Regulation of the unfolded protein response, endoplasmic reticulum stress and autophagy pathways in the host response to coronavirus infection

Abstract

Coronaviruses (CoV) are a group of RNA viruses that have continuously posed risks to humans’ health and economy. Since the first recorded major CoV outbreak of SARS-CoV in 2001, using modern scientific methods, tremendous effort and resources have been devoted to better understanding the transmission of these viruses to humans and the associated disease manifestations. Like all known viruses, CoV require and utilize intracellular signaling pathways to replicate in the host cell. In doing so, CoV have apparently evolved diverse strategies to modulate these systems to initiate the formation of novel pathogenic intracellular structures, including so-called double membrane vesicles (DMVs), which collectively have been termed replicative organelles (ROs). DMVs and ROs are essential for successful CoV replication. Their formation, together with massively increased synthesis of viral protein and RNA components, leads to pronounced activation of affected cells. Among the cellular processes modulated early and most strongly during CoV infection are those related to the endoplasmic reticulum (ER) stress response and autophagy. Activation of ER signaling pathways by CoV results in ER stress and triggering of the unfolded protein response (UPR). The UPR is an adaptive process by which the cell attempts to restore normal ER function through activation of protein kinases, transcription factors, and downstream genetic programs. This involves a number of well-characterized molecular sensor and effector molecules anchored in the ER membrane, which include the protein kinases PERK and IRE1α and the chaperone BiP (GRP78, HSPA5). Autophagy refers to a highly dynamic process by which the cell can degrade larger molecular complexes, organelles, or invading pathogens and recycle or destroy the resulting macromolecules. As part of the autophagy process, a double membrane structure (the autophagosome) is formed, which interacts with the proteins LC3B and p62/SQSTM1 (sequestosome1) to coordinate basal, as well as selective, autophagy flux in response to various stressors, including viral infections. In this work, the central questions were investigated to what extent (i) CoV specifically modulate the ER stress and autophagy system and whether (ii) new starting points for antiviral strategies can be derived from this. To this end, a series of specific molecular but also proteome-wide analyses were performed in pharmacologically and genetically perturbed cellular model systems. Small molecule inhibitors were used to examine the effect of inhibiting PERK, IRE1α, or both protein kinases on CoV replication as well as on activation of the UPR and host response. One of the major findings of this approach was that inhibition of PERK resulted in a reduction in replication of HCoV-229E and MERS-CoV, but not SARS-CoV-2. HCoV-229E infection of Huh7 cells resulted in induction of the transcription factor ATF3 and suppression of BiP. PERK inhibition abolished both of these effects and led to a reduction in phosphorylation of serine 52 of the translation-inhibitory factor eIF2α. In contrast, inhibition of IRE1α resulted in only a small reduction in HCoV-229E replication but completely prevented virus-induced ATF3 induction. Arteficial, "chemical" activation of the UPR by the natural product thapsigargin resulted in strong inhibition of replication of all three CoV tested with a half-maximal effective concentration (EC50) in the low nanomolar range with concomitant low cytotoxicity. Furthermore, thapsigargin treatment of CoV-infected cells was associated with partial abrogation of global HCoV-229E-induced inhibition of protein biosynthesis and resulted in almost complete inhibition of CoV-induced selective and basal autophagy flux. Subsequently, a genetic approach based on the CRISPR-CAS-9 system was used to further investigate the roles of PERK, IRE1α, and ATF3 in CoV replication. Silencing of PERK resulted in a decrease in HCoV-229E replication, whereas suppression of IRE1α or ATF3 showed very little or no effect on virus replication. In addition, mass spectrometric techniques were used to comparatively examine changes in the proteome in CoV-infected and thapsigargin-treated cells. Bioinformatic analyses of the differentially expressed proteins resulted in the identification of a number of specific metabolic processes, signaling pathways, and factors that may be responsible for the antiviral effect of thapsigargin, including ER-associated degradation (ERAD) and ER quality control (ERQC) pathways. In the final part of the work, a proximity-based, proteome-wide interaction screen was performed to reveal the intracellular binding partners of ATF3 in the context of HCoV-229E infection. This led to the identification of a number of components of the immune system and mitochondrial homeostasis that are potential ATF3-dependent regulated effector molecules. In summary, the results obtained in this work provide insight into the CoV-specific activation patterns of the ER stress factors PERK, IRE1α, BiP, and ATF3 and their role in CoV replication and host response. The extensive evidence for antiviral efficacy of thapsigargin establishes chemical activation of the ER stress system as a novel antiviral therapeutic principle against enveloped RNA viruses such as HCoV-229E, MERS-CoV, and SARS-CoV-2 and identifies thapsigargin as a new prototype of compounds with multimodal host-directed antiviral activity based on molecular mechanisms of action.