Male 6- to 7-week-old C57BL/6 mice, and genetically deficient (^−/−) for IL-6, IL-23p19, and IL-17RA, were obtained from our Isogenic Breeding Unit and maintained under specific pathogen-free conditions in microisolator cages in the animal housing facility of the Department of Biochemistry and Immunology, School of Medicine of Ribeirão Preto, São Paulo University, Ribeirão Preto, Brazil. Mice were supplied with sterilized food and water ad libitum. Experiments were conducted according to the ethical principles of animal research adopted by the Brazilian College of Animal Experimentation and approved by the Ethical Commission in Animal Research (protocol 095/2010).

The virulent yeast strain 18 of P. brasiliensis (Pb18) was cultured (7 days at ~36°C) in Brain Heart Infusion (BHI) agar medium (Oxoid Basingstoke, Hampshire, England) supplemented with gentamicin (100 µg/mL) and 5% fetal bovine serum (FBS; Gibco BRL, Life Technologies, Inc.). After, Pb18 colonies were grown for 24 h (37°C, 150 rpm) in F12 Coon’s modified medium (Sigma-Aldrich, St. Louis, MO, USA) containing gentamicin (100 µg/mL). The Pb18 yeast cells were then harvested and washed in sterile phosphate-buffered saline (PBS; pH 7.2), and the fungal viability was determined by fluorescein diacetate-ethidium bromide staining (39). Only high viability suspensions (≥90%) were used in this study. For in vivo infection, the concentration of fungal cells was adjusted to 1 × 10⁷ yeasts/mL in PBS and 100 µL of this solution (1 × 10⁶ cells) was intravenously (i.v.) inoculated in each mice. In vitro fungal stimulation was performed with a multiplicity of infection (MOI) of 1.

The spleen from uninfected male C57BL/6 mice were removed, and single-cell suspensions were obtained after red blood cells lysis (1 min) with buffer containing NH₄Cl (0.16 M, 9 parts) and Tris-HCl (0.17 M, 1 part). Splenocytes (1 × 10⁶ cells/well) were cultured for 5 days with 1 × 10⁶ Pb18 yeasts cells in RPMI 1640 supplemented with 5% FBS (Gibco), 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, l-glutamine (2 mM), 100 U of penicillin/mL, and 100 µg of streptomycin/mL (all from Sigma-Aldrich). Lung lobules were excised from uninfected and Pb18-infected mice, washed in PBS, minced with scissors, and enzyme digested at 37°C for 30–35 min in 1 mL of digestion buffer [RPMI 1640, 2 mg/mL collagenase IV (Sigma) and 1 mg/mL DNase (Sigma)]. Tissue fragments were further homogenized using a 1-mL pipette, crushed through a 50-µm pore size nylon filter (BD Biosciences, San Jose, CA, USA) and then centrifuged (1,300 rpm, 10 min, 4°C). Next, red blood cells were eliminated using lysis buffer, and remaining cells were washed in PBS, centrifuged, and resuspended in RPMI 1640 containing 5% FBS. For splenocytes and pulmonary cells activation, 1 × 10⁶ cells/well were cultured for 4 h with PMA (50 ng/mL), ionomycin (500 ng/mL), and brefeldin A (5 mg/mL), followed by staining for extracellular and intracellular markers and analysis by flow cytometry.

The expression of CD4, CD8, CD19, CD49b, γδ, F4/80, IL-6, and IL-17A was assessed in splenocytes or pulmonary cells by flow cytometry as previously reported (40, 41). The antibodies were conjugated to different fluorochromes (BD Biosciences, eBiosciences, San Diego, CA, USA, and Santa Cruz Biotechnologies, Santa Cruz, CA, USA). For intracellular cytokine staining, cells were previously permeabilized using PBS containing 1% FBS, 0.1% sodium azide, and 0.2% saponin. Data acquisition was performed using a FACSCanto II flow cytometer and FACSDiva software (BD Biosciences). Data were plotted and analyzed using FlowJo software (Tree Star, Ashland, OR, USA).

The left lung (42) from uninfected and Pb18-infected mice were removed, weighed, homogenized in sterile PBS-containing protease inhibitor (Complete, Roche), and centrifuged (1,300 rpm, 10 min, 4°C). Supernatants were collected and stored at −20°C. The levels of IL-17A, IL-6, IL-23, TNF-α, and IFN-γ were measured by enzyme-linked immunosorbent assay (ELISA) according to manufacturers’ recommendations (BD Pharmingen, San Jose, CA, USA). The reading was held in eMax ELISA reader (Molecular Devices, Sunnyvale, CA, USA) at 450 nm.

The amount of viable yeast cells in the lungs from Pb18-infected mice (3–7 animals per group) was determined after 15 and 30 days postinfection (dpi). Briefly, lung lobules were aseptically collected, weighed, and homogenized using a sterile tissue grinder (IKA^®-Werke, Deutschland, Germany). Each homogenate was diluted 1:10 in sterile PBS and plated (100 µL) on BHI agar supplemented with 5% FBS and gentamicin (100 µg/mL). Plates were incubated at 35–37°C for 7 days, and the amount of CFU per gram of tissue was calculated.

Comparative histopathology was conducted with excised lung lobules from C57BL/6, IL-6^−/−, IL-23^−/−, and IL-17RA^−/− mice at 30 dpi. Tissue was fixed with 10% formalin for 24 h and embedded in paraffin. Next, lung sections (5 µm) were stained by standard procedures with hematoxylin and eosin for lesion analysis, Grocott methenamine silver stain for fungal cell labeling, or silver impregnation (Gomori method) to demonstrate reticulum fibers. Sections were examined with HLIMAGE++ 97 Application Western Vision Software using an optical microscope coupled to a digital camera. Granulomas were visually scored in mature or immature structures according to cellular and reticulin organization and distribution, as described before (43, 44). The percentage of mature granulomas and the granulomatous diameter were calculated. Reticulin measurement was performed in the halo surrounding the granulomas as previously described (45).

Lungs from C57BL/6, IL-6^−/−, IL-23^−/−, and IL-17RA^−/− mice were immersed in OCT medium (Sakura Finetek), snap-frozen in liquid nitrogen, and stored at −80°C until analysis. Tissue sections (5 µm) were submitted to an immunohistochemical reaction. Briefly, slides were incubated with anti-mouse IgG as control or rabbit IgG anti-mouse inducible nitric oxide synthase (iNOS) enzyme (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted 100 times in PBS 0.01% saponin. At the end, the reaction was followed by Mayer’s hematoxylin counterstaining. We measured immunostained areas using Image J software as previously described (41). Briefly, the range of positivity was defined using the IHC Tool box. Next, the images were converted to 8-bit, and the grayscale was converted to binary (black and white). The threshold was adjusted, and the labeled areas became the black portions. Finally, the percentage of stained area was analyzed.

Bone marrow cells isolated from femurs and tibias from 7-week-old C57BL/6, IL-6^−/−, IL-23^−/−, and IL-17RA^−/− naive mice were cultured (7 days, 37°C, 5% CO₂) in RPMI 1640 medium supplemented with 20% FBS and 30% L-929 cell conditioned media, as previously described (46). After differentiation, cells were harvested and infected with P. brasiliensis (MOI of 0.04).

Naive T CD4⁺ cells were purified from C57BL/6 lymph nodes by using the CD4⁺ MicroBeads Isolation Kit (Miltenyi Biotec) in conjunction with an AutoMacs separator. Purified naive T cells were cultured for 5 days with P. brasiliensis-infected BMDMs from C57BL/6 or IL-6^−/− mice (MOI of 0.04). Next, the amount of T CD4⁺IL-17A⁺ was analyzed by flow cytometry.

After adoptive transference of C57BL/6 BMDMs (1 × 10⁶, i.v.) into IL-6^−/−, IL-23^−/−, and IL-17RA^−/− animals, mice were infected with 1 × 10⁶ P. brasiliensis yeast cells, and the CFU was determined at day 30 as described above.

The differences observed between uninfected or Pb18-infected wild-type (WT) group (C57BL/6) and IL-6-, IL-17RA-, or IL-23-deficient mice were analyzed by applying the ANOVA test followed by multiple comparison using the Bonferroni method (GraphPad Prism, GraphPad Software version 5.0, San Diego, CA, USA). Values are expressed as mean ± SEM. All values were considered significant when p < 0.05.

To evaluate whether fungal cells modulate the IL-17A production, splenocytes from C57BL/6 mice were isolated, cultivated in the presence of or not of P. brasiliensis, and the frequency of IL-17A-producing cells was determined 5 days later. Our results showed that the stimulation of spleen cells with viable Pb18 increased the IL-17A expression (Figure 1A), and the frequency of IL-17A-producing CD4 T cells (5.5-fold, p < 0.05; Figure 1B) compared to medium alone. Next, to evaluate whether P. brasiliensis modulates the Th17 profile in vivo, C57BL/6 mice were infected with Pb18 yeast cells, and the IL-17A production was determined in the lungs at day 30 postinfection. As expected, we observed that Pb18-infected mice exhibited a significant increase (~9.5-fold) in IL-17A production compared to uninfected mice (Figure 1C). In addition, at 30 dpi, there was an increase in the number of IL-17A-producing cells (~4.5-fold; Figure 1D).

Despite the fact that most of the evidence describes IL-17A as a cytokine secreted by T cells, part of the IL-17A is produced by different cell populations, including innate immune cells (47). To determine the major source of IL-17A in the lungs during the experimental PCM, we evaluated the frequency and absolute number of Th17, Tc17, γδ⁺IL-17A⁺, CD49b⁺IL-17A⁺ [natural killer (NK)], and CD19⁺IL-17A⁺ (B cells). Initially, we excluded the doublets; delineated a gate comprising lymphocytes, monocytes, and neutrophils; and then a subgate in IL-17-positive cells, which allowed us to evaluate the percentage of each cell subtype (Figure 1E). The Th17 population was significantly increased in comparison with other IL-17-producing cells (p < 0.05; Figure 1E). For better and clear results, we next used a second gate strategy, which consisted of singlets selection in forward scatter height (FSC-H) against forward scatter area (FSC-A) (Figure 1F, upper left panel). Next, lymphocytes were gated inside the previous population in side scatter area (SSC-A) against FSC-A (Figure 1F, upper right panel), and all analyzed cell populations were gated inside the lymphocyte population. Our data clearly showed the predominance of IL-17A-producing CD4⁺ cells, both in percentage (~5%, Figures 1F,G) or absolute number (~9.2 × 10⁴ cells; Figure 1G), compared with other IL-17A⁺ cells subtypes: CD8 (~1.2%, ~2.5 × 10⁴ cells), γδ (~0.7%, ~1.7 × 10⁴ cells), NK (~0.3%, ~0.7 × 10⁴ cells), and B lymphocytes (~0.1%, ~0.3 × 10⁴ cells) (Figures 1F,G). After evaluating the number and frequency of IL-17-producing neutrophils, we could not see an increase of this cell subtype in the lungs from Pb18-infected mice (data not shown). These findings suggest that Th17 cells are the main source of IL-17A during P. brasiliensis infection.

The Th17 subset differentiation requires the presence of IL-6, and its maintenance is sustained by IL-23 (13, 14, 16). Thus, we evaluated the production of these cytokines in the lungs from naive and Pb18-infected mice at 15 and 30 dpi. Our data showed that at both time points, the IL-6 and IL-23 levels were significantly increased when compared to uninfected mice (Figure 2A). To confirm these findings, IL-6^−/− and IL-23^−/− mice were infected with P. brasiliensis yeast cells, and the IL-17A levels were evaluated in the pulmonary tissue after 15 and 30 days. We found that the deficiency of IL-6 or IL-23 impaired the IL-17A-production, which was ~2.5 times smaller than C57BL/6 mice at the same period of infection (p < 0.05; Figure 2B). Moreover, when compared to WT mice, we found that the frequency and absolute number of IL-17A-producing CD4⁺ T cells were decreased (p < 0.05) in the lung tissue from IL-6^−/− and IL-23^−/− mice at 30 dpi (Figure 2C). These data establish the importance of IL-6 and IL-23 in promoting the late Th17-immune response during the P. brasiliensis experimental infection.

To verify whether P. brasiliensis-induced Th17-associated cytokines are important to control the fungal growth and pulmonary infection, we analyzed the amount of CFU and the histopathology of lung tissue from Pb18-infected IL-6-, IL-23-, and IL-17RA-deficient mice. Our data revealed that all deficient mice showed impaired fungal control at both 15 and 30 days after P. brasiliensis inoculation when compared to C57BL/6 mice (p < 0.05; Figure 3A). Likewise, the Grocott stain at 30 dpi corroborated a higher fungal load in the lungs of all deficient mice when compared to infected C57BL/6 mice (Figure 3B), suggesting that Th17-associated cytokines, as IL-6 and IL-23, are important to control the fungal growth during the experimental PCM in mice. To better assess the inflammatory process developed during P. brasiliensis infection, we analyzed the histology of pulmonary tissue at 30 dpi. The C57BL/6 mice showed bigger and mature granulomas characterized by epithelioid and giant cells surrounded by a peripheral mononuclear layer (Figures 3C,E,F). In contrast, most of the granulomas from IL-6^−/− and IL-17RA^−/− mice exhibited an intermediate size and immature profile with few lymphocytes forming the peripheral ring, associated with a diffuse inflammatory process (Figures 3C,E,F). Specifically, the IL-23 deficiency induced the formation of numerous well-formed granulomas, but with a smaller diameter compared to all Pb18-infected groups (Figures 3C,E,F).

Reticulin fibers are composed of type III collagen; besides very delicate and fine, they form a firm layer linking the connective to the surrounding tissue, being considered a main element of alveolar organization (48). We observed a well-organized and dense reticulin ring contouring the granulomas in the lungs from Pb18-infected C57BL/6 mice (Figures 3D,G). In addition, the diminished number of compact granulomas in IL-6^−/− and IL-17RA^−/− pulmonary tissue was associated with a defective reticulin halo accumulation and formation (Figures 3D,G). Moreover, a tiny reticulin layer surrounded the small mature and immature granulomas in the lungs from IL-23-deficient mice at 30 days post-Pb18 infection (Figures 3D,G).

Since TNF-α is important to the granuloma formation (36, 37), we also evaluated the expression of this cytokine in the lung tissue from Pb18-infected C57BL/6, IL-6^−/−, IL-23^−/−, and IL-17RA^−/− mice. Our data showed that P. brasiliensis infection increased the TNF-α production at both evaluated time points (Figure 3H). There were no differences in TNF-α levels in the various noninfected knock-out mice (data not shown). At 15 dpi, all groups showed similar TNF-α production. However, when compared to C57BL/6 mice, the levels of TNF-α were reduced in all deficient mice at 30 dpi (p < 0.05; Figure 3H). These results suggest that IL-6, IL-17A, and in less extent, IL-23 modulate the formation of organized and compact granulomas during the experimental P. brasiliensis infection.

Several studies have demonstrated that the Th1-type cytokine IFN-γ is implicated in resistance during P. brasiliensis infection, inducing cell migration and activation (37, 49). Then, we next evaluated the importance of IL-6, IL-17A, and IL-23 in modulating the IFN-γ-production during the experimental PCM. All naive knock-out mice showed similar IFN- γ levels (data not shown), and these levels were increased after the P. brasiliensis infection (Figure 4A). At both evaluated time points, 15 and 30 dpi, all deficient mice showed a reduction of IFN-γ levels in the pulmonary tissue in comparison to C57BL/6 group (Figure 4A). These findings suggest that Th17-associated cytokines are involved in the IFN-γ production during the experimental PCM in mice.

It was previously demonstrated that IFN-γ is able to induce fungicidal activity in macrophages through nitric oxide (NO) production (50, 51), which is indirectly measured by iNOS expression. At 15 dpi, there was increased iNOS expression inside the granulomatous structure from C57BL/6 mice, in comparison with all analyzed deficient mice at the same time point (Figures 4B,C).

Together, these results suggest that the deficiency of Th17-associated cytokines (IL-6 and IL-23) modulates the IFN-γ production and iNOS expression and dampens the fungal death during an experimental PCM model.

We first showed that ~38% of F4/80⁺ cells (macrophages) from splenocytes cultured with Pb18 yeast cells are able to produce IL-6 (Figure 5A). Gate strategy consisted of doublets exclusion in FSC-H against FSC-A (data not shown). Next, monocytes were gated in SSC-A against FSC-A (Figure 5A, left panel), and then the IL-6-positive cells were selected (Figure 5A, middle panel). Finally, the IL-6⁺F4/80⁺ population was delimited (Figure 5A, right panel). We next evaluated the role of IL-6 in Th17 differentiation during Pb18 infection. The BMDMs from C57BL/6 and IL-6^−/− mice were co-cultured with naive CD4⁺ lymphocytes isolated from C57BL/6 WT mice, in the presence of or not of Pb18 yeast cells. We observed that the frequency and absolute number of Th17 cells induced by WT BMDMs were significantly higher than that of IL-6^−/− BMDMs (p < 0.05; Figures 5B,C, respectively). Overall, these data suggest that the IL-6 produced by F4/80⁺ macrophages induces Th17 cells during the P. brasiliensis infection.

To verify the relevance of Th17-associated cytokines over the macrophage fungicidal activity during P. brasiliensis infection, BMDMs from C57BL/6 mice were transferred to IL-6-, IL-23-, and IL-17RA-deficient mice, and the fungal burden was evaluated after 30 dpi. To confirm that transferred cells arrive in the lungs, we labeled BMDMs from GFP⁺ mice with F4/80, and at day 7 after transference, we tracked these cells. We verified a frequency of ~7.3% of F4/80⁺GFP⁺ cells in the pulmonary tissue (data not shown). The IL-6^−/− and IL-17RA^−/− transferred mice showed diminished number of Pb18 yeast cells in comparison with same genotypes without transfer (p < 0.05; Figure 6A). However, the adoptive transfer of BMDMs did not change the fungal burden in IL-23^−/− mice (Figure 6A).

Since resistance to P. brasiliensis infection is characterized by the presence of mature granulomas, we evaluated the influence of macrophages in the granuloma formation. Corroborating our previous findings (Figure 3C), the Pb18-infected IL-6- and IL-17RA-deficient mice showed diminished capacity to mount organized granulomas at 30 dpi when compared to C57BL/6 group (Figures 6B,C). Surprisingly, the adoptive transfer of WT BMDMs restored the ability of IL-6^−/− and IL-17RA^−/− mice to form mature granulomas in the lung tissue (Figures 6B,C). We could not observe histological differences in Pb18-infected IL-23-deficient mice transferred or not with C57BL/6 BMDMs, which continued showing small well-formed granulomas in the lungs after transference (Figures 6B,D).

Taken together, our data suggest that Th17-associated cytokines are involved in the modulation of immune response and granuloma formation during experimental PCM.