Th cell responses induced by Aspergillus fumigatus have been extensively investigated in mouse models. However, the requirements for differentiation and the characteristics of A. fumigatus–induced human Th cell subsets remain poorly defined. We demonstrate that A. fumigatus induces Th1 and Th17 subsets in human PBMCs. Moreover, we show that the cytokine IL-22 is not restricted to a specific Th subset, in contrast to IL-17A. The pattern recognition and cytokine pathways that skew these Aspergillus-induced Th cell responses are TLR4- and IL-1–, IL-23–, and TNF-α–dependent. These pathways are of specific importance for production of the cytokines IL-17A and IL-22. Additionally, our data reveal that the dectin-1/Syk pathway is redundant and that TLR2 has an inhibitory effect on Aspergillus-induced IL-17A and IL-22 production. Notably, blocking complement receptor (CR)3 significantly reduced Aspergillus-induced Th1 and Th17 responses, and this was independent on the activation of the complement system. CR3 is a known receptor for β-1,3-glucan; however, blocking CR3 had significant effects on Th cell responses induced by heat-killed Aspergillus conidia, which have minimal β-glucan expression on their cell surface. Collectively, these data characterize the human Th cell subsets induced by Aspergillus, demonstrate that the capability to produce IL-22 is not restricted to a specific T cell subset, and provide evidence that CR3 might play a significant role in the adaptive host defense against Aspergillus, although the ligand and its action remain to be elucidated.
The primary line of defense against Aspergillus fumigatus is mediated by neutrophils and other cells of the innate immune system (1). In addition to the innate immune response, adaptive Th cell responses also play a crucial role during invasive aspergillosis. The Th1 response is associated with a protection in invasive aspergillosis (2). However, conflicting data are reported regarding the role of Th17 responses in the host defense against A. fumigatus. A protective role for IL-17 is described by Werner et al. (3, 4) who showed that dectin-1−/− mice have decreased IL-17 production and subsequently reduced A. fumigatus clearance. In contrast, other studies show that IL-17 promotes inflammation and reduces resistance to the fungal infection (5, 6). Although these studies have investigated the role of IL-17 in mice, our group has previously reported that the human host response against A. fumigatus is mainly driven by the Th1 response, rather than by the Th17 response (7). Additionally, human mononuclear cells have been shown to express a primarily Th1-biased cytokine profile in response to stimulation with Aspergillus (8), and when Th2 responses are being suppressed an enhanced protective Th1 response develops in mice (9). Furthermore, mice are resistant to A. fumigatus when proper IL-12–dependent Th1 responses are induced (9, 10).
Even though various reports have focused on Th1 and Th17 responses against A. fumigatus, a role for the cytokine IL-22, which is a characteristic cytokine of the Th17 response, has not been addressed in human host responses. Recently, a protective role for IL-22 was demonstrated in the early host defense against A. fumigatus in a murine model of invasive pulmonary aspergillosis (11). Furthermore, the induction of IL-22 contributes to lung pathology in a murine model of allergic bronchopulmonary aspergillosis (12). In the host defense, IL-22 is primarily responsible for the induction of antimicrobial peptides (13) and is mainly produced by CD4+ T cells, NK cells, and NKT cells (13–19). Within the CD4+ population, IL-22 is mainly produced by cells of the Th17 lineage (13), but also by T cells that are specialized in the production of IL-22 and TNF-α, named Th22 cells (19).
Still, little is known about which cells mainly produce IL-17 (IL-17A), IFN-γ, and/or IL-22 and which recognition pathways and cytokines play a role in the induction of these cytokines in response to A. fumigatus in humans. In this study, we investigated the A. fumigatus–induced characteristic Th cytokines IL-17, IL-22, and IFN-γ in PBMCs to elucidate which human cells primarily produce these cytokines and which pattern recognition receptors (PRRs) and cytokines are involved in the induction of these cytokines in response to A. fumigatus.
Materials and Methods
Healthy volunteers and patients
Blood samples from healthy controls and patients were obtained after written informed consent. Three patients with homozygous Y238X mutations in exon 6 of CLEC7A gene (the gene encoding Dectin-1) provided blood samples. In these patients, diminished dectin-1 expression and failure to induce a cytokine response to β-glucan was demonstrated previously (20).
β-1,3-glucan immunofluorescence of HI Aspergillus
To determine β-1,3-glucan expression after heat inactivation of Aspergillus, the HI conidia and hyphae were incubated for 30 min with mouse anti–β-1,3-glucan (Biosupplies, Bundoora, VIC, Australia). Subsequently, the Aspergillus was washed and Abs directed to β-glucan that were bound to Aspergillus were secondarily stained by goat anti-mouse Alexa 488 (InvivoGen), according to the protocol supplied by the manufacturer. Immunofluorescence was observed at ×400 magnification using a Zeiss Axio imager M1 fluorescence microscope, equipped with an MRm camera (Carl Zeiss, Sliedrecht, The Netherlands).
Venous blood was drawn into 10 ml EDTA tubes, and PBMCs were isolated as described previously (21). In brief, blood was diluted in PBS (1:1) and fractions were separated by Ficoll (Ficoll-Paque Plus; GE healthcare, Zeist, The Netherlands) density gradient centrifugation. Cells were washed twice with PBS and resuspended in RPMI 1640 culture medium (Life Technologies/Invitrogen, Breda, The Netherlands) supplemented with 10 μg/ml gentamicin, 10 mM l-glutamine, and 10 mM pyruvate (Life Technologies). The cells were counted using a particle counter (Beckmann Coulter, Woerden, The Netherlands) and the concentration was adjusted to 5 × 106/ml.
CD4/CD56 cell depletion
To deplete the CD56 or CD4 cells from isolated PBMCs, cells were labeled using magnetic beads coated with anti-CD56 or anti-CD4 (MACS; Miltenyi Biotec, Bergisch Gladbach, Germany). Subsequently, the cells were depleted over a depletion column according to the protocol supplied by the manufacturer. As control for the isolation procedure, PBMCs were also washed over the columns without labeling with magnetic beads.
PBMCs were plated in 96-well round-bottom plates (Corning) at a concentration of 2.5 × 106/ml in a volume of 200 μl. They were either not stimulated or stimulated with 1 × 107/ml HI conidia or hyphae for 24 h or 7 d at 37°C and 5% CO2. All stimulations were performed in medium containing 10% human serum, which was obtained from a serum pool of healthy volunteers.
PRRs were inhibited in PBMCs by preincubation for 1 h with specific inhibitors. LPS derived from Bartonella quitana was used to block TLR4 at a final concentration of 20 ng/ml (23). B. quitana LPS was extracted and purified as described previously (24). Mouse anti-humanTLR2 (eBioscience, Halle-Zoersel, Belgium) and control mouse IgG1 (eBioscience), anti-human β2 integrin (anti-complement receptor 3, CR3), and control goat IgG (R&D Systems Minneapolis, MN) were used in a final concentration of 10 μg/ml. Laminarin was provided by Prof. David Williams of Tennessee University and was used in a final concentration of 50 ng/ml to inhibit dectin-1. Syk kinase inhibitor was purchased from Calbiochem (Merck, Darmstadt, Germany) and was used in a concentration of 50 nM. To check the blockade of the PRRs, PBMCs were stimulated with the TLR4 ligand LPS (10 ng/ml) from Escherichia coli serotype O55:B5 (Sigma-Aldrich, St. Louis, MO) or with the TLR2 ligand Pam3Cys (1 μg/ml) (EMC Microcollections, Tübingen, Germany). Inhibition of dectin-1 and Syk was validated by stimulation with HI Candida albicans (1 × 106/ml). After 24 h stimulation at 37°C and 5% CO2, IL-1β was measured by ELISA. All blockades resulted in a significant reduction of cytokine production (Supplemental Fig. 1A, 1B).
The cytokine pathways of IL-1, IL-23, and TNF-α were investigated using supplementation of the cultures with recombinant human (rh)IL-23 (50 ng/ml) and rhTNF-α (10 and 100 ng/ml) (R&D Systems). IL-1R signaling was blocked by its natural receptor antagonist (Ra) IL-1Ra (10 μg/ml) (Amgen, Thousand Oaks, CA), and IL-23 was blocked with mouse anti-human IL-23p19 (10 μg/ml) (R&D Systems). Soluble TNFR1 (sTNFR1; Enbrel) and human anti-human TNF-α (Humira) were provided by Renoud Marijnissen and Dr. Marije Koenders (Department of Rheumatology, Radboud University Nijmegen Medical Center, The Netherlands) and were used to block TNF-α in a final concentration of 100 μg/ml.
IL-17A, IL-22, IFN-γ, IL-1β, and IL-23 were measured using commercially available ELISAs (R&D Systems or eBioscience) according to the protocols supplied by the manufacturers.
Intracellular IL-17, IL-22, and IFN-γ flow cytometry
Following 7 d stimulation, PBMCs were stimulated 4–6 h with PMA (50 ng/ml) (Sigma-Aldrich), ionomycin (1 μg/ml) (Sigma-Aldrich), and GolgiPlug (BD Biosciences, Breda, The Netherlands) according to the protocols supplied by the manufacturers. Cells were stained extracellular using PE-Cy7–conjugated anti-CD4 (BD Biosciences), PE-Cy7–conjugated anti-CD8 (BioLegend, San Diego), or PE-Cy7–conjugated anti-CD56 (Beckman Coulter) Ab. Subsequently, the cells were fixed and permeabilized with Cytofix/Cytoperm solution (eBioscience) according to the protocol supplied by the manufacturer. Following permeabilization the cells were stained intracellularly with Alexa 647–conjugated anti–IL-17 (BD Biosciences), PE-conjugated anti–IL-22 (R&D Systems), and FITC-conjugated anti–IFN-γ (eBioscience) according to the protocols supplied by the manufacturers. The cells were measured on an FC500 flow cytometer (Beckman Coulter) and the data were analyzed using CXP analysis software v2.2 (Beckman Coulter).
Differences in IL-17, IL-22, and IFN-γ production and the percentage of CD4+ cells between the medium and Aspergillus-stimulated samples were analyzed with the Mann–Whitney U test. Data of stimulations with and without inhibitors of PRRs, cytokines, or cytokine inhibitors were subjected to statistical analysis with the Wilcoxon signed rank test. A p value of <0.05 was considered statistically significant. All experiments were performed at least twice and data represent cumulative results of all experiments performed and are presented as means ± SEM unless indicated otherwise. Data were analyzed using GraphPad Prism v5.0. The proportional Venn diagram was drawn using the eulerAPE application v2.0.3 (25, 26).
The proinflammatory adaptive cytokine response to A. fumigatus
We investigated the capacity of A. fumigatus conidia and hyphae to induce the cytokines IL-17A, IL-22, and IFN-γ. Stimulation with conidia and hyphae induced a significant production of the Th1 cytokine IFN-γ in human PBMCs, whereas IL-17A was induced in low amounts. Both conidia and hyphae also induced IL-22 in human PBMCs (Fig. 1A).
To determine which cell populations expand after stimulation with Aspergillus, we performed flow cytometry analysis. We initially focused on CD4+ cells, because Th cells are generally considered to be the main producers of IL-17A, IL-22, and IFN-γ. PBMCs that were not stimulated with conidia showed relatively small populations of IL-17A+ and IL-22+ CD4 cells (1.2% [SD 0.7%] and 1.7% [SD 0.9%], respectively), whereas the IFN-γ+ CD4 cell population was 10.7% (SD 3.9%). Stimulation with conidia induced a significant expansion of the IL-17A+, IL-22+, and IFN-γ+ CD4 T cell population (Fig. 1B). By gating on the Aspergillus-induced IL-17A+, IL-22+, and IFN-γ+ cells we observed that most IL-17A+ and IL-22+ cells were CD4+ cells. However, approximately half of the IFN-γ+ cells were negative for CD4 (Fig. 1C). Subsequently, extracellular staining with the NK cell marker CD56 and the cytotoxic T cell marker CD8 revealed that a significant proportion of the IFN-γ+ population was CD56+ and CD8+ (Fig. 1D). The remainder of the IL-17A+ cells were positive for CD8+ only. No CD56+IL-17A+ cells were found (Fig. 1D). However, despite the fact that a fraction of the IL-22+ cells was positive for CD8 or CD56, still ∼9% of the IL-22+ cells were negative for all of the investigated surface markers (Fig. 1D). Interestingly, we found that the populations of CD8+ and CD56+ cells that were IFN-γ+ were already present in unstimulated PBMCs and did not expand upon stimulation with A. fumigatus (Fig. 1E).
CD4 cells are the primary cellular source of Aspergillus-induced IL-17A, IL-22, and IFN-γ
To demonstrate that CD4+ cells are the major contributors to Aspergillus-induced IL-17A, IL-22, and IFN-γ, we depleted CD4+cells from PBMCs and compared this with normal PBMCs. Depletion of CD4+ cells resulted in a complete loss of IL-17A production. Furthermore, production of IL-22 and IFN-γ was also significantly reduced, in most donors, to undetectable levels (Fig. 1F). Because several reports indicate that NK cells can produce IL-22 (14, 16–18), we also investigated the contribution of CD56+ cells. Depletion of CD56+ cells had no significant effect on IL-17A, IL-22, and IFN-γ production (Fig. 1G).
IL-22 is not restricted to a specific Th cell subset
Because CD4+ cells were the major population that was IL-17A+, IL-22+, and IFN-γ+, we investigated the phenotypic diversity of these CD4+ T cells by focusing on the expression of single or multiple cytokines. Stimulation with HI conidia resulted in the induction of IL-17/IL-22 and IL-22/IFN-γ double-positive CD4 cells (Fig. 2A), whereas we found a relatively small percentage of IL-17/IFN-γ double-positive CD4+ T cells. Strikingly, the cytokine IL-22 was not specifically expressed in a certain subset, but it could be found in both the IL-17+ and IFN-γ+ T cell populations. Moreover, Aspergillus induces IL-17/IL-22/IFN-γ triple-positive CD4+ T cells (Fig. 2A). To determine whether the IL-22+ cells in our experiments match the phenotype of previously described Th22 cells, we assessed whether these cells coexpressed TNF-α, a characteristic of Th22 cells (15, 19, 27). Indeed, most IL-22+ cells coexpressed TNF-α (Fig. 2B). Additionally, we investigated how IL-1 and TNF-α signaling can influence the IL-17 and IL-22 populations. Blocking IL-1 resulted in significantly decreased IL-17+ and IL-22+ populations and a trend toward decreased IL-17/IL-22+ cells was observed, whereas blocking TNF-α with sTNFR1 only significantly decreased the number of IL-22+ cells (Fig. 2C).
Differential roles for TLR2 and TLR4 in Aspergillus-induced T cell responses
We investigated the role of the TLRs in the induction of the IL-17, IL-22, and IFN-γ responses to A. fumigatus, as TLR2 and TLR4 have been associated with the recognition of A. fumigatus (28–32). PBMCs were preincubated with B. quintana LPS to block TLR4, and afterward stimulated with conidia or hyphae. Preincubation with B. quintana LPS alone did not result in any cytokine induction in PBMCs (Supplemental Fig. 1C). Blockade of TLR4 resulted in significantly reduced IL-22 production in conidia-stimulated PBMCs and reduced IL-17 and IL-22 production in hyphae-stimulated PBMCs, whereas IFN-γ production was not affected by TLR4 blockade (Fig. 3A). Blocking TLR2 resulted in an upregulation of IL-17 and IL-22 production by conidia-stimulated PBMCs and a trend toward upregulation of IL-22 in hyphae-stimulated PBMCs (Fig. 3B).
Redundant role for dectin-1/Syk signaling in Aspergillus-induced T cell responses
Dectin-1 has been associated with the defense against Aspergillus in numerous reports (3, 4, 33–38). Gessner et al. (11) described that dectin-1–deficient mice lack the ability to induce IL-22 early in infection, which causes a decreased ability to induce antimicrobial peptides, resulting in an increased fungal burden and mortality. To see whether the dectin-1/Syk signaling pathway is involved in the induction of IL-22 production by human T cells, PBMCs were stimulated with conidia or hyphae in the presence of laminarin, a dectin-1 inhibitor. Blocking dectin-1 with laminarin did not result in any significant effect on hyphae or conidia-induced IL-17A, IL-22, or IFN-γ (Fig. 4A). However, there was one donor with unusual high IL-17 production in response to hyphae, which we excluded from analysis. Note that the IL-17 production in this donor was almost completely blocked with laminarin.
We investigated whether Aspergillus conidia and hyphae expressed β-1,3-glucan on the surface by immunofluorescence staining. We found that β-1,3-glucan was only minimally present on a few conidia, which is in line with reports demonstrating that resting conidia do not express β-glucan (34). In contrast, hyphae abundantly exposed β-glucans (Fig. 4B). To confirm that dectin-1/Syk signaling does not play a significant role in the induction of IL-17A, IL-22, or IFN-γ by Aspergillus, PBMCs of three patients with the homozygous mutation Y238X in dectin-1 (lacking dectin-1 expression on the membrane of the cell) (20) were stimulated with conidia and hyphae. PBMCs from these patients demonstrated either a similar or higher cytokine response to Aspergillus compared with PBMCs from a healthy control (Fig. 4C).
Syk kinase is the downstream signaling kinase of dectin-1 and other C-type lectin receptors (34). Although hyphae-induced IL-17 production tended to decrease upon blocking of dectin-1, inhibition of Syk kinase demonstrated no difference in IL-17A, IL-22, or IFN-γ production by Aspergillus-stimulated PBMCs (Fig. 4D).
The role for CR3 for Aspergillus-induced T cell responses
Recently, CR3 was demonstrated to recognize β-glucan, which leads to downstream signaling and activation of granulocytes (39). Moreover, CR3 has also been associated with phagocytosis of pentraxin 3–opsonized Aspergillus (40). We investigated whether CR3 was involved in the induction of IL-17A, IL-22, and IFN-γ by A. fumigatus. Blockade of CR3 resulted in a significant reduction of IL-17A, IL-22, and IFN-γ responses induced by conidia. In response to hyphal stimulation, only IL-17A and IL-22 were significantly inhibited by blocking of CR3 (Fig. 4D). Because the primary role of CR3 is recognition of C3-opsonized structures, we investigated whether active serum complement was required for Aspergillus-induced IL-17A, IL-22, and IFN-γ. When human serum was HI prior to stimulation, no differences in IL-17A, IL-22, and IFN-γ induction were observed (Fig. 4E).
The role of IL-1 and IL-23 in A. fumigatus–induced IL-17 and IL-22
The predominant cytokine produced by human PBMCs in response to Aspergillus was IL-22, whereas the IL-17A production was relatively low (Fig. 1A). To elucidate the reason for this low IL-17A release by Aspergillus-stimulated PBMCs, we focused on the cytokines that regulate the Th17 response. Both IL-1β and IL-23 have been associated with the induction and maintenance of the Th17 response in humans (41). Additionally, a critical role for IL-23 in the induction of IL-22 was observed in the early response against A. fumigatus (11). Interestingly, IL-23 induction by PBMCs stimulated with conidia and hyphae was below the detection limit of the ELISA, whereas IL-1β production was detectable (Fig. 5A). To investigate whether the absence of IL-23 production is responsible for the low production of IL-17 in response to Aspergillus, IL-23 was blocked with an anti-human IL-23p19 Ab. Although not statistically significant, compared with stimulation with the isotype control, IL-17A production was reduced to undetectable levels, whereas IL-22 production was not affected (Fig. 5B). Addition of IL-23 significantly increased IL-17A but not IL-22 production (Fig. 5C). Blocking IL-1 signaling with IL-1Ra reduced both IL-17A and IL-22 induction significantly (Fig. 5D). These results suggest that IL-17A production requires both IL-1β and IL-23, whereas IL-22 production requires only IL-1β.
Role of TNF-α in A. fumigatus–induced IL-17 and IL-22
Although IL-1β and IL-23 play key roles in the induction of Th responses by A. fumigatus, it cannot be excluded that other cytokines also play a role in the induction of the cytokines IL-17 and IL-22. To investigate the role of TNF-α in A. fumigatus–induced IL-17 and IL-22, we stimulated PBMCs with HI conidia in the presence of rTNF-α. Both IL-17 and IL-22 production were dose dependently increased in the presence of TNF-α when compared with stimulation with Aspergillus alone (Fig. 6A). To further investigate the role of TNF-α we stimulated PBMCs with HI conidia in the presence of sTNFR1 (Enbrel) or human anti-human TNF-α (Humira), two drugs that are used to block TNF-α signaling in IL-17 related diseases like rheumatoid arthritis (42). Both blockers significantly reduced the IL-17 and IL-22 response (Fig. 6B).
In this study, we demonstrate that CD4+ T cells are the main producers of IL-17, IL-22, and IFN-γ in human PBMCs upon Aspergillus stimulation. Interestingly, CD4 Th cells that are capable of producing IL-22 after Aspergillus stimulation are not a distinct characteristic subset, but they can have a Th1, Th17, or Th22 signature. Similar to IL-17, the production of IL-22 induced by Aspergillus is dependent on TLR pathways. Whereas the TLR4 pathway contributes to the production of these cytokines, TLR2 has an inhibitory effect on Aspergillus-induced IL-17 and IL-22 production. In contrast, the TLR2 and TLR4 pathways did not modulate IFN-γ production by human PBMCs that are stimulated with Aspergillus. An unanticipated finding was that the blockade of the dectin-1/Syk pathway or absence of the dectin-1 receptor did not significantly affect IL-17, IL-22, or IFN-γ production induced by Aspergillus. In the present study, CR3 is the only receptor that is important for induction of all three cytokines (IL-17, IL-22 and IFN-γ) and therefore the CR3 pathway could specifically play an important role in the host defense against Aspergillus. Furthermore, we provided evidence that IL-17 and IL-22 responses induced by Aspergillus are dependent on both the IL-1 and TNF pathway.
Using flow cytometric analysis of PBMCs and depletion of CD4 or CD56 cell subsets, we determined that the cytokines IL-17, IL-22, and IFN-γ are primarily produced by CD4+ T cells. Within the CD4+ T cell populations diverse intracellular cytokine expression was detected. We detected IL-17 single-positive and IL-17/IL-22 double-positive CD4 cells that match the classically described Th17 cells (13) and IFN-γ+ CD4 cells that match the classical Th1 type. Half of the IFN-γ+ cells were CD4+ cells, whereas the rest of the IFN-γ+ cells were negative for this Th cell marker. We were able to demonstrate that a significant number of NK cells express IFN-γ, which is in line with a previous report that demonstrated that NK cells play an important role in the antifungal response to A. fumigatus by releasing IFN-γ (43). However, in contrast to this, depletion of CD56+ cells did not alter Aspergillus-induced IFN-γ responses. Furthermore, the IFN-γ+CD56+ cells did not expand upon stimulation with Aspergillus. These data suggest that, in the setting of PBMCs, NK cells do not contribute to the IFN-γ response to Aspergillus. We found that after stimulation with Aspergillus IL-17/IFN-γ double-positive CD4+ cells were present in low numbers. Therefore, our results demonstrate that the polarization of Th responses induced by Aspergillus in vitro is different from that induced by the commensal fungus C. albicans (44). It was demonstrated that C. albicans induces a T cell polarization with high numbers of IL-17/IFN-γ double-positive Th cells. In this study, we demonstrate that Aspergillus stimulation leads to low numbers of this T cell subset. This is of interest, as these double-positive cells have been linked to autoimmunity (44).
We also observed IL-22 single-positive cells that might fit the description of Th22 cells (15), as these cells coexpressed TNF-α. Additionally, large numbers of IFN-γ/IL-22 double-positive CD4 cells and even IL-17/IL-22/IFN-γ triple-positive CD4 cells are present. This is to our knowledge the first study that demonstrates that Aspergillus can induce single IL-22+ and IFN-γ/IL-22 double-positive cells. It remains to be elucidated whether these specific subsets contribute to pathology or protection. A study in mice revealed an important role for IL-22 in the early defense against Aspergillus (11), whereas another study demonstrates contribution to pathology in an allergic aspergillosis model (12). In the present study, we observed that the proinflammatory adaptive cytokine response to A. fumigatus is dominated by IL-22, and the primary cellular sources were CD4+ cells. In contrast to the study performed by Gessner et al. (11), our model revealed that IL-22 production by Aspergillus-stimulated PBMCs was independent of IL-23. We found that the innate cytokines IL-1 and TNF-α played a crucial role in the induction of IL-22 and the induction of the IL-22+ CD4 subset, which is in agreement with earlier studies (45). In line with our results, it was previously demonstrated that the IL-22 response also dominated over the IL-17 response in patients with pulmonary tuberculosis (46). Moreover, IL-22 was shown to be important in the defense against multiple pulmonary pathogens including Mycobacterium tuberculosis (14, 46, 47), Klebsiella pneumoniae (48), and influenza A virus (16). Because the host defense against these pulmonary pathogens seems to strongly rely on IL-22, we hypothesize that IL-22 might play a pivotal role in the human anti-Aspergillus pulmonary host defense.
Although TLR2 and TLR4 play an important role in innate immune responses against A. fumigatus (28–32), the role of these receptors in the induction of the adaptive immune responses such as the IL-22 response against Aspergillus remains to be established. In this study, we demonstrate that TLR4 plays a role in the induction of IL-17 and IL-22 in response to Aspergillus. In contrast, blockade of TLR2 resulted in higher IL-17 and IL-22 production. Therefore, this might indicate that TLR2 plays a role in an inhibitory pathway for the induction of IL-17 and IL-22. These findings are in line with several studies that report an anti-inflammatory role for TLR2 through Th2 skewing (49, 50), the induction of anti-inflammatory cytokines by TLR2, and the TLR2-dependent induction of regulatory T cells (51). Interestingly, TLR2 is a negative regulator of the Th17 response in a murine pulmonary infection model with Paracoccidioides brasiliensis (52), supporting a general role of TLR2 as a negative regulator of Th17 polarization of fungal infections.
The fact that the β-glucan receptor dectin-1 and its downstream kinase Syk play a redundant role in the induction of IL-17, IL-22, and IFN-γ by hyphae, which abundantly express β-1,3-glucan, is rather unanticipated, as the dectin-1 pathway plays an important role in the host defense against invasive aspergillosis (35, 53). In line with our results, it has been demonstrated that IL-17 production by PBMCs with the dectin-1 Tyr238X polymorphism is not different from PBMCs that do not have this single nucleotide polymorphism (35). Notably, this single nucleotide polymorphism influences innate proinflammatory cytokines produced by PBMCs in response to A. fumigatus (35, 53). In mice, dectin-1 deficiency results in an increased susceptibility to A. fumigatus (3), which has been linked to disability to induce protective cytokines such as IL-17 (3) and IL-22 (11). Although it is evident that dectin-1 plays a role in the induction of IL-17 and IL-22 in mice and that it plays a role in the innate host defense against invasive aspergillosis in humans, our studies suggest that IL-17, IL-22, and IFN-γ production by CD4+ T cells in the human host response to Aspergillus conidia is not predominantly mediated by dectin-1.
CR3 is a β2 integrin (CD11b/CD18) that is expressed by monocytes and neutrophils (54, 55). This receptor can recognize self-molecules such as complement, but it can also recognize pathogen-associated molecular patterns from pathogens such as LPS from E. coli (56). CR3 plays a role in phagocytosis and induction of cytokine responses (57). In this study, we show that CR3 is involved in modulating Th cytokine responses induced by A. fumigatus. To date, the modulation of Th17 and Th1 responses induced by fungi has not been linked to CR3, and future studies are required to further characterize its role in fungal infection. Interestingly, CR3 has been shown to bind β-glucan, which can have modulatory effects on the immune response, suggesting that fungal components can modulate proinflammatory Th responses (Th1 and Th17) through CR3 (58). Immunofluorescent staining of β-glucan revealed a very low expression on Aspergillus conidia, and therefore β-glucan seems to be unlikely the main CR3 ligand in the recognition of Aspergillus.
In this study, we compared the IL-1β and IL-23 cytokine profiles produced by PBMCs stimulated with A. fumigatus and related this to the induction of IL-17 and IL-22. Aspergillus induced relatively low IL-17 levels, which was rather surprising because A. fumigatus induced a significant number of IL-17+ CD4 cells. These observations are in line with a previous report, which demonstrates that A. fumigatus is a poor inducer of IL-17 and is even capable of inhibiting the IL-17 response by interfering with the tryptophan metabolism (7). One possible explanation for the low IL-17 production could be that Aspergillus does not induce a significant IL-23 response, as IL-23 production was undetectable in A. fumigatus stimulations. Interestingly, supplementation of IL-23 to the Aspergillus-stimulated PBMCs boosted the IL-17 response, which is in line with an earlier report that showed that IL-23 can augment IL-17 responses in Aspergillus-infected mice (3). Although, Aspergillus did not induce IL-23 in PBMCs, it was found earlier that dendritic cells induce high levels of IL-23 upon stimulation with Aspergillus and that these dendritic cells can polarize toward both Th1 and Th17 responses (37, 59). However, when we used monocyte-derived dendritic cells instead of PBMCs in our experimental setup we were not able to detect any IL-23 or IL-1β in response to Aspergillus (data not shown). We further demonstrate that Aspergillus-induced IL-17 and IL-22 responses were dependent on IL-1. Notably, although IL-1β was present and the Th17 response was almost completely dependent on the IL-1 pathway, Aspergillus did not abundantly induce IL-17/IFN-γ+cells. This is in contrast with a previous study that suggests that IL-1β is responsible for the development of IL-17+IFN-γ+ Th cells in response to the fungal pathogen C. albicans (44). These data provide evidence that next to IL-1, another pathway, such as the IL-23 pathway, must be triggered to induce IL-17+IFN-γ+ cells.
Another striking observation is the role of TNF in modulating proinflammatory Th17 responses induced by A. fumigatus. Blocking TNF lowered the IL-17 production in response to A. fumigatus; moreover, the effects were most prominent on IL-22 production. These observations are in line with previous reports, which demonstrated that IL-22 production can be triggered by TNF-α (15, 60, 61), and is in line with the observation that anti–TNF-α treatment can reduce IL-17 levels in patients (42). So far, the role of TNF-α in pathogen-induced IL-22 production has not been described. This observation could be highly relevant given the widespread use of anti-TNF therapy. Notably, the use of anti-TNF (62–64) and polymorphisms in TNFR1 or decreased TNFR1 mRNA expression (64, 65) have been linked to increased susceptibility to invasive aspergillosis, emphasizing the importance to identify the impact of TNF blockade on anti-Aspergillus host responses, such as lowering IL-22 production in response to A. fumigatus as described in the present study.
In conclusion, we demonstrate that the Th cytokines IL-17, IL-22, and IFN-γ in response to A. fumigatus are primarily produced by CD4 T cells, and that these cytokines are not limited to a specific subset but can be produced by a variety of polarized Th cell subsets. TLR4 and CR3 and the cytokines IL-1, IL-23, and TNF-α are of specific importance for Aspergillus-induced IL-17 and IL-22, whereas we observed that the dectin-1/Syk pathway is redundant for this response. Collectively, these findings contribute to a better understanding of the human adaptive host defense against A. fumigatus, and we provide new knowledge that contributes to the development of targeted adjunctive immunotherapeutic regimens in patients with invasive aspergillosis.
F.L.v.d.V. was supported by a Veni grant of the Netherlands Organization for Scientific Research, as well as by a grant from the Nijmegen Center for Molecular Life Sciences. M.G.N. was supported by a Vici grant of the Netherlands Organization for Scientific Research.
The online version of this article contains supplemental material.
Abbreviations used in this article:
pattern recognition receptor
The authors have no financial conflicts of interest.