Asbestos modulates thioredoxin-thioredoxin interacting protein interaction to regulate inflammasome activation

Abstract

Background: Asbestos exposure is related to various diseases including asbestosis and malignant mesothelioma (MM). Among the pathogenic mechanisms proposed by which asbestos can cause diseases involving epithelial and mesothelial cells, the most widely accepted one is the generation of reactive oxygen species and/or depletion of antioxidants like glutathione. It has also been demonstrated that asbestos can induce inflammation, perhaps due to activation of inflammasomes.

Methods: The oxidation state of thioredoxin was analyzed by redox Western blot analysis and ROS generation was assessed spectrophotometrically as a read-out of solubilized formazan produced by the reduction of nitrotetrazolium blue (NTB) by superoxide. Quantitative real time PCR was used to assess changes in gene transcription.

Results: Here we demonstrate that crocidolite asbestos fibers oxidize the pool of the antioxidant, Thioredoxin-1 (Trx1), which results in release of Thioredoxin Interacting Protein (TXNIP) and subsequent activation of inflammasomes in human mesothelial cells. Exposure to crocidolite asbestos resulted in the depletion of reduced Trx1 in human peritoneal mesothelial (LP9/hTERT) cells. Pretreatment with the antioxidant dehydroascorbic acid (a reactive oxygen species (ROS) scavenger) reduced the level of crocidolite asbestos-induced Trx1 oxidation as well as the depletion of reduced Trx1. Increasing Trx1 expression levels using a Trx1 over-expression vector, reduced the extent of Trx1 oxidation and generation of ROS by crocidolite asbestos, and increased cell survival. In addition, knockdown of TXNIP expression by siRNA attenuated crocidolite asbestos-induced activation of the inflammasome.

Conclusion: Our novel findings suggest that extensive Trx1 oxidation and TXNIP dissociation may be one of the mechanisms by which crocidolite asbestos activates the inflammasome and helps in development of MM.

Figure 1

Crocidolite asbestos exposure modulates Trx1 levels in mesothelial cells. (A) LP9 cells were exposed to crocidolite asbestos at 75 × 10⁶ μm²/cm² for 8 and 24 h. Glass beads (GB) at the same surface area concentration were used as an inert particulate control. RNA was extracted from the samples and used to prepare cDNA, which was quantified by qRT-PCR (*p < 0.05 compared to untreated controls (NC)). (B) Western blot analysis of LP9 cells for total thioredoxin protein after exposure to crocidolite asbestos, β-actin was used as a loading control. (C) densitometric analysis of (B) using Quantity One software (n =2 per group) .

Figure 2

Crocidolite asbestos exposure causes oxidation of Trx1 in human mesothelial cells. (A) LP9/h-TERT cells were exposed to crocidolite asbestos at 75 × 10⁶ μm²/cm² for 8 and 24 h and 40 μg of protein was run on a 15% native gel and a redox Western blot analysis was performed as described in the Methods section on the immobilized proteins. (B) Quantitation of the blot in (A). (C) Cells were exposed to two concentrations of crocidolite asbestos (15 × 10⁶ μm²/cm² and 75 × 10⁶ μm²/cm²), as well as chrysotile asbestos (75 × 10⁶ μm²/cm²) and a redox Western analysis was performed on the lysates (*p < 0.05 compared to null controls (n = 2 per group).

Figure 3

Inhibition of thioredoxin reductase by DNCB and pretreatment of cells with DHA rescues asbestos-induced oxidation of Trx1. (A) LDH assay to assess lytic cell death after pretreatment of LP9 cells with DNCB and exposure to asbestos (data is presented as a percentage of the lytic control). (B) Effect of DNCB on asbestos-induced Trx1 oxidation. Cells were pretreated with 10 μM DNCB for an hour and then exposed to asbestos for 8 h. Cell lysates were then derivatized with IAA and analyzed for oxidation of Trx1 by redox Western blot and densitometry of redox Western analysis of Trx1 oxidation state was performed. (C) Effect of DNCB pretreatment on chrysotile asbestos-induced oxidation of Trx1 (D) Analysis of apoptosis in response to DNCB pretreatment and asbestos exposure for 8 h as measured by Apostain technique. (E) LP9 cells were pretreated with 1 mM DHA for an hour and exposed to asbestos for 8 h. Thereafter, the oxidation state of Trx1 was assessed by redox Western blot analysis (*p < 0.05 compared to null controls; †p < 0.05 compared to crocidolite asbestos exposure alone (Croc 75 n = 2 per group).

Figure 4

Over-expression of Trx1 increases cell survival and ameliorates asbestos-induced ROS generation in LP9 cells. (A) LP9 cells were exposed to two doses of crocidolite asbestos (15 × 10⁶ μm²/cm² and 75 × 10⁶ μm²/cm²) for 24 h and incubated with NBT for 45 min at 37°C. The absorbance of the solubilized formazan formed after incubation with NBT was then read at 630 nm to determine ROS levels after asbestos exposure. (B) Over-expression of Trx1 in LP9 cells using the pCMV-SPORT6 plasmid was confirmed by qRT-PCR 48 and 72 h after transfection. (C) Trx1 transfected cells were exposed to crocidolite asbestos for 2 h and incubated with NBT for 45 min to determine ROS levels. Solubilized formazan was measured spetrophotometrically at 630 nm on a plate reader. (D) Analysis of the oxidation state of Trx1 after asbestos exposure of Trx1 over-expressing LP9 cells was determined by redox Western blot analysis and densitometry of the blot was performed (n = 2 per group). (E) LP9 cells transfected with the Trx1 over-expressing plasmid, pCMV-SPORT6 were exposed to crocidolite asbestos for the times indicated. Cells were then trypsinized and counted to estimate cell survival (*p < 0.05 compared to control; †p < 0.05 compared to Trx1 OE). Cell survival and NTB graphs are the average results of 3 experiments.

Figure 5

Asbestos-induced inflammasome priming and activation is attenuated by NAC. (A) LP9 cells pretreated with 2 mM NAC were exposed to 5 ug/cm² asbestos for 48 h and changes in NLRP3 mRNA levels were assessed by qRT-PCR. (B) Inflammasome activation was assessed by Western blot analysis of the media supernatants from cells exposed to asbestos with and without pretreatment with NAC. Immobilized proteins on the nitrocellulose membrane were probed for the presence of active caspase-1 (p20 fragment) (*p < 0.05 compared to null control; † compared to Croc 75 alone; n = 2 per group).

Figure 6

Knockdown of TXNIP decreases inflammasome activation. (A) LP9/hTERT cells (90% confluent) were transiently transfected with siTXNIP or siControl siRNA for 48 h and knockdown of TXNIP expression was assessed by qRT-PCR. (*p < 0.05 compared to siControl) (B) siControl and siTXNIP transfected cells were exposed to crocidolite asbestos for 48 h and the medium were collected, concentrated and analyzed for the presence of caspase-1 (p20) by Western blot analysis. (C) The transcript levels of ERK 1/2 as well as TXNIP were verified in shERK2 HMESO cells by qRT-PCR in the presence and absence of 5 μM Dox along with priming of the inflammasome (*p < 0.05 compared to shCon alone; †p < 0.05 compared to shCon + Dox; n = 2 per group). (D) Inflammasome activation was assessed by Western blot analysis for the p20 fragment of caspase-1 in the media supernatants after treatment with Dox (*p < 0.05 compared to shCon alone; †p < 0.05 compared to shCon + Dox; n = 2 per group). (E) LP9 cells were pretreated with 40 μM of the caspase-1 inhibitor VI (cas-1 inh) prior to exposure to asbestos for 48 h and cells were counted to determine survival (*p < 0.05 compared to control; †p < 0.05 compared to Croc 75 alone). (F) Western blot analysis for Cas-1 p20 fragment in medium supernatant from LP9 cells pretreated with the caspase-1 inhibitor prior to exposure to asbestos (n = 3 per group).

Figure 7

Role of ROS and antioxidants in asbestos-induced activation of the NLRP3 inflammasome. A simplified schema showing how increased ROS or decreased GSH as a result of asbestos exposure can cause oxidation of Trx1 and release of TXNIP. TXNIP thus released binds to NLRP3 and activates it as represented by caspase-1 activation. NAC on the other hand reduces ROS and elevates GSH levels resulting in inhibition of activation of NLRP3.