Date Published: August 16, 2018
Publisher: Public Library of Science
Author(s): Guoxin Ni, Zhe Ma, Blossom Damania, Matthew J. Evans.
As the first line of host defense, the innate immune system utilizes germline-encoded receptors named pattern-recognition receptors (PRRs) to detect invading pathogens. PRRs recognize conserved molecular structures of pathogens known as pathogen-associated molecular patterns (PAMPs) to initiate immune responses that counteract pathogen infection. The immunostimulatory feature of exogenous nucleic acids, such as viral DNA and RNA, has been known for more than half a century, but the mechanism by which they function as an immune stimulant remained unclear for a long time. The past two decades have witnessed tremendous progress in understanding the signaling mechanisms of innate immune networks and established the retinoic acid inducible gene-I (RIG-I)/melanoma differentiation associated gene 5 (MDA5)–mitochondrial antiviral-signaling protein (MAVS) axis and cyclic GMP-AMP synthase (cGAS)–stimulator of interferon genes (STING) axis as the major sensing pathways for cytosolic RNA and DNA, respectively . However, emerging evidence indicates that, in addition to its well-established role in sensing cytosolic DNA, the cGAS–STING pathway is also involved in restricting RNA virus infection, suggesting that there exists crosstalk between the innate sensing of cytosolic DNA and RNA.
cGAS binds to cytosolic double-stranded DNA (dsDNA) from various sources, including bacteria, DNA viruses, and retroviruses, in a sequence-independent but length-dependent manner [1, 2]. Following the binding of dsDNA, cGAS catalyzes the production of a second messenger known as cyclic guanosine monophosphate (GMP)-adenosine monophosphate (AMP) (cGAMP) in the presence of GTP and ATP, which subsequently binds to the adaptor protein STING on the endoplasmic reticulum (ER) membrane . Aside from cGAMP, STING also directly senses other cyclic dinucleotides (CDNs), which are secreted by some bacteria . After binding to CDNs, the STING dimer undergoes a dramatic trafficking process from the ER to the Golgi complex and eventually to perinuclear compartments to form large punctate structures where it is degraded . STING recruits TANK binding kinase 1 (TBK1) and activates transcription factors interferon regulatory factor 3 (IRF3) and nuclear factor-κB (NF-κB), which then translocate into the nucleus to induce the transcriptional activation of type I interferons (IFNs) and other inflammatory cytokines, thus establishing an antiviral state in infected and uninfected neighboring host cells  (Fig 1). Numerous DNA viruses have been reported to activate the cGAS–STING pathway, and cGAS or STING deficient mice are more susceptible to lethal infection after exposure to many DNA viruses, including herpes simplex virus 1 (HSV-1), vaccinia virus, and murine gammaherpesvirus 68 (MHV68) . Infection of retroviruses such as human immunodeficiency virus (HIV) generates RNA:DNA hybrids and dsDNA in the cytosol that can also activate the cGAS–STING pathway [6, 7]).
The question of whether cGAS and STING are also engaged in antiviral responses to RNA viruses has been asked since the very beginning. A quick answer to this question would be yes because studies have shown that deficiency of cGAS or STING in cells or mice greatly facilitated replication of several RNA viruses, such as vesicular stomatitis virus (VSV), Sendai virus (SeV), dengue virus (DENV), and West Nile virus (WNV) [8–11]. However, exactly how cGAS and STING are involved in RNA virus-induced immune responses is largely unknown. Although cGAS was reported to bind dsRNA, this interaction did not lead to the production of cGAMP . Moreover, cGAS deficiency does not affect SeV-induced IFNβ production . Therefore, although cGAS restricts replication of some RNA viruses, it is not required for RNA virus-induced type I IFN responses. One recent study reported that DENV infection led to mitochondria damage and release of mitochondrial DNA into the cytosol, which then activated the cGAS–STING pathway to potentiate host defense responses . This finding provides a possibility that cGAS might play an indirect role in restricting RNA virus infection.
Considering the central role of cGAS and STING in the innate DNA sensing pathway, it’s not surprising to find that many DNA viruses have evolved effective strategies to antagonize the function of cGAS and STING in order to facilitate their replication in host cells  (Fig 1). Binding of viral DNA is the first step of the DNA sensing pathway. The tegument protein open reading frame 52 (ORF52) of several gammaherpesviruses, including Kaposi’s sarcoma-associated herpesvirus (KSHV), Epstein–Barr virus (EBV), and MHV68, was found to interact with cGAS and disrupt its binding to viral DNA, thus inhibiting activation of the cGAS–STING pathway . Another KSHV protein, latency-associated nuclear antigen (LANA), was also reported to interact with cGAS and antagonize the cGAS–STING-dependent signaling . STING seems to be an even more favorable target of many DNA viruses, probably because of its essential role in transducing signaling from not only cGAS but also other DNA sensors . HSV-1 regulatory protein infected cell protein 27 (ICP27) interacts with STING and TBK1 and thereby prevents the phosphorylation of IRF3 by TBK1 . Another HSV-1 protein ICP0 was reported to promote degradation of interferon-γ-inducible protein 16 (IFI16), which detects HSV-1 DNA in human fibroblasts, thereby blocking its activation of downstream STING-dependent signaling . Interestingly, HSV-1 appears to induce infected cells to export exosomes containing STING . KSHV viral interferon regulatory factor 1 (vIRF1) was also found to interact with STING and prevent its interaction with TBK1, thereby inhibiting TBK1-mediated phosphorylation and activation of STING . Two oncoviruses, human papillomavirus 18 (HPV18) and human adenovirus 5 (hAd5), have been shown to inhibit STING activity using their viral oncoproteins E7 and E1A, respectively . Moreover, the Hepatitis B virus (HBV) polymerase was found to bind STING and attenuate its K63-linked polyubiquitination and function .
Although it is clear that cGAS is essential in cytosolic recognition of DNA viruses, its role in restricting RNA virus infection is still inconclusive. On the other hand, it is now apparent that STING is required for host responses against both DNA and RNA viruses. Studies of mechanisms employed by STING to restrict RNA virus infection have only just begun. The study which unraveled STING’s function in protein translation inhibition has shed light on our understanding of how the cGAS–STING pathway functions in RNA virus restriction . However, the underlying mechanisms still need to be further elucidated. Several outstanding questions remain: How does RIG-I/MDA5 transduce a signal to STING after sensing viral RNA? Are there any other cofactors involved in this process? It is known that STING activation requires binding of CDNs. However, cytosolic RNA does not activate cGAS to generate cGAMP, and STING does not undergo any posttranslational modifications or trafficking during RNA virus infection. This begs the question as to how STING is activated to initiate translation inhibition. Furthermore, what is the strategy that STING uses to inhibit translation if it is eIF2α-independent? Answers to these questions and others will further expand our understanding of how the cGAS–STING pathway deploys immune defenses to detect and eliminate viral infection and how viruses evade or inhibit activation of this pathway. Lessons learned from this will greatly facilitate the development of new vaccines and antivirals for the prevention and treatment of infectious diseases.