Silica induces NLRP3 inflammasome activation in human lung epithelial cells
© Peeters et al.; licensee BioMed Central Ltd. 2013
Received: 14 September 2012
Accepted: 29 January 2013
Published: 12 February 2013
Skip to main content
© Peeters et al.; licensee BioMed Central Ltd. 2013
Received: 14 September 2012
Accepted: 29 January 2013
Published: 12 February 2013
In myeloid cells the inflammasome plays a crucial role in innate immune defenses against pathogen- and danger-associated patterns such as crystalline silica. Respirable mineral particles impinge upon the lung epithelium causing irreversible damage, sustained inflammation and silicosis. In this study we investigated lung epithelial cells as a target for silica-induced inflammasome activation.
A human bronchial epithelial cell line (BEAS-2B) and primary normal human bronchial epithelial cells (NHBE) were exposed to toxic but nonlethal doses of crystalline silica over time to perform functional characterization of NLRP3, caspase-1, IL-1β, bFGF and HMGB1. Quantitative RT-PCR, caspase-1 enzyme activity assay, Western blot techniques, cytokine-specific ELISA and fibroblast (MRC-5 cells) proliferation assays were performed.
We were able to show transcriptional and translational upregulation of the components of the NLRP3 intracellular platform, as well as activation of caspase-1. NLRP3 activation led to maturation of pro-IL-1β to secreted IL-1β, and a significant increase in the unconventional release of the alarmins bFGF and HMGB1. Moreover, release of bFGF and HMGB1 was shown to be dependent on particle uptake. Small interfering RNA experiments using siNLRP3 revealed the pivotal role of the inflammasome in diminished release of pro-inflammatory cytokines, danger molecules and growth factors, and fibroblast proliferation.
Our novel data indicate the presence and functional activation of the NLRP3 inflammasome by crystalline silica in human lung epithelial cells, which prolongs an inflammatory signal and affects fibroblast proliferation, mediating a cadre of lung diseases.
Crystalline silica (SiO2) is the second most common mineral in the earth’s crust and is the major component of sand, rock and mineral ores. In occupational and environmental settings, microscopic mineral particles in airborne dust, generated by wind, manufacturing or demolition are inhaled and can get taken up by epithelial cells lining the respiratory tract, initiating and sustaining inflammatory responses at high concentrations. Prolonged exposure in the workplace may lead to the development of silicosis, which can be irreversible and is characterized by the development of progressive pulmonary fibrosis . Crystalline silica exists in many different polymorphs, but those of particular concern are the naturally occurring polymorphs quartz, cristobalite and tridymite.
The Nacht Domain- Leucine-Rich Repeat-, and PYD-containing Protein 3 (NLRP3) inflammasome has recently been recognized as an innate immune signaling receptor important in mediating cell responses to various endo- and exogenous signals [2–4]. This multi- protein platform interacts with the apoptosis-associated speck-like protein PYCARD/ASC, which contains a caspase-1 recruitment domain. Detection of a danger signal such as crystalline silica particles in macrophages, activates the inflammasome complex leading to binding and activation of pro-caspase-1, which in turn causes the cleavage and secretion the of pro-inflammatory cytokines interleukin – 1 beta (IL – 1β) and interleukin – 18 (IL – 18) to their active forms [5–7]. IL-1, a very potent and pivotal mediator of inflammatory responses induced by silica exposure [8, 9] affects virtually all tissue types and has been implicated in the pathophysiology of human and experimental silicosis [10, 11]. Additionally, caspase-1 expression and activation is required for unconventional secretion of IL-33, IL-1α, basic fibroblast growth factor (bFGF) and high mobility group box 1 (HMGB1) although these proteins are not substrates of caspase-1 [12–15].
Until recently, it was assumed that the initial responses of lung tissue to silica were orchestrated mainly by cells of the innate immune system, such as monocytes, macrophages, neutrophils, and dendritic cells. Purified primary human monocyte-derived macrophages, peripheral blood mononuclear cells, and mouse bone marrow-derived macrophages release IL-1β after inflammasome activation by silica [6, 7]. NLRP3 knockout mouse models develop reduced pulmonary inflammation and less abundant collagen deposition compared to wild-type animals following silica administration . Importantly, in these studies the localization of the active inflammasome in lung tissue was not shown.
Non-myeloid cells can also be important guardians for the detection of danger signals, fulfilling tasks usually performed by resident macrophages. Many groups have demonstrated NLRP3 presence and functional activity in non-myeloid cells by various agonists. For example, nanoparticles activate the NLRP3 inflammasome, leading to IL-1β secretion in primary human keratinocytes . Recently, expression of the NLRP3 inflammasome in human airway epithelium following in vivo particulate matter exposure has been shown although its functional significance in lung disease was unknown . Because the lung epithelial surface is one of the largest primary barriers to environmental exposures and the initial site of impingement of respirable silica, we hypothesized that bronchial epithelial cells are an important target of inflammasome activation. This activation may fuel cross-talk between neighboring fibroblasts, endothelial cells, as well as cells of the immune system which in turn release secondary mediators and initiate or mediate fibrogenesis.
Non-tumorigenic human bronchial epithelial cells (Ad12-SV40 immortalized) BEAS 2B (ATCC, Manassas, VA) were grown and maintained in Dulbecco's Minimal Essential Medium (DMEM)/F12 containing 10% Fetal Bovine Serum (FBS) (CellGro® Mediatech inc, Manassas, VA), with penicillin (50 U/ml), streptomycin (100 μg/ml) (Invitrogen, Carlsbad, CA), hydrocortisone (100 μg/ml), insulin (2.5 μg/ml), transferrin (2.5 μg/ml) and selenium (2.5 μg/ml) (Sigma, St. Louis, MO). Culture flasks and plates (BD, Franklin Lakes, NJ) were pre-coated with a mixture of fibronectin (Sigma, St. Louis, MO) (0.01 mg/ml), bovine collagen type I (0.03 mg/ml) (Invitrogen, Carlsbad, CA) and bovine serum albumin (0.01 mg/ml) (Sigma, St. Louis, MO), in DMEM/F12 media for 24 h at 37°C . Prior to exposures, medium was aspirated and replaced with reduction medium containing 0.5% FBS. In selected experiments BEAS-2B cells were primed with 5 μg/mL LPS for 4 h prior to silica exposure. Particle uptake was blocked by administration of 0.5 μg/mL cytochalasin D for 1h prior to silica exposure.
Primary normal human bronchial epithelial cells (NHBE-17917, Lonza, Clonetics®) were cultured and maintained in BEGM® (Lonza, Clonetics®. (Switzerland)) according to the manufacturer’s protocol.
The MRC-5 (CCL-171) cell line, a human fetal lung fibroblast cell line, was obtained from the ATCC and maintained in Eagle's Minimum Essential Medium (Gibco) supplemented with L- Glutamine (200 mM, Invitrogen), 100 U/ml penicillin, 100 μg/ml streptomycin, and 0.5% heat-inactivated fetal calf serum (Gibco) and non-essential amino acids (MP Biomedicals). For addition of conditioned media, MRC-5 cells were serum starved for 24 h in Eagle's Minimum Essential Medium (Gibco) supplemented with L- Glutamine (200 mM, Invitrogen), 100 U/ml penicillin, 100 μg/ml streptomycin, and 0.5% heat-inactivated fetal calf serum (Gibco) and non-essential amino acids (MP Biomedicals).
THP-1 cells obtained from ATCC were grown in RPMI 1640 medium containing 10% fetal bovine serum with penicillin (50 U/ml), streptomycin (100 μg/ml) and 2 mM L-glutamine at 37°C. Ten ng/mL PMA was used to differentiate THP-1 cells for 24-36 h prior to experiments.
Cristobalite silica particles (C & E Mineral Corp., King of Prussia, PA) were UV-irradiated over night to inactivate possible contaminating endotoxin. Silica particle suspensions (1mg/mL) were sonicated for 15 min, aspirated 5 times through a 23 gauge needle and added to cell cultures. Throughout the studies presented in this paper, we utilized several particle doses based on their surface area characteristics and toxicity . Glass beads (1–4 μm diameter), obtained from Particle Information Services, Inc. (Kingston,WA) were incorporated as a negative control based on particle surface area metrics.
siRNA against NLRP3 (ON-TARGET plus SMARTpool L-017367–00-0005: GGAUCAAACUACUCUGUGA, UGCAAGAUCUCUCAGCAAA, GAAGUGGGGUUCA GAUAAU, and GCAAGACCAAGACGUGUGA) and the ON-TARGET plus GAPDH Control Pool (human); were purchased from Dharmacon, (Thermo Scientific, USA). In BEAS-2B cells siRNA transfections were performed with Lipofectamine 2000 and 100nM siRNA 24 h prior to stimulations. THP-1 cells were differentiated with 10 ng/mL PMA over night after which 100nM siRNA was transfected, using Lipofectamine 2000.
Total RNA was prepared using an RNeasy® Plus Mini Kit according to the manufacturer’s protocol (Qiagen, Valencia, CA) and 1 μg was reverse-transcribed with random primers using the AMV Reverse Transcriptase kit (Pro-mega, Madison, WI). Primers for human NLRP3 were fw:GAT CTTCGCTGCGATCAACAG and rev:CGT GCATTATCTGAACCCCAC. Primers used to determine caspase-1 mRNA levels were fw:TTTCCGCAAGGTTC GATTTTCA and rev:GGCATCTGCGCTCTACCATC. HPRT was used as the housekeeper gene (fw:GACCGGT TCTGTCATGTCG, rev:ACCTGGTTCATCATCACTAA TCAC). Primer sequences were taken from Primerbank. Fold changes in the expression of genes of interest were calculated using the ΔΔCt method. Duplicate assays were performed with all samples.
Cells were lysed in a buffer containing 20 mM Tris, 150 mM NaCl, 1% Nonidet P-40, 1 mM DTT, 1% Protease Inhibitor Cocktail and 1% Phosphatase Inhibitor Cocktail. Total protein content was determined by the Bio-Rad DC Protein Assay kit (Bio-Rad, Hercules, CA), and 20-40 μg of protein was loaded onto polyacrylamide gels. After transfer of proteins to a nitrocellulose membrane, primary antibodies against caspase-1 (Santa Cruz Biotechnology, Santa Cruz, CA), β-actin (Cell Signaling Technologies), IL-1β (Cell Signaling Technologies), NLRP3 (Santa Cruz Biotechnology) or HMGB1 (Cell Signaling Technologies) were applied. After three washes with TBS-Tween, HRP conjugated secondary antibodies were detected by chemiluminescence according to the manufacturer’s instructions. Densitometry was performed on scanned immunoblot images using the QuantityOne software (Bio-Rad Laboratories). Supernatants were concentrated using Amicon Ultra centrifugal filters (Millipore) and analyzed as described above.
Caspase-1 activity was measured using a caspase-1 activity kit (R & D) following the manufacturer’s protocol.
The levels of IL-1β (Biolegend) and bFGF (Biolegend) in cell culture media were measured using human ELISA kits. HMGB1 was determined using a direct ELISA protocol. The primary antibody (ab18256 Abcam) was used at a 1/1000 dilution for 16 h at 4°C. The secondary antibody was a biotin-conjugated swine anti-rabbit used at a 1/1000 dilution. A blocking step was performed using 5% BSA for 1 h. Concentrations of IL-1β, bFGF and HMGB1 were established via extrapolation from a standard curve of the appropriate recombinant protein.
BEAS-2B cells were treated with silica suspensions for 24 h. Supernatants were collected and centrifuged to remove suspended particulates after which conditioned media was added to MRC-5 cells at a 1 to 4 ratio. At indicated time points, MRC-5 cells were fixed with ice cold 70% ethanol for 5 min. After aspiration, 5N NaOH was added followed by DNA content measurement at 260 nm.
All experiments were performed 3 times. Data were analyzed by one-way analysis of variance (ANOVA) using the Student Neuman-Keul’s test to adjust for multiple pair-wise comparisons between treatment groups, or the Student's t test where appropriate. Data from multiple experiments were averaged and expressed as mean values ± SEM. Differences with p-values < 0.05 were considered statistically significant.
Selection of NLRP3 inflammasome and IL-1 family member related genes
Fold Change ↑
Interleukin 1, alpha
Interleukin 1 receptor-like 1
Interleukin 1, beta
NLR family, pyrin domain containing 1
NLR family, pyrin domain containing 3
Interleukin 1 receptor-like 2
Interleukin 18 receptor 1
Interleukin 1 receptor accessory protein
Fold Change ↑
Interleukin 1 family, member 9
Thioredoxin interacting protein
Interleukin 1 family, member 5 (delta)
Interleukin 1 receptor-like 1
Caspase 1 (interleukin-1 beta convertase)
Fibroblast growth factor 2 (basic)
Interleukin 1, alpha
We conducted parallel experiments in BEAS-2B and THP-1 macrophages using the same dose of silica to investigate IL-1β levels and the effect of LPS priming on these cell types comparatively. Basal levels of secreted IL-1β are higher in THP-1 cells compared to BEAS-2B (28.3 ± 2.7 vs 6.29 ± 0.5 pg/ml respectively). Silica induced IL-1β levels in BEAS-2B cells to 8.67 ± 0.5 pg/ml. In THP-1 cells silica alone also induced IL-1β release, to a level of 191.0 ± 6.7 pg/ml, about 22-fold higher compared to in BEAS-2B. LPS had no significant effect by itself on IL-1β levels in either cell type. In BEAS-2B there was no synergistic effect of priming with LPS prior to silica treatment, whereas this was the case in THP-1 cells. HMGB1 levels in culture medium were also approximately double in THP-1 cells compared to BEAS-2B cells, under baseline conditions as well as after silica stimulation. bFGF levels on the other hand were comparable in unstimulated cells, but induction was greater in bronchial epithelial cells (Additional file 2: Table S1). The release of IL1β, HMGB1 and bFGF induced by silica was also NLRP3-dependent in THP-1 cells as indicated in Additional file 3: Figure S2.
In myeloid cells, the inflammasome receptor complex is recognized as one of the cornerstones of the long-sought intracellular surveillance system . The NLRP3 (NALP3) complex contributes to innate immune defenses against pathogen- and danger-associated moieties such as exogenous or endogenous crystalline molecules [5–7, 21].
Our present study assessed the importance of the bronchial epithelial cells as a target for cristobalite silica-induced inflammasome activation. Thereby, classical and non-canonical outputs of NLRP3 assembly and caspase-1 activation might contribute to the development of silicotic nodules and fibrosis. This study illuminated increases in transcription of components of the NLRP3 multiprotein platform which leads to caspase-1 activation in response to cristobalite silica in bronchial epithelial cells.
It has been postulated that the release of active IL-1 needs at first a signal provided by Toll-like receptor (TLR) engagement that results in gene transcription and pro-IL-1β accumulation, a process sometimes described as “priming”; and a second signal through NLR signaling, which results in caspase-1 activation and subsequently processing of this cytokine into mature bioactive IL-1β. Nevertheless, it is still controversial whether there is a direct role for TLR signaling in the activation of the inflammasome because of contrasting results regarding the requirement of a “two-hit-model” in the assembly of the multiprotein complex. Epithelial cell monolayers, like some myeloid cells, do not always require LPS priming in contrast with the well documented ‘two-hit-model’ and therefore may provoke a constant phase of alarm when bombarded with cristobalite particles. It is therefore not surprising to find only modest but significantly increased levels of the most potent pyrogen IL-1β, in whole cell lysates or medium of cell cultures after treatment with silica without priming with LPS. Although in the comparative experiments using THP-1 macrophages silica induced a 22-fold higher IL-1β release compared to the response in BEAS-2B cells, many more epithelial cells are present in the lungs compared to macrophages. Together with the observations that ambient particulate matter PM10 alone could induce increases of detectable cleaved subunits of IL-1β in bronchial epithelial cells  and described IL-1β release from bronchial epithelial cells after cigarette smoke and rhinovirus without priming with LPS [22, 23], bronchial epithelial cells can be considered danger sensing cells contributing to small but relevant levels of biologically active IL-1β.
The nuclear-damage-associated molecular pattern molecule (DAMP), HMGB1 was originally identified as one of the nuclear proteins mediating gene transcription as a chromatin binder . In secreted form, it is thought to be a cytokine-like molecule derived from various cell types, and acting as an alarmin. Furthermore, it induces production of pro-inflammatory cytokines . Activated inflammasomes are involved in the release of HMGB1 through unconventional protein secretion, or as a soluble molecule from cells during pyroptosis . HMGB1 is involved in innate and specific immune responses and contributes to acute lung injury in bleomycin-induced fibrosis in mice . The same study also demonstrated that HMGB1 stimulates proliferation of human fibroblasts without effects on apoptosis and directly participated in fibrogenesis. Additionally, HMGB1 seems to protect the host from pathogen-induced mortality  and has the ability to attract stem cells to areas inflammation, thus promoting their regeneration [28, 29]. Here we show that inflammasome activation mediates unconventional HMGB1 release from BEAS-2B and THP-1 cells following early interactions with silica and as a consequence further amplifying the reaction cascade.
bFGF is released after tissue injury and during inflammation and is produced by lung epithelial cells, macrophages and endothelial cells . It plays a crucial role during fibrosis as well as angiogenesis [31, 32]. We recently showed higher transcript levels and bFGF protein secretion in cristobalite silica treated BEAS-2B cells compared to untreated controls . Here, we extended these findings by demonstrating inflammasome dependence of release of bFGF from epithelial cells and macrophages suggesting an early role for the inflammasome in initiating fibrogenesis.
In the previous study, we demonstrated uptake of cristobalite silica in bronchial epithelial cells . Therefore, we investigated whether inflammasome activation was dependent on uptake of particles as shown by others in myeloid cells [5–7, 21]. Decreased levels of bFGF and HMGB1 release occurred in medium after cytochalasin D pre-treatment, indicating a pivotal role for internalization of silica in inflammasome activation and release of pro-inflammatory and fibrotic mediators. It remains to be elucidated how particulates are internalized and how this triggers inflammasome assembly and activation.
Finally, possible effects of inflammasome activation in epithelial cells on the lung microenvironment were examined. It was found that lung epithelial cells after internalization of silica secrete substances, including bFGF that induce fibroblast proliferation in vitro. The fact that the inhibition of particle uptake and inflammasome inactivation in bronchial epithelial cells dramatically blunted fibroblast proliferative response is a novel observation of direct relevance to fibrogenesis.
Data presented here are the result of in vitro experiments and it remains to be investigated whether bronchial epithelial cells in vivo are a major site of silica-induced inflammasome activation and consequent signaling. Recently, it was published that human bronchial epithelial cells express functional NLRP3  following PM10 exposure in a mouse model. Also, they demonstrated NLRP3-dependent airway neutrophilia and increases in dendritic cell numbers in intrathoracic lymph nodes, indicating the influence of NLRP3 on different aspects of immune cell signaling.
Collectively, our data indicate that bronchial epithelial cells are much more active than previously assumed. The NLRP3 inflammasome is activated in a one-step fashion, leading to downstream occurring events that are important in relaying the danger signal to the microenvironment contributing to the pathological condition of silicosis.
These novel data show the presence and functional activation of the NLRP3 inflammasome by cristobalite silica in human lung epithelial cells that mediates a cadre of lung diseases. Moreover, these findings show that inflammasome activation by silica in epithelial cells is linked to fibroblast proliferation.
This work was supported by an unrestricted grant from the Weijerhorst Foundation. The authors thank the Vermont Cancer Center DNA Analysis facility (Vermont Genetics Network) at the University of Vermont (Burlington, VT) for assistance with qRT-PCR. We would like to acknowledge Max Macpherson, Arti Shukla, Jos van der Velden, Amy Guala and Renske Krijgsman for technical expertise and help. We also thank Jennifer Díaz for editorial assistance with this manuscript.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.