Silica particle-induced NLRP3 inflammasome activation and DNA damage in respiratory epithelium

Sammanfattning: Respirable crystalline silica particles (CSi) can persist and induce inflammation and DNA damage in human lungs. Previous studies have used complex model systems to characterize CSi-induced DNA damage, involving co-cultures of epithelial cells and macrophages and long-term exposures, necessitating alternative models. Here, the respiratory epithelium was used in monocultures and the results were confirmed in short-term in vivo studies. In paper I, the role of CSi-induced NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome was studied. siRNA, small molecule inhibitors, and knockout-cells (NLRP3 KO cells) were used, and all experimental approaches indicated a role for NLRP3. Phosphorylation of NLRP3 (Ser 198) and co-localization with mitochondria occurred within 5-10min and therefore seemed critical. During this time frame there were also clear indications of DNA damage, but the early DNA damaging effect could not be correlated to increased levels of reactive oxygen species (ROS). The DNA damaging effects was prevented by cell treatments which decreased NLRP3 expression or assembly. Notably, in this study, lower doses of CSi were used and the DNA damage was detected earlier than in previous efforts to describe silica toxicity. Mice were exposed intra-nasally with 0.025 mg CSi. Increased levels of γH2AX and pCHK2 were detected within 10min in isolated lung homogenates, thus confirming an early DNA damage also in vivo. Additional markers were analyzed in serum, BAL and lung tissue. These markers indicated early inflammatory responses already at 5 min. One of the markers investigated was autotaxin (ATX) secretion, an immediate response to silica exposure that was followed up in paper II. In paper II, the role of ATX was studied. As release was detected as early as 5min in serum and BAL after CSi inhalation, we expected ATX to have a so far uncharacterized role in lung toxicity. Two small molecular inhibitors of ATX enzyme activity prevented both CSi-induced mitochondrial depolarization and DNA damage. Extracellular ATX activity forms lysophosphatidic acid (LPA), and it was found that addition of LPA to the experimental system gave rise to DNA damage. LPA may also activate Rac1, and in further studies we documented Rac1 to be activated at 3min of exposure. A Rac1 inhibitor prevented depolarization and DNA damage. Rac1 is known to increase ROS levels. Therefore, antioxidants were tested and found to prevent DNA damage. In in vivo studies the expression of ATX in lung tissue was examined by confocal microscopy. An increased expression was seen in apical parts of bronchial epithelia at 5min, and culminated at 60min, Protrusions from apical epithelium plasma membranes increased with time, an effect that can explain the increase of ATX in BAL, which was seen in the first study. A parallel increase in staining for 53BP1 was also detected, confirming the occurrence of double strand breaks (DSBs) in genomic DNA. This thesis provides data indicating a novel mechanism for CSi-induced DNA damage in respiratory epithelial cells. It occurs rapidly and has not previously been explored for risk assessment purposes. The data suggest that CSi in contact with the plasma membrane activates ATX release and that this causes Rac1 activation, NLRP3 phosphorylation and mitochondrial translocation. These effects are followed by an accumulation of DNA damage, including DSBs. Clear effects are seen within 5-10min and can be recapitulated in mice inhaling CSi. Low doses were used in these studies and several of the endpoints and the in vivo model should be suitable for testing low-dose DNA damaging effects of CSi. Such experimental studies are important as they may reveal cancer causing effects of CSi at lower exposure levels than previously anticipated. ATX stands out as a suitable biomarker for CSi effects. Thus, ATX is rapidly secreted in BAL from target cells and presumably also in plasma. It seems critical for an early and perhaps carcinogenic DNA damaging effect and it has previously been associated with cancer development.