Abstract
During adaptive immunity to pathogens, dendritic cells (DCs) capture, kill, process, and present microbial Ags to T cells. Ag presentation is accompanied by DC maturation driven by appropriate costimulatory signals. However, current understanding of the intricate regulation of these processes remains limited. Cryptococcus gattii, an emerging fungal pathogen in the Pacific Northwest of Canada and the United States, fails to stimulate an effective immune response in otherwise healthy hosts leading to morbidity or death. Because immunity to fungal pathogens requires intact cell-mediated immunity initiated by DCs, we asked whether C. gattii causes dysregulation of DC functions. C. gattii was efficiently bound and internalized by human monocyte-derived DCs, trafficked to late phagolysosomes, and killed. Yet, even with this degree of DC activation, the organism evaded pathways leading to DC maturation. Despite the ability to recognize and kill C. gattii, immature DCs failed to mature; there was no increased expression of MHC class II, CD86, CD83, CD80, and CCR7, or decrease of CD11c and CD32, which resulted in suboptimal T cell responses. Remarkably, no increase in TNF-α was observed in the presence of C. gattii. However, addition of recombinant TNF-α or stimulation that led to TNF-α production restored DC maturation and restored T cell responses. Thus, despite early killing, C. gattii evades DC maturation, providing a potential explanation for its ability to infect immunocompetent individuals. We have also established that DCs retain the ability to recognize and kill C. gattii without triggering TNF-α, suggesting independent or divergent activation pathways among essential DC functions.
Introduction
Dendritic cells (DCs) are the most efficient APCs for eliciting adaptive immune responses against pathogens (1). DCs excel at presentation of Ag by providing optimal costimulation to T cells after the process of DC maturation (2). Immature DCs (iDCs) are relatively quiescent in their ability to stimulate effector T cells, yet they exhibit efficient Ag capture and antimicrobial activity (2). After appropriate stimulation by a pathogen, DCs alter their morphology and phenotype consistent with enhanced Ag-presenting functions and become “mature DCs” (3–5). During the process of pathogen binding, internalization, phagosome processing, degradation, and killing, there are multiple opportunities for signaling to occur, resulting in DC maturation that facilitates the canonical Ag-presenting role of these cells. Indeed, once the pathway leading to DC maturation is initiated, phagocytosis and antimicrobial activity are suppressed (6, 7).
The initial interaction of DCs with pathogens is through pattern recognition receptors (PRRs) such as C-type lectin receptors, scavenger receptors, TLRs, retinoid-inducible gene-like receptors, nucleotide oligomerization domain-like receptors, and DNA-sensing receptors (8, 9). Extracellular binding to PRRs results in internalization of the organisms into phagosomes (10). Once internalized, the phagosomes are acidified and undergo phagolysosomal fusion (10, 11). Phagolysosomes then provide an environment conducive to killing and for appropriate Ag processing. Of interest, DCs show decreased capacity to degrade organisms compared with macrophages, presumably to preserve Ag integrity for presentation to T cells (12, 13). During this process, microbial products stimulate DCs to increase effector molecules responsible for antimicrobial activity and costimulatory molecules associated with DC maturation (14). Consequently, it may seem counterintuitive that a pathogen can be taken up and killed by iDCs but fail to activate the essential process of DC maturation leading to appropriate costimulation.
The process of DC maturation is characterized by morphologic, phenotypic, and functional changes. DCs alter their irregular morphology by decreasing coarse cytoplasmic projections associated with Ag capture, and develop a round shape with fine projections that enhance T cell interactions (2, 15). As they mature, DCs become less efficient at Ag capture and more efficient at Ag presentation (6), which is partly due to decreased surface expression of phagocytic PRRs including mannose receptor (MR) and CD32 (16, 17). Maturing DCs also increase their surface expression of MHC, to enhance Ag presentation to T cells (2, 18). In addition to MHC and Ag, T cells also require costimulatory signals for full activation and to avert anergy (19, 20). Indeed, mature DCs increase surface expression of the costimulatory molecules CD86, CD80, and CD40 to a greater extent than macrophages (21). Subsequently, mature DCs traffic to the local lymph nodes through increased expression of CCR7 and decreased expression of CCR6 (22). Trafficking allows the DCs to come into contact with a vast repertoire of naive Ag-specific T cells, an event not observed with macrophages. Efficient processing and antimicrobial activity not only supply appropriate Ag, but also provide stimulation for the maturing process (14). Unfortunately, knowledge is limited regarding the interplay of receptors, signaling pathways, and molecules that are specifically responsible for orchestrating these changes in DC functions.
Elimination of many fungal pathogens requires intact T cell immunity (23–31). It is also clear that DCs are the most effective APCs in eliciting cell-mediated immune responses against these fungi (9, 18, 32–34). Monocyte-derived DCs have contributed substantially to our current understanding of DC biology and potential vaccine strategies against microbes including fungi (35, 36). Indeed, in vivo studies found that protective TH1 immune responses against Cryptococcus neoformans required recruitment and differentiation of monocytes into mature DCs (37). Although these studies do not exclude a role for other DC subtypes, they do highlight the importance of this lineage of DCs during fungal host immunity.
Cryptococcus gattii is an emerging fungal pathogen on the Pacific Northwest of North America causing debilitation and death in hosts with no previous immune suppression (38). The emergence of C. gattii, which originated on Vancouver Island, Canada, is caused by infection with hypervirulent C. gattii, characterized by strain R265 (genotype VGIIa). Typically, infection consists of pneumonia and meningoencephalitis, resulting in 8–20% mortality rate despite the best available antifungal treatments (38–40). The observation that C. gattii infects healthy adults indicates that C. gattii is a primary fungal pathogen (38). Consequently, there is much interest in understanding the host immune response against C. gattii.
Although C. gattii is genetically most closely related to C. neoformans, C. neoformans preferentially infect immune-suppressed individuals (41), whereas C. gattii infect immune-competent individuals. It follows that host defense to C. gattii is most relevant to the primary fungal pathogens such as Coccidioides immitis, Histoplasma capsulatum, and Blastomyces dermatitidis. Primary fungal pathogens all have unique mechanisms of immune evasion to overcome normal host defenses. Whereas C. immitis, H. capsulatum, and B. dermatitidis use mechanisms of immune evasion other than subverting DC activation and appropriate T cell responses (32, 33, 42), the importance of cell-mediated immunity in cryptococcal host defense led us to ask whether C. gattii might evade DC functions that are responsible for eliciting Ag-specific T cell responses.
We hypothesized that C. gattii might impair the function of DCs leading to suboptimal cell-mediated immunity and defective pathogen clearance. To explore this hypothesis, we assessed DC binding and internalization of C. gattii using flow cytometry and optical microscopy. The ability of DCs to advance the C. gattii–containing phagosome to the phagolysosome was assessed using lysosomal-associated membrane protein-1 (LAMP-1) labeling, and anticryptococcal activity was examined using a quantitative CFU assay. Surface phenotypic analysis was used to assess the ability of C. gattii to induce DC maturation, and the ability of these DCs to provide T cell costimulation was examined using dilution of CFSE in responding T cells. To explore the mechanism(s) required for DC maturation, intracellular TNF-α expression and TNF-α secreted by DCs exposed to C. gattii was assessed by intracellular cytokine analysis and by Luminex bead array, respectively. Loss-of-function and gain-of-function studies were used to determine the role of TNF-α in DC responses to C. gattii. Finally, CFSE dilution or intracellular expression of Ki67 in responding T cells was used to assess the capacity of TNF-α to restore DC costimulation in response to C. gattii.
Materials and Methods
C. gattii species
The hypervirulent C. gattii strain, R265 (VGIIa), and less virulent R272 (VGIIb) strains from the Vancouver Island outbreak were kindly supplied by Dr. Jim Kronstad (University of British Columbia). C. gattii was stored at −80°C in 25% glycerol. For use in culture, frozen samples of C. gattii were cultured on Sabouraud dextrose agar (BD Biosciences, Sparks, MD) slants, followed by growth to stationary phase in Sabouraud dextrose broth at 32°C for 72 h. Aliquots from the stationary culture were then reconstituted in Sabouraud dextrose broth and grown to exponential phase over 24 h. For the C. gattii live-cell studies, the cultures were quantified and washed in RPMI complete media (RPMI 1640, 1% penicillin-streptomycin, 1% sodium pyruvate, and 1% nonessential amino acids [Life Technologies, Grand Island, NY]) before addition to culture systems. C. gattii was killed by incubating in a 60°C water bath for 60 min and washed in RPMI complete media before addition with DCs. Unless otherwise stated, studies were completed using heat-killed organisms.
Monocyte isolation, T cell isolation, and monocyte-derived DC differentiation
PBMCs were isolated from the blood of healthy donors (in compliance with the University of Calgary Conjoint Health Research Ethics Board Protocol #23363) by Ficoll Paque Plus (GE Healthcare Biosciences, Uppsala, Sweden) separation using centrifugation for 30 min at 650 × g. The cells were then washed three times in HBSS (Life Technologies), quantified, and labeled with CD14+ MACS microbeads (Miltenyi Biotec, Auburn, CA) as per the manufacturer’s protocol. Separations were performed using MACS 25 LS columns with a Manual MACS Cell Separator (Miltenyi Biotec). iDCs were derived from CD14+ cells cultured over 6–8 d in T-75 culture flasks (Corning, Corning, NY) at 1 × 106− fraction was labeled with MACS Pan T cell (negative) isolation kit (Miltenyi Biotec) as per manufacturer’s protocol. Purity of CD14+ cells was >85% and T cells >95% as assessed by flow cytometry using CD14-, CD3-, CD19-, CD11c-, CD56-, and CD16-specific Abs (BD Biosciences, Mississauga, ON). The viability of the CD14+ cells and T cells was >98% by trypan blue (Life Technologies) exclusion as assessed by optical microscopy. Unless otherwise stated, viability for all other experiments was assessed and maintained >95%. For all experiments, unless indicated, the iDCs and T cells were washed and cultured in RPMI complete media with 5% heat-inactivated human A/B serum (CELLect, Fairlawn, NJ) at 1 × 106 cells/ml in Costar low-attachment flat-bottom plates (Corning). Serum was heat-inactivated in a 60°C water bath for 20 min. In addition, all human cells were incubated at 37°C with 5% CO2 unless otherwise indicated. Finally, all flow cytometric studies were conducted using the Guava EasyCyte flow cytometer (version Cytosoft 5.3, Guava Technologies; Millipore, Danvers, MA), and data were analyzed using FlowJo software version 8.8.6 (Tree Star, Stanford, CA).
Phenotypic characterization of monocytes and monocyte-derived DCs
To determine the change in phenotype of monocytes differentiated into iDCs, the monocytes and iDCs were washed in FACS buffer followed by labeling with specific Abs CD1a (Beckman Coulter, Mississauga, ON, Canada), CD14, CD11a, CD11b, CD11c, MHC class II (MHC II), CD86, CD83, CD80, CD16, CD32, MR, CD3, CD19, CD56, or isotype controls (BD Biosciences). The surface labeling of these molecules was assessed using flow cytometry. The gating strategy included the entire forward/side scatter population.
Binding and internalization assessed by flow cytometry
To observe binding and internalization by flow cytometry, we labeled C. gattii with FITC (FITC-R265) (Sigma-Aldrich, St. Louis, MO). The organisms were cultured to exponential phase of growth and killed by heating at 60°C for 1 h. C. gattii were washed in 1× PBS followed by incubation in the presence of 1 μg/ml FITC per 108 cells for 10 min at 22°C, which labels both surface and intracellular amine groups of heat-killed organisms. The organisms were then washed three times in 1× PBS. iDCs with and without FITC-R265 were cultured at a ratio of 1:10 (DC/FITC-R265). At the various times, samples from the cultures were washed with 1× PBS, incubated for 10 min in the presence or absence of 62.5 μg/ml trypan blue (Life Technologies), and assessed for fluorescence by flow cytometry with initial DC gating by forward/side scatter. For some internalization studies, DCs were preincubated for 30 min in the presence or absence of 2 μg/ml cytochalasin D (Sigma-Aldrich), followed by washing in RPMI complete medium three times before the addition of R265. The cells were then assessed by flow cytometry.
Binding and internalization assessed by optical microscopy
To observe binding and internalization by optical microscopy, we left heat-killed C. gattiicytokmDCs). Samples were withdrawn from the cultures at various times, and cells were fixed and permeabilized with Cytofix/Cytoperm, as per manufacturer’s protocol (BD Biosciences). The cells were then labeled with PECy5-conjugated anti–LAMP-1 (CD107a) Ab or isotype control (BD Biosciences), as well as 0.1% DAPI (Sigma-Aldrich) to label the nuclei. The cells were mounted in ProLong Gold antifade reagent (Invitrogen Molecular Probes, Eugene, OR) on SuperfrostD*/plus microscope slides (Fisher Scientific, Waltham, MA), using No. 1.5 glass coverslips (Corning).
The immunofluorescent specimens were imaged with a combination of fluorescence and differential interference contrast (DIC) microscopy using a DeltaVision system (Applied Precision, Issaquah, WA), based on an Olympus IX70 microscope (Mississauga, ON, Canada). Standard multiband filter sets combined with a Sedat multiband dichroic filter were used for wavelength discrimination (multiband set 89000; Chroma Technologies, Rockingham, VT). Images were recorded with a PlanApo 60×/1.42 NA objective at room temperature.
Matched unlabeled cells and isotype-labeled controls were also imaged to ensure the fluorescence signal was not due to autofluorescence and/or nonspecific binding, respectively. In addition, the FITC-labeled C. gattii were also examined for autofluorescence and possible bleed through into the channels used to visualize the LAMP labeling. To assess internalization and association with LAMP-1 vesicles, we acquired three-dimensional optical stacks by recording images at 0.2- to 1.0-μm intervals. The image stacks with 0.2-μm spacing were deconvolved with a full iterative deconvolution SoftWoRx 5.0 (version 3; Applied Precision). Image processing and analysis was carried out using ImageJ software (version 14.3) (43). To assess internalization, we analyzed three-dimensional stacks in the xz and yz directions using orthogonal views. The organisms were identified by DIC and FITC fluorescence, and the DC cytoplasm by DIC and LAMP-1 labeling (44). Organisms were considered internalized by three-dimensional visualization if they were localized within the cytoplasm. LAMP-1 association with C. gattii–positive phagosomes required a ring of increased LAMP-1 fluorescence intensity immediately surrounding the phagosomes, which constituted a positive score.
Anticryptococcal activity assay
To assess anticryptococcal activity of DCs, we incubated iDCs or iDCs that had been cultured for 48 h with maturation cytokines (cytokmDC) at various effector (DC)/target (C. gattii) ratios for 24 h, with live C. gattii organisms in 96-well Costar flat-bottom plates (Corning). At time zero and after incubation, serial dilutions were performed and CFUs quantified. Viability of DCs assessed after incubation with C. gattii was >90%.
DC maturation by surface phenotype
iDC were incubated in the presence or absence of different strains of C. gattii at a ratio of 1:10 (DC/C. gattii), unless otherwise stated. iDC were cultured with strain R265 to produce R265-matured DCs (R265mDCs), or strain R272 to produce R272-matured DCs (R272mDCs), or 1 μg/ml LPS (List Biological, CA) to produce LPS-matured DCs (LPSmDCs), or maturation cytokines to produce cytokmDCs, or maturation cytokine plus R265 to produce cytok+R265mDCs, or maturation cytokine plus R272 to produce cytok+R272mDCs, or LPS plus R265 to produce LPS+R265mDCs. At 24 and 48 h, the cells were washed in FACS buffer followed by labeling with Abs specific for MHC II, CD86, CD83, CD80, CCR7, MR, CD32, CD11c, or isotype controls (BD Biosciences). The surface expression of these molecules, indicated by the fluorescence, was assessed using flow cytometry following initial gating by forward/side scatter.
For other studies, the DCs were permeabilized with Cytofix/Cytoperm, as per manufacturer’s protocol (BD Biosciences) and C. gattii were labeled with mouse anti-Cryptococcus 18B7 Ab (a kind gift from Dr. Arturo Casadevall) followed by anti-mouse Alexa Fluor 555 (Invitrogen Molecular Probes) secondary Ab. The DCs were then washed in FACS buffer, and surface labeling of CD86 or isotype control completed. DCs were assessed by flow cytometry with initial gating by forward/side scatter.
Cytokine analysis
Intracellular labeling of cytokines was performed on DCs after 6 h of incubation with and without the respective stimulants including no stimulant (iDC), R265 (R265mDC), maturation cytokines (cytokmDC), or LPS (LPSmDC). Brefeldin A (10 μg/ml; Sigma-Aldrich) was added at the start of each incubation. In brief, after incubation, the DCs were permeabilized with Cytofix/Cytoperm, as per manufacturer’s protocol (BD Biosciences) and labeled with PE anti-human TNF-α Ab or isotype control (BD Biosciences). The cells were then assessed for fluorescence using flow cytometry, with initial forward/side scatter gating of DCs. In addition, iDCs, R265mDCs, and DCs stimulated with 10 μg/ml exoenzyme S (45) and 10 μg/ml polymyxin B (Sigma-Aldrich) (ExoSmDC) were cultured over 48 h, and supernatants were assessed for TNF-α by Luminex Bead Array (EVE Technologies, University of Calgary, Calgary, AB, Canada).
Proliferation of T cells
T cells (50 × 106) were labeled with 5 μM CFSE (Millipore, Hayward, CA) for 10 min at 37°C, followed by washing three times in 1× PBS. CFSE-labeled T cells (1 × 106/ml) were resuspended and added to media alone, strain R265 alone, iDCs, cytokmDCs, R265mDCs, iDCs with the C. gattii added at the same time as the T cells, and cytok+R265mDCs, at a ratio of 1:10:10, DCs to C. gattii to T cells. The cultures were then stimulated with and without 1 μg/ml staphylococcal enterotoxin B (SEB) (Toxin Technology, Sarasota, FL). At various times, cells were washed in FACS buffer and fixed in 1% formalin (EMD, Gibbstown, NJ). The cells were then assessed for CFSE fluorescence by flow cytometry after initial gating of T cells by forward/side scatter.
In other experiments, various groups of DCs (see earlier) were added to C. gattii and allogeneic T cells at a 1:10:10 (DC/C. gattii/T cell) ratio. On day 5, the cells were washed in FACS buffer and labeled with PECy5 anti-human CD4, PECy5 anti-human CD8, or isotype control (BD Biosciences). The cells were made permeable with Cytofix/Cytoperm (BD Biosciences) and labeled with FITC anti-human Ki67 (Dako, Glostrup, Denmark) or isotype control (BD Biosciences). The cells were then assessed for fluorescence by flow cytometry, after initial gating of T cells by forward/side scatter.
Statistical analysis
For all experiments, the data are the means of at least three independent experiments using different donors on different days unless otherwise stated. Error bars indicate ±SEM. Statistical analysis tests include two-tailed paired Student t test, one-way and two-way ANOVA with Bonferroni multiple comparison test, and one-way ANOVA with Tukey multiple comparison test when appropriate. For all experiments, *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001 indicate statistically significant differences. All statistical analysis was completed on Prism 5 for Mac OS X software (version 5c).
Results
Human DCs bind and internalize C. gattii
Previous studies have demonstrated that efficient protective T cell responses against fungi such as H. capsulatum (32), Candida albicans (34), Aspergillus fumigatus (46), and C. neoformans (9, 47) initially require binding and internalization of the fungus by iDCs. We produced iDCs from monocytes using standard protocols (48). When compared with their monocyte precursors, these nonadherent and loosely adherent cells had characteristic decreased expression of CD14, increased CD1a and CD11c (48), with corresponding decreases in CD11a and CD11b, indicating differentiation into iDCs (Table I). There was also an increase in expression of MHC II and the costimulatory molecule CD80 (48), which is characteristic of this transition of monocytes to iDCs, and an increase in the expression of phagocytic receptors, CD16, CD32, and MR (Table I) (48).
Strain R265 was labeled with FITC and binding assessed by flow cytometry to determine whether iDCs bound to C. gattii. As early as 30 min, 66% of iDCs bound to strain R265 with no further increase over 24 h (Fig. 1A). To reveal the internalized organisms, we used trypan blue to quench fluorescence of extracellular C. gattii (49). Very few DCs had internalized organisms at 30 min, but a significant increase in internalization was noted between 30 min and 24 h (3.5 ± 3.3 versus 28 ± 2.0%; Fig. 1A). iDCs were preincubated with cytochalasin D to assess the contribution of actin-dependent phagocytosis (49). Cytochalasin D–treated iDCs did not significantly block association of organisms, suggesting that actin is not involved in binding of C. gattii (Fig. 1B). However, there was a significant decrease in the percentage of iDCs that had internalized R265 when compared with the absence of cytochalasin D treatment (10.0 ± 5.5 versus 28.0 ± 2.0%; Fig. 1B), indicating that internalization of R265 requires actin-mediated phagocytosis. Thus, these data suggest that there is no defect in association or uptake of C. gattii by iDCs that would explain defective T cell responses.
DCs associate with and internalize C. gattii. iDCs were cultured with FITC-labeled R265 (FITC-R265) and assessed for association or internalization by DCs using flow cytometry (A, B) or optical microscopy (C, D). (A) iDCs in the presence of FITC-R265 at 30 min and 24 h were assessed in the absence of trypan blue quenching (associated) and in the presence of trypan blue (internalized). (B) As in (A) at 24 h in the presence and absence of cytochalasin D. (C) The percent iDCs or cytokmDCs that had internalized R265 at 30 min and 24 h. (D) The percent iDCs or cytokmDCs that had internalized 1–3 or >3 R265 per DC at 24 h. More than 25 DCs with associated or internalized organisms were observed per experiment, including z-stack image analysis of each. Statistical analysis using repeated-measures one-way (A, B) and two-way (C, D) ANOVA both with Bonferroni multiple comparison posttest. *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001.
Internalization of R265 by iDCs was confirmed using optical microscopy. There was a statistically significant increase in the percentage of iDCs with internalized organisms at 24 h when compared with 30 min (44.0 ± 3.5 versus 7.7 ± 3.7%; p < 0.001; Fig. 1C). Upon activation of DCs their phagocytic capacity decreases (6). Therefore, as a control, DCs were incubated with maturation cytokines (cytokmDC) along with the R265. At 24 h, the percentage of cytokmDCs that had internalized R265 was significantly lower than that of the iDCs (Fig. 1C). Also, of the DCs that had internalized small numbers of organisms, cytokmDCs predominated, whereas of the DCs that had internalized large numbers of organisms, iDCs predominated (Fig. 1D). These observations extend prior observations that mature DCs have reduced phagocytosis when compared with iDCs to now include C. gattii (2, 6). It further indicates that the defect in immunity against C. gattii is not due to DC failure to bind or internalize the organism.
Efficient Ag presentation of microorganisms requires specific sorting of internalized endosomes during phagosome maturation (50, 51). These processes are precursors to degradation that leads to anticryptococcal activity and Ag processing of intracellular organisms (10). In particular, phagosomes that fuse with LAMP-1+ lysosomes (phagolysosomes) house the machinery to process Ag (50). Pathogenic organisms such as Mycobacterium tuberculosis have been found to evade this essential mechanism (52). Therefore, to determine whether C. gattii is internalized into late phagolysosomes, iDCs were cultured with R265 and assessed by fluorescence microscopy. R265 was observed to associate with LAMP-1+ endosomes at 24 h (Fig. 2A), and significantly more R265 were present in LAMP-1+ phagosomes at 24 h compared with 30 min (Fig. 2B). Together, these data indicate that phagosomes containing C. gattii progress to a late phagolysosomal compartment providing an appropriate environment for Ag processing.
DCs internalize C. gattii and traffic to late phagolysosomes. (A) Single optical section images of iDCs cultured with FITC-labeled R265 (green), PECy5 anti–LAMP-1 (red), DIC, and the overlay image with LAMP-1 (red), R265 (green), and DAPI nuclear stain (blue) at 24 h (scale bars, 8 μm). Images presented have not been manipulated: autofluorescence and spectral bleed-through account for <6% of the signal. (B) Percent internalized FITC-R265 within LAMP-1+ vesicles at 30 min and 24 h for iDCs and 24 h for cytokmDCs, as assessed by two-dimensional visualization of z-stack images. Values are from at least 25 internalized organisms in each of at least three 24-h experiments and 10 organisms at 30 min. Statistical analysis was performed using a two-tailed paired Student t test comparing iDCs at 30 min to 24 h and comparing iDCs with cytokmDCs. *p ≤ 0.05.
Human DCs exhibit anticryptococcal activity against C. gattii
Having shown that C. gattii is internalized to late phagolysosomal compartments, we did experiments to determine whether the organisms were killed as a precursor to Ag processing. Association of DCs with other fungi such as C. albicans, H. capsulatum, and C. neoformans stimulates the production of effector molecules such as superoxides or lysosomal hydrolases, which are responsible for fungal death (10, 32, 34). To determine whether DCs have antifungal activity to C. gattii, we cultured the highly virulent strain R265 and the less virulent strain R272 in the presence and absence of various ratios of DCs and assessed for CFU over 24 h (39). When strain R265 or R272 was cultured in the presence of iDCs, there was a significant dose-dependent reduction in the CFU, indicating that iDCs possess anticryptococcal activity against C. gattii (Fig. 3A, 3B). In addition, there was no significant difference in the anticryptococcal activity of DCs against strains R265 and R272, suggesting that differences in anticryptococcal activity do not explain the difference in virulence between these two strains. As expected, cytokmDCs also had significantly less anticryptococcal activity against both R265 and R272 compared with iDCs (Fig. 3C, 3D) (7). These experiments showed that DCs are fully capable of recognizing C. gattii, which leads to signaling that causes phagolysosomal maturation and anticryptococcal activity.
DCs acquire anticryptococcal activity against C. gattii. R265 and R272 were cultured with and without iDCs or cytokmDCs at various E:T ratios, over 24 h. The cultures were then assessed for CFU. (A and B) CFU for R265 and R272 cultures, respectively, in the presence and absence of increased numbers of iDCs. Data are the means of quadruplicate wells and are representative of one of at least three independent experiments with different donors. (C and D) Fold decrease in R265 or R272 CFU (R265 or R272 alone/test CFU), respectively, in the presence and absence of iDCs or cytokmDCs, at a 1:100 (Cryptococcus/DC) ratio. Data are the mean of at least three individual experiments with different donors. (A–D) Statistical analysis using one-way ANOVA with Tukey multiple comparison posttest. *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001.
Human DCs have dysregulated maturation in response to C. gattii
In addition to acquiring and processing the organism, DCs receive signals from pathogens resulting in DC maturation that is required for efficient Ag presentation (2, 16, 53). This transformational change is induced by infectious stimuli that include many fungal pathogens such as C. albicans (54), A. fumigatus (55), C. neoformans (18), and Coccidioides posadasii (56), and nonpathogenic species or their products including Ganoderma lucidum (57), Cordyceps sinensis (58), Clitocybe nebularis (59), Hericium erinaceus (60), Malassezia furfur (61), and Saccharomyces cerevisiae (62). Moreover, although a pathogenic fungus has never been demonstrated to inhibit DC maturation, isolated fungal products have been shown to inhibit DC maturation (63, 64). To determine whether DCs mature in the presence of C. gattii, we incubated DCs in the presence or absence of strain R265 or R272 and assessed for surface expression of MHC II, CD86, CD80, CD83, CCR7, MR, CD32, and CD11c. No significant change in surface expression of any of the maturation-specific molecules (with the exception of MR at 48 h with R265) was observed in the presence of R265 or R272, compared with the cytokmDC+ controls, at 24 or 48 h (Fig. 4A–C). Studies were also conducted with heat-killed R265 at a 1:10 ratio over 72 h and showed no change in surface expression of CD86 (data not shown). As additional controls, no absorption of the labeling Abs by C. gattii was noted (data not shown), and no significant change in the surface expression of CD45, which remains constant during the process of DC maturation, was observed (data not shown).
Altered state of DC maturation induced by C. gattii. iDCs were cultured for 48 h in the presence of R265 or R272 (R265mDCs or R272mDCs), or with maturation cytokines (cytokmDCs), as a positive control. The cells were then labeled with fluorescent-conjugated Abs and assessed by flow cytometry. The geometric mean fluorescence intensity (gMFI) was used to quantify the fold increase in surface expression (test DC/iDC). (A–C) Heat-killed C. gattii with ratio of 1:10 (DC/C. gattii). Bar graphs include (A) MHC II and costimulatory molecules CD86 and CD80, along with CD83; (B) phagocytic receptors MR, CD32, and CD11c; and (C) chemokine receptor CCR7. (A–C) Statistical analysis using two-tailed paired Student t test comparing stimulated with control (iDC). DCs cultured as described over 48 h (D) with various ratios of heat-killed R265 (1:1, 1:10, 1:25, and 1:50) or live R265 (1:1, 1:3, 1:7, 1:33, and 1:67 starting inocula; DC to C. gattii) with fold increase in surface expression of CD86 and MR; (E) with heat-killed organisms in the presence or absence of various concentrations of heat-inactivated serum (10, 20, and 50%); and (F) with heat-killed organisms and 10% heat-inactivated (HI) or non–heat-inactivated (non-HI) serum. Statistical analysis (D–F) was performed using repeated-measures ANOVA with Bonferroni multiple-comparison posttest. *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001.
To explore possible adverse effects of heat-killing C. gattii on the exposure of ligands recognized by DCs, we also conducted experiments with live R265 (Fig. 4D). No significant increase in surface expression of CD86 was observed when live C. gattii was compared with heat-killed organisms, indicating that heating the organisms did not disrupt a DC maturation-specific ligand. During these studies, the expression of MR decreased in a dose-dependent manner in response to both live and heat-killed organisms (Fig. 4D). The observed change in MR expression with no change in CD86 expression further supported a dysregulation in DC maturation in response to C. gattii.
DC maturation can be stimulated when organisms are opsonized with Ag-specific Ab (65). However, although natural Ab that is present in nonimmune serum can enhance phagocytosis, it has been shown to inhibit or have no effect on DC maturation (66, 67). To examine the role of natural Ab, we conducted experiments with various concentrations of pooled human serum. Despite the presence of the FcR CD32 (Fig. 4B), there was no change in the expression of DC CD86 over a broad range of serum concentrations (0–50%; Fig. 4E), suggesting that natural Abs do not influence DC maturation.
DC complement receptors have been previously shown to be important for activation of phagocytosis, but their role in DC maturation was dispensable (68). To explore a possible role for complement, we compared heat-inactivated sera with non–heat-inactivated serum. No significant change in the expression of CD86 was observed with non–heat-inactivated compared with heat-inactivated serum at 10 (Fig. 4F), 20, or 50% (data not shown). These data indicate that the mechanism by which DCs fail to mature in the presence of C. gattii is not due to the absence of natural Abs or complement. Together, these studies show that despite recognition, internalization, and killing of the organisms by DCs, the DC signals received in response to C. gattii induced altered DC maturation characterized by decreased expression of MR without increased expression of costimulatory molecules.
C. gattii–matured DCs induce suboptimal T cell responses
Having shown altered expression of costimulatory molecules, we performed experiments to determine whether the defective DC maturation was associated with functional consequences. For this purpose, the ability of T cells to enter into the cell cycle in response to the superantigen SEB was assessed, as this response is sensitive to the density of MHC and costimulatory receptors, and excludes the effects of Ag processing and presentation by MHC (69, 70). To this end, CFSE-labeled T cells were cultured alone, in the presence of syngeneic iDCs, cytokmDCs, R265mDCs, R265 alone, or R265 plus iDCs and stimulated with low concentrations of SEB. The T cell response was assessed at 48 h because SEB induces DC maturation at later times, masking the maturation state of prestimulated cells (71). When costimulation was provided by various populations of DCs, significantly less T cell proliferation occurred when the accessory cell activity was supplied by R265mDCs compared with cytokmDCs (Fig. 5). Indeed, the percentage of proliferating T cells was not significantly different when R265mDC APCs were used as the source of costimulation compared with when iDCs were used (Fig. 5). These observations were also consistent over a series of DC-to-T cell ratios (Fig. 5). The T cell viability was assessed over 48 h and remained >94%, suggesting that the mechanism of low T cell proliferation in response to R265mDCs was not due to a decrease in viability. Overall, the failure of T cells to proliferate efficiently in response to R265mDCs extends the observation that C. gattii–stimulated DCs remain functionally immature.
DCs matured with C. gattii induce suboptimal T cell responses. T cells were stained with CFSE and stimulated with 1 μg/ml SEB in the presence of no APCs, R265 with no APCs, iDCs, cytokmDCs, R265mDCs, or R265 with iDCs. The percentage of T cells with >2-fold reduction in CFSE fluorescence was assessed by flow cytometry as an indication of one or more rounds of proliferation at 48 h. The line graph shows the percent proliferating T cells from a single donor stimulated with SEB in the presence of multiple DC/T cell ratios (1:10, 1:20, 1:40, and 1:80) under each condition. The bar graph shows the mean of the percent proliferating T cells from three independent experiments with different donors stimulated with SEB in the presence of a DC/T cell ratio of 1:10 under each condition. Statistical analysis using repeated-measures ANOVA with Bonferroni multiple-comparison posttest. **p ≤ 0.001, ***p ≤ 0.0001.
C. gattii–matured DCs do not produce TNF-α
DCs recognized C. gattii, leading to internalization, phagosome processing, and anticryptococcal activity. However, the presence of C. gattii produced an altered state of DC maturation that was functionally consistent with an immature phenotype. Because current knowledge is limited regarding coordination of signals that are responsible for connecting phagocytosis, phagosome maturation, growth inhibition, and Ag processing and how they relate to the maturation state of DC, we sought to define a mechanism. Studies have shown that DC maturation is dependent on TNF-α production, whereas the other functions may occur independent of TNF-α (17, 21, 72–74). To understand this dysfunction, we determined intracellular expression of TNF-α in response to various stimuli. iDCs were incubated in the maturation protocols with R265 (R265mDCs), LPS (LPSmDCs), and maturation cytokines (cytokmDC), in the presence of brefeldin A, which blocks vesicular trafficking from the endoplasmic reticulum to the Golgi, allowing accumulation of TNF-α within the DC. The percentage of DCs with intracellular expression of TNF-α was then measured by flow cytometry. LPSmDCs served as a positive control, showing an increase in the percentage of DCs with TNF-α production (Fig. 6A). As expected, cytokmDCs do not produce TNF-α (Fig. 6A); rather, the TNF-α is added to the culture media to stimulate DC maturation directly. No increase in the percentage of DCs with intracellular production of TNF-α was observed in the presence of R265 (Fig. 6A). Viability of cells was maintained >90% throughout the experiment. Additional experiments were performed to determine whether TNF-α was released by DCs in response to C. gattii in the absence of brefeldin A. Whereas the positive control Pseudomonas aeruginosa exoenzyme S induced TNF-α secretion (Fig. 6B) (75), no TNF-α was detected in the culture supernatants of R265mDCs (Fig. 6B). Moreover, TNF-α production correlated with DC maturation as assessed by expression of MHC II and costimulatory molecules (Fig. 6C). Thus, despite binding, internalization, phagolysosomal processing, and anticryptococcal activity, C. gattii evaded signaling leading to TNF-α production, which was, in turn, associated with an inability of DCs to fully mature and stimulate T cell responses.
DCs matured with C. gattii do not increase production of TNF-α. (A) iDCs were cultured in the absence or presence of R265 (R265mDCs), maturation cytokines (cytokmDCs), or LPS (LPSmDC) each in the presence of brefeldin A. The cells were then made permeable and labeled with anti–TNF-α Ab at 6 h and assessed by flow cytometry. The percentage of TNF-α+ cells under each condition is shown. (B) iDCs were cultured in the absence or presence of R265 (R265mDCs) or with exoenzyme S (10 μg/ml) and polymyxin B (10 μg/ml) (ExoSmDCs) over 48 h. The supernatants were then measured for the presence of TNF-α by Luminex Bead Array and quantified as indicated on the bar graph. (C) iDCs were matured for 48 h in the absence or presence of 1 μg/ml LPS (LPSmDCs) or R265 (R265mDCs). The cells were then labeled with anti–MHC II, CD86, and CD83, and assessed by flow cytometry as indicated in Fig. 4. Values represent the fold increases in gMFI (test DC/iDC). (A–C) Statistical analysis was performed using a two-tailed paired Student t test of stimulated compared with control (iDCs). *p ≤ 0.05, **p ≤ 0.001.
C. gattii–matured DCs undergo normal maturation upon exogenous addition or induction of TNF-α
Our studies demonstrated that iDCs fail to produce TNF-α and failed to mature in response to C. gattii. Consequently, we asked whether DC maturation might be restored by an exogenous source of TNF-α or by stimulating TNF-α production using LPS. To this end, iDCs were matured in the presence of strain R265 or R272 alone (R265mDCs or R272mDCs) or R265 or R272 with LPS or maturation cytokines (R265+LPSmDCs, cytok+R265mDCs, or cytok+R272mDCs). Under these conditions, DCs increased their cell surface expression of MHC II, CD86, CD83, CD80, and CCR7, and decreased expression of MR and CD32 compared with C. gattii–matured DCs (Fig. 7), indicating that the response could be restored. In addition, there was no significant decrease in CD86 expression when cytok+R265mDCs were compared with cytokmDCs (Fig. 7E), also suggesting that strain R265 did not inhibit cytokine-induced maturation. We also considered the possibility that C. gattii might inhibit only those DCs that were associated with organisms. To test this, we fluorescently labeled C. gattii and added them to DCs undergoing maturation with cytokines. Expression of CD86 on the mDCs associated with R265 was compared with mDCs that were not associated with R265. No significant change in the expression of CD86 was observed on cytok+R265mDCs that were associated with R265 compared with the cytok+R265mDCs that were not (Fig. 7E). Therefore, we were unable to demonstrate active inhibition of DC maturation by C. gattii. Together, these data indicate that recovery of DC maturation in the presence of C. gattii may be achieved by the addition of TNF-α either supplied in the form of maturation cytokines or induced through LPS. Therefore, these findings confirm the significance of TNF-α in the process of DC maturation and raise the possibility that TNF-α might functionally restore defective T cell responses to C. gattii.
Recovery of a mature DC state in the presence of C. gattii. iDCs were cultured in the absence or presence of C. gattii (R265mDCs or R272mDCs), with maturation cytokines (cytokmDCs), with C. gattii and maturation cytokines (cytok+R265mDCs or cytok+R272mDCs), or C. gattii and LPS 1 μg/ml (LPS+R265mDCs), over 48 h. The DCs were then labeled with fluorescent-conjugated Abs and assessed by flow cytometry, as described in Fig. 4. The fold increase in gMFI (test DC/iDC) for (A) MHC II and costimulatory molecules CD86 and CD80 along with CD83, (B) phagocytic receptors MR and CD32, (C) chemokine receptor CCR7, and (D) CD83 for cytok+R272mDCs and LPS+R265mDCs is shown. (E) iDCs cultured as described, but with dual labeling of R265 and CD86 followed by flow cytometry. Bar graphs indicate the fold increase in gMFI (test DC/total iDC) representing surface CD86 expression. Test conditions included the total population of DCs under each condition (total DCs), the DCs not associated with R265 under each condition (R265− DCs), and DC positive for R265 for culture conditions with R265 (R265+ DCs). Statistical analysis using repeated-measures ANOVA with Bonferroni multiple-comparison posttest. *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001.
TNF-α restores the ability of C. gattii–matured DCs to induce T cell responses
To determine whether maturation cytokines restored the function of DCs, T cell responses were examined in the presence of DCs that had been matured with strain R265 with (cytok+R265mDCs) or without maturation cytokines (R265mDCs). As before, the response to SEB was assessed, which is dependent on the level of expression of MHC II as well as costimulatory molecules (69, 70). As previously observed, cytokmDCs enhanced T cell responses, whereas R265mDCs did not (Fig. 8A). Addition of maturation cytokines to R265 (cytok+R265mDCs) caused DCs to produce a significant increase in T cell responses compared with DCs that had been matured in the presence of R265 alone (R265mDC), confirming a functional recovery of DC maturation (Fig. 8A).
Recovery of T cell responses by DCs matured in the presence of C. gattii, correlates with the state of DC maturation. (A) T cells were labeled with CFSE and stimulated with 1 μg/ml SEB in the presence of iDCs, cytokmDCs, R265mDCs, or cytok+R265mDCs. Dilution of CFSE was assessed as an indication of proliferation (one or more rounds) at 48 h, using flow cytometry. The bar graph represents the percent proliferating T cells cultured with a DC/T cell ratio of 1:10, under each condition. (B) T cells were incubated alone, in the presence of allogeneic iDCs, R265mDCs, or cytok+R265mDCs. On day 5, the cells were surface labeled with PE-Cy5–conjugated Abs specific for CD4 and CD8 or an isotype control. The cells were then permeabilized and colabeled with FITC-conjugated Abs specific to Ki67. Bar graphs represent the percent CD4+ Ki67+ and CD8+ Ki67+ cells. (C) T cells were stained with CFSE and cultured alone or with R265 alone, syngeneic iDCs, cytokmDCs, R265mDCs, or cytok+R265mDCs. Dilution of CFSE was assessed as an indication of proliferation (one or more rounds) on day 4, using flow cytometry. Bar graph representing quantification of the percent of proliferating T cells with a DC/T cell ratio of 1:10 with each condition. (A–C) Statistical analysis using repeated-measures ANOVA with Bonferroni multiple-comparison posttest. *p ≤ 0.05, **p ≤ 0.001, ***p ≤ 0.0001.
It has previously been shown that both CD4+ T cells and CD8+ T cells play important roles in immunity to fungi (23–31). To test the ability of maturation cytokines to restore DC maturation and stimulate both CD4+ and CD8+ T cell responses, we stimulated allogeneic T cells with cytok+R265mDCs, and expression of the proliferation-specific nuclear protein, Ki67, was used as a measure of the MLR. These experiments demonstrated a significant increase in Ki67 expression in both CD4+ and CD8+ T cells when cultured with allogeneic cytok+R265mDCs compared with allogeneic R265mDCs (Fig. 8B) (1, 76), confirming functional recovery of both CD4+ and CD8+ T cell responses.
Recovery of DC maturation, which induced both allogeneic and superantigen-mediated T cell responses, raised the possibility that DC maturation might recover C. gattii–specific responses. Therefore, to determine whether DC maturation induced by maturation cytokines was sufficient to stimulate C. gattii–specific T cell responses, autologous T cells were stimulated with R265mDCs and compared with cytok+R265mDCs. T cells alone and T cells incubated with iDCs or R265 alone failed to respond (Fig. 8C). However, significantly more T cells proliferated in response to cytok+R265mDCs than in response to R265mDCs, indicating that maturation cytokines recover DC maturation and, subsequently, T cell proliferative responses to C. gattii (Fig. 8C). Moreover, significantly more T cells proliferated in response to cytok+R265mDCs than in response to cytokmDCs, suggesting that the response is specific to C. gattii Ag (Fig. 8C). Together, these data indicate that the addition of maturation cytokines results in recovery of T cell proliferative response against C. gattii.
Discussion
In these studies, we have made five observations that are important to our understanding of DC biology and host defense against the emerging primary fungal pathogen, C. gattii: 1) iDCs associate with and internalize C. gattii into late phagolysosomes and exhibit anticryptococcal activity against C. gattii; 2) DC maturation by C. gattii strain R265 (R265mDC) exhibit a dysregulated phenotype that causes suboptimal T cell activation and proliferation; 3) the altered maturation status of R265mDC is associated with a failure to produce TNF-α, and DC maturation could be recovered by addition of maturation cytokines that contain TNF-α or by stimulation with LPS that induces TNF-α production by DCs; 4) recovery of mature DCs reduced Ag capture, phagosome maturation, and anticryptococcal activity; and 5) recovery of DC maturation in response to cytok+R265mDC resulted in a significant increase in T cell activation and proliferation compared with R265mDC.
Little is known about the host immune response to C. gattii. However, because C. gattii is a primary human pathogen, we reasoned that it might somehow circumvent normal immunity. Given that DCs are the sentinels required for effective T cell–dependent immunity against fungi (9, 32–34), we asked whether the DC recognition or response to C. gattii might somehow be aberrant, which would result in abnormal T cell–mediated responses.
DCs initially associate with pathogens through surface PRRs, whereby association results in DC activation (9, 18). Organisms such as Klebsiella pneumoniae evade DC binding through the production of an external polysaccharide capsule that masks the inherent pathogen-associated molecular patterns recognized by PRRs, resulting in decreased binding and internalization of the organism (77). In our studies, however, it is evident that DCs were able to bind the encapsulated C. gattii and internalize the organism, which was a process dependent on actin polymerization (49), indicating that the polysaccharide capsule of C. gattii was not inhibiting recognition or activation of these DC processes. Although some reports have suggested a strict requirement for opsonization (47), in our studies and those of others (18), specific opsonization was not required for binding of C. gattii by human monocyte-derived DCs.
Once internalized, other organisms, such as Mycobacterium tuberculosis, evade the host immune response both through failure to activate antimicrobial mechanisms and avoiding phagosome maturation (52). In our studies, however, the R265mDCs and R272mDCs had significant anticryptococcal activity, indicating that appropriate DC effector mechanisms were in place against both strains. These observations are consistent with the ability of DCs to exhibit antifungal activity against other primary fungi such as H. capsulatum (32). Moreover, the phagosomes containing R265 underwent maturation as assessed by association of R265 phagosomes with the lysosomal protein, LAMP-1, indicative of phagolysosomal fusion (50). These observations indicate that DCs not only recognize the organism, they also internalize C. gattii to appropriate compartments required for growth inhibition and Ag processing.
The hallmark of DCs is their ability to efficiently present Ag to T cells. To accomplish this, the cells undergo DC maturation (16). Increased surface expression of MHC in the context of Ag along with costimulatory molecules (CD86 and CD80) is the hallmark of this process (2, 18, 21). Together, these molecules are required for efficient activation of T cells. Mature DCs also show alterations in the surface expression of chemokine receptors, such as increased CCR7 expression that is associated with trafficking of DCs toward the vast repertoire of T cells that reside in the lymph nodes (22). In our studies, however, no significant increase in expression of these molecules was noted on the surface of DCs that were incubated with either live or heat-killed C. gattii (R265mDCs and R272mDCs) compared with LPSmDCs or cytokmDCs, suggesting that DCs fail to mature in the presence of C. gattii. Mature DCs also decrease their phagocytic capacity, in part through decreased surface expression of phagocytic receptors (2). Consistent with this, MR, which is important in recognition of fungi by DCs, was observed to significantly decrease in the presence of live or heat-killed C. gattii, although there was no concomitant decrease in CD32 (9, 18, 78). In addition, despite the observed changes in MR expression, R265mDCs do not decrease their phagocytic capacity to that observed with cytok+R265mDCs, suggesting that they are not fully matured. Finally, failure to express costimulatory molecules had functional consequences; R265mDCs failed to provide optimal costimulation for T cell proliferation. Together, then, these data indicated a dysregulation in DC maturation in the presence of C. gattii that was functionally consistent with an immature state, which may be a host evasion strategy used by C. gattii.
Previous studies had shown that Abs and complement were important for DC binding and antimicrobial activity against Cryptococcus species (47). These data were consistent with other observations using opsonized Staphylococcus aureus or apoptotic cells as stimulants. However, these studies also showed that natural Abs and complement were dispensable for DC maturation or TNF-α production that is a precursor to DC maturation (66, 68, 73, 74). In our studies, we found no evidence that natural Abs or complement from human sera were required for binding, internalization, or antifungal activity by DCs against C. gattii. Furthermore, the presence of Abs and complement did not recover DC maturation in the presence of C. gattii, indicating that the mechanisms by which C. gattii evaded DC maturation occurred independent of these host molecules.
We were intrigued by the observations that DCs could recognize a microbe as danger and become activated, resulting in the induction of phagosome maturation and antimicrobial activity without stimulating DC maturation. Previous studies have identified other organisms that avoid DC maturation to evade effective immunity (79–81). However, these studies either did not investigate binding, internalization, phagosome maturation, and antimicrobial activity or found that there was also a defect in these mechanisms along with DC maturation, suggesting that the signaling mechanisms may be linked. In our studies, the DC response to C. gattii appears to be unique because it clearly shows a disconnection between the mechanisms involved in these processes. Of interest to DC biology, we have identified a model system that can be used experimentally to further our understanding of DC-specific molecular mechanisms that are required for their function as APCs.
The current literature indicates that TNF-α is required for DC maturation (73, 74). Pathogens, such as HSV, evade DC maturation by failing to induce proinflammatory cytokines including TNF-α (79). TNF-α production, which is associated with DC maturation, also correlates with a reduction in antimicrobial activity by DCs (7), suggesting that the process of maturation is linked to a reduction in antimicrobial activity. However, little is known about the role of TNF-α in phagosome maturation or Ag processing, and the current evidence is confined to the role of TLRs in induction of TNF-α (51, 82). In our studies, despite phagosome maturation and killing, there was no increase in the production of TNF-α by R265mDCs. Moreover, the addition of R265 with recombinant TNF-α or with LPS that stimulates production of TNF-α resulted in recovery of DC maturation and T cell responses. We confirmed that the mechanism of enhanced DC maturation with cytok+R265mDCs was not due to an increase in binding, internalization, phagosome maturation, or antimicrobial killing leading to Ag processing and presentation. Rather, recovery of the response was due to DC maturation alone. These studies suggest that separate signals are required for killing and DC maturation, but once DC maturation is initiated, it reciprocally regulates uptake, killing, and Ag processing.
An important goal of DC biology is to develop vaccine and therapeutic strategies. In these studies, we have identified a DC-specific evasion strategy by C. gattii that has not been previously shown to play a role in subversion of host immunity by primary fungal pathogens such as H. capsulatum, C. immitis, and B. dermatitidis or by other Cryptococcus species that infect immunocompromised individuals. We also show that this response could be recovered by the addition of maturation cytokines or LPS, and that C. gattii did not actively inhibit DC maturation induced by maturation cytokines. In doing so, we have also observed enhanced T cell responses to cytok+R265mDCs, when compared with R265mDC alone. We also observed enhanced CD4 and CD8 T cell allogeneic responses against cytok+R265mDCs compared with R265mDCs, indicating that C. gattii does not block these specific responses. Together, these observations are consistent with previous studies including potential cancer therapies or HIV treatments that target DC-based vaccine strategies (36, 83, 84). In this case specifically, the potential use of either ex vivo or in vivo TNF-α induction of DC maturation may be used for vaccine strategies to improve outcomes for patients with C. gattii infections.
In summary, we have investigated a C. gattii–specific evasion mechanism associated with DC Ag-presenting function. In this way, we have found that DCs have dysregulated DC maturation because of the absence of TNF-α production that is also consistent with suboptimal T cell responses, together describing a C. gattii–specific evasion mechanism. In doing so, we have also determined that there was a disconnect between the processes of Ag capture, phagosomal maturation, and antimicrobial activity and DC maturation, as only the former occurs in the absence of TNF-α. In summary, our studies offer increased knowledge for the field of DC biology that may be used as potential therapeutics against C. gattii and other intracellular pathogens that evade DC Ag-presenting function.
Disclosures
The authors have no financial conflicts of interest.
Acknowledgments
We thank Laurie Kennedy and Laurie Robertson for technical assistance with flow cytometry, and Jen Amon from the Live Cell Imaging Core Facility, Snyder Institute for Chronic Disease (Calgary, AB, Canada), for valuable assistance with immunofluorescence microscopy.
Footnotes
This work was supported by grants from the Lung Association–Alberta and North West Territories (to C.H.M.). C.H.M. is the Jessie Boden Lloyd Professor of Immunology. S.M.H. is supported by a studentship from the Lung Association–Alberta and North West Territories and Queen Elizabeth II scholarships. Microscopy was supported by the Live Cell Imaging Facility funded by an equipment and infrastructure grant from the Canadian Foundation for Innovation and the Alberta Science and Research Authority.
Abbreviations used in this article:
- cytokmDC
- maturation cytokine-matured dendritic cell
- cytok+R265mDC
- maturation cytokine plus R265-matured dendritic cell
- cytok+R272mDC
- cytokine plus R272-matured dendritic cell
- DC
- dendritic cell
- DIC
- differential interference contrast
- gMFI
- geometric mean fluorescence intensity
- iDC
- immature DC
- LAMP-1
- lysosomal-associated membrane protein-1
- LPSmDC
- LPS-matured DC
- MHC II
- MHC class II
- MR
- mannose receptor
- PRR
- pattern recognition receptor
- R265mDC
- R265-matured DC
- R272mDC
- R272-matured DC
- SEB
- staphylococcal enterotoxin B.
- Received September 27, 2012.
- Accepted April 23, 2013.
- Copyright © 2013 by The American Association of Immunologists, Inc.
References
In this issue
