rPCN full ) administration on the course of murine PCM. The protocols were designed based on prophylactic administration of the plant lectin ArtinM in mice, which succeeded in conferring resistance to murine leishmaniasis [33] and PCM [34]. In our study, rPCN full was administered to groups of mice at various schedules and time intervals. We found that prophylactic administration of rPCN full conferred protection against PCM, and was associated with the development of Th1immunity and a clear reduction of the disease severity. All the rPCN full administration protocols led to significantly lower frequencies of pulmonary granulomas and decreased fungal burdens. The control mice, which received only vehicle, had high fungal loads and extensive granulomatous lesions that disseminated to the liver and spleen. The pattern of disease in mice treated with rPCN full was very similar to that exhibited by mice treated with ArtinM [34] or P10 vaccine [35]. This pattern was also similar to the mild form of human PCM, which is easily cured by anti-fungal therapy [36,37].
Mild forms of human PCM are consistently demonstrated to be associated with Th1-biased immunity, a response that is considered essential for conferring resistance in animal models of fungal diseases [13,[38][39][40]. In a different way, Th2 immunity is associated with severe and disseminated forms of fungal diseases. These features are well established not only for the experimental models of paracoccidioidomycosis [13,41], but also for cryptococcosis [42], and candidiasis [38,43]. By comparing the cytokine concentrations in lung homogenates from the control and rPCN full -treated mice, we could demonstrate that, among the groups that were treated with rPCN full , the mice that received a single injection on day 10 or day 3 before the challenge had the highest levels of Th1 cytokines and higher NO levels. Surprisingly, no exacerbated inflammatory lesions were observed in any of the rPCN full -treated animals, a fact that can be justified by the high levels of IL-10 in their lungs. The cytokine levels allow us to hypothesize that rPCN full interacts with glycosylated receptors on the surface of immune cells and triggers the release of IL-12, as was demonstrated in mice treated with ArtinM, because it is well known that IL-12 stimulates NK and TCD4 cells to produce high levels of IFN-c, which, in turn, induces high levels of NO and other proinflammatory mediators. They improve the fungicidal activity of macrophages [12], which is essential to confer resistance to P. brasiliensis infection. It was previously demonstrated that in vivo depletion of IFN-c leads to exacerbation of lung infection and systemic dissemination of fungal cells [44]. In contrast, TNF-a increases the permeability of the vascular endothelium, macrophage accumulation in the infected tissues, the development of epithelioid granulomas and restriction of fungal spreading [45][46][47]. Granuloma formation as well as the local immune response played by phagocytic cells and sustained granulomatous inflammation depend on TNF-a synthesis and are necessary to restrict fungal growth and dissemination [48,49]. Increased pulmonary levels of IL-10 were detected in animals that received rPCN full on days 10 and 3 before challenge, and in animals that received rPCN full only on day 10 before challenge. Although IL-10 is considered a major mediator that facilitates the infection [50], it is also able to prevent inflammatory injuries due to exacerbated Th1 immunity [51]. This mechanism also seems to be the case for the P. brasiliensis-infected mice, following inoculation with rPCN full , reported herein, or with ArtinM, as reported by Coltri [52].
Finally, our results support the relevance of the fungal protein paracoccin for the fate of experimental PCM. Prophylactic administration of a recombinant form of paracoccin resulted in a drastic reduction of pulmonary lesions and fungal burden compared to control mice. Our findings bring new insights into modulation of the immune response during P. brasiliensis infection with recognizable implications in the clinical investigation of PCM.