An antibiotic triggers cell stress responses and Epstein Barr virus lytic activation

Epstein–Barr virus (EBV) is a human gammaherpesvirus, which infects most of the human population worldwide, and is one of the first human viruses found to be associated with cancer. EBV has a latent lifecycle characterized by limited gene expression, and a lytic lifecycle where infectious virions are produced. Latent infection is responsible for causing various types of cancers, and inducing the EBV lytic cycle has been suggested as a new approach for treating these types of malignancies. The unfolded protein response (UPR) is one of the major pathways which leads EBV to enter the lytic replication cycle. Thus several UPR inducers such as bortezomib are widely used for activating EBV in various cancer cell lines in laboratory. Recent investigations have demonstrated that clofoctol, an antibiotic drug that has anti-proliferative activities in prostate cancer cell lines, activates the UPR. In order to investigate whether clofoctol activates the UPR and leads to the viral activation, we tested the drug on various types of EBV-positive cancer cell lines, including the BX1-Akata Burkitt lymphoma cell line, lymphoblastoid cell line (LCL), SNU719 gastric cancer cell line and C666 nasopharyngeal cancer cell line. Here, we show that clofoctol induces EBV lytic activation in a variety of cancer cell lines at a clinically achievable concentration. The effect was similar to that seen with bortezomib, but was more rapid and worked in broader range of the cell lines. Viral immediate early lytic gene Zta, early lytic gene Bmrf1 and late lytic gene gp350 expression were increased in RNA level. Increased Zta protein and EBV DNA copy number were confirmed in BX1-Akata cells. However, the expression of late protein gp350 was not detected after clofoctol treatment. A Raji infection assay using EBV expressing GFP showed that clofoctol treatment does not lead to infectious virion production. Similar to a previous report with prostate cancer cell lines, clofoctol induces all three pathways of UPR – PERK, IRE1 and ATF6. Activation of the PERK branch was confirmed by demonstration of increased eIF2α phosphorylation, and increased ATF4 and CHOP expression. Activation of IRE1 and ATF6 pathways was demonstrated by an increase in XBP1 spliced RNA and Bip RNA, respectively. With modulation of the PERK branch by shRNA-mediated knockdown and a PERK-specific inhibitor, we found that the PERK pathway mediates EBV lytic activation by clofoctol as inhibited PERK led to reduced Zta expression. Activated PERK leads to eIF2α phosphorylation which in turn inhibits most translation, but allows some RNAs including ATF4 to be translated. In our results, Zta protein synthesis was enhanced with PERK activation and eIF2α phosphorylation by clofoctol treatment, suggesting Zta translation overcomes the block to translation mediated by eIF2α phosphorylation. Furthermore, we found the PERK pathway differentially regulates viral lytic gene expression at the protein level. Clofoctol treatment alone triggers expression of the immediate early protein Zta. However, ZTA expression was reduced by treatment with a PERK inhibitor. In contrast, expression of the viral late lytic protein gp350 was not increased by clofoctol alone but was increased by the combination of clofoctol and PERK inhibitor. This result suggests that clofoctol activates other branches of the UPR and that these also are important in driving EBV lytic infection. eIF2α phosphorylation by PERK is also a part of the integrated stress response (ISR) and there are more kinases that induce the phosphorylation of eIF2α, including HRI, PKR, and GCN2. In order to demonstrate that activation of all eIF2α kinases results in EBV lytic induction, we performed pharmacologic inductions of four kinases and evaluated the effect on virus in BX1-Akata cells. The specific activation of each of the eIF2α kinases triggered EBV Zta RNA expression and increased GFP expression in BX1-Akata cells. Our results show that clofoctol activates EBV lytic gene expression, and that this expression is partially mediated by the PERK pathway of UPR and ISR. Inhibition of global protein translation by the PERK pathway allows immediate early Zta protein synthesis but blocks late protein gp350 synthesis and virion production. We also found that other ISR stimuli can lead to viral lytic induction. These results suggest the possibility of new approaches to modulate viral gene expression for therapeutic purposes as well as providing new insights into physiologic stimuli that may trigger EBV reactivation in vivo