There have been sufficient reports on the recognition of DNA viruses by cGAS-STING signal. It has been demonstrated that cGAS-STING induces type I IFN production and further inhibits cytomegalovirus (CMV) replication in primary human endothelial cells (41). In the central nervous system, the activation of the cGAS-STING pathway suppresses herpes simplex virus 1 (HSV-1) replication in mice microglial cells (42). Moreover, the replication of hepatitis B virus (HBV) is inhibited due to activation of the cGAS-STING pathway in both human liver cell lines and in vivo mouse models (43). Another study also found that high-level expression of STING restricts susceptibility to HBV by mediating type III IFN induction (44). African swine fever virus (ASFV) is a complex, cytoplasmic double-stranded DNA (dsDNA) virus currently expanding worldwide. The cGAS-STING pathway is efficiently activated during NH/P68 attenuated strain infection, producing large amounts of IFN-β to inhibit ASFV replication. In contrast, the virulent Armenia/07 virus blocks the synthesis of IFN-β by impairing STING activation during infection (45). However, with further research, cGAS-STING has also been confirmed to play an important role in the response to RNA virus infection. In 2013, Schoggins et al. used an ectopic expression system to verify that cGAS also widely inhibits several RNA viral infections (36). During the human immunodeficiency virus (HIV) infection, cGAS senses its RNA-DNA hybrid and dsDNA, inducing IFN production to inhibit virus replication via the cGAS-STING pathway (46). During the SARS-CoV-2 infection, virus spike (S) protein induced cell fusion and then damaged nuclei to form micronuclei. The micronuclei are sensed by cGAS and lead to the activation of STING, which further induces type I IFN production (37).

As induction of type I IFN mediated by the cGAS-STING axis is crucial for host antiviral responses, viruses have evolved various strategies to antagonize this signalling pathway for immune evasion ( Table 1 ). Numerous evasion mechanisms and immunomodulators have been identified in DNA viruses that target cGAS-STING signalling. It has been found that herpesviruses employed multiple strategies to antagonize the cGAS-STING pathway for immune evasion. The herpesvirus family includes HSV, CMV, varicella zoster virus (VZV), human herpesvirus (HHV), and Epstein Barr Virus (EBV), which are all DNA viruses. Human CMV (HCMV) tegument protein UL82 was reported to impair the translocation of STING from the ER to perinuclear microsomes and inhibit the recruitment of TBK1 and IRF3 to STING (58). Moreover, HCMV US9 was confirmed to disrupt STING oligomerization and STING-TBK1 association and block IRF3 nuclear translocation (59). The HSV-1 protein ICP27 interacts with the STING-TBK1 complex to inhibit IRF3 phosphorylation (53). The tegument proteins UL41 and UL46 of HSV-1 directly degrade cGAS mRNA or inhibit TBK1 activation, respectively (54, 55). Similarly, the murine CMV (MCMV) protein m152 was able to prevent the trafficking of STING from the ER to the endoplasmic reticulum-Golgi intermediate compartment (ERGIC), therefore inhibiting the interaction between STING and TBK1 (56). Pseudorabies virus (PRV) belongs to the alphaherpesvirus subfamily, which is also known as suid herpesvirus 1 or Aujeszky’s disease virus and infects a broad range of vertebrates. A recent study showed that PRV tegument protein UL13 functions as a suppressor of STING-mediated signalling to inhibit IFN production and antiviral response via recruitment of E3 ligase RING-finger protein 5 (RNF5) to induce K27-/K29-linked ubiquitination and degradation of STING (57). ASFV also uses different viral proteins to target the cGAS-STING pathway, inhibiting IFN production and escaping the innate immunity of the host. So far, it has been found that MGF360-15R (pA276R), pDP96R, pE120R, pI215L, pMGF505-7R and L83L protein encoded by ASFV target different adaptor proteins of the cGAS-STING pathway to inhibit type I IFN production (49, 60). In conclusion, maintaining high levels of IFN by ensuring the cGAS-STING activity is critical for host resistance to viral infection. Although cGAS-STING is considered the most potent signalling pathway to induce IFN, Kiran et al. found that JEV-induced type I IFN is cGAS-STING-independent (28). Most researchers believe that TLR and RLR are the main factors that induce IFN production. Interestingly, increased viral load was observed in a cGAS-depleted environment when IFN-β levels were still high. It suggested that the abundance of IFN-β transcripts was not sufficient alone to restrict viral replication. Therefore, there might be additional antiviral approaches regulated by the cGAS-STING signal. With the deepening of research, multiple functional roles and specific mechanisms of cGAS-STING during viral infections were identified, especially its effects on inflammation, oxidative stress and cell death.

Recognition of viruses by PRRs causes the interaction of many adaptor molecules, which in turn initiate inflammatory signalling, including the NF-κB pathway, the JAK-STAT pathway, and the inflammasome pathway. The NF-κB pathway is thought to be the regulatory centre of the inflammatory response process. The NF-κB signalling pathway is involved in a variety of stress responses during viral infection, which in turn mediates various transcriptional processes and ultimately induces pro-inflammatory cytokine production. The SARS-CoV-2 infection causes varying degrees of respiratory symptoms and results in lung damage or even death in a significant number of cases. These severe cases are associated with high levels of pro-inflammatory cytokines and low antiviral responses (61). A recent study reported that in SARS-CoV-2 infected cells, the TBK1 and IRF3 pathways are blocked by several viral proteins. The SARS-CoV-2 infection causes mitochondrial stress/damage, DNA damage, cell death and leakage of mitochondrial DNA. These DNA activate the cGAS-STING axis and induce NF-κB activation to drive inflammatory immune response (21). cGAS-STING is recognized as a potential target for the treatment of SARS-CoV-2. And several STING-targeting drugs can attenuate the inflammatory response. The HIV/SIV (Simian immunodeficiency virus) research study showed that its Vpx proteins efficiently inhibit cGAS–STING-induced NF-κB signalling but not IRF3 activation, which further induces the production of several pro-inflammatory cytokines (62). In addition. ASFV protein pD345L has been found to suppress cGAS/STING-induced NF-κB activation (63). It is well known that NF-κB is the predominant regulator of inflammation and cGAS-STING can drive NF-κB activity during viral infections (21). Therefore, the role of cGAS-STING signalling in mediating inflammatory responses deserves more attention.

NLRs also have powerful effects on inflammation induction. It has been proved that several NLRs, including NLRP1b, NLRP3, NLRC4, NLRP6 and NLRP12, are involved in the formation of inflammasome and regulate innate antiviral immunity. When viruses invade cells, NLRs recognize viral nucleic acids or endogenous molecules released from damaged or dying cells. Then, NLRs oligomerize and recruit pro-caspase-1 with or without ASC to form inflammasomes. In the inflammasome complex, caspase-1 can activate self-cleavage, and the activated caspase-1 cleaves pro-IL-1 and pro-IL-18 for their maturation and release. These mature pro-inflammatory cytokines then exert their antiviral function. IFN and pro-inflammatory cytokines are produced and function simultaneously during the host antiviral responses. Importantly, balance type I IFN production and inflammasome activation pathways are essential for immune homeostasis. Upon infection with HSV-1 or cytosolic DNA stimulation, STING engages with NLRP3, facilitating inflammasome activation via dual mechanisms (64). On one hand, STING recruits NLRP3 and promotes the localization of NLRP3 in the endoplasmic reticulum, thus promoting the formation of an inflammasome. On the other hand, STING interacts with NLRP3 to attenuate NLRP3 polyubiquitination associated with K48 and k63, thereby promoting inflammasome activation. It is widely known that the assembly of the NLRP3 inflammasome leads to the activation of caspase-1, which further results in the production of several pro-inflammatory cytokines. Caspase is the important link between inflammasome and inflammatory cytokines. Wang et al. found that caspase-1 interacted with cGAS to inhibit IFN production in DNA virus infection (65). This study also demonstrated that deficiency in inflammasome signalling enhanced host resistance to DNA viruses in vitro and in vivo. Moreover, this regulatory role also extended to other inflammatory caspases, including Caspase-4, 5, and 11 (65). These Caspases cut cGAS in conditions of non-canonical inflammasome activation. ZIKV, an RNA virus, has been reported to promote NLRP3 inflammasome activation to benefit its infection by stabilizing caspase-1 to suppress cGAS-mediated type I IFN signalling (31). The detailed mechanism is that the non-structural protein NS1 of ZIKV recruits the host deubiquitinase USP8 to cleave K11-linked poly-ubiquitin chains from caspase-1 at Lys134 to inhibit the proteasomal degradation of caspase-1. The enhanced stabilization of caspase-1 by NS1 promotes the cleavage of cGAS to inhibit the recognition of releasing mitochondrial DNA and then suppress type I IFN signalling. In addition, the activation of human caspase-3, an apoptotic caspase, has been demonstrated to cleave cGAS at D319, IRF3 at D121/125 and MAVS at D429/490, thus making apoptotic cells immunologically silent and negatively regulating DNA or RNA virus-induced cytokine production (66). Currently, there are few studies on the interaction between cGAS-STING and inflammasome signalling in viral infection, but the available evidence already suggests that the interplay between the cGAS-STING pathway and inflammasome complex affects IFN, inflammation and cell death. Therefore, this aspect deserves more attention.

Oxidative stress is an important pathological factor causing tissue damage, aging, tumours, and cardiovascular diseases. Under normal circumstances, oxidation and antioxidation are maintained in a balanced state. The oxidative and antioxidant systems in the body are disordered when harmful substances stimulate the organism. Excessive production of highly reactive molecules such as ROS and reactive nitrogen species (RNS) leads to the inhibition of antioxidant capacity, which tilts the equilibrium toward oxidation, resulting in oxidative stress. Oxidative stress is always associated with viral infections. Viral infection-induced ROS generation triggers oxidative stress in the organism and mediates apoptosis, which in turn mediates ROS and causes extensive damage, aggravating the disease process (67). For example, oxidative stress is a major characteristic of asthma and chronic obstructive pulmonary disease (COPD), and rhinovirus infection can make their condition worse. Oxidative stress attenuated the antiviral capacity of bronchial epithelial cells in asthma and COPD patients. Furthermore, oxidative stressor H₂O₂ could down-regulate the expression of epithelial cellular PRR TLR3 and antioxidants (SOD1 and SOD2), which suggested that ROS might have reduced the host’s antiviral capacity and promoted viral infection (68). But in some other studies, to a certain degree, oxidative stress activates the antioxidant defence system and autophagy in the tissues and organs, which help to scavenge some of the ROS and induce stress defence (69). Oxidative stress-induced ROS can also activate autophagy and apoptosis through various specific mechanisms, which induce cell death and inhibit virus replication. Moreover, H₂O₂ has been confirmed to regulate autophagy by inhibiting the autophagy-related gene (ATG) 4, which affects the lipidation of light chain 3 (LC3) and the degradation of pathogens (70). In addition, Latent Membrane Protein 1, a major EBV protein, facilitates ROS production, causes DNA damage and induces autophagy initiation (71). These studies suggest oxidative stress affects viral infection by directly regulating viral survival or indirectly affecting virus infection via apoptosis and autophagy.

It is widely known that the invading DNA virus will activate the cGAS-STING pathway, inducing type I IFN production and causing a range of innate immune responses. In recent studies, it has been found that STING is an upstream regulator of cellular oxidative stress. It is possible to regulate the level of lipid peroxidation and ROS by activating the cGAS-STING downstream signal ISG15. ISG15 is a member of the ISG family that induces IFN expression, contributes to “protein ISGylation”, and interferes with ubiquitin modifications. STING can negatively regulate the ubiquitin-proteasome system through ISG15, resulting in increased interferon-mediated ROS (72, 73). Indeed, IFN-mediated protein ISGylation regulates the ubiquitin-proteasome system to increase cellular ROS. Furthermore, glutathione peroxidase (GPX), an antioxidant molecule, attenuates oxidative stress by reducing H₂O₂ to water, which is also inhibited by ISG15. STING knockdown elevates glutathione peroxidase (GPX) activity via inhibition of ISG15. Recently, Hayman et al. also found the knockdown of STING down-regulated expression of ISG15 and ROS-related genes, including HECT domain and RCC1-Like Domain-Containing Protein 5 (HERC5), kruppel-like factor 4 (KLF4), and dual oxidase 2 (DUOX2) (73–76). These results suggest that STING is an upstream regulator of the intracellular oxidation processes. However, it is worth noting that some other studies believe that oxidative stress is an important inducement of cGAS-STING activation. During HSV-1 infection, GPX4 is indispensable for cGAS-STING activation. Actually, GPX4 inactivation leads to cellular lipid peroxidation, which decreases host innate antiviral immune responses and promotes virus replication via inhibition of the cGAS-STING signalling axis (22). Mechanistically, GPX4 inactivation did not affect the binding of viral DNA to cGAS but suppressed the trafficking of STING to the Golgi apparatus by facilitating STING carbonylation at C88. Another interesting study showed ROS promoted the replication of murine gammaherpesvirus-68 (MHV68), a close genetic relative of KSHV and EBV. ROS suppressed the production of IFN in a STING-dependent manner (77). ROS inhibits STING dimerization by oxidizing Cysteine 147 on murine STING during MHV68 infection. Redox modification of STING is an important regulatory mechanism of STING activity during viral infection. It is generally known that viral infection usually leads to oxidative stress in host cells, including SARS-COV-2, influenza virus and Hepatitis C virus (HCV) (15, 78–80). Oxidative stress is closely related to mitochondrial dysfunction, which triggers mitochondrial DNA (mtDNA) damage and DNA leakage, activating the cGAS-STING pathway ( Figure 1 ) (81). It remains uncertain whether oxidative stress is the cause or the consequence of cGAS-STING signalling activation during viral infection. And whether oxidative stress induces STING activity or inhibits STING activation is also controversial. Meanwhile, there are few reports about the direct interaction between the cGAS-STING signal and oxidative intermediates. Therefore, more research is needed to explore their relationship. However, it must be admitted that the alteration of the levels of oxidative stress affects the cGAS-STING pathway and host antiviral immunity.

To accommodate the diverse needs of metabolism, intracellular substances are constantly synthesized and degraded to maintain homeostasis. Autophagy is an evolutionarily conserved metabolic process of eukaryotic cells that degrades or recycles intracellular proteins and organelles and plays a key role in activating and regulating early immune responses during viral infection (82). PRR signals interact with autophagy adaptor proteins to regulate a series of immune responses, which effectively eliminates pathogenic microorganisms. For example, activation of the TLR-MYD88/TRIF pathway can disrupt the interaction between B cell lymphoma-2 (BCL-2) and Beclin-1, which induces autophagy (83). The recognition of VSV and SeV by TLR7 requires the transport of cytosolic viral replication intermediates into the lysosome. ATG5 deletion would reduce TLR7-mediated IFN production (84). Many studies have suggested that autophagy can degrade viral components, particles, and host factors, which functions as an effective innate antiviral mechanism. HCV non-structural 5A (NS5A) protein, which is crucial for HCV replication, can be degraded in autophagosomes. Autophagy helps to remove HCV in the presence of ER protein Scotin (85). Autophagy facilitates selectively degrading the HIV-1 transactivator Tat, inhibiting viral transcription and virion production in CD4⁺ T cells (86). There are also many other viruses, such as hepatitis B virus, porcine epidemic diarrhoea virus, and ZIKV, that are restricted by autophagy (87–89). On balance, viral infection is detected by multiple signalling pathways, and further triggers the activation of immune defences via autophagy.

Earlier studies have mainly focused on the mechanism of IFN induction by the cGAS-STING pathway. This signalling pathway plays antiviral effects by regulating antiviral gene transcription. Of interest, a growing number of researchers find cGAS-STING signalling axis regulates virus clearance more immediately through autophagy. Of which, the main function of STING in combatting HSV-1 infection seems to be attributed to autophagy activation rather than type I IFN production (90). A study found a mouse model harbouring a serine 365-to-alanine (S365A) mutation in STING remained resistant to HSV-1, despite the loss of STING-induced IFN activity. It seems that the activation of autophagy, triggered by STING, is contingent upon CTT and TBK1, yet remains uninfluenced by IRF3. Therefore, understanding the molecular mechanism of autophagy regulation by the cGAS-STING pathway is crucial. Saitoh et al. first found that the dsDNA of pathogenic microorganisms could induce co-localization of STING, ATG9a and LC3, which are important autophagy proteins (91). Subsequently, STING was identified as an essential factor that triggered autophagy under the stimulation of microbial DNA, which could degrade pathogens by delivering them to autophagosomes (92). Furthermore, a study showed that cGAS could directly bind to the coiled-coil domain of Beclin-1, which is a pivotal protein for autophagy initiation (93). As a result, this interaction inhibits the synthesis of cGAMP and IFN and promotes autophagy-mediated cytosolic DNA degradation by releasing Rubicon from the Beclin-1 complex. Notably, Gui et al. explained the mechanism of STING-mediated autophagy without TBK1 activation and IFN induction (24). When pathogenic microorganisms infect cells, cGAS recognizes cytosolic DNA and synthesizes cGAMP, which further binds to STING. As a result, STING translocates to the ERGIC by interaction with SEC24C. Then, ERGIC acts as a membrane source for LC3 lipidation, promoting autophagosome formation that degrades the DNA virus. In many invertebrates, such as drosophila and sea anemone Nematostella vectensis, their STING only participates in autophagy induction but not IFN response (24, 94). These research suggest that autophagy induction is an evolutionarily conserved function of the cGAS-STING signalling axis which predates the emergence of the IFN signalling. Additionally, the structural analysis showed that STING had a conserved LIR domain which was exposed to the cytoplasm by conformational changes upon activation (95). Consequently, the exposed LIR domain could directly interact with LC3 to activate autophagy, leading to the degradation of STING itself and p-TBK1. This finding also showed STING could directly link immune activation to autophagy. So quite a few researchers believe that autophagy induction via STING trafficking is a primordial function of the cGAS pathway, which has long been thought that its primary function is to induce type I IFN production. Further studies show that STING orchestrates endoplasmic reticulum stress and the unfolded protein response via a novel UPR motif within the cyclic-dinucleotide-binding (CBD) domain. This motif exerts a negative regulatory effect on the Akt/tuberous sclerosis complex (TSC)/mammalian target of the rapamycin (mTOR) pathway, thereby amplifying canonical autophagy (25, 96). Several other studies also revealed the mechanisms of STING-mediated noncanonical autophagy (97, 98). Activated STING translocates from the endoplasmic reticulum to the ER-Golgi intermediate compartment and Golgi apparatus, contingent upon the coat protein II (COP II) complex and Arf GTPases. The ERGIC serves as a membrane reservoir for LC3 lipidation and the genesis of autophagosomes. Different from canonical autophagy, STING-elicited noncanonical autophagy operates independently of upstream autophagy modulators, including unc-51-like kinase 1 (ULK1), Beclin-1, and ATG9a, yet relies on downstream autophagy regulators such as ATG5 and ATG16L1 (24).

Viruses have also evolved unique mechanisms to ensure their survival by influencing autophagy processes and cGAS-STING signalling. ASFV MGF505-7R, MGF505-11R and L83L proteins promote autophagy-lysosomal degradation of STING, thereby blocking the phosphorylation of the downstream signalling molecules TBK1 and IRF3 and impairing type I IFN production (51, 60, 99). PCV2 infection can induce cGAS degradation via the autophagy-lysosome pathway (100). Mechanically, PCV2 infection triggers the phosphorylation of cGAS at S278 through the PI3K/Akt pathway. This phosphorylation of cGAS promotes the K48-linked poly-ubiquitination of cGAS which interacts with autophagy receptor p62 for autophagic degradation in autolysosome. As a result, the autophagic degradation of cGAS inhibits cGAMP and IFN-β production, which further impair hosts’ innate antiviral responses. Similarly, HBV X protein also can inhibit type I IFN production by boosting ubiquitination and autophagic degradation of cGAS (101). Except for cGAS, another autophagy receptor Coiled-coil domain containing 50 (CCDC50) associates with and targets STING for autophagic degradation (102). The MIU motifs of CCDC50 can recognize K63-polyubiquitinated STING and facilitate the conveyance TBof K63-polyubiquitinated STING to LC3B-marked autophagosomes, subsequently initiating autophagic degradation in a p62-independent way. During HSV-1 infection, the absence of CCDC50 promotes IFN and pro-inflammatory cytokines production and inhibits HSV-1 replication. These results suggest the autophagic degradation of cGAS-STING signalling during infections has a significant impact on type I IFN production and viral replication ( Figure 2 ).

Although many studies have been conducted on viral evasion of STING-induced IFN-mediated antiviral function, investigations about viral evasion of STING-induced autophagy-mediated antiviral function remain notably limited. Recently, an interesting study found that the bat STING can only induce autophagy and antiviral activity but not IFN induction (103). SARS-CoV-2 ORF3a constitutes a distinctive viral protein capable of interacting with STING, consequently disrupting the STING-LC3 association and impeding cGAS-STING-mediated autophagy, whilst preserving IRF3-Type I IFN induction. This novel functionality of ORF3a, different from targeting autophagosome-lysosome fusion, is a selective impediment of STING-mediated autophagy, thereby promoting viral proliferation. In addition, the interaction between the TBK1-IRF3-IFN pathway downstream of cGAS-STING and autophagy in viral infections is very complex. During infection, excessive accumulation of STING will trigger a strong inflammatory reaction, leading to deleterious effects on the host (93, 104). When the cGAS-STING pathway is activated, TBK1-IRF3 signalling downstream of STING will phosphorylate p62 at S403, which has a remarkably high affinity for ubiquitinated STING. As a result, the ubiquitinated STING is degraded in autophagosomes in an IRF3‐dependent manner (105). Moreover, another research group also found that TBK1 could phosphorylate selective autophagy receptors optineurin (OPTN), NDP52, and TAX1BP1 linking ubiquitinated cargo to autophagic membranes (106). As is known to us all, type I IFN participates in activating JAK/STAT and PI3K/Akt pathways, which are always involved in autophagy induction (107). Type I IFN does not induce autophagy in STAT-deficient cells (108). PI3K/Akt signalling axis inhibits autophagy by activating mTORC1 and inhibiting the expression of forkhead box O (FOXO) and autophagy-related genes. At later time points, negative regulators of the PI3K/AKT/mTOR pathway are induced, inhibiting mTORC1 activity and inducing autophagy (107). Therefore, the TBK1‐IRF3-IFN axis plays a crucial role in activating and regulating the host’s immune and autophagy. At present, there are still some problems plaguing us. Some researchers believe cGAS-STING-mediated autophagy plays an antiviral role (90, 103, 109). But other research groups suggest that cGAS-STING-mediated autophagy contributes to inhibiting the antiviral function of the host by degrading cGAS/STING directly or by degrading key proteins downstream of the cGAS-STING pathway ( Table 2 ) (51, 100, 117, 118). As a result, this process inhibits IFN production. The next question that needs to be solved is how to control the target of autophagic degradation in viral infection.

There is adequate evidence that the state of the endoplasmic reticulum also influences a variety of selective autophagy, including mitophagy and ER-phagy (119, 120). To further expand our understanding of the effects of cGAS-STING on autophagy and oxidative stress, we focus on the endoplasmic reticulum. The endoplasmic reticulum is a continuous membrane system widely distributed in the cytoplasm. It mainly performs the functions of intracellular material transport, glucose and lipid metabolism, and protein processing. In addition, ER also provides a membrane structure for the formation of autophagosomes and peroxisomes. Many viruses use the ER as a replication site, where they synthesize proteins, replicate genomes, and assemble virion (121). ER stress is usually triggered by calcium homeostasis disequilibrium, unfolded protein (UPR) accumulation and lipid dysregulation (122). ER stress is also considered to be a potential cause of mitophagy and ER-phagy (123, 124). Accumulation of viral proteins in ER can also induce ER stress (125). ER stress initiates UPR-mediated protein degradation pathways, apoptosis and autophagy in host cells, inhibiting or degrading the accumulation of viral proteins to maintain cellular homeostasis. Generally, different viruses selectively activate the PERK (proline-rich extensin-like receptor kinase)-eIF2α (eukaryotic translation initiation factor 2) pathway, IRE1α (inositol-requiring enzyme 1α)-XBP1 (X-box binding protein-1) pathway or ATF6 (activating transcription factor 6) pathway, leading to ER stress. Transmissible gastroenteritis virus (TGEV) can activate the PERK-eIF2α signalling pathway and subsequently diminish the synthesis of viral proteins by decreasing protein translation efficiency (126). The HCV negatively regulates ER stress via the IRE1α-XBP1 pathway, increasing the synthesis of viral proteins and facilitating viral infection (127). Influenza A virus promotes viral replication by inhibiting ER stress response factor XBP1 and limiting host protein production to alleviate ER stress (128). However, the UPR remains to be a double-edged sword during viral infection. Some viruses regulate UPR to promote survival by activating other cellular responses. For example, duck enteritis virus (DEV) can activate ER stress and autophagy in a PERK-eIF2α/IRE1α-XBP1 dependent manner. Inhibiting the expression of PERK and IRE1 helps to suppress autophagy and DEV replication (129).

During viral infection, the viral DNA can be recognized by the cGAS and activates the cGAS-STING pathway, triggering a series of immune and cellular responses to protect the body, including ER stress (23). Notably, the inactivated STING is located on the outer membrane of the ER, and the migration of the activated STING and the activation of the STING-TBK1-IRF3 signal always occur simultaneously with ER stress. Several recent studies have shown a partial overlap between ER stress signals and the cGAS-STING signalling axis. During pathogenic microbial infection, phosphorylation of PERK was significantly impaired in STING-deficient macrophages. STING gain-of-function mutant N154S induces chronic ER stress by disrupting Ca2+ homeostasis. A newly identified STING CTD motif is involved in mediating ER stress in an IFN-independent manner (96). Similarly, deletion of the Ca²⁺ sensor STIM1 leads to spontaneous activation of the STING-TBK1-IRF3 pathway, which results in type I IFN-mediated ER stress (130). Moreover, higher levels of PERK phosphorylation were induced at times of the expression of STING (25). Furthermore, co-immunoprecipitation assay suggests that STING and PERK can interact directly and promote the removal of pathogenic microorganisms. Additionally, down-regulating the expression of PERK or IRE-1 inhibits STING activity (131). These research imply a link between ER stress and cGAS-STING signal. The interrelationship between downstream signalling molecules of cGAS-STING and ER stress was also illustrated in several reports. The activation of IRF3 by STING is initiated by ER stress (132). The signal that triggers the phosphorylation of IRF3 is derived from the ER. ER stress triggered the phosphorylation of IRF3 at S386 in an XBP1-independent manner, promoting IRF3 nuclear translocation (133). Moreover, ER stress can mobilize the ER-resident STING and facilitate the co-localization of STING and TBK1. Another research found that XBP1 splicing and IRF3 phosphorylation depend on the presence of STING (132). There is also evidence suggesting that several genes, including tyrosine kinase 2 (TYK2), STAT2 and IRF9, take part in IFN-induced ER stress, but the specific mechanism is still unclear (134). As yet, studies on the interaction between ER stress and the cGAS-STING pathway are mainly focused on metabolic and autoimmune diseases. More research is needed to further understand their role in viral infections.