Recombinant paracoccin reproduces the biological properties of the native protein and induces protective Th1 immunity against Paracoccidioides brasiliensis infection

Abstract

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.

Figure 1. Cloning strategies for cloning the paracoccin ORF for expression.

The left panel shows the standard strategy, with PCR amplification of the largest exon, restriction endonuclease digestion, and cloning into the expression vector pGEX-4T-1. The positions of the exons are displayed on the map of the gene. Genomic DNA template was extracted from P. brasiliensis strain Pb18. Agarose gel electrophoresis (mid-left) shows the corresponding band amplified by PCR. The exon 4 amplicon was cloned by BamHI and EcoRI digestion. The right panel shows the strategy for synthesis of the predicted paracoccin sequence fused with the 5′-UTR elements for transcription in the vector pUC57. Green arrow, T7 promoter; black box, lacO (lac operator); and red box, the phage T7 trailer sequence for ribosome binding.

Figure 2. Electrophoretic profile of rPCN_exon4 and rPCN_full.

Panel A: Induction time/response rPCN_exon4 detected in the lysates of E. coli. The time elapsed since IPTG induction is shown in hours. Panel B: The bound material to glutathione-sepharose was separated by 12% SDS-PAGE Coomassie blue staining revealed a single band (lane 1), which was recognized by specific antibodies against GST (lane 2) and PCNprep (lane 3). Panel C: rPCN_full was purified and evaluated for its ability to bind to N-acetylglucosamine. The bound material was separated by 10% SDS-PAGE under reducing conditions, and then stained with Coomassie blue. Lane 1, material before refolding; Lane 2, material after refolding (a single band was detected with an apparent molecular mass of 28 kDa). Molecular markers were a mixture of pre-stained proteins (Fermentas). Panel D: The rPCN_full band was recognized by an anti-paracoccin antibodies (lane 2) and not by antibodies from pre-immune sera.

Figure 3. Biological and enzymatic properties of rPCN_exon4 and rPCN_full.

Panel A: production of TNF-α, Panel B: production of NO by induced murine macrophages following in vitro stimulation with PCNprep and rPCN_exon4 or PCNprep and rPCN_full. Cells were harvested from the peritoneal cavity of C57BL/6 mice and induced with thioglycollate. Adherent cells were incubated for 48 h with different recombinant proteins (0.25 mg/mL), medium (negative control), or LPS+IFNγ (positive control). The standard deviation was calculated based on tests performed in triplicate. The activity of the samples was compared to that of the medium alone. Panel C: The PCNprep, rPCN_exon4 and rPCN_full were assayed for NAGase activity. A colorimetric assay was performed in spectrophotometer set at λ = 405 nm. The standard deviation was calculated by analysis of experiments performed in triplicate. Panel D: Binding of rPCN_full to laminin. Different amounts of biotinylated recombinant protein were incubated with laminin (250 ng) coated in the microplate wells. The binding of the biotinylated protein was detected with a neutravidin-peroxidase conjugate and a chemiluminescent substrate. Luminescence readings are reported as relative luminescence units (RLU). Panel E: Inhibition of rPCN_full binding to laminin by sugars. Different concentrations of GlcNAc, d-glucose, and d-galactose were pre-incubated with the recombinant protein (100 ng), and the mixture was then added to the laminin-coated wells. The margin of error was calculated by analysis of triplicate experiments. Each sample with sugar was compared to a sample without sugar.

Figure 4. Anti-rPCN_full antibody reactivity on the yeast cell surface.

Fluorescence labeling with anti-rPCN_full (Panel A–D) was evenly distributed over the yeast cell surface, with more intense labeling in some budding regions. Panel D is the merge of panels B and C.

Figure 5. Fungal burden, granuloma incidence, and histopathology in infected mice that received prophylactic administration of rPCN_full.

Mice were infected (i.v.) with 10⁶ P. brasiliensis yeast cells and analysis was performed on day 30 after infection. Panel A: Pulmonary CFU recovery. Each group of five mice was either not treated (injected with vehicle [PBS]) or prophylactically treated according to protocol G1, G2, G3, or G4. Panel B: Morphometric analysis of lung sections in terms of granulomas/mm² of tissue, and according area of granuloma. Bars depict the means and SD. *P<0.05 versus the PBS group. Panel C: Lung histopathology of infected mice that received prophylactic administration of rPCN_full. Mice were infected (i.v.) with 10⁶ P. brasiliensis yeast cells and analysis was performed on day 30 after infection. The panels show representative lung sections from mice that were infected and not treated (PBS), or infected and prophylactically treated according to protocol 1 (G1), protocol 2 (G2), protocol 3 (G3), and protocol 4 (G4).

Figure 6. Prophylactic administration of rPCN_full increases proinflammatory cytokines and NO production.

Mice were infected (i.v.) with 10⁶ P. brasiliensis yeast cells and analysis was performed 30 days after infection. IL-12p40 (A), IL-12p70 (B), TNF-α (C), IFN-γ (D), IL-4 (E), IL-10 (F), and NO (G) levels were measured in lung homogenates. Data are reported as the mean and SD of five mice per group and the experiments were performed in triplicate. *P<0.05 versus the PBS group.