Understanding STING-mediated cytokine release and antitumor immunity in malignant pleural mesothelioma cell lines
OA Version
Citation
Abstract
INTRODUCTION: Malignant Pleural Mesothelioma (MPM) is an aggressive cancer of the serosal lining linked to asbestos carcinogenesis. Prognosis is poor with current life expectancy of 18-31 months after diagnosis with the potential to improve with treatment. Current standard of care is platinum based chemotherapy; however, our research indicated that MPM would be an ideal tumor type in which to deploy stimulator of interferon genes (STING) agonists (Knelson et al., manuscript in revision). When activated, the STING pathway initiates an innate immune response through the release of pro-inflammatory cytokines and chemokines (Su et al., 2019). Stimulation of this pathway can create a potential therapeutic target to boost antitumor immunity (Berger et al., 2019). Our research thus far has indicated that both MPM cell lines and patient specimens express high levels of STING and have positive responses to treatment with STING agonists. However, the effectiveness of the treatment varies across cell lines and patient specimens, providing an opportunity to understand mechanisms in patient subsets and develop clinical biomarkers.
OBJECTIVE: We took a mechanistic approach to understanding the STING pathway to identify potential blocks in the cascade in cells that were unresponsive to STING agonists despite their high-level of STING expression. STING signaling is initiated by dsDNA interacting with cGAS to create second messenger cyclic dinucleotide cGAMP which activates STING and several downstream pathway components including: TBK1, IRF3, type 1 IFNs, and STAT1 to release pro-inflammatory cytokines and chemokines (Wan et al., 2020). We hypothesized that there was a block in the cascade preventing activation of one of the downstream phosphorylated steps in non-responsive cell lines resulting in a decreased secretion of CXCL10, a STING effector chemokine, after treatment with a STING agonist. In order to investigate systematically, we specifically looked at whether cell lines with low CXCL10 had a signaling block downstream of STING and whether cell lines with low CXCL10 release demonstrate impaired cytokine exocytosis.
METHODS: CXCL10 ELISA was conducted to observe cytosolic and secreted levels of CXCL10 protein in various MPM cell lines to determine if there was a decrease in production or exocytosis. Western blot analysis was conducted after treatment with several STING agonists as well as transfection with poly (dC:dG) to observe the protein levels of various basal and phosphorylated steps in the STING pathway. Additionally, PCR analysis of MPM cell lines treated with STING agonists was performed to identify if there was an issue in CXCL10 trafficking.
RESULTS: Western Blot analysis indicated that all MPM cell lines had similar basal levels of protein for STING and TBK1 but began to differ at IRF3. Unresponsive cell lines had decreased basal protein levels of IRF3 and STAT1. Western blot analysis of phosphorylated and total protein steps showed no pIRF3 and decreased total IRF3 and pSTAT1 in unresponsive cell lines. ELISA analysis found that both cytosolic and secreted levels of CXCL10 were increased in responder cell lines compared to non-responders. PCR analysis found similar patterns in gene expression of CXCL10 and IFN.
CONCLUSION: The data collected indicate that there is a block in the STING pathway cascade between TBK1 and IRF3 inhibiting the pathway in some cell lines. This block is affecting downstream steps such as pSTAT1, IFN and CXCL10. Further research using RNA sequencing analysis can identify if there are similar patterns in patient specimens. This information can help identify patient subsets in which STING agonist treatment will benefit the most. Further research can attempt to overcome this block by independently targeting downstream IF3 activation or by using other pathways that target similar cytokines such as toll-like receptor (TLR) signaling pathways that are not dependent on IRF3.