Human LP9 mesothelial cells, an hTERT-immortalized cell line that phenotypically and functionally resembles normal human mesothelial cells, were obtained from Dr. James Rheinwald (Brigham and Women’s Hospital, Harvard University, Boston, MA). All cells were incubated at 37°C in 5% CO2 and grown to 80–90% confluency as described previously. The physical and chemical characterization of the National Institute on Environmental Health Sciences (NIEHS) reference sample of crocidolite asbestos has been reported previously. The NIEHS chrysotile reference sample was used for asbestos fiber comparisons. After sterilization under UV light overnight, particulates were suspended in Hank’s balanced salt solution (HBSS) at 1 mg/ml, sonicated for 15 min in a water bath sonicator, and triturated five to ten times through a 22-gauge needle. A volume of this suspension was added to cells in medium to achieve the desired final concentration of 75 × 106 μm2/cm2 dish surface area, a concentration known to cause apoptosis and compensatory proliferation of surrounding rat pleural mesothelial and murine alveolar type II epithelial cells[3, 17]. Glass beads (Polysciences Inc, Warrington, PA) were used as a non-pathogenic particle control. Dehydroascorbic acid (DHA) and 2,4-Dinitro-1-chlorobenzene (DNCB) were purchased from Sigma (St. Louis, MO). Trx1 expression vector (pCMV-SPORT6) and pcDNA (empty vector control) used for over-expression studies were obtained from Dr. Nicholas Heintz.
Western blot analyses
Cells grown in 60 mm culture dishes were washed 3× with ice-cold phosphate buffered-saline (PBS), collected in lysis buffer (20 mM Tris pH 7.6, 1% Triton X-100, 137 mM NaCl, 2 mM EDTA, 1 mM Na3O4V, 10 mM NaF, 1 mM DTT, 1 mM phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, and 10 μg/ml aprotinin) and incubated on ice for 30 min. Lysates were centrifuged at 14,000 rpm for 15 min at 4°C. Supernatants were collected, and protein concentrations were determined using the Bradford assay (Bio-Rad, Richmond, CA). Cell lysates (40 μg protein per lane) were resolved by one-dimensional SDS-PAGE and transferred to nitrocellulose membranes according to standard procedures. Equal loading of protein was verified by β-actin (Abcam, Cambridge, MA). Membranes were washed in Tris-buffered saline (TBS), and blocked for 2 h with TBS-Tween (TBST) containing 10% nonfat milk, then incubated with rabbit anti-human Trx1 antibody (Abcam, Cambridge, MA) at 1:5000 dilution in TBST containing 5% nonfat milk and 0.01% sodium azide overnight at 4°C. Membranes were washed four times with TBST for 15 min each time prior to incubation with secondary antibody. Western blots shown are representative blots with their accompanying densitometric analysis.
Determination of the redox state of thioredoxin
The redox state of thioredoxin in response to crocidolite asbestos exposure was determined using the redox Western blot method as previously described. Briefly, cells were lysed in 6 M guanidine HCl buffer (6 M Guanidine HCl; 50 mM Tris/Cl pH 8.3; 3 mM EDTA, 0.5% Triton X – 100; 10 μg/ml aprotonin and 10 μg/ml leupeptin) containing 50 mM iodoacetic acid (IAA) for alkylation of the thiol groups of thioredoxin. Cells were incubated in lysis buffer at 37°C for 30 min in the dark. Excess IAA was removed by spinning lysates on Amicon centrifugal concentrating columns with a 10,000 nominal molecular weight limit (NMWL) (EMD Millipore, Billerica, MA). In order to exchange the IAA containing buffer, the lysates were washed 3 times with a HEPES buffer at pH 7.4 and the concentrated lysates were collected in fresh collection tubes by inverting the columns in the tubes. After protein determination by the Bradford method (Bio-Rad, Richmond CA), 40 μg of protein was loaded onto a 15% non–reducing native polyacrylamide gel using a 1× Tris Glycine (pH 8.8) running buffer. The electrophoresis was carried out at 75 V for approximately 3.5 h. The redox gel was then washed in 50 mM Tris (pH 8.3) for 5 min and then equilibrated in 1× transfer buffer by washing in buffer 3 times for 5 min each. Thereafter, the proteins were transferred onto a nitrocellulose membrane by wet transfer at 100 V for 2 h. The nitrocellulose membrane was then blocked with 10% milk in 1× Tris buffered saline with Tween 20 (TBS-T) for 2 h at room temperature (RT) and incubated in anti-Trx1 primary antibody (1:5000 in 5% milk/TBS-T, Abcam, Cambridge, MA) overnight at 4°C. Goat anti–rabbit secondary antibody conjugated to horseradish peroxidase (1:2000 in 1X TBS-T, Jackson ImmunoResearch Laboratories Inc. West Grove, PA) was used and visualization was done by enhanced chemiluminescence reagents (Amersham Pharmacia Biotech, Piscataway, NJ) on X-ray film. Blots were quantified using Quantity One software (Bio-Rad, Richmond, CA). Distribution of redox states of Trx1 was determined as band intensity of reduced or oxidized Trx/(reduced + semi-oxidized + fully oxidized) as described by Watson et al., (2003).
Real-time quantitative PCR (qRT-PCR)
Total RNA was prepared using an RNeasy plus mini kit according to the manufacturer’s protocol (Qiagen, Valencia, CA) as described previously. Total RNA (1 μg) was reverse-transcribed with random primers using the Promega AMV Reverse Transcriptase kit (Promega, Madison, WI) according to the recommendations of the manufacturer. To quantify gene expression, the cDNA was amplified by TaqMan Real Time Q-PCR using the 7900HT SequencePrism Detector (Applied Biosystems, Foster City, CA). Duplicate assays were performed with RNA samples isolated from at least four independent experiments. Fold changes in gene expression were calculated using the delta-delta Ct method. The values obtained from cDNA and hypoxanthine phosphoribosyl transferase (HPRT) controls helped determine relative gene expression levels for the gene locus investigated. The Assay on Demand primers and probes used were purchased from Applied Biosystems.
Since exposure of cells to chrysotile asbestos had no effect on the oxidation state of Trx1, all subsequent experiments were performed using crocidolite asbestos. All references to asbestos relate to crocidolite asbestos unless otherwise specified.
Dehydroascorbic acid (DHA) pretreatment
In order to investigate whether the ROS generated by asbestos exposure was responsible for the extensive oxidation of thioredoxin, cells where pretreated with the ROS quencher, dehydroascorbic acid (1 mM) for 1 h before exposure to asbestos for 8 h.
Antioxidant pretreatment of cells by N-acetylcysteine (NAC)
To investigate the role of the asbestos-induced ROS on inflammasome activation, cells were pretreated with 2 mM NAC for 20 h as previously described (Shukla et al., 2004). Briefly, cells were grown to 90% confluency in 60 mm dishes and serum starved by replacing complete medium with 0.5% fetal bovine serum (FBS) supplemented medium for 6 h prior to addition of NAC diluted in HBSS at pH 7.4 (cells were maintained in reducing medium for the entire duration of NAC pretreatment). After pretreatment with NAC, cells were exposed to asbestos for 48 h. Thereafter, inflammasome priming was assessed by qRT-PCR of NLRP3 transcript levels while activation was analyzed by Western blot analysis of caspase-1, p20 fragment.
Treatment with 2,4-Dinitro-1-chlorobenzene DNCB
To obtain a final concentration of 10 μM in culture, 2,4-Dinitro-1-chlorobenzene (DNCB) (an irreversible alkylating inhibitor of TR) was dissolved in DMSO with a final DMSO content of 0.2% which was determined to be non-cytotoxic in previous experiments. For all experiments, cells were pretreated with DNCB for 1 h prior to exposure to asbestos.
Lactate dehydrogenase (LDH) activity assay
To determine the cytotoxic effects of DNCB and asbestos exposure on LP9 cells, an LDH assay was performed using the LDH kit from Promega, (Madison WI) according to the manufacturer’s direction. Briefly, 50 μl of media were collected from each dish in triplicate into a 96 well plate using cells lysed by the addition of 0.9% Triton X-100 as a positive control. To each of these wells, 50 μl of LDH substrate buffer was added and the reaction was incubated on a rotary shaker at room temperature for 30 min in the dark. Thereafter, the reaction was terminated by the addition of 50 μl stop buffer to each well. Any bubbles present were broken with a hypodermic needle and the plate was read spectrophotometrically at 490 nm in a 96 well plate reader. Cytotoxicity was expressed as a percentage of LDH released relative to the lysis control.
Detection and quantitation of apoptosis
To determine whether modulation of Trx1 protein levels and oxidation state altered cell death in human mesothelial cells, detection of apoptosis was performed using the ApoStain technique as described previously (Shukla et al., 2003). In brief, cells were grown on glass cover slips and exposed to asbestos with or without DNCB (10 μM) for 8 h. The cover slips were then processed to determine the numbers of apoptotic cells and total cell numbers per field. Five random fields were evaluated at a magnification of 400× on each cover slip.
Assessment of pyroptosis by asbestos
Since asbestos causes inflammasome activation, as measured by caspase-1 activation, we were interested in learning if asbestos-induced cell death may be due in part to pyroptosis (caspase-1 dependent cell death) or not. For this purpose we pretreated LP9 cells with a specific caspase-1 inhibitor (40 μM Caspase-1 inhibitor VI (zYVADfmk), EMD Biosciences, Billerica, CA) for 1 h and subsequently with asbestos for 24 h. The number of viable cells was determined after trypsinization and counting of cells on a hemocytometer at the end of the experiment. Media supernatants were also analyzed for the levels of the p20 subunit of active caspase 1 by Western blot analysis.
Trx1 over- expression
Cells at 90% confluence were transfected with pcDNA (empty vector control, 4 μg DNA per 60 mm dish) and human Trx1 over expression vector (pCMV-SPORT6, 4 μg DNA per 60 mm dish) using Lipofectamine 2000 (10 μl) (Life Technologies, Grand Island, NY), following the manufacturer’s protocol. The efficiency of Trx1 protein over expression was determined by qRT-PCR after 48 and 72 h.
Knockdown of TXNIP
LP9 cells that were 90% confluent were transfected with either ON-TARGET plus smart pool human TXNIP siRNA (siTXNIP) or ON-TARGET plus non-targeting siRNA (siControl) from Dharmacon (Fisher Scientific, Pittsburgh, PA) using Lipofectamine 2000 (Life Technologies, Grand Island, NY) diluted in a final volume of 500 μl Optimem medium (Life Technologies, Grand Island, NY), as previously described. All siRNA were reconstituted to 20 μM before transfection and stored at -20°C until use. The magnitude of TXNIP knockdown was assessed by qRT-PCR.
To confirm observations of the role of TXNIP in inflammasome activation, a human mesothelioma cell line (HMESO) in which the extracellular signal regulated kinase 2 had been stably knocked down (shERK2) was used. These cells have been previously reported (GSE21750) to have down-regulated expression of TXNIP (several fold). To activate the inflammasome, shERK2 HMESO cells and corresponding control cells stably transfected with non-targeting shRNA (shCon) were treated with 5 μM doxorubicin (Dox) as previously described. After 48 h of treatment the shERK2 and shCon HMESO cells, treated with or without Dox, medium supernatants were harvested by centrifuging medium at 300 × g for 7 min in the cold (4°C) to remove cellular debris. The resulting supernatants, stored in 1 ml aliquots, were concentrated using Amicon centrifugal concentrating columns with a 10,000 nominal molecular weight limit (NMWL) (EMD Millipore, Billerica, MA). 4× sample buffer was added to the concentrated supernatant to a final concentration of 1× before being electrophoresed at 100 V for 2 h on a 15% SDS PAGE gel. Immunobloting was performed as described above with caspase-1 p20 antibody (Cell Signaling, Danvers MA) to detect caspase-1 activation.
All data were analyzed by one-way ANOVA compared with the respective control group and a Neuman-Keuls post test for multiple comparisons or the Student’s ‘t’test. Results are presented as the mean ± SEM. All experiments were repeated at least twice or more. Comparisons with a p value ≤ 0.05 were considered statistically significant. Statistical analyses were conducted using Graph Pad Prism v6 software.