Endotoxin-free buffers and reagents were scrupulously used in all experiments. Curdlan, Zymosan, and Laminarin were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Aspergillus fumigatus β-glucan preparations were isolated as previously described (14). To ensure that all glucan preparations were free of endotoxin prior to use in culture, Curdlan, Zymosan, and A. fumigatus glucans were vigorously washed 10 times with distilled physiological saline, incubated rotating overnight with polymyxin B (Sigma, St. Louis, MO, USA) at 4°C, then vigorously washed again with distilled physiological saline. The final preparations were assayed for endotoxin with the limulus amebocyte lysate method using Pyrosate Rapid Endotoxin Detection Kit (Associates of Cape Cod, East Falmouth, MA, USA) and found to consistently contain less than 0.25 EU/ml. Glucans were pulse sonicated 10 times using a Branson digital sonifier (VWR Scientific, Radnor, PA, USA) at 35% amplitude immediately before addition to the cultures. The Erk 1/2 inhibitor (PD98059), SYK inhibitors (Piceatannol and R406), and NF-κB inhibitor (Bay11-7085) were all obtained from Calbiochem, Inc. (San Diego, CA, USA). AP-1 inhibitor (SR 11302) was from R&D Systems, Inc. (Minneapolis, MN, USA). Caspase inhibitors (Ac-YVAD-CMK and Ac-YVAD-CHO) were purchased from Cayman Chemical (Ann Arbor, MI, USA), VX-765 was from AdooQ Bioscience (Irvine, CA, USA), and Rapamycin was purchased from Selleck Chemicals (Houston, TX, USA). KCl and Adenosine 5′-triphosphate (ATP) were from Sigma and oxidized ATP (oxATP) and MCC950 were from EMD Millipore (Billerica, MA, USA). Phosphorothioate-protected CpG oligonucleotide (5′-TCGTCGTTTTGTCGTTTTGTCGTT-3′) ODN 2006 and A. fumigatus oligonucleotides; AF1 (5′-TCGTCGTTGTCGTCGTC-3′) and AF2 (5′-TCGTCGTTGTCGTC-3′) were commercially synthesized by Integrated DNA Technologies, Inc. For most of the experiments, we used CpG ODN 2006 (CpG) unless otherwise specified. Antibodies recognizing the inflammasome components caspase-1 and IL-1β were purchased from Santa Cruz Inc. (Dallas, TX, USA), while NLRP3, ASC, and the mTOR pathway antibodies; mTOR, phospho-mTOR, S6K, and phospho-S6K were from Cell Signaling Technology, Inc. (Danvers, MA, USA). The neutralizing antibody for Dectin-1 was purchased from AbD Serotec (Raleigh, NC, USA) and Isotype control antibody was from R&D Systems, Inc. (Minneapolis, MN, USA). All other reagents were obtained from Sigma-Aldrich (St. Louis, MO, USA) unless specified otherwise.

All methods were carried out in accordance with relevant guidelines and regulations. Human B-lymphocytes were isolated as previously described (7). Briefly, B-lymphocytes were isolated from acid citrate dextrose anticoagulated blood obtained from de-identified healthy volunteer platelet donors in accordance with the current regulations by the AABB and the US Food and Drug Administration (15) and by the Mayo Clinic Institutional Review Board (IRB). Ethics committee approval was not required according to the local and national and guidelines. Cells were isolated using RossetteSep B-cell enrichment cocktail according to the manufacturer’s protocol (StemCell Technologies, Vancouver, BC, Canada). The enriched B-lymphocyte population was repeatedly observed to contain an average of 93.1% ± 2.3% B-lymphocytes. Sorting of human CD19+ B-lymphocytes into naïve (CD27−) and memory (CD27+) B-lymphocytes was performed using mouse anti-human CD27-APC-Vio770 and mouse anti-human CD19-PE antibodies from Miltenyi Biotec (San Diego, CA, USA), on a FACSAria II SORP flow cytometer running FACSDiva v 6.1.3 software.

Total cellular RNA was isolated from B-lymphocytes using RNeasy Plus Universal Mini Kit (Qiagen) according to the manufacturer’s instructions. For cDNA synthesis, total RNA concentrations and purity were determined using a Nanodrop ND-1000 spectrophotometer, and 1 µg RNA was used in a 20 µl reaction mixture using a Verso cDNA Synthesis Kit (Thermo Scientific). Quantitative real-time PCR was performed in 10 µl reaction in a 96-well plate using 2 µl of diluted cDNA with SYBR Premix Ex Taq (Clontech Laboratories, Inc.) on a ViiA7 Real-Time PCR Detection System (Life Technologies). The data were analyzed with ViiA7 software, version 1.2.4 (Life Technologies). Relative transcript expression of IL-1β was determined using the comparative Ct method, and was normalized to GAPDH transcripts of the same cDNA samples. The results were expressed as % of the target gene relative to that of GAPDH and plotted as the mean ± SEM. Experimental reproducibility was confirmed using three biological replicates from independent experiments which used B-lymphocytes from three different donors. The primers used for amplification were: IL-1β, 5′-ATGCACCTGTACGATCACTG-3′ and 5′-ACAAAGGACATGGAGAACACC-3′; GAPDH, 5′-ACATCGCTCAGACACCATG-3′ and 5′-TGTAGTTGAGGTCAATGAAGGG-3′.

B-lymphocytes (4 × 10⁵ cells/well in 96-well plates) were cultured with 1 µg/ml of CpG (ODN 2006), or with 1 µg/ml AF1, 1 µg/ml AF2, 200 µg/ml of Curdlan (Curd), 200 µg/ml Zymosan (Zym), 200 µg/ml or 200 µg/mlA. fumigatus β-glucan (AspG) in culture medium for 24 h for IL-1β and other cytokines and 5 days for IgM unless otherwise indicated. Cell supernatants were then analyzed for IL-1β, TNFα, IL-6, MMP7, and IgM production using human DuoSet ELISA kit for MMP-7 from R&D Systems, Inc. (Minneapolis, MN, USA), BD OptEIA human ELISA kit for IL-1β and Ready-Set-Go ELISA kit for TNFα, IL-6, and IgM from Thermo Fisher (Rochester, NY, USA). ELISAs were performed according to each manufacturer’s instructions. Inhibitors [50 and 100 µM Caspase-1 inhibitors (VX-765 and YVAD.CMK), 50 mM KCl, 50 µM oxATP, 100 µM MCC950, 40 µM Piceatannol, 5 µM R406, 10 µM Bay11-7085, 10 µM SR11302 or 1 mg/ml Laminarin] and neutralizing antibodies (5 µg/ml anti-Dectin-1 IgG or mouse IgG) were incubated with B-lymphocytes for 1 h prior to addition of CpG or β-glucan preparations. Inhibitors were dissolved in dimethyl sulfoxide unless otherwise indicated. Solvents were added to all treatments to assure that solvents concentrations are the same in all treatment conditions.

Caspase-1 activity was determined by a Caspase-1 Fluorometric Assay (BioVision, Mountain View, CA, USA), according to the manufacturer’s instruction. Briefly, 10⁷ B-lymphocytes were stimulated with Curdlan and CpG in the presence of the caspase inhibitor YVAD-CMK, incubated at 37°C for 24 h and lysed with cell lysis buffer on ice for 10 min. Then, equal volumes of 2× reaction buffer and YVAD-AFC were added to the 200 µg of lysates in a 96-well plate and incubated for 2 h at 37°C. The samples were read on a Softmax microplate reader (Molecular Devices, Sunnyvale, CA, USA) with a 400 nm excitation filter and 505 nm emission filter. The caspase-1 activity is expressed relative to the unstimulated control.

Cell viability was assessed using the XTT Cell Proliferation Kit II (Roche Molecular Biochemicals, Indianapolis, IN, USA) according to the manufacturer’s protocol. This assay measures the conversion of sodium-3′-[1-(phenylaminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzenesulfonic acid hydrate (XTT) to a formazan dye through electron coupling in metabolically active mitochondria using the coupling reagent N-methyldibenzopyrazine methyl sulfate. Only metabolically active cells are capable of mediating this reaction, which is detected by absorbance of the dye at 450–500 nm. Briefly, 50 µl of the XTT labeling mixture was added to the 100 µl of growth medium containing B-lymphocytes and different concentrations of the inhibitors. The XTT labeling mixture was added in parallel samples 24 h after the addition of the various inhibitors. A set of blanks was also included that did not contain cells and was treated identically as the normal samples. In addition, a set of solvent controls was also included for each inhibitor. Absorbance was measured at 6 h after XTT addition. All treatments were performed in three replicates. Only inhibitor concentrations that elicited less than 20% net toxicity were used in these assays.

Total cellular proteins were obtained from B-lymphocytes following the described culture conditions. Briefly, the cells were washed with cold PBS twice and lysed in RIPA buffer (50 mM Tris–HCl pH 7.4, 15 mM NaCl, 0.25% deoxycholic acid, 1% NP-40) freshly supplemented with 1 mM phenylmethylsulfonyl fluoride, mammalian protease inhibitor mixture, 10 mM Na Fluoride, and 1 mM Na Orthovanadate. Cells were kept on ice for 15 min and then the lysates were centrifuged at 12,000 × g for 10 min at 4°C. The resultant soluble supernatant contained total cellular protein. Protein concentrations were determined with the Bio-Rad protein assay (Hercules, CA, USA) using BSA as the standard. Equal amounts of total cellular proteins were separated on SDS-10% polyacrylamide gels with Precision Plus Protein Dual Color Standards (Bio-Rad; Hercules, CA, USA) being used as the molecular weight standards. Proteins were transferred to Immobilon-P membranes (Millipore, Bedford, MA, USA). Membranes were then blocked at room temperature for 30 min with 1% BSA/TBST (TBS, pH 7.4, 1% BSA, 0.1% Tween 20) and incubated overnight at 4°C in a blocking solution containing primary antibodies at the appropriate dilutions. After washing with TBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h. Immunoreactive bands were detected with SuperSignal West PicoChemiluminescent Substrate from Thermo Scientific (Rockford, IL, USA). Actin was used as loading control for the cell lysates and a non-specific protein for the cell supernatant.

All data are presented as the mean ± SEM, from at least three independent experiments from different biological donors, unless otherwise specified. An experimental run included at least three different donors. The data were first analyzed using one-way ANOVA, the differences between the individual groups were compared using multiple comparisons post-test unless otherwise indicated. Statistical analysis was performed using GraphPad Prism Version 5 (GraphPad Software, La Jolla, CA, USA).

B-lymphocytes respond to fungal β-glucans and bacterial DNA in a T-cell independent manner by releasing cytokines and chemokines important for the acute inflammatory response (8, 16). We have previously shown that β-glucan-stimulated peripheral human B-lymphocytes are a source of IL-8, IL-6, and TNFα (7). Additionally, supernatant from β-glucan-stimulated B-lymphocytes was able to significantly increase neutrophil chemotaxis suggesting the contribution of B-lymphocytes in the acute inflammatory process (7). To further understand the role of B-lymphocytes in fungal defense and since IL-1β has been shown to participate in fungal clearance and neutrophil recruitment (17, 18), freshly isolated peripheral B-lymphocytes from normal donors were stimulated with either β-glucan (Curdlan) or CpG (CpG ODN 2006) and assessed for IL-1β expression. RNA levels of IL-1β were significantly elevated in β-glucan-stimulated cells but not in those stimulated with CpG (Figure 1A). Protein levels of IL-1β were next confirmed in the cell supernatants by ELISA. As expected, β-glucan-stimulated cells secreted significant amounts of active IL-1β in a dose and time responsive manner (Figures 1B,C). The addition of β-glucan to CpG did not enhance IL-1β stimulation (Figure 1D). Furthermore, IL-1β secretion was not restricted to curdlan β-glucans as other β-glucan preparation such as zymosan and β-glucan from Aspergillus were also potent inducers of IL-1β (Figure 1E). Similarly, failure to induce IL-1β secretion was not restricted to CpG ODN 2006, since two different CpG motifs present in A. fumigatus DNA (AF1 and AF2) (4) also failed to stimulate IL-1β secretion (Figure 1F). The differential response of B-lymphocytes to β-glucan and CpG was intriguing, though similarly observed by our group in the setting of IL-8 and IgM (7). Our published studies demonstrated that β-glucan-induced IL-8 secretion had no effect on IgM while CpG stimulated IgM secretion did not participate in IL-8 secretion (7).

The NLRP3 inflammasome is crucial for the processing of biologically inactive IL-1β (pro-IL-1β) into the mature and biologically active form (IL-1β). It was therefore important for us to determine its role in β-glucan-mediated IL-1β secretion. Protein expression of NLRP3, ASC, caspase-1, and pro-IL-1β as well as caspase-1 activity were therefore assessed and found to increase significantly upon β-glucan stimulation (Figures 2A,C). CpG stimulation did not induce pro-IL-1β, a required step needed for mature IL-1β production (Figure 2B). However, despite the lack of increase in pro-IL-1β, CpG also resulted in activation of the NLRP3 inflammasome (Figures 2B,C).

To further demonstrate that Caspase-1 and NLRP3 were indeed important in β-glucan-mediated IL-1β secretion and since primary human B-lymphocytes are not suitable for lentivirus infection or transfection with other forms of interfering RNA, IL-1β was measured in the cell supernatant of stimulated cells in the presence of different Caspase-1 and NLRP3 inhibitors. As shown in Figures 2D–F, the use of the caspase-1 inhibitors VX765 and YVAD-CMK, KCL, oxATP, and the specific NLRP3 inhibitor, MCC950 (19), resulted in significant decreases of IL-1β secretion, confirming the participation of the canonical NLRP3 inflammasome pathway in β-glucan-mediated IL-1β secretion. Absence of cell toxicity was confirmed for all the inhibitors as shown in Supplement S1 in Supplementary Material.

Since prior data suggested that in NLRP3 deficient mice IgM secretion and antibody production are impaired (8, 20–22), we hypothesized that NLRP3 may regulate IgM secretion. We and others have additionally shown that peripheral B-lymphocytes secrete IgM in response to CpG (ODN 2006) (7, 23) (Supplement S2 in Supplementary Material). To better understand the role of NLRP3 activation in CpG-stimulated B-lymphocytes we further investigated if IgM was also triggered in response to AF1 and AF2. As shown in Figure 3A IgM was similarly induced by all forms of CpG tested. Next, we evaluated IgM levels after CpG stimulation in the presence of different concentrations of MCC950. Interestingly, the levels of IgM decreased in the presence of MCC950 in a dose-dependent manner (Figure 3B; Supplement S3A in Supplementary Material). To ensure that this effect was specific to inhibition of the NLRP3 inflammasome, IL-6, MMP-7, and TNFα were also assessed in the presence of MCC950. As shown in Figure 3C, none of the other cytokines and metalloproteases were affected by the presence of the inhibitor further supporting the specific role of the NLRP3 inflammasome in IgM secretion. Additionally, IgM release was also impaired in the presence of KCL, oxATP, and caspase inhibitors (Figures 3D,E; Supplement S3A in Supplementary Material), all well-known agents that affect the functioning of the NLRP3 inflammasome.

Since the inflammasome was activated by both β-glucan and CpG and both are known to signal through very different PRRs, we next explored the specific signaling pathways that lead to IL-1β and IgM secretion upon β-glucan and CpG, respectively.

β-Glucans are known to signal predominantly through the C-lectin receptor Dectin-1 (6, 7, 24, 25) while CpG is the natural ligand for TLR9 (4, 26). Hence, we first explored the contribution of Dectin-1 in β-glucan-induced IL-1β signaling. B-lymphocytes were pre-incubated with either laminarin (a soluble β-glucan known to bind to Dectin-1 acting as a competitive inhibitor) or a specific Dectin-1 blocking antibody prior to β-glucan stimulation. IL-1β levels were significantly reduced in the presence of both, laminarin and the Dectin-1 antibody, confirming the role of Dectin-1 in β-glucan-mediated IL-1β secretion (Figures 4A,B). β-glucan-mediated IL-1β was also dependent on Syk and the transcription factors NF-κB and AP-1 (Figures 4C,D).

TLR9 is the main receptor for CpG motifs found in bacterial and fungal DNA (4, 26) and it is known that stimulation by CpG also involves mTOR (4, 6, 27–29). In contrast, β-glucan stimulation does not seem to require mTOR (6). Thus, to additionally understand the potential involvement of mTOR activation in IgM and IL-1β regulation, phosphorylation of mTOR and other mTOR-related proteins were assessed after cells were stimulated for different period of time with CpG or β-glucan. mTOR and other mTOR-related proteins (S6K, S6, and 4EBP1) tested were phosphorylated upon CpG but not after β-glucan stimulation with the exception of 4EBP1 which seemed to be phosphorylated by both (Figure 5A; Supplement S3B in Supplementary Material). β-glucan activation of 4EBP1 was inhibited in the presence of PD98059, a specific ERK1/2 inhibitor, and not by Rapamycin, a well-known mTOR inhibitor, suggesting that 4EBP1 activation, in β-glucan-stimulated cells, was mTOR-independent and required ERK1/2 (Supplement S4 in Supplementary Material), consistent with prior observations (30).

As our data suggested that NLRP3 regulates IgM via mTOR, we measured IgM levels after CpG-stimulation in the presence and absence of Rapamycin. The use of Rapamycin resulted in significant reduction of IgM thus confirming the role of mTOR signaling in IgM regulation (Figure 5B; Supplement S3A in Supplementary Material). The levels of IgM in the presence of Rapamycin were comparably reduced to the levels of IgM after MCC950. In contrast, β-glucan stimulation of IL-1β, while reduced in the presence of MCC950, was not affected by the use of Rapamycin confirming that this is an mTOR-independent process (Figure 5C). Interestingly, Rapamycin inhibition of the NLRP3 inflammasome seemed to affect Caspase-1 activation as shown in Figure 5D. Similar observations were made for MCC950, consistent with prior published observations (31).

Peripheral B-lymphocytes can be divided into naïve and memory cells. In order to further understand the subtype of B-lymphocytes responsible for the secretion of IL-1β and IgM, and since it is clearly established that naïve and memory cells produce different cytokine patterns upon stimulation (32, 33). We isolated B-lymphocytes by negative selection and then sorted them into CD20⁺CD27⁻ (naïve) and CD20⁺CD27⁺ (memory) cells. The two different subtypes of B-lymphocytes were independently stimulated with either β-glucan or CpG. IL-1β and IgM levels in the cell supernatant were then measured by ELISA. As Figure 6A shows, we observed that IL-1β was mostly secreted by memory cells whereas IgM production was greater from the sorted memory cells and significantly higher than naïve cells (p < 0.001), but was still seen at significant levels in the naïve cells upon CpG stimulation. Co-stimulation of the BCR receptor resulted in an increase in IgM secretion but it did not have a significant effect on IL-1β secretion (Figure 6B).