The malate synthase of Paracoccidioides brasiliensis is a linked surface protein that behaves as an anchorless adhesin

Background The pathogenic fungus Paracoccidioides brasiliensis is the agent of paracoccidioidomycosis (PCM). This is a pulmonary mycosis acquired by inhalation of fungal airborne propagules that can disseminate to several organs and tissues leading to a severe form of the disease. Adhesion and invasion to host cells are essential steps involved in the internalization and dissemination of pathogens. Inside the host, P. brasiliensis may use the glyoxylate cycle for intracellular survival. Results Here, we provide evidence that the malate synthase of P. brasiliensis (PbMLS) is located on the fungal cell surface, and is secreted. PbMLS was overexpressed in Escherichia coli, and polyclonal antibody was obtained against this protein. By using Confocal Laser Scanning Microscopy, PbMLS was detected in the cytoplasm and in the cell wall of the mother, but mainly of budding cells of the P. brasiliensis yeast phase. PbMLSr and its respective polyclonal antibody produced against this protein inhibited the interaction of P. brasiliensis with in vitro cultured epithelial cells A549. Conclusion These observations indicated that cell wall-associated PbMLS could be mediating the binding of fungal cells to the host, thus contributing to the adhesion of fungus to host tissues and to the dissemination of infection, behaving as an anchorless adhesin.


Background
Paracoccidioidomycosis (PCM), the most important systemic mycosis in Latin America, is a chronic granulomatous disease that affects about 10 million people.
Paracoccidioides brasiliensis, a thermally dimorphic fungus pathogen, is the pulmonary infective agent [1,2]. This initial interaction appears to govern the subsequent mechanisms of innate and acquire immunity, which result in localized infection or overt disease [3].
The mechanisms of adherence and invasion have been studied extensively in pathogenic bacteria [4], and in pathogenic fungi such as Candida albicans [5], Histoplasma capsulatum [6] and Aspergillus fumigatus [7], and P. brasiliensis [8][9][10]. Fungi are non-motile eukaryotes that depend on their adhesive properties for selective interaction with host cells [11]. Adherence molecules are fundamental in pathogen-host interaction; during this event, the fungal cell wall is in continual contact with the host and acts as a sieve and reservoir for molecules such as adhesins [12]. The ability of P. brasiliensis to adhere to and invade nonprofessional phagocytes or epithelial cells has been recognized in previous studies [13][14][15]. Some P. brasiliensis adhesins such as gp43 [10], glyceraldehyde-3-phosphate dehydrogenase (GAPDH) [16], a 30 kDa protein [9], and triosephosphate isomerase (TPI) [17] have been described. Evidence for extracellular localization of some glycolytic enzymes lacking secretion signals at cell-wall anchoring motifs has been reported for some pathogens [18,19]. In addition malate synthase (MLS) is also described as an adhesin on Mycobacterium tuberculosis [20].
The glyoxylate cycle and its key enzymes isocitrate lyase (ICL) and MLS play a crucial role in the pathogenicity and virulence of various fungi such as the human pathogens A. fumigatus [21], Cryptococcus neoformans [22] and C. albicans [23,24], the bacterium M. tuberculosis [25][26][27] as well as the phytopathogenic fungus Magnaporthe grisea [28] and the necrotropic wheat pathogen Stagonospora nodorum [29]. A relevant role for the glyoxylate cycle in the viability and growth of fungi inside macrophages and, consequently, in the development of a disseminated fungal infection has been postulated [21]. ICL and MLS have also been considered a therapeutic target for the development of novel antifungal compounds, since there are no human orthologues. In P. brasiliensis, the enzyme MLS (PbMLS) participates in the glyoxylate pathway, which enables fungus to assimilate two-carbon compounds from the tricarboxylic acid cycle and in the allantoin degradation pathway of the purine metabolism, which allows the fungus to use nitrogen compounds [30].
Here it is demonstrated that PbMLS is the first fungal MLS localized on the cell surface which interferes with the infection process.

Expression, purification and production of polyclonal antibody to PbMLSr
The cDNA encoding PbMLS was subcloned into the expression vector pET-32a to obtain recombinant fusion protein. The protein was not present in crude extracts of non-induced E. coli cells carrying the expression vector (Fig. 1A, lane 1). After induction with IPTG, a 73 kDa recombinant protein was detected in bacterial lysates (Fig.  1A, lane 2). The six-histidine residues fused to the N terminus of the recombinant protein were used to purify the protein from bacterial lysates by nickel-chelate affinity. The recombinant protein was eluted and analyzed by SDS-PAGE (Fig. 1A, lane 3) and His-, Trx-, and S-Tag were removed by cleavage with the enterokinase (Fig. 1A, lane  4). An aliquot of the purified recombinant protein was used to generate rabbit polyclonal anti-PbMLSr antibody. Western blot confirmed the positive reaction of antibody with the fusion protein (Fig. 1B, lane 1) identifying a protein of 73 kDa. The cleaved recombinant protein was detected as a species of 60 kDa (Fig. 1B, lane 2).

Detection of PbMLS on cell wall extracts, culture filtrate, crude extract and peroxisomal fraction
To determine the subcellular distribution of PbMLS, cell wall-enriched, secreted, and peroxisomal fractions purified from P. brasiliensis yeast cells were obtained. Crude extract proteins, SDS-extracted cell wall proteins, and extracted cell wall proteins from yeast cells were subjected to SDS-PAGE analysis, blotted onto nylon membrane and reacted to polyclonal anti-PbMLSr antibody. PbMLS was detected in crude extract (Fig. 1B, lane 3), and in SDSextracted cell wall proteins (Fig. 1B, lane 4), but was not detected in extracted cell-wall proteins (Fig. 1B, lane 5). PbMLS activity was evaluated in SDS-extracted cell wall and in crude extract, showing specific activity of 2131.2 U/ mg and 2051.28 U/mg, respectively. No cross-reactivity to the pre-immune rabbit serum was evidenced with the samples (Fig. 1C).
To determined whether PbMLS was secreted to the medium, proteins were extracted from culture filtrates harvested from P. brasiliensis which had been growing for 24 and 36 h (Fig. 1D, lanes 1 and 2, respectively), 7 days (Fig. 1D, lane 3), and 14 days (Fig. 1D, lane 4). The proteins were subjected to SDS-PAGE analysis, blotted onto nylon membrane and reacted to polyclonal anti-PbMLSr antibody. PbMLS was detected in all these preparations (Fig. 1D, lanes 1 to 4). No signal was detected in medium free of cells (Fig. 1D, lane 5). PbMLS activity was evaluated in culture filtrate showing specific activity of 1305.3 U/ mg. No cross-reactivity to the pre-immune rabbit serum was evidenced with the samples (Fig. 1E). Altogether, these results suggest that PbMLS binds weakly to the cell wall and is actively secreted in P. brasiliensis.
Since PbMLS has the AKL tripeptide, a peroxisomal/glyoxysomal signal which targets PTS1 [31], the presence of the protein was investigated in this cellular compartment. Peroxisomal and mitochondrial fractions purified of P. brasiliensis were obtained. The proteins were subjected to SDS-PAGE analysis, blotted onto nylon membrane and reacted to the polyclonal anti-PbMLSr antibody. PbMLS was detected in the peroxisomal fraction (Fig. 1F, lane 1) confirming the localization of PbMLS in this organelle. Since PbMLS has not been found in mitochondria, the mitochondrial fraction was used as the negative control (Fig. 1F, lane 2).

Cellular localization of PbMLS by confocal microscopy
To observe the cellular location of PbMLS, P. brasiliensis yeast cells were grown in rich medium and visualized by laser confocal microscopy. The expression of PbMLS was highly positive in the budding cells ( Fig. 2 B, C and 2F) but was usually negative (Fig. 2 B and 2C) or weakly positive ( Fig. 2 D) in the mother cells. Although reactivity was evident inside the cytoplasm of budding cells, it was much more intense on the cell surface (Fig. 2 F). The patterns and intensities of the fluorescence spectra of two regions of interest (ROI) are shown in Figure 2 G. The localization of PbMLS was also evaluated on P. brasiliensis yeast cells grown in medium containing acetate or glucose as the sole carbon source. Yeast cells accumulated PbMLS in the presence of acetate (Fig. 3 B) or glucose ( Fig.  3 D), but the quantity of PbMLS was higher when the fungus was cultivated in the presence of acetate. This disparity was exemplified by the fluorescence spectra ( Fig. 3 E), representative of two ROIs indicated by arrows 1 and 2 ( Fig.  3 B and 3D). No cross reaction was observed with the preimmune serum (data not shown).

Binding of PbMLSr to extracellular matrix proteins (ECM) and the reactivity to sera of PCM patients
The ability of the PbMLSr to bind to ECM proteins was evaluated by Far-Western blot assays. PbMLSr binds to fibronectin, type I and IV collagen, but not to laminin as shown in Fig. 4A, lanes 1, 2, 3 and 4, respectively). Nega- Localization of Pb MLS by confocal laser scanning microscopy in P. brasiliensis yeast cells Figure 2 Localization of Pb MLS by confocal laser scanning microscopy in P. brasiliensis yeast cells. Differential accumulation of PbMLS on the surface of budding cells is easily seen in B, C and F. Images A and E represent the differential interference contrast (DIC) of images B and F, respectively. Image C corresponds to a three-dimensional reconstruction of an immunofluorescent tomographic image showing the accumulation of PbMLS only on the budding cells and not in the mother. This is also observed in images B and F. Image G displays the fluorescence pattern and intensity of two regions of interest (ROI) specified by arrows 1 and 2 in image F, indicating that the fluorescence is more intense on the cell surface (2) than in the cytoplasm of budding cells (1). Image D shows a mother cell positive to PbMLS on the cellular surface and the formation, in culture, of budding cells also expressing PbMLS. tive controls were obtained incubating PbMLSr with the secondary antibody in the absence of ECM or PbMLSr with ECM only (Fig. 4A, lanes 5 and 6, respectively). The reaction between PbMLSr and the antibody anti-PbMLSr was used as a positive control (Fig. 4A, lane 7). The binding between PbMLS and ECM compounds was also evaluated by ELISA assay. The results reinforced that PbMLSr binds to fibronectin, type I and IV collagen (Fig. 4B). Negative controls were performed using PbMLSr (Fig. 4B) or ECM only (data not shown). The positive control was performed using anti-PbMLSr, anti-laminin, anti-fibronectin, anti-colagen I or anti-colagen IV antibody (data not shown).

Localization of Pb MLSr
PbMLSr was reacted with three sera of patients with PCM and one serum from a healthy individual in immunoblot assays (Fig. 4C). Strong reactivity was observed with the PCM-patient sera (

Binding of PbMLSr to pneumocytes
The ability of PbMLSr to bind to A549 cells was evaluated. PbMLSr was biotinylated and incubated with A549 cells. After lyses, proteins from A549 cells were electrophoresed by SDS-PAGE and blotted onto a membrane to perform Western blot with anti-PbMLSr antibody. A positive signal was detected from lysed pulmonary A549 cells treated with biotinylated PbMLSr (Fig. 4D, lane 1). The negative control was obtained using supernatant of A549 cells untreated with biotinylated protein (Fig. 4D, lane 2).

The inhibitory effect of PbMLSr and anti-PbMLSr antibody on the interaction of P. brasiliensis cells with pneumocytes
The infection index was determined by interactions between P. brasiliensis yeast cells and A549 pneumocytes, as shown in Figure 5. P. brasiliensis yeast cells were treated Localization of Pb MLS by confocal laser scanning microscopy in P. brasiliensis yeast cells growing in different carbon sources Figure 3 Localization of Pb MLS by confocal laser scanning microscopy in P. brasiliensis yeast cells growing in different carbon sources. The same groups of cells grown in the presence of potassium acetate (images A and B) or glucose (images C and D) as the sole carbon source are shown, side by side, using differential interference contrast microscopy (DIC) and confocal immunofluorescence. In both situations, the accumulation of PbMLS was restricted to the budding cells. The graph in E displays, comparatively, the immunofluorescence patterns and intensities of two regions of interest (ROI 1 and 2), corresponding to arrows 1 and 2. The data indicate that, under the same labeling conditions, the budding cells cultivated on potassium acetate accumulate PbMLS more intensely on the cell surface than those grown on glucose.
with the anti-PbMLSr antibody before interaction with pneumocytes or pneumocytes were treated with PbMLSr before interaction with P. brasiliensis. The controls (nontreated cells) were used to calculate the percentages of total infection. The interaction was analyzed by flow cytometry. Ten thousand events were collected to analysis as monoparametric histograms of log fluorescence and list mode data files. When P. brasiliensis yeast cells treated with anti-PbMLSr antibody were incubated with A549 cells, a decrease in infection was observed after 2 h and 5 h of incubation (Fig. 5A). Similarly, after treatment of A549 cells with PbMLSr, infection was reduced after 2 h and 5 h of incubation when compared to the values for non-treated cells (Fig. 5B). Controls were performed by incubating the pneumocytes with rabbit pre-immune serum or BSA before the addition of A549 cells or yeast cells ( Fig. 5A and 5B, respectively).

Discussion
Our studies showed that PbMLS is a multifunctional protein; besides its enzymatic role as described by Zambuzzi-Carvalho [30], it could participate in the adherence process between the fungus and host cells through its ability to bind fibronectin, type I and type IV collagen. PbMLS was detected in crude extract, cell wall and culture filtrate of P. brasiliensis, which is confirmed by activity assay. Taken together, our results suggest that PbMLS is actively secreted by P. brasiliensis. In the same way, M. tuberculosis MLS has been consistently identified in the culture filtrates of midlog phase M. tuberculosis cultures [32][33][34].
Adherence molecules are important in pathogen-host interactions. They operate as intercellular adhesion molecules (ICAM) or substrate adhesion molecules (SAM), contributing to cell-cell or cell-ECM adherences, respectively, and are usually exposed on the cellular surface. Successful host tissue colonization by fungus is a complex event, generally involving a ligand (adhesin) encoded by the pathogen and a cell or ECM receptor. The pathogen could interact with three types of host component: secreted cell products, host cell surface, or ECM proteins, such as types I and IV collagen, fibronectin, fibrinogen, and laminin [35]. The adhesin potential of PbMLS was demonstrated through Far-Western blot, ELISA and binding assays. These showed that the recombinant protein recognized the ECM proteins, fibronectin and types I and IV collagen, as well as pulmonary epithelial cells. This event indicates that PbMLS can play a role in the interaction of the fungus with host components. Studies have reported the capacity of P. brasiliensis for adhesion and invasion [9,15]. This is the first glyoxylate cycle enzyme identified on the fungal surface and released extracellularly which possesses the ability to bind to ECM proteins. The definition of PbMLSr as a surface-exposed ECM-binding protein, with an unknown mechanism for secretion from the cell or sorting proteins to cellular membrane, suggests that PbMLSr is compatible with anchorless adhesions [36,20]. In these types of adhesions, proteins are reassociated on the cellular surface after being secreted to execute their biological functions [36]. The presence of PbMLS in the culture filtrate harvested after 24 and 36 h, and 7 and 14 days of growth confirmed that it is truly a secreted protein. The presence of PbMLS in SDS-extracted cell-wall protein fraction indicates that PbMLS is associated with the cell surface through weak interactions. Taken together these results provide evidence that PbMLS may be transported out of the cell through the cell wall to be localized on the outer surface of the cell.
Reports have described the presence of some enzymes of the glycolytic pathway on the cell surface in P. brasiliensis as well as in other pathogens [16][17][18][19]37,38]. The presence of these housekeeping enzymes in unusual locations often correlates with their ability to perform alternative functions such as adherence/invasion of the host cells [38,18]. The ability of anti-adhesin antibodies to confer protection by blocking microbial attachment to host cells is being explored as a vaccination strategy in several microbial diseases [39][40][41][42][43]. The identification of the PbMLS as a probable adhesin has several implications. Understanding the consequences of the binding of PbMLS to host cells will lead to improved understanding of the initial events during infection. Further insights into the role of the PbMLS in the host-pathogen interaction could contribute to the design of novel therapeutic strategies for PCM control.
Although PCM infection starts by inhalation of airborne propagules of the mycelia phase, as conidia, which reach the lungs and differentiates into the yeast phase [2], we performed experiments just with yeast cells since this is the phase found inside the host. Is important emphasize that Pbmls transcript is also present in the mycelium phase as described [44,45].
The results of confocal laser scanning microscopy demonstrated differences in the accumulation of PbMLS among P. brasiliensis cells grown in different carbon sources. Accumulation of PbMLS was also higher in P. brasiliensis yeast cells than in the mycelial phase (data not shown). These findings were reinforced by the results of Felipe et al. [44], which suggested that the glyoxylate cycle is up-regulated in yeast cells [46]. Yeast cells grown on potassium acetate accumulated more PbMLS on the cell membrane than yeast cells grown on glucose. These results are in agreement with those obtained by Zambuzzi-Carvalho et al. [30] where the Pbmls transcript level was higher in yeasts cells grown in a two-carbon source than in cells grown on glucose only. The high intensity of ROI found in budding cells, mainly in the cellular membrane, suggests that the PbMLS is metabolically relevant and mainly synthesized by young cells (budding cells). It is unknown whether PbMLS plays any part in the differentiation and/or maturity processes of P. brasiliensis budding cells [45,47]. In fact, the glyoxylate pathway provides metabolic versatility for Candida albicans to utilize alternate substrata for development and differentiation and is involved in the formation of the filamentous state from the single cell state [23]. This process may help Laccaria bicolor grow toward the host with the aggressiveness required for mycorrhiza formation [48].

Conclusion
The results showed the presence of PbMLS in the culture filtrate of yeast cells (parasitic phase), its surface location in P. brasiliensis and its binding to ECM in Far-Western blot and ELISA assays and to A549 cells membranes. The reduction in the adherence of P. brasiliensis to A549 cells by anti-PbMLSr suggests that PbMLS could contribute to active fungal interaction and disease progression in humans through its ability to act as a probable adhesin. In addition, the absence of conventional secretion or cell wall anchoring motifs defines PbMLS as a probable anchorless adhesin that could contribute to virulence by promoting P. brasiliensis infection and dissemination.

Heterologous expression and purification of the PbMLS recombinant (PbMLSr)
The cDNA encoding to PbMLS was obtained by Zambuzzi-Carvalho et al. [30] (GenBank accession number:AAQ75800). EcoRI and XhoI restriction sites were introduced in oligonucleotides to amplify a 1617 bp cDNA fragment of the Pbmls, which encodes a predicted protein of 539 amino acids. The PCR product was subcloned into the EcoRI/XhoI sites of the pET-32a(+) expression vector (Novagen, Inc., Madison, Wis.). The resulting plasmid was transferred to Escherichia coli BL21 C41 (DE3). Bacteria transformed with the pET-32a-MLS were grown in LB medium supplemented with ampicillim (100 μ/mL) at 37°C until reaching the optical density of 0.6 at 600 nm. Synthesis of the recombinant protein was then initiated by adding isopropyl-β-D-thiogalactopyranoside (IPTG) (Sigma-Aldrich, St. Louis, MO) to a final concentration of 1 mM to the growing culture and the bacterial extract was pelleted and resuspended in phosphate buffered saline (1 × PBS). After induction, the cells were incubated for 2 h at 37°C with shaking at 200 rpm. Cells were harvested by centrifugation at 10,000 × g for 5 min at 4°C. The supernatant was discarded and the cells were resuspended in 1 × PBS buffer. E coli cells were incubated for 60 min with lysozyme (100 μg/mL). After addition of 1% v/ v Sarcosyl at 4°C, the cells were lysed by extensive sonication. The sample was centrifuged 8,000 × g for 15 min at 4°C and 2% v/v Triton was added to the supernatant containing the soluble protein fraction. His-tagged PbMLSr was purified using the Ni-NTA Spin Kit (Qiagen Inc., Germantown, MD) and the tags were subsequently removed by the addition of EKMax™ Enterokinase (GIBCO™, Invitrogen, Carlsbad, CA).

Antibody production
The purified PbMLSr was used to produce anti-PbMLSr polyclonal antibodies in New Zealand rabbits. The immunization protocol consisted of an initial injection of 300 μg of purified recombinant protein in complete Freund's adjuvant and two subsequent injections of the same amount of the antigen in incomplete Freund's adjuvant. Each immunization was followed by an interval of 14 days. After the fourth immunization, the serum containing the anti-PbMLSr polyclonal antibody was collected and stored at -20°C.

Western blotting analysis
SDS-PAGE was performed in 12% polyacrylamide gels according to Laemmli [49]. The proteins were electrophoresed and stained with Coomassie brilliant blue or transferred to a nylon membrane and checked with Ponceau S to determine equal loading. PbMLS, as well as PbMLSr, were detected with the polyclonal antibody raised against the recombinant protein (diluted 1: 4000). After reaction with alkaline phosphatase anti-mouse immunoglobulin G (IgG) or alkaline phosphatase antihuman IgG, the reaction was developed with 5-bromo-4chloro-3-indolylphosphate-nitroblue tetrazolium (BCIP-NBT).

Cell wall protein extractions
Yeast cells were frozen in liquid nitrogen and disrupted using a mortar and pestle. The procedure was carried out until complete cell rupture, verified by microscopic analysis, and by the failure of cells to grow on Fava Netto's medium. Ground material was lyophilized and resuspended in 25 μL Tris buffer (50 mM Tris-HCl, pH 7.8) for each milligram of dry weight, as previously described [50]. The supernatant was separated from the cell wall fraction by centrifugation at 10,000 × g for 10 min at 4°C. The crude extract was kept and a new protein extraction was performed with the Tris buffer as described above. The cell wall was extensively washed in solutions with decreasing concentrations of 1 M NaCl to remove any extracellular or cytosolic protein contaminants that could adhere to the walls through electrostatic forces. Isolated cell walls were treated with SDS-extraction buffer (50 mM Tris-HCl, pH 7.8, 2% w/v SDS, 100 mM Na-EDTA, and 40 mM -mercaptoethanol) to extract cell surface-associated proteins, i.e. proteins loosely associated with the cell surface through non-covalent interactions or disulfide bridges (SDS-SW). The proteins from the cell wall and from crude extract were quantified according to Bradford [51].

Preparation of culture filtrate proteins
The culture filtrate were processed as described previously [52], with modifications. Briefly, after 24 and 36 h, and 7 and 14 days of growth at 37°C with gentle agitation, the culture supernatant were removed from the cells by filtration and the culture filtrate was dialyzed and dried by lyophilization. The protein content of the concentrated culture filtrate was quantified according to Bradford [51].

Preparation of Peroxisomal Fraction
The Peroxisome Isolation Kit (Sigma-Aldrich) was used in the preparation of crude peroxisomal fraction from cell cultures P. brasiliensis Pb01 (~2 × 10 8 cells) by differential centrifugation followed by density gradient centrifugation. Briefly, spheroplasts were obtained at 30°C by lysing the cell wall in 400 U of lyticase (Sigma) for 24 h. Sphero-plast membranes were disrupted using a grinder and pestle. After centrifugation for 10 min, the crude peroxisomal fraction was obtained. The organelles were isolated by density gradient centrifugation to separate the enriched peroxisomes fraction from the purified mitochondrial fraction using the Peroxisome Isolation Kit.
The presence of peroxisomes was determined by measuring the activity of the peroxisomal enzyme marker catalase (Catalase Assay Kit) (Sigma-Aldrich). Separation of peroxisomes from mitochondria was determined by measuring the activity of the mitochondrial enzyme marker, cytochrome c oxidase (Cytochrome c Oxidase Assay Kit) (Sigma-Aldrich). In addition, peroxisomal membrane proteins were detected and their degree of enrichment in the purified fraction was determined by immunoblot using anti-PbMLSr.

Affinity ligand assays
Far-Western blot assays were carried out as previously described [53]. PbMLSr underwent SDS-PAGE and was blotted onto nylon membrane. Blotted protein was assayed for laminin, fibronectin, type I and type IV collagen, or for PCM patients' sera as follows. After being blocked for 4 h with 1.5% (w/v) BSA in 10 mM PBS-milk and then washed three times (for 10 min each time) in 10 mM PBS-T, the membranes were incubated with laminin (20 μg/mL), fibronectin (20 μg/mL), or type I and IVcollagen (30 μg/mL), diluted in PBS-T with 2% BSA for 90 min, and then washed three times (for 10 min each time) in PBS-T. The membranes were incubated for 18 h with rabbit anti-laminin, anti-fibronectin, anti-type I collagen or anti-type IV collagen antibodies in PBS-T with 2% BSA (diluted 1:100). The blots were washed with PBS-T and incubated with peroxidase-labeled goat anti-rabbit immunoglobulin (diluted 1:1000). The blots were washed with PBS-T and the reactive signals were developed with hydrogen peroxide and diaminobenzidine (Sigma-Aldrich) as the chromogenic reagent. The positive control was obtained by incubating the PbMLSr with the polyclonal anti-PbMLSr antibody (diluted 1:500), and the reaction was developed as described above.

ELISA analysis
ELISA was carried out as previously described by Mendes-Giannini et al. [8] with modifications. Briefly, Polypropylene 96-well microtiter ELISA plates were sensitized with extracellular matrix (ECM) proteins (10 μg/mL), overnight at 4°C. After blocking with 2% w/v BSA, 10% v/v SFB and 1% w/v milk, the incubation was followed with PbMLSr (5 μg/mL) for 2 h at 37°C in triplicate wells. The reaction was developed using buffer citrate pH 4.9 conjugated with o-phenylenediamine as chromogenic substrate. Negative controls were performed using PbMLSr or ECM only. Positive controls were performed using anti-PbMLSr, anti-laminin, anti-fibronectin, anti-colagen I or anti-colagen IV antibody. The absorbance was measured at 490 nm and the results were analyzed by using Software Microcal ™Origin ™ software version 5.0 Copyright © [54].
Inhibition assay of P. brasiliensis interaction with epithelial cells using PbMLSr and anti-PbMLSr antibody A549 pneumocytes were incubated for 1 h at 37°C with PbMLSr (50 μg/mL), diluted in 10 mM PBS. After this incubation period, the cells were washed three times in PBS and 10 6 yeast forms of P. brasiliensis were added. Incubation was performed for 2 and 5 h at 37°C to allow invasion and internalization, respectively, as described previously [9,15,13]. Four control experiments were performed using A549 cells not preincubated with PbMLSr; P. brasiliensis yeast cells not preincubated with the anti-PbMLSr antibody; pneumocytes preincubated with BSA (25 μg/mL) and P. brasiliensis yeast cells preincubated with rabbit pre-immune serum. The percentage of infected cells was obtained by flow cytometry (BD FACS-Canto) (BD Biosciences, Hialeah, FL). An adhesion index was created by multiplying the mean number of attached yeast cells per pneumocyte by the percentage of infected cells. The infection index (adherence plus internalization) was determined by the number of total fungi interacting with the epithelial cells 5 h after addition of the yeast cells, as previously described [15,13]. The mean and S.D. of at least three independent experiments were determined. Statistical analysis was calculated by using ANOVA (F test followed by Duncan test). P values of 0.05 or less were considered statistically significant.

Biotinylation of protein
PbMLSr was biotinylated with the ECL protein biotinylation kit (GE Healthcare, Amersham Biosciences) as recommended by the manufacturer. Monolayers of A549 cells were incubated with the biotinylated proteins at 37°C overnight and washed with PBS to remove unbound protein. Next, double-distilled water was added and the cells were incubated for 4 h at 25°C to obtain total lysis. The lysates were centrifuged at 1,400 × g for 5 min, and the supernatant underwent electrophoresis by SDS-PAGE. Proteins in the gel were blotted onto a nylon membrane; membrane strips were incubated with blocking buffer for 4 h at 25°C. Incubation for 1 h with streptavidin-HRP followed. A control containing PbMLSr was revealed with the Catalyzed Signal Amplification (CSA) System kit (DAKO). The negative control was developed with the supernatant of A549 cells after lyses (without incubation with the biotinylated protein).

Confocal analysis
The cellular localization of the PbMLS was performed as described by Batista et al. [55] and Lenzi et al. [56] for confocal laser scanning microscopy (CLSM). Briefly, the cells growing in different sources of carbon were fixed in 4% paraformaldehyde for 1 h, washed and centrifuged. After permeabilization with Triton X-100, the cells were washed in PBS and incubated in blocking solution (2.5% BSA, 1% skim milk, 8% fetal calf serum) for 20 min (Fernandes da Silva, 1988). The diluted (1:100) primary antibody anti-PbMLSr was added overnight at 4°C. After washing three times with PBS, the cells were incubated with secondary antibody (Alexa Fluor 488 anti-rabbit Molecular Probes 1:700) for 1 hour. Before mounting in 90% glycerol in PBS, adjusted to pH 8.5, containing antifading agent (pphenylenediamine 1 g/L) (Sigma-Aldrich), the cells were stained with Evans blue (1/10000 in 0.01 M PBS). The specimens were analyzed by laser confocal microscopy (LSM 510-META, Zeiss).

Flow cytometry assay analysis
All flow cytometry analyses were performed on a BD FAC-SCanto (BD Biosciences) using an air-cooled argon-ion laser tuned to 488 nm and 115 mW. The flow rate was kept at approximately 10,000 events (cells), and green fluorescence was amplified logarithmically. Ten thousand events were collected as monoparametric histograms of log fluorescence, as well as list mode data files. The data were analyzed by FACSDiva Software (BD Biosciences) and Origin Software [54].

Enzymatic activity
MLS activity was determined as described by Zambuzzi- Carvalho (2009) [30]. Briefly, the enzymatic assay was carried out at room temperature. 25 mg samples were added to 500 ml assay buffer containing 5 mM acetyl-CoA (20 ml) and water to a volume of 980 ml. The reaction had the optical densities read at 232 nm until stabilization. The enzymatic reaction was started by the addition of 100 mM glyoxylate (20 ml). The method is based on the consumption of acetyl-CoA at 232 nm. The activity was calculated considering that one unit at 232 nm is defined as 222 nmols/mg of acetyl-CoA. The specific activities were given in U/mg protein, with U being equal at nmol/min.

Statistical analysis
Results are expressed as the mean ± SD of the mean of three independent experiments. Statistical analysis was performed using ANOVA (F-test followed by Duncan test). P-values of 0.05 or less were considered statistically significant.