Involvement of the capsular GalXM-induced IL-17 cytokine in the control of Cryptococcus neoformans infection

Cryptococcus neoformans is an opportunistic fungus that can cause lethal brain infections in immunosuppressed individuals. Infection usually occurs via the inhalation of a spore or desiccated yeast which can then disseminate from the lung to the brain and other tissues. Dissemination and disease is largely influence by the production of copious amounts of cryptococcal polysaccharides, both which are secreted to the extracellular environment or assembled into a thick capsule surrounding the cell body. There are two important polysaccharides: glucuronoxylomannan (GXM) and galactoxylomannan, also called as glucuronoxylomanogalactan (GXMGal or GalXM). Although GXM is more abundant, GalXM has a more potent modulatory effect. In the present study, we show that GalXM is a potent activator of murine dendritic cells, and when co-cultured with T cells, induces a Th17 cytokine response. We also demonstrated that treating mice with GalXM prior to infection with C. neoformans protects from infection, and this phenomenon is dependent on IL-6 and IL-17. These findings help us understand the immune biology of capsular polysaccharides in fungal pathogenesis.

induced transforming growth factor (TGF)-β in the macrophage cell line RAW 264.7 29 . Mice infected with encapsulated strains were unable to induce T-helper (Th) 1 cytokines such as IL-2 and interferon (IFN)-γ, inducing a significant accumulation of IL-10 that was not observed in the mice infected with an acapsular mutant. These results suggest that yeasts containing GXM on their surface limit the development of a Th1-type protective response in an inhibitory process in which IL-10 plays a critical role 28,30,31 . Our group recently showed that GXM does not induce the release of neutrophil extracellular traps (NETs) by human neutrophils and that in the presence of GXM, stimulated human neutrophils block NET release 32 . In addition to these immunomodulations, GXM can also induce apoptosis in different systems. Monari and colleagues 33 demonstrated that FasL expression in murine macrophages induces apoptosis in activated T cells through processes involving intrinsic and extrinsic pathways 24,33,34 . It has also been shown that GXM can induce apoptosis in macrophages through a mechanism that involves an increase in Fas and FasL 29 .
The majority of studies on the immunomodulatory effects of capsular polysaccharides from C. neoformans have been performed with GXM, but the possible roles of GalXM as an immunomodulatory molecule remain unclear. Reports have increased in recent years suggesting this polysaccharide may also have important immunomodulatory activities. Chaka and colleagues 35 showed that GalXM could induce the production of tumor necrosis factor (TNF)-α in peripheral blood mononuclear cells (PBMCs) 35 . The production of nitric oxide through the expression of inducible nitric oxide synthase and the release of TNF-α induced by GalXM have also been described 29 . Unlike the action of GXM on the production of NETs, Rocha and colleagues 32 have shown that stimulation with GalXM or with acapsular fungus CAP67 (which lacks GXM in the polysaccharide capsule) is sufficient for the induction of NETs by human neutrophils 32 .
These observations suggest GXM and GalXM have different immunomodulatory activities. In addition, GalXM can induce apoptosis in different cells of the immune system. Pericolini and colleagues 36 showed that GalXM induced apoptosis in memory T cells in rheumatoid arthritis patients 36 . Villena and colleagues 29 also demonstrated that GalXM's induction of apoptosis in the RAW cell line was mediated by Fas/FasL interactions, and the effect was ~50 times greater than that observed for GXM 29 . GalXM-mediated cell death might enhance the suppressive effect of GXM and contribute to suppression during cryptococcosis [37][38][39] . Furthermore, Moyrand et al. demonstrated the importance of GalXM in the virulence of C. neoformans by compromising its biosynthesis via UDP-Glc epimerase (uge1Δ) and UDP-Gal transporter (ugt1Δ) 40 .
In this report, we provide the first description of the in vitro modulation of dendritic cell activation and evaluate the participation of purified capsular polysaccharides. Our results showed that GalXM induced the production of IL-6, TGF-β, and IL-17 in co-cultures of stimulated dendritic cells and T lymphocytes, a characteristic profile for a Th17 response. The GalXM also induced a protective effect in vivo where the number of colony-forming units (CFUs) showed reductions in the fungal loads in different organs. Our data also show that GalXM induces a protective response in mice infected with C. neoformans, suggesting an effect mediated by Th17 lymphocytes during the early stages of infection.

Capsular constituents GXM and GalXM induce the maturation of murine dendritic cells and the production of cytokines. Dendritic cells are professional antigen-presenting cells in which T cells can initi-
ate the primary and secondary responses [41][42][43] . Immature dendritic cells reside inside non-lymphoid organs where they actively capture and process antigens. Their maturation can be induced by contact with proinflammatory cytokines and by the recognition of microbial products migrating to the lymphoid organs, where the initiation of the adaptive immune response occurs 44 . To evaluate the involvement of the purified capsular constituents from C. neoformans on dendritic-cell maturation, the cells were incubated for 24 h in the presence or absence of 50-μg/ mL GXM or GalXM. Both capsular constituents induced the increased expression of major histocompatibility complex class II (MHC II) and CD86 (surface structures characteristic of mature dendritic cells), with GalXM eliciting a more intense increase in marker expression than GXM (Fig. 1A,B). To eliminate potential LPS contamination of the polysaccharides, the experiments were performed with polymyxin B (Fig. 1A,B).
We also evaluated the cytokine profile produced by dendritic cells stimulated by the capsular constituents purified from C. neoformans. The results showed that both capsular polysaccharides (GXM and GalXM) induced a significant increase in the production of IL-10 ( Fig. 1C). Furthermore, both capsular polysaccharides induced increased IL-12p40 cytokine production in the dendritic cells. Figure 2 shows the production of IL-12p40 at different concentrations of GXM and GalXM. Although some authors have used higher than 50 μg/mL concentrations 45,46 , our data show that small concentrations of GXM or GalXM (Fig. 2) result in the same level of IL-12p40 production as when larger concentrations (up to 250 μg/mL) are used. These experiments showed that the 50 μg/ mL dosage did not compromise the results obtained and promoted the increase in IL-12p40 production indicative of dendritic cell activation.
Increased lymphoproliferative response induced by GXM or GalXM capsule constituents. Because the capsular constituents were able to induce the maturation of dendritic cells, we evaluated the effect of these GXM-or GalXM-stimulated cells on lymphocyte activation in vitro. The proliferative response of the total mesenteric lymph node (MLN) cells was significantly increased (Fig. 3B). When the MLN cells were exposed to ConA (5 μg/mL), the proliferative response was even higher (Fig. 3B).
As previously demonstrated by Pericolini and colleagues 37 , when interacting directly with human T lymphocytes and the Jurkat lineage, GalXM induces apoptosis via the activation of caspase 8 in the cells. We performed the annexin V/propidium iodide apoptosis assay with flow cytometry to detect phosphatidylserine expression on the cell surface. Neither the GXM-nor GalXM-activated dendritic cells induced cell death by apoptosis in CD4 + T cells when co-incubated for 24 or 48 hours (Fig. 3C). We also evaluated whether capsular GXM or GalXM interacting with MLN cells could induce an increase in the proliferative response in the absence of dendritic cells or induce apoptosis as previously published by Pericolini and colleagues 37 . The capsular GXM did not induce a lymphoproliferative response directly, a result in agreement with Yauch and colleagues 46 . Moreover, GalXM did induce an increase in the lymphoproliferative response at concentrations of 10 or 50 μg/mL ( Supplementary Fig. S1), as previously described by Pericolini and colleagues 37 . These results show that the capsular constituents have different mechanisms of action, but acting together they may lead to a suppression of the immune response 29,37 . Production of IFN-γ by CD4 + T cells in cultures with murine dendritic cells pre-treated with GalXM. As the MLN cells showed an impressive lymphoproliferative response when co-cultured with dendritic cells previously stimulated with both capsular constituents, we decided to evaluate the possible implications for the differentiation of T-cell subpopulations. CD4 + T cells were purified by negative selection to produce 92.14% CD4 + -positive cells (data not shown). Dendritic cells stimulated with GXM co-cultivated with the CD4 + T cells were unable to induce differentiation to the Th1 or Th2 subpopulations, represented respectively by the cytokines IFN-γ and IL-4 ( Fig. 4A,B). However, when dendritic cells stimulated with GalXM were co-cultured with T lymphocytes, a significant increase in IFN-γ production was observed (Fig. 4A). This result is in agreement The control corresponds to dendritic cells without stimulus. The graph represents the mean fluorescence intensity (MFI). The result represents one of three independent experiments. The asterisks indicate ***P < 0.001, **P < 0.01, and *P < 0.05 compared to the control, and the hash symbol (#) indicates P < 0.01 compared to LPS without the addition of polymyxin B. To test cytokine production after stimulation, the culture supernatants were collected, and the production of cytokines IL-12p40 and IL-10 were determined with an ELISA assay. The control corresponds to dendritic cells without stimulus. The results represent one of four independent experiments. The asterisks indicate ***P < 0.001 and **P < 0.01 in relation to the control. Dendritic cells stimulated with the capsular polysaccharide GalXM in co-culture with CD4 + T cells induced differentiation to the Th17 subtype. A significant question evaluated in this work was the differentiation of T cells to the Th17 subpopulation. The importance of Th17 cells has been well characterized in autoimmune-disease models, and they are important in host defense against various infections caused by bacteria and fungi, particularly on mucosal surfaces [52][53][54][55][56][57] . Therefore, we evaluated the role of capsular GXM and GalXM in the induction of Th17 cells. In addition to inducing the production of IFN-γ (Fig. 4A), the GalXM-activated dendritic cells surprisingly induced the production of the TGF-β, IL-6, and IL-17 cytokines in the co-cultured CD4 + T cells (Fig. 5A,C). We also found that in the presence of GalXM, dendritic cells could secrete IL-23 ( Supplementary Fig. S2A), but not IL12p70 ( Supplementary Fig. S2B). These results strongly suggest that dendritic cells activated by the GalXM capsular constituent induce the differentiation of CD4 + T cells into the Th17 subpopulation.

Capsular GalXM induces the removal of fungal cells during C. neoformans infection. The
cytokines IL-17 and IL-23 have been described as necessary factors for the protective immune response in murine pulmonary histoplasmosis 58 , and the absence of cytokines IL-6 and granulocyte colony stimulating factor increase susceptibility to Candida albicans infection in the murine model 59 . These findings suggest that Th17 effector cells could be involved in immunoprotection during fungal infections. Based on this information, we next evaluated the role of the GalXM during C. neoformans infection in vivo. Our results showed that treatment with GalXM 24 h prior infection with C. neoformans significantly decreased the number of CFUs in the brain (Fig. 6A) and spleen ( Fig. 6B) 14 days after infection. In the lung, which is the first site of infection, the number of CFUs was approximately 3-fold lower than that in the control (Fig. 6C). Figure 6D shows that CFUs from the GalXM-treated animals had a similar shape, size, and texture as those from the control. These results demonstrate that capsular GalXM induces the removal of fungal cells in these organs, and the effect remains 30 days after infection ( Supplementary Fig. S3). To evaluate whether this decrease in the number of CFUs was a mechanism conserved in other strains of mice, we performed the same infection assay with Balb/c mice. The treatment with capsular GalXM decreased the CFUs in the lung relative to the control 14 days after infection ( Supplementary  Fig. S4) and in the same way as was seen in the C57BL/6 mice (Fig. 6C). IL-17 cytokine is essential for the control of murine cryptococcosis. To understand the protection mechanisms used by GalXM during the early stages of C. neoformans infection in the murine model, we investigated whether the Th17 phenotype was involved in the protection mediated by this capsular constituent during experimental cryptococcosis. Cells belonging to the Th17 subtype produce IL-17, IL-17F, IL-21, and IL-22. The cytokine TGF-β is required to initiate the differentiation process, and IL-6 has been identified as a critical cofactor for Th17 cell differentiation 52,60 . Based on this information and the previously reported modulatory activities mediated by GalXM, we first evaluated the production of IL-6 in a GalXM-stimulated bronchoalveolar lavage. The results confirmed that GalXM induced the production of high concentrations of IL-6 from the second day through the last day of observation (Fig. 7). We then analyzed the role of IL-6 during C. neoformans infection. Our results suggest IL-6 is an essential cytokine for the protective immune response against murine cryptococcosis. In the absence of IL-6 (IL-6 −/− mice), the number of CFUs increased approximately 50-fold over the control (Fig. 8). These observations were confirmed by an analysis of lung histological sections ( Fig. 9A-F), where a high number of fungal cells were seen in the lung alveoli of the IL-6 −/− mice compared to the control animals, suggesting an exacerbation of the infection in the IL-6 −/− mice. We also observed important morphological changes in the lungs of the IL-6 −/− mice infected with C. neoformans relative to the wild-type controls. To evaluate the involvement of GalXM in the IL-17 cytokine-mediated response, we performed the GalXM treatment 24 h before infection and evaluated the number of CFUs in the lungs of animals 30 days after infection. We found that in the absence of IL-6 or IL-17, the capsular constituent GalXM no longer had a protective effect (Fig. 10), suggesting the immunoprotective mechanism of capsular GalXM during a C. neoformans infection is Th17 dependent.

Discussion
In this work, we evaluated the immunomodulatory role of the capsular polysaccharides GXM and GalXM from C. neoformans in dendritic cells that are, together with alveolar macrophages, important in regulating the innate immune response against lung infection by C. neoformans [61][62][63][64] . Initial results using dendritic cells as a model showed that GXM and GalXM induced the increased expression of MHC II and CD86, which are molecules characteristic of mature dendritic cells. In addition, we observed a positive release of cytokine IL-12p40 after the maturation of dendritic cells. The release of IL-12p40 in the presence of different concentrations of the capsule constituents indicated that small concentrations of the two polysaccharides stimulate the dendritic cells to produce IL-12p40, suggesting they contribute to dendritic cell maturation and may induce efficient T-cell responses. These hypotheses are supported by results obtained by Wozniak and colleagues 65 , who showed that dendritic cells isolated from the lungs of mice infected with encapsulated C. neoformans had an increase in the expression of the CD80, CD86, and MHC II maturation markers. These results agree with the observation that CD4 + T cells in bronchoalveolar infiltrate coincide with the increased expression of two subunits (p40 and p35) of the IL-12 cytokine and with the presence of IFN-γ in the pulmonary homogenate of mice 14 days after infection with C. neoformans 66,67 . It has been shown that acapsular C. neoformans is easily phagocytosed by immature dendritic cells, inducing increases in MHC I and II and an increase in the expression of CD40 and CD83 45 . The addition of monoclonal antibodies against GXM facilitated the maturation of dendritic cells, promoting the ingestion of the fungal cells. This maturation allowed for the efficient activation of T cells, their differentiation, and an increase in the production of IFN-γ 45,68 . These authors also investigated the effect of the polysaccharide capsule on genetic expression in dendritic cells by examining the interaction with an acapsular mutant (CAP59) of C. neoformans. The results showed that a large number of genes involved in dendritic cell maturation were positively regulated after treatment with the acapsular yeast that was not seen with the capsular cells. However, the author observed a completely different pattern of gene expression in the cells treated with an encapsulated strain 69 . Together, these results suggest that GXM interferes with the activation and maturation of dendritic cells, indicating the fungus may impair the capability for an efficient T-cell response. However, it is important to note that the acapsular yeasts of the CAP59 strain produce GalXM 70,71 and that these molecules are involved in the maturation of dendritic cells and T-cell activation. To confirm our hypothesis, we verified the ability of GXM-or GalXM-stimulated dendritic cells to activate T lymphocytes. Our results showed that both the GalXM-and GXM-stimulated dendritic cells were able to induce the proliferation of MLN cells in the presence or absence of ConA. However, unlike the GalXM polysaccharide, GXM was unable to induce the proliferation of MLN cells directly, that is, without the participation of dendritic cells. These data are consistent with the work of Yauch and colleagues 46 , who observed that at high concentrations (300-1000 μg/mL), the GXM capsular component inhibited the proliferation of activated T cells in the presence or absence of dendritic cells. It has also been shown that the inhibition of T cell proliferation was not due to apoptosis or necrosis, but rather due to the inhibition of the IL-2 produced by these cells. This suggests that GXM may have a direct effect on T-cell proliferation in late cryptococcal infections when GXM is present at higher concentrations due to its accumulation in tissues 72 .
The Th1 and Th2 cytokines are involved in infection caused by C. neoformans. Although cytokines associated with the Th1 profile are essential for natural immunity, cytokines associated with the Th2 profile do not confer protection in mice 31,73 . It is well described in the literature that the increased expression of Th1 cytokines is related Figure 6. Treatment with GalXM induces the removal of fungal cells from the organs of C57BL/6 mice infected with C. neoformans. Mice were pretreated with PBS or capsular GalXM (250 μg/mL intratracheally) 24 h prior to an intratracheal injection with 10 6 C. neoformans cells. After 14 days, the mice were euthanized and the brain, spleen, and lungs were recovered. Viable fungal cells were quantified in petri dishes containing agar from the respective homogenized organs: (A) brain, (B) spleen, and (C) lung. (D) Illustrative picture of a lung count. The results are expressed in CFUs (dilutions: 1:2 for brain and spleen; 1:100 for lung). The results represent one of three independent experiments (n = 5 mice/group). The asterisks indicate ***P < 0.001 and **P < 0.01 relative to the control. to protection against cryptococcosis through increased phagocytic capacity and macrophage fungicidal activity, resulting in the control of the infection [74][75][76] . Experiments using IFN-γ −/− mice show an increased fungal load 77 . In addition, an IL-4 −/dependent Th2 response exacerbates pulmonary infection 78,79 , and IL-13 (another Th2 cytokine) contributes to fatal allergic inflammation during murine infection with C. neoformans 80 .
In our studies, GalXM-stimulated dendritic cells induced the production of IFN-γ, but not IL-4, in CD4 + T cells. As already mentioned, GalXM also induces the release of TNF-α in PBMCs, monocytes, and murine macrophages 29,35,81 . These data suggest that GalXM may induce the production of Th1 cytokines in different cell types. However, GXM-stimulated dendritic cells were unable to induce IFN-γ or IL-4 production in CD4 + T cells. The GXM capsular polysaccharide has been described as being associated with several immunoregulatory effects and could be considered an immunosuppressive molecule 29,32,82,83 , as its immunosuppressive activity has been well characterized in the literature in recent years 25,29,32,33,83,84 .
Kleinschek and colleagues 85 showed that the Th17-mediated response is dependent on IL-23 and contributes to protection against infection by C. neoformans. In the murine model for C. neoformans infection, IL-12 p40-and p35-deficient mice were described as developing a Th2 response associated with high levels of susceptibility 86 . These data strongly suggest that IL-12 plays an essential role in immunity against C. neoformans 87 . In addition, Figure 8. GalXM loses the ability to induce the removal of fungal cells in the lung of C. neoformans-infected IL-6 −/− mice. Mice were pretreated with PBS or capsular GalXM (250 μg/mL intratracheally) 24 h prior to an intratracheal injection with 10 6 C. neoformans cells. After 30 days, the mice were euthanized, and the lungs were recovered. Viable fungal cells were quantified in petri dishes containing agar from the homogenized organ. The results are expressed in CFUs (dilutions: 1:10 for wild type; 1:1000 for IL-6 −/− ). The results represent one of four independent experiments (n = 5 mice/group). The asterisk indicates *P < 0.05 relative to the control. The hash symbol (#) indicates P < 0.001 compared to the wild-type mice. IL-17 secretion by splenocytes from C. neoformans-infected mice was observed in resistant (IL-13 −/− ) mice, but not in susceptible mice expressing IL-13 80 .
The proliferation of yeast inside macrophages has been characterized as significantly reduced after treatment with the Th1 cytokines IFN-γ, TNF-α, and IL-17 73 . Surprisingly, our results showed that co-cultures of GalXM-stimulated dendritic cells and CD4 + T lymphocytes efficiently induced the release of IL-17 and IFN-γ, indicating that this capsular constituent may lead to a Th17 response. These data are supported by Annunziato and colleagues 88 , who found Th17 cells producing IFN-γ in the gut of individuals with Crohn's disease, which is characterized by a dysregulated Th1/Th17 immune response. Another important study demonstrated that murine C. neoformans infection results in the induction of the Th1 response with the release of the IL-17 cytokine, classical macrophage activation, and infection resolution 89 . In 2009, Lin and colleagues 90 reported that mice vaccinated with Als3p (Candida albicans adhesin) for 14 days who then underwent the removal of their splenocytes/ lymph nodes and were subsequently re-stimulated with Als3p for 5 days showed an increased production of Th1 (CD4 + IFN-γ + ), Th17 (CD4 + IL-17 + ), and Th1/Th17 (CD4 + IFN-γ + IL-17 + ) cytokines. Moreover, our results showed an increase in the production of cytokines that are essential for the Th17 profile such as the production of TGF-β required for the onset of the phenotype and IL-6, a cytokine characterized as a critical cofactor for the differentiation of Th17 60,91 . We also demonstrated that the production of IL-23 by GalXM-stimulated dendritic cells amplifies and/or stabilizes the Th17 phenotype 52,92 .
To date, we know that the capsular constituent of C. neoformans GalXM can induce the maturation of dendritic cells and that these cells can activate T lymphocytes and drive the response to the Th17 phenotype with IFN-γ production, which could be called the Th1/Th17 profile. Some studies have suggested that the Th17 subpopulation, that could be involved in immunoprotection in the fungal-infection model 93 . Recently, Meya and colleagues 94 have demonstrated that stimulation in patients with ex vivo IFN-γ induced a high production of monocytes producing TNF-α and IL-6. Thus, the presence of IFN-γ would favor a differentiation for a Th17 profile. However, a better approach to IFN-γ production is needed in our study model. In addition to the studies of C. neoformans infection, the importance of IL-17 has also been described in other fungal-infection models 58,[95][96][97] . Based on this, we decided to examine whether the response to the Th17 subpopulation induced by capsular GalXM-stimulated dendritic cells is related to protection in the murine model during experimental cryptococcosis, despite Zelante and colleagues 98 having demonstrated that in fungal infections with C. albicans and Aspergillus fumigatus, an IL-23-and IL-17-mediated response subverted the inflammatory neutrophil program, resulting in severe inflammatory tissue pathology associated with the infections. Surprisingly, our data showed that if treated with GalXM prior to infection, the fungal load was removed, and the effect lasted for the 30-days post infection period that we observed.
The Th1 and Th17 responses resulted in the removal of yeast from the lungs, but they did not prevent the systemic spread of the highly virulent C. neoformans strain H99. This suggests that the Th2 response induced by the C. neoformans H99 strain is a virulence mechanism in this strain and that the use of Th2 profile-deficient mice (IL-13 −/− and IL-4 −/− ) leads to a strong response involving the Th1 and Th17 profiles 76 . However, other mechanisms of virulence could be responsible for the strong central nervous system tropism 76,99 . The mechanisms observed by the authors were not enough to prevent the death of mice that presented with severe pathology in the brain. In our work, we suggested that the capsular polysaccharide GalXM could lead to a protective response mediated by Th1 and Th17 subpopulations, but there are probably other mechanisms involved in preventing the death of mice infected with C. neoformans that deserve to be studied. Our results suggest that the capsular polysaccharide GalXM is involved in the protection of mice infected with C. neoformans. The data showed that the IL-6 cytokine was essential for the protective immune response against murine cryptococcosis, as demonstrated by the large increase in CFUs in IL-6 −/− mice, the high number of fungal cells found within the lung alveoli and the morphological changes in the lung relative to the control. These data suggest that the protective mechanism used by GalXM during C. neoformans infection is IL-6 dependent and may involve the Th17 phenotype. The importance of IL-6 during C. neoformans infection has also been confirmed in previous work. For example, C. neoformans infection was exacerbated in IL-6 −/− and IL-12 −/− mice, confirming the hypothesis that Th17 and Th1 responses are involved in natural resistance to infection and the induction of a protective response 31,100 . Administering IL-6 exogenously 24 and 3 hours before intracerebral infection with viable C. neoformans cells resulted in a significant reduction in the number of CFUs in the brain and blood, and it increased the survival of infected mice 101 . In addition, PBMCs pre-stimulated in vitro for 12 days with heat-killed C. neoformans before the addition of viable encapsulated yeasts produced high concentrations of IL-6 compared to PBMCs that were pre-stimulated with feasible fungi 102 . We also showed that the capsular constituent of GalXM induced an increase in IL-6 production by alveolar macrophages and co-cultures of dendritic cells and CD4 + T cells.
Finally, we observed the deleterious effect of a lack of the cytokine IL-17 103,104 , which, like IL-6, is important for the resolution and control of C. neoformans infection.
It is known that the variations in the extent and position of O-acetylation on the backbone of GXM induces antigenic mutability, which gives rise to the classification of C. neoformans strains into five serotypes (A, B, C, D and AD) 105 . Variations related to the capsular polysaccharide GalXM are not yet well characterized. But, it is possible to occur in the constitutions of capsular polysaccharides, and this phenomenon happens mainly by the variation of the culture conditions used and depending on the culture condition used, is a phenomenon often observed in fungi 106,107 . De Jesus et al. 108 demonstrated that differences in sugar composition, physicochemical properties and serological reactivity may occur among GalXM preparations of C. neoformans grown under varied culture conditions.
An interesting example mentioned above, the O-acetyl groups are substitutes for polyssacharide capsular GXM and play an important role in the modulation of host immune protectors. For example, the higher virulence observed in C. gatti as compared to C. neoformans depends on different types of O-acetylation of GXM capsule 109 .
Given the bulky nature of O-acetyl groups and their positioning on the GalXM side chains, we suspect that the extent of O-acetylation may affect immunoregulatory properties of GalXM.
Taken together, our data demonstrated that the capsular polysaccharide GalXM induced maturation of dendritic cells. The presence of this capsular polysaccharide also increased expression of the CD86 and MHC II molecules on the cell surface, followed by IL-23 cytokine production. In addition, GalXM-treated dendritic cells induced T cell proliferation and production of IL-6, IFN-γ and IL-17, suggesting a bias to Th17 response. The properties of GalXM observed in dendritic cells and T cells allowed evaluating their involvement during cryptococcosis. Mice that received GalXM capsular 24 h prior to C. neoformans infection were protected as demonstrated by reduced lung, brain and splenic fungal load, suggesting significant involvement of the Th17 subtype in the infection. Mice. Six to eight week-old Balb/c, C57BL/6, C57BL/6 IL-6 −/− (IL-6 deficient), and C57BL/6 IL-17 −/− (IL-17 deficient) mice were purchased from the University of São Paulo. The mice were kept in polypropylene boxes in a temperature-controlled environment between 23 °C and 25 °C and had free access to water and standard feed during the experiments.

Isolation and purification of C. neoformans capsular polysaccharides (GalXM and GXM).
Isolation and purification procedures were adapted from protocols mentioned before 19,29 . The GalXM was purified from culture supernatant from C. neoformans CAP67 mutant strain grown as described above. Supernatants were separated from the cells by centrifugation at 1200 × g for 30 min at 4 °C, concentrated by ultrafiltration using a 10 kDa-cutoff spiral cartridge (Amicon/Millipore, USA), and submitted to overnight precipitation at 4 °C after addition of three volumes of ethanol. The precipitate containing GalXM and mannoprotein was dissolved in distilled water, filtered through a 0.45 μm filter and lyophilized. The freeze-dried mixture of GalXM and mannoprotein was dissolved in 10 mL of 5 mM sodium acetate buffer pH 5.2, containing 1 mM calcium chloride, 1 mM magnesium chloride, 1 mM manganese chloride and 0.15 M sodium chloride (Con-A buffer), and applied to a XK-26/40 column packed with 70 mL ConA-Sepharose 4B (GE Healthcare, Sweeden), equilibrated in the same buffer and connected to an HPLC system (AKTApurifier GE Healthcare, Sweeden) at a flow rate of 0.25 mL/min. The column was washed with 5 column volumes (CV) of Con-A buffer to elute the GalXM. The flow through and washes were collected as fractions of 10 mL and assayed with phenol-sulfuric acid reaction 111 . Fractions containing GalXM were pooled, concentrated, dialyzed against distilled water and lyophilized.
The GXM capsular polysaccharide was isolated from C. neoformans B3501 wild type strain and purified by differential precipitation with CTAB 12 . Briefly, capsular polysaccharides isolated from the culture supernatant by precipitation with ethanol were dissolved in 0.2 M NaCl, CTAB (3 mg/mg of polysaccharides) was added slowly. Then, a solution of 0.05% CTAB was added and the GXM was selectively precipitated. The precipitate was collected by centrifugation and washed with 2% acetic acid in ethanol then in 90% ethanol. The precipitate was dissolved in 1 M NaCl and three volumes of ethanol were added to precipitate the GXM. The precipitate was centrifuged, washed with 2% acetic acid in ethanol and then with ethanol, dissolved in water and lyophilized. To eliminate potential LPS contamination, 10 mg of GalXM or GXM preparations were dissolved in LPS-free water and further purified through chromatography on a column of Polymyxin B-Agarose (Sigma) equilibrated with LPS free-water. LPS-free GalXM or GXM were eluted with 12 mL of water, recovered and lyophilized.

Dendritic cells.
Dendritic cells were obtained as described by Lutz and colleagues 112 . The tibia and femur of C57BL/6 mice were dissected and the marrow removed using a syringe containing RPMI medium. The cells were homogenized, counted, and adjusted to 2 × 10 6 cells/plate in RPMI medium supplemented with 2 mM L-glutamine, 5 × 10 −5 M β-mercaptoethanol, 1 mM sodium pyruvate, 1 mM non-essential amino acids, 10% SFB, and 20 ng/mL rmGM-CSF (recombinant granulocyte macrophage mouse growth factor; R&D Systems). A 10-mL suspension was transferred to a petri dish and incubated at 37 °C in a 5% CO 2 atmosphere for 10 days. Fresh rmGM-CSF (20 ng/mL) was added every third day of culture.
Histopathological studies. The mice were euthanized after 30 days of infection, and the lungs, brain, and spleen were removed and fixed in 10% buffered formalin for 24 h. The organs were embedded in paraffin, sliced (5-6 μm thick) with a Leica microtome, and stained with the mucicarmine technique. The stained sections were analyzed under a Nikon Eclipse E400 microscope.
Colony forming units (CFUs). The lungs, brains, and spleens were macerated in 1-mL sterile PBS and 10-fold serial dilutions were plated in duplicate on Sabouraud dextrose-agar plates. Neat and 1:10 dilutions of freshly extracted blood were similarly plated. The plates were allowed to grow for 72 h at 30 °C before counting the visible colonies to determine the total CFUs per organ. Statistical analysis. All calculations were performed using GraphPad Prism v5.0. Results in the figures are expressed as the mean and standard deviation (SD). The number (N) of animals per group is indicated in the figure legends. Data were compared by analyses of variance (ANOVA) followed by a Dunnett or Tukey post-test. Significant differences are indicated for *P < 0.05 **P < 0.01, and ***P < 0.001.