The STING Signaling: A Novel Target for Central Nervous System Diseases

Abstract

The canonical cyclic GMP-AMP (cGAMP) synthase (cGAS)-Stimulator of Interferon Genes (STING) pathway has been widely recognized as a crucial mediator of inflammation in many diseases, including tumors, infections, and tissue damage. STING signaling can also be activated in a cGAS- or cGAMP-independent manner, although the specific mechanisms remain unclear. In-depth studies on the structural and molecular biology of the STING pathway have led to the development of therapeutic strategies involving STING modulators and their targeted delivery. These strategies may effectively penetrate the blood–brain barrier (BBB) and target STING signaling in multiple central nervous system (CNS) diseases in humans. In this review, we outline both canonical and non-canonical pathways of STING activation and describe the general mechanisms and associations between STING activity and CNS diseases. Finally, we discuss the prospects for the targeted delivery and clinical application of STING agonists and inhibitors, highlighting the STING signaling pathway as a novel therapeutic target in CNS diseases.

Graphical Abstract

STING Signaling Pathways

The cGAS-STING Pathway

The cGAS, first reported in 2012 (Sun et al. 2013), is a cytosolic DNA sensor that identifies exogenous DNA (from dying cells, exosomes, DNA viruses, retroviruses, and bacteria) and mitochondrial DNA (mtDNA) (Galluzzi et al. 2018). In mammals, cGAS contains a long amino-terminal unstructured region followed by a nucleotidyltransferase domain, which is activated by the binding of the sugar-phosphate backbone of double-stranded DNA (Sun and Hornung 2022). Once activated, cGAS induces conformational changes that lead to the cyclization of ATP and GTP into the second messenger, 2′3′- cGAMP (Gui et al. 2019; Zhang et al. 2019). Subsequently, the second messenger cGAMP translocates to the endoplasmic reticulum (ER) to activate the stimulator of interferon genes (STING). Then STING transfers to ER-Golgi intermediate compartments and forms tetramers by a higher-order oligomerization (Kwon and Bakhoum 2020). Activated STING plays a crucial role in engaging both the canonical and non-canonical NF-κB signaling pathways. In the canonical signaling pathway, STING recruits the IκB kinase (IKK) complex, which consists of IKKα, IKKβ, and NEMO. This complex phosphorylates IκBα, leading to the release of the RELA (p65)/p50 dimer, which then translocates to the nucleus to initiate the transcription of pro-inflammatory genes (Abe and Barber 2014). On the other hand, in the non-canonical signaling pathway, STING activates NF-κB-inducing kinase (NIK, mitogen-activated protein kinase kinase kinase 14,MAP3K14). NIK phosphorylates IKKα, which processes p100 into p52. This processing allows the RELB/p52 dimer to activate different transcriptional programs (Hou et al. 2018). STING is also phosphorylated by tank-binding kinase 1 (TBK1), which recruits interferon (IFN) regulatory factor 3 (IRF3) to induce expression of type I IFN (IFN-I) and other cytokines (Liu et al. 2015; Sun and Hornung 2022). ER withdrawal is necessary for trafficking-mediated STING degradation. STING is transported to the vesicles and rapidly turns off downstream signaling, and then STING vesicles are assigned to Rab7-positive endolysosomes to degrade (Gonugunta et al. 2017) (Fig. 1).

Fig. 1.

The cGAS-STING signaling pathway. The cGAS is activated following identifying exogenous or endogenous DNA, which catalyzes ATP and GTP into cGAMP. cGAMP binds to the endoplasmic reticulum adaptor protein STING to initiate the signaling cascade. Subsequently, activated STING transfers to endoplasmic reticulum-Golgi intermediate compartments and to the Golgi apparatus, where it recruits IKK to mediate canonical NF-κB targets, and also recruits TBK1 in parallel, which phosphorylates STING and IRF3 to induce type I IFN and other cytokines. Meanwhile, STING also activates NF-κB-inducing kinase (NIK, mitogen-activated protein kinase kinase kinase 14,MAP3K14) mediated non-canonical NF-κB pathway. Finally, STING is transported to the vesicles and speedily turns off downstream signaling, subsequently STING vesicles are assigned to endolysosomes to degrade

The cGAS- or cGAMP-Independent STING Pathway

The cGAS—STING pathway is the canonical mode of STING activation. However, not all STING signaling is activated by cGAS. Following cGAMP—triggered translocation, Armadillo—like helical domain—containing protein 3 interacts with STING at the Golgi and mediates the recruitment of phosphatidylinositol 4—kinase beta (PI4KB) to synthesize PI4P, which is involved by PI4P—binding proteins in the trafficking of STING from the Golgi to endosomes. Abnormally elevated cellular PI4P has been shown to lead to cGAS-independent STING activation (Fang et al. 2023). STING interacts with the lysosomal membrane protein Niemann–Pick C1 and is recruited to the lysosome for degradation. However, in the absence of Niemann–Pick C1, the lysosomal degradation of STING is blocked leading to increased STING signaling, which occurs independently of cGAS (Chu et al. 2021). Beyond NPC1, additional factors regulate lysosomal STING degradation. For example, the E3 ubiquitin ligase TRIM29 directly ubiquitinates STING, promoting its lysosomal degradation, while TRIM29 deficiency stabilizes STING and amplifies its signaling (Zhang et al. 2020; Xie et al. 2023). These findings highlight that dysregulated lysosomal homeostasis broadly impacts STING activity through multiple mechanisms.

Other DNA sensors that detect lysosomal-localized DNA and activate STING signaling have been described (Motani et al. 2015). For example, DNA binding protein IFI16 and DNA damage response factors ATM and PARP-1 can also induce cGAS-independent STING activation (Dunphy et al. 2018), while DNA-dependent protein kinase initiates cGAS-independent IFN-I responses (Taffoni et al. 2023). Furthermore, the STING/IRF7 pathway has been proved to mediate DNA vaccine immunogenicity independent of cGAS (Suschak et al. 2016). Effector memory T cells trigger dsDNA breaks and induce cGAS-independent STING signaling to drive innate inflammation (Meibers et al. 2023). Functionally, Brucella abortus, influenza A virus, and streptococcal M protein have all been shown to trigger STING but not cGAS to induce host protection (Holm et al. 2016; Costa Franco et al. 2018; Movert et al. 2018). To date, activation of STING in a cGAS- and cGAMP-independent manner remains poorly understood.

STING in Brain Tumors

Anti-tumor Effects

STING signaling plays an important role in brain tumors, and functional cytoplasmic DNA sensing via cGAS-STING pathway is lacking in the vast majority of neuroblastoma cell lines (Wolpaw et al. 2022). STING mainly enhances the signal of IFN-I in the tumor microenvironment and exerts anti-tumor effects. IFN-I mainly comes from CD11b⁺ brain-infiltrating leukocytes in de novo gliomas in mice (Ohkuri et al. 2014; Chen et al. 2022). Both PP2Ac/STRN4 and PTEN regulates STING/IFN-I signaling in the tumor microenvironment (Ho et al. 2023; Mondal et al. 2023; Dogan et al. 2024). Hypoxia promotes the release of extracellular vesicles in glioblastoma cells and miR-25/93 expression in the extracellular vesicle, subsequently, extracellular vesicle carrying miR-25/93 were absorbed by macrophages, resulting in impaired STING/IFN-I signaling (Tankov et al. 2024), Manganese can exacerbate ionizing radiation-induced signaling damage (He et al. 2023b).

Adaptive immune cell infiltration (T cells, dendritic cells, and natural killer cells) is also an important pathway through which STING exerts its effects. Temozolomide enhances DNA damage repair by upregulating GBP3 and STING and its function is enhanced in the presence of PTEN protein (Xu et al. 2022; Yildirim et al. 2024). The combination of CD47 blockade and temozolomide enhances the antigen cross-presentation and activation of STING in antigen-presenting cells, leading to more effective T-cell initiation (von Roemeling et al. 2020). Not only N-Myc inhibition and Tumor Treating Fields (Layer et al. 2017; Chen et al. 2022) but also beta- FAM131B-AS2 inhibition, mangostin and HA-MSA2, can also increase T cell activation and recruitment through STING signaling (Chellen et al. 2024; Wang et al. 2024c; Yang et al. 2024b). In addition, OMA1, GNB4 silencing, and H3.3-G34 mutations can inhibit tumor proliferation, migration, and invasion by activating STING signaling (Haase et al. 2022; Gao and Yang 2024; Zhu et al. 2024).

In animal experiments, STING agonists (IACS-8779 and 8803) demonstrate improved efficacy, tolerability, and extended median survival. Clinically, STING agonists such as ADU-S100 (MIW815) and MK-1454 have advanced into phase I/II trials for advanced solid tumors. However, monotherapy showed limited efficacy, likely due to immunosuppressive tumor microenvironments and poor intratumoral drug delivery. Recent combinatorial strategies (e.g., STING agonists with PD-1 inhibitors or radiotherapy) achieved enhanced antitumor responses in trials (NCT03172936, NCT03010176), underscoring the need for optimized delivery systems and rational therapeutic combinations (Wang et al. 2024a). In humans, STING is markedly increased in high-grade gliomas with high recurrence risk (Zhong et al. 2024), PCBP2 upregulation suppresses cGAS/STING signaling in glioma patients(Chen et al. 2022). However, STING regulators may exhibit species-specific effects (Boudreau et al. 2021; Najem et al. 2024). For instance, murine STING has been recognized as a molecular target of ASA404 (also known as 5,6-dimethylxanthenone-4-acetic acid (DMXAA)) and binds directly to ASA404 resulting in potent anti-tumor activity through activation of TBK1 and IRF3 signaling pathways. However, human STING does not bind to and interact with ASA404 to activate TBK1–IRF3 signaling (Conlon et al. 2013, Bähr et al. 2017). Besides, ASA404 as a vascular disrupting agent has been shown to cause necrosis and hemorrhage, as well as reduce tumor growth and proliferation in subcutaneous glioma, but has no relevant activity in situ brain tumors (Bähr et al. 2017). These findings highlight the potential of STING as an anti-tumor target but underscore the challenges in translating preclinical results to human therapies.

Targeted Delivery System

The BBB is a major obstacle limiting the distribution and targeting efficiency of therapeutic drugs to the brain. Studies addressing this issue have demonstrated that nanoparticles loaded with STING agonists significantly enhance STING-mediated remodeling of the immune microenvironment, potentiate anti-tumor immunity, and inhibit tumor growth in gliomas, glioblastoma multiforme, and neuroblastoma (Wang-Bishop et al. 2020; Bielecki et al. 2021; Wang et al. 2022a; Yang et al. 2022). Currently, various targeted delivery systems have been developed to improve brain tumor therapy, including dendritic cell-tumor hybrid cell-derived chimeric exosomes (Bao et al. 2024), Trojan horse nanocapsule (Zhou et al. 2022), antimiR-25/cGAMP nano complexes (Petrovic et al. 2024), copper-coordination driven brain-targeting nano assembly (Chen et al. 2024), vitamin D3-inserted lipid hybrid neutrophil membrane biomimetic multimodal nanoplatforms (He et al. 2024), a manganese-based nanodriver (Zhang et al. 2024), a pH-responsive and guanidinium-rich nanoadjuvant (Lu et al. 2025), sequential release hydrolipo system (Yu et al. 2025), and bridging-lipid nanoparticle (Zhang et al. 2023a).

These findings indicate that nanoparticles show promising efficacy as targeted delivery systems for brain tumor therapy (Fig. 2). Owing to the BBB, many drugs and modulators fail to reach the brain at therapeutic concentrations, leading to suboptimal treatment outcomes. By contrast, nanoparticles loaded with therapeutic agents can effectively cross the BBB and deliver precise treatment to affected areas. While further research is needed to develop safer and more effective nanomaterials, potential toxicity concerns—such as reproductive toxicity, metabolic dysregulation, and abnormal gene expression—must also be thoroughly investigated. The mechanisms of STING agonists are discussed in detail in "Microglia and Neuroinflammation".

Fig. 2.

Schematic diagram of nanoparticles penetrating the BBB. The pathway of nanoparticles passing through BBB to reach the brain: (a) transcellular pathway, (b) paracellular pathway, (c) endocytosis, and (d) receptor-mediated endocytosis

Tumor Brain Metastasis

Protocadherin 7 promotes the assembly of connexin 43-based carcinoma–astrocyte gap junctions, which allow metastatic cancer cells in the brain to transfer cGAMP into astrocytes to activate the STING pathway and release inflammatory cytokines including IFN-I and tumor necrosis factor (TNF). These cytokines subsequently activate NF-κB and STAT1 signaling in brain metastatic cells to promote cancer cell survival, growth, and proliferation (Chen et al. 2016, 2017). Chromosome segregation errors lead to chromosomal instability and a preponderance of micronuclei. Micronuclear rupture gives rise to the release of genomic DNA into the cytosol, and subsequent activation of the cGAS-STING pathway and downstream non-canonical NF-κB pathway. Tumor cells displaying chromosomal instability may chronically activate innate immune pathways to boost invasion and metastasis of cancer cells via disruption of the disruption of epithelial responses to cytosolic DNA (Bakhoum et al. 2018). Perhaps, whether STING activation leads to tumor inhibition or promotion may be related to activation intensity or cell types (Gulen et al. 2017). Cancer metastasis is the principal reason for cancer-associated deaths due to the lack of effective treatments. Thus, there is a critical need to increase the survival rates of patients with cancer metastasis, for which STING may hold therapeutic potential.

STING in Stroke

Microglia and Neuroinflammation

Microglia, the primary resident immune cells in the CNS, perform continuous immune surveillance, even in a resting state (M0). Upon stimulation, they polarize into pro-inflammatory (M1) or anti-inflammatory (M2) phenotypes (Li et al. 2022). In the CNS, STING is predominantly expressed in microglia. Following subarachnoid hemorrhage, cGAMP is transferred from neuron to microglia, activatING microglial IFN-I responses (Chang et al. 2024). Microglial STING expression is markedly upregulated after ischemic stroke and subarachnoid hemorrhage. Mechanistically, STING activation polarizes microglia toward the M1 phenotype, whereas inhibition of cGAS or STING (e.g., using A151, C-176, or sulfasalazine) promotes an M2 shift, exerting anti-inflammatory effects (Kong et al. 2022; Shi et al. 2022; Li et al. 2024b; Liu et al. 2024c).

After a subarachnoid hemorrhage, microglial STING recruits and activates TBK1. TBK1 subsequently undergoes autophosphorylation and phosphorylates STING, forming a feedforward loop that sustains TBK1-STING signaling. This axis inhibits AMPK activity, driving microglial activation and M1 polarization (Peng et al. 2020). Interestingly, miR-340-5p directly inhibits STING expression and TBK1 phosphorylation to attenuate neuroinflammation (Song et al. 2022). Studies have also indicated that inhibiting STING signaling after cerebral ischemic stroke not only leads to a microglial polarization switch to the M2 phenotype, microglia-mediated synapse engulfment and improves damaged neurological function (Jiang et al. 2021; Wu et al. 2024) but also alleviates NLRP3 inflammasome-mediated brain injury and microglial pyroptosis (Li et al. 2023).

The mtDNA released into the microglial cytoplasm following cerebral ischemia/reperfusion injury has been demonstrated to activate STING and subsequently facilitate microglial polarization towards the M1 phenotype by downstream pathways, including IRF3 and NF-κB (Kong et al. 2022; Li et al. 2024a). Beyond mtDNA, histone deacetylase 3 and ceramide have also been shown to activate the cGAS-STING pathway to promote the formation of a pro-inflammatory microenvironment. Mechanistically, the former promotes the transcription of cGAS through deacetylation of NF-κB p65 in the microglia, leading to the upregulation of cGAS-STING-mediated neuroinflammation (Liao et al. 2020; Huang et al. 2025). Conversely, α-lipoic acid, FGF2, and iron Q-pretreated mesenchymal stem cells suppress neuroinflammation by inhibiting cGAS-STING (Lin et al. 2024; Ma et al. 2024; Yang et al. 2024a). Downregulating cGAS/STING/NF-κB p65 axis through targeting the JAK2-STAT3 pathway ameliorates neuroinflammation and neuronal senescence (Zhang et al. 2023b). Furthermore, m6A modification increases cGAS mRNA stability to upregulate the STING/NF-κB pathway resulting in aggravation of microglia-mediated neuroinflammation, while fat mass and obesity-associated protein can abrogate these detrimental effects in an m6A-dependent manner (Yu et al. 2023). Collectively, microglia are central to neuroinflammation and post-stroke neuroprotection, making them promising therapeutic targets.

Oxidative Stress, Autophagy and Apoptosis

STING inhibition reduces NLRP3 inflammasome activity and inflammatory cytokine production, ameliorating neuroinflammation, oxidative stress, neuronal apoptosis, microglial pyroptosis, and neurobehavioral deficits after cerebral venous sinus thrombosis (Ding et al. 2022). The cGAS-STING pathway exacerbates inflammation, autophagy, apoptosis, and oxidative stress via NCOA4-mediated ferritinophagy, worsening cerebral ischemia/reperfusion injury (Li et al. 2023; Zhao et al. 2024; Sui et al. 2025). Ginsenoside Rd inhibits ferroptosis to alleviate early brain injury through the cGAS/STING/DHODH pathway (Jiang et al. 2025). Similarly, 25-hydroxycholesterol regulates the mTOR pathway by reducing STING activity, which inhibits autophagy, apoptosis, oxidative injury, and inflammation, and eventually attenuates cerebral ischemia/reperfusion injury (Lin et al. 2021). The neuroprotective factor, Activin A, markedly suppresses cGAS-STING-mediated autophagy rather than mTOR-dependent autophagy by activating the PI3K-PKB pathway, and thus alleviates neuronal injury after ischemic stroke (Liu et al. 2023a). In a neonatal hypoxia–ischemia rat model, activating cGAS/STING signaling via LINE-1-derived cytosolic DNA was found to contribute to delayed neurodegeneration, with cortical neurons more affected than hippocampal neurons (Gamdzyk et al. 2020). STING agonists (DMXAA and CMA) significantly decrease oxygen and glucose deprivation-induced cell death in vitro through a canonical STING-TBK1-IRF3 axis. Blocking the cGAS-STING pathway promotes remyelination, thereby facilitating post-stroke functional recovery within an extended treatment window (Maimaiti et al. 2024). Likewise, tilorone as a DNA intercalator stabilizes cytosolic DNA and induces a protective IFN-I response via the same pathway (Kundu et al. 2022).

Vascular Destruction and BBB Breakdown

Neutrophils accumulate in the brain after stroke, releasing neutrophil extracellular traps (NETs), which induce the activation of STING and downstream signaling molecules including TBK1, IRF3, and IFN-Iβ, resulting in impaired vascular remodeling and revascularization. Peptidylarginine deiminase 4 is integral for the formation of NETs, and its inhibition has been manifested to reduce the STING-mediated IFN-I response, thereby reducing the breakdown of the BBB and enhancing vascular regeneration and repair (Kang et al. 2020). NETs also promote tissue plasminogen activator-induced brain hemorrhage via activating the cGAS-STING signaling and IFN-I response, which eventually leads to the destruction of the vasculature and integrity of the BBB. These effects can be reversed following treatment with DNase I or by inhibiting PAD4 (Wang et al. 2021) Smart liposomal nanocarriers targeting NETs and cGAS-STING mitigate ischemic penumbra and improve stroke outcomes (Sun et al. 2023). Maintaining the integrity of the vasculature and BBB is essential for CNS homeostasis. Collectively, targeting NETs or the cGAS-STING signaling is a prospective therapy to protect against the destruction of blood vessels and the BBB after stroke.

STING in Degenerative Diseases

Human Neurodegenerative Diseases

Elevated STING and cGAMP expression has severally been observed in the substantia nigra pars compacta of PD patients (Hinkle et al. 2022) and in the spinal cord samples of ALS patients (Yu et al. 2020). The cGAS-STING- induced IFN-I was found to be increased together with type III IFN in the peripheral blood and plasma of A-T patients (Hartlova et al. 2015; Gul et al. 2018). In addition, the STING-IFN-β axis was shown to be significantly downregulated in patients with relapsing–remitting multiple sclerosis (Masanneck et al. 2020). Meanwhile, activation of the cGAS-STING pathway in microglia can be detected in human AD and aged mouse brain. Furthermore, studies have demonstrated that the application of STING inhibitor H-151 can effectively inhibit the activation of the cGAS-STING pathway in 5 × FAD mice (Xie et al. 2023). Interestingly, IFN-β therapy led to a reduction in IFI16 and STING expression levels in a STING-dependent manner in these patients perhaps due to a negative feedback mechanism (Helbi et al. 2020). In summary, STING and its associated signaling pathways are altered in human neurodegenerative diseases, suggesting that targeting STING has great therapeutic potential and deserves further clinical and experimental research.

Mitochondrial Dysfunction

Mitochondrial dysfunction is a hallmark of degenerative diseases. Impaired DNA repair and mitochondrial oxidative stress lead to mtDNA damage and reduced mitochondrial membrane potential during neurodegeneration. Subsequently, more mtDNA to the cytoplasm results in activating the cGAS/STING/IRF3 pathway and pathological inflammatory response, which leads to synaptic loss and neurodegeneration. Exogenous melatonin has been shown to inhibit these events (Jauhari et al. 2020). Melatonin has also been found to ameliorate myopathic changes induced by activating the ALDH2-cGAS-STING-TBK1 pathway and promoting mitophagy (Wang et al. 2020). METTL14 suppresses ferroptosis and mitochondrial dysfunction in dopaminergic neurons by regulating the m6A modification of TRAF6 through the cGAS-STING Pathway (Shao et al. 2024). In contrast, metformin restores mitochondrial function to delay astrocyte senescence by Mfn2-dependent cGAS-STING (Wang et al. 2024b).

Loss of STING activity in parkin mutant flies have been shown to suppress muscle defects and ameliorate disrupted mitochondrial morphology (Moehlman et al. 2023), while parkin and Pink1 mitigate STING-induced inflammation (Sliter et al. 2018). Lee et al. reported that loss of STING or Relish was insufficient to generate the above effects (Lee et al. 2020). therefore, in the light of current evidence, it remains unknow whether STING contributes to behavioral or mitochondrial phenotypes in parkin or Pink1 mutants.

Inflammation

Neuroinflammation is thought to accompany and accelerate degenerative processes. cGAS-STING signaling has been discovered to promote neuroinflammation and neurodegeneration in various diseases, for example, Alzheimer’s disease (AD), Huntington’s disease, and amyotrophic lateral sclerosis (ALS) (Sharma et al. 2020; Tan et al. 2022; Gulen et al. 2023; Nelson and Xu 2023). In Parkinson’s disease (PD), STING activation not only induces neuroinflammation and degeneration of dopaminergic neurons via activation of downstream IFN-I, interferon regulatory factor 7, or inflammasome but also induces α-synuclein pathology (Szego et al. 2022; Zhou et al. 2025). Microglial DNase II deficiency initiates neuroinflammation, synapse loss and increased Abeta levels and tauopathy by activating the cGAS-STING-IFN-I pathway, (Li et al. 2025). Both microglial cGAS deficiency and STING inhibition not only can effectively alleviate MPTP-induced neuroinflammation, dopaminergic neurotoxicity, and neurodegeneration in PD (Ma et al. 2023; Wang et al. 2023) but also can alleviate microglial dysfunction and preserve intercellular communication in AD (Chung et al. 2024; He et al. 2025).

STING is also regulated by upstream signaling molecules in neurodegenerative diseases. For example, C9ORF72 inhibits STING- and JAK-STAT-mediated inflammation in ALS and frontotemporal lobar degeneration (McCauley et al. 2020; Pang and Hu 2023). Conversely, Tau activates microglia to facilitate brain inflammation through the PQBP1-cGAS-STING pathway (Jin et al. 2021). Moreover, SGK1, 1-BP exposure or high-fat diet not only exerts pro-inflammatory effects via the cGAS-STING-, NLRP3- and NF-κB-mediated inflammatory pathways but also exacerbates the degeneration and necroptosis of midbrain dopaminergic neurons (Kwon et al. 2021; Song et al. 2024). IL-6 and impaired proteasomes aggravate neuroinflammation and Semen Strychni pulveratum and vomicine alleviate neuroinflammation both by regulating STING signaling (Dwamena et al. 2024; Liu et al. 2024a; Zhan et al. 2025). Taken together, the above studies manifest that regulating STING signaling to improve neuroinflammation has significant potential in the treatment of neurodegenerative diseases.

Ataxia-Telangiectasia Mutated (ATM)

Ataxia-telangiectasia (A-T), a rare recessively inherited disorder that leads to progressive neurological decline, is caused by biallelic mutations in the ATM gene, which encodes a PI3K-like kinase. ATM kinase is not only involved in regulating the response to DNA double-strand breakage but is also a potential target for radiosensitization (Chiu et al. 2023). DNA damage that is not properly repaired results in increased cytosolic DNA and subsequent cGAS-STING pathway activation, which in turn exacerbates the inflammatory microenvironment (Quek et al. 2017). ATM deficiency leads to cytoplasmic DNA accumulation, STING activation and microglia overactivation (Song et al. 2019), while inhibition of ATM has been proved to activate the STING signaling, thereby enhancing the effects of anti-tumor and cancer treatments (Hu et al. 2021; Gao et al. 2023b).

In A-T fibroblasts, it displays elevated cGAS-STING signaling, and IFN-stimulated gene expression, but normal levels of IFN gene expression (Haj et al. 2023). The rapid senescence of ATM-deficient lung fibroblasts is driven by the synergistic effect of the cGAS-STING, p53, and p38 MAPK pathways to cope with persistent DNA damage (Haj et al. 2025). Suppression of the cGAS-STING signaling can mitigate neuropathology and premature senescence in A-T brain organoids (Aguado et al. 2021). Together, these studies indicate that the cGAS-STING signaling pathway provides a new therapeutic target for managing inflammation, premature senescence, and other related symptoms in A-T patients.

Others

Glial cells have a significant impact on neurodegeneration. For example, inducing IFN-I by STING during chronic neurodegeneration influences the microglial phenotype and exacerbates disease progression (Nazmi et al. 2019), Furthermore, deletion of microglial cGAS markedly suppresses plaque formation and prevents amyloid-β-induced cognitive impairment (He et al. 2023a), and STING inhibition can normalize the Abeta phagocytic capacity of microglia and alleviate cellular senescence (Yuan et al. 2025). Increased STING expression has been reported in the striatal astrocytes of multiple system atrophy patients (Inoue et al. 2021). Mechanistically, deletion of astrocytic cGAS has been shown to ameliorate astrocyte senescence and neurodegeneration (Jiang et al. 2023). Chemerin/chemokine-like receptor 1/STING pathway is involved in astrocyte recruitment to Abeta plaques (Liu et al. 2024b).

Increased mtDNA in the cytosol and defects in the degradation of activated STING have been associated with PD pathogenesis (Hancock-Cerutti et al. 2022). In addition, the mtDNA-STING-NLRP3/IL-1β axis has been shown to initiate neutrophil infiltration into the cerebra and induce impaired cognition in AD (Xia et al. 2023). Withaferin A has been proved to exert neuroprotective effects by activating DJ1 and Nrf2, and inhibiting STING in PD (Zhao et al. 2021). Strikingly, activation of the cGAMP-STING-IRF3 pathway induces the expression of triggering receptor expressed on myeloid cells 2 (TREM2), which regulates microglial polarization and suppresses AD (Xu et al. 2019). TOLLIP is a STING stabilizing agent that can prevent the degradation of STING through direct interactions (Pokatayev et al. 2020).

Drug Therapy

To date, several drugs have been shown to play a role in treating neurodegenerative diseases through STING signaling. Their common mechanism of action is to regulate the inflammatory response. Supplementation with the nicotinamide adenine dinucleotide (NAD⁺) precursor nicotinamide riboside was found to normalize elevated STING levels resulting in decreased pro-inflammatory cytokines, DNA damage, and cellular aging in AD (Hou et al. 2021). Analogously, tetrahydroxy stilbene glycoside, the main active ingredient of the Chinese herb Polygonum multiflorum, was shown to reduce M1 type microglia and pro-inflammatory cytokine in AD via inhibition of the cGAS-STING signaling pathway (Gao et al. 2023a). Consistent with the STING inhibitor H-151, dimethyl fumarate can downregulate inflammatory cytokines, granzyme, and chemokines by NF-κB and cGAS-STING to target inflammation and autoimmunity in sporadic ALS (Zamiri et al. 2023). The neuroprotective effects of silibinin include ameliorating STING-mediated neuroinflammation through attenuating ferroptosis damage in sporadic AD and STING-IRF3-IFN-β pathway in PD (Liu et al. 2021, 2023b). These findings indicate that targeting STING signaling may be beneficial in the development of novel treatments for neurodegenerative diseases.

STING in Intracranial Infection

Herpes Simplex Virus Type 1 Infection

IFN-I can protect the brain from viral infections, and the critical regulatory factor DDOST is indispensable for producing anti-viral IFN-I mediated by STING (Tu et al. 2022). Following herpes simplex virus type 1 (HSV-1) infection, IFN-I is mainly generated by microglia in a cGAS-STING-dependent manner. Furthermore, cGAS- or STING-deficient mice display increased viral loads and are susceptible to herpes simplex encephalitis (HSE) (Reinert et al. 2016). The STING agonist DMXAA has been shown to reduce viral replication by increasing IFN-I production (Ceron et al. 2019). HSV-1 infection of microglia induced IFN-I expression at low-to-medium viral loads, but cGAS/STING-dependent apoptosis at high viral loads (Reinert et al. 2021). While IFN-I plays a central role in HSV-1 antiviral responses, emerging evidence highlights IFN-independent mechanisms of STING in viral clearance. Notably, STING activation can directly induce autophagy-mediated degradation of viral particles by recruiting autophagy-related proteins (e.g., LC3) independent of IFN signaling (Gui et al. 2019). This pathway complements IFN-I-mediated immunity and may explain residual antiviral activity observed in IFNAR-deficient models (Gui et al. 2019). These dual roles (IFN-dependent and IFN-independent) collectively position STING as a multifaceted hub for antiviral defense against neurotropic viruses like HSV-1.

STING is crucial for control of HSV-1 after central infection, but dispensable for survival after peripheral infection (Parker et al. 2015). MAM domain containing 2 (MAMDC2) interacts with STING via its first MAM domain and increases the polymerization of STING, subsequently activating downstream TBK1-IRF3 signaling to enhance the IFN-I-based innate anti-viral response (Wang et al. 2022b). In contrast, TRIM18 recruits protein phosphatase 1A to dephosphorylate and inactivate TBK1, thereby preventing TBK1 from interacting with upstream adaptors and STING, and ultimately leading to the suppression of IFN-I mediated anti-viral effects during viral infections (Fang et al. 2022). HSV-1 VP1-2 results in deubiquitination of STING, which inhibits downstream IRF3 activation and IFN-I expression, thus facilitating brain infection and potential HSE progression (Bodda et al. 2020). In addition, resveratrol, an antitoxin produced by many plants when stimulated, ameliorates HSV-1-induced neuroinflammation, microglial M1 polarization, and viral encephalitis by STING/NF-κB pathway (Huang et al. 2024).

Other Infections

Infection with the Japanese encephalitis virus leads to the upregulation of various viral sensors, IFN regulatory factors, and STING. STING then induces anti-viral immune responses mediated through the activation of IFN-I. Moreover, STING expression affects the intracellular viral load (Nazmi et al. 2012; Sharma et al. 2021). However, in Rasmussen’s encephalitis, STING, IFN-β, and IFI16 are produced at relatively lower levels, and anti-viral innate immunity is insufficiently activated (Wang et al. 2022c). Microbial activation of STING signaling requires TRIF, which interacts directly with STING to enhance STING dimerization and intermembrane translocation. Thus, STING agonists are unable to induce genes in the absence of TRIF (Wang et al. 2016).

Following West Nile virus encephalitis, ELF4 combines with STING to induce IFN-I production and anti-viral immunity (You et al. 2013). Consistent with these reports, knockout of STING in bone marrow-derived macrophages leads to reduced IFN-I expression levels, which alters the CD4⁺/CD8⁺ T cell ratio and increases viral susceptibility in the CNS to causing CNS diseases (McGuckin Wuertz et al. 2019). Furthermore, Zika virus infection has been shown to activate NF-κB signaling and downstream dSTING expression to alleviate brain infections in Drosophila (Liu et al. 2018). In general, these studies expound that targeting the STING signaling pathway and STING-mediated IFN-I response is a beneficial therapeutic strategy to attenuate intracranial viral infection.

STING in Traumatic Brain Injury

IFN-I is associated with the occurrence and development of traumatic brain injury (TBI) neuropathology, and its canonical activation by cGAS-STING signaling is a key component of neuroinflammation, cognitive impairment, neural tissue damage, and autophagy dysfunction (Abdullah et al. 2018; Barrett et al. 2021; Fritsch et al. 2022; Fryer et al. 2024; Packer et al. 2024). STING signaling in peripheral immune cells and microglia drives an early inflammatory response, but CD300LF⁺ microglia can reduce neuroinflammation by STING inhibition (Fritsch et al. 2024; Lu et al. 2024). Following TBI, aberrant ER stress activates neuronal STING signaling and the release of excess IFN-I, which leads to white matter injury due to T cell infiltration and microglial activation (Sen et al. 2020). Consistent with previous studies, STING agonist DMXAA has been shown to intensify IFN-I-associated inflammation and gliosis (Wangler et al. 2022), while intranasal administration of atomic-let-7i was found to effectively suppress its downstream target STING to ameliorate cognitive function (He et al. 2022).

STING signaling also affects programmed cell death following TBI. Inhibiting the formation of NETs not only attenuates neuroinflammation, neurological deficits, and neuronal apoptosis through the STING-dependent IRE1α/ASK1/JNK pathway (Shi et al. 2023) but also mitigates NLRP1-associated neuronal pyroptosis through the STING/IRE1a pathway (Cao et al. 2023). Moreover, STING also activates NLRP3-mediated pyroptosis and augments pyroptosis-related neuroinflammation (Zhang et al. 2022). Ferroptosis is a novel type of iron-dependent programmed cell death. Moderately intensive treadmill exercise has been proved to ameliorate ferroptosis, cognitive deficits, and neurodegeneration through inhibition of the STING pathway (Chen et al. 2023). Strikingly, gender differences have been reported in cases of repeated mild TBI. Female mice were reported to display higher DNA damage, and lower expression of the cGAS-STING signaling and senescence protein p21 than males (Schwab et al. 2022). In conclusion, further research should focus on the specific mechanisms underlying STING signaling to improve and treat TBI.

Discussion and Conclusion

In recent years, STING signaling has received increasing attention. As a member of the cGAS/DncV-like nucleotidyltransferase family, cGAS detects abnormal DNA signals in the cytoplasm and catalyzes the synthesis of cyclic dinucleotide cGAMP to activate the downstream adapter protein STING (Kranzusch 2019). This is the canonical pathway for STING activation. However, STING can also be activated in a cGAS- and cGAMP-independent manner, although less is known about this signaling pathway. STING is an ER transmembrane protein, and its translocation to the Golgi apparatus is a hallmark of its activation. Thus, further research is required to elucidate the specific mechanisms mediating the translocation of STING from the ER to the Golgi apparatus, as well as to understand how STING is effectively sorted into lysosomes for degradation.

Targeting STING signaling pathways is a growing field of research. In particular, STING agonists and inhibitors (Tables 1, 2) have been the main focus of studies due to the regulatory role of STING in many diseases. STING agonists include native cyclic dinucleotides, cyclic dinucleotides derivatives, and non-nucleotide agonists (Wu et al. 2020) while STING inhibitors include nitrofurans, indole derivatives, oxo-tetrahydro-isoquinoline carboxylic acids, multisubstituted benzamides or benzenesulfonamides, cyclopeptides, dimeric benzimidazoles, heterobicyclic derivatives, miscellaneous compounds and STING degraders (Shen et al. 2022) (Fig. 3). Due to the presence of the blood–brain barrier, 98% of small-molecule drugs and almost all large-molecule drugs are restricted from entering the brain (Li et al. 2021), which limits the application of STING agonists and inhibitors in CNS diseases. Nanoparticle-based drug delivery is one of the most effective and promising ways of delivering drugs into inaccessible regions such as the brain since it protects the therapeutic drugs while simultaneously delivering them effectively to impaired areas (Saraiva et al. 2016). Previous studies have confirmed the efficacy of nanoparticles in crossing the BBB (Dong 2018). Indeed, nanoparticles loaded with STING modulators have been shown to display therapeutic effects on various diseases including cancers, SARS-CoV-2 infections, and Middle East respiratory syndrome coronavirus (Lin et al. 2019; Dosta et al. 2023; Zhang et al. 2023c). In the past few decades, researchers have explored many specific pathways leading to central nervous system diseases and developed targeted drugs and molecular formulations. Nevertheless, its clinical treatment progress is slow, and one of the main challenges is the obstruction of the BBB. Due to the presence of the BBB, many drugs cannot reach the brain or reach the brain at a very low concentration that cannot achieve effective therapeutic concentrations, resulting in limited clinical use of these drugs in the treatment of CNS diseases. Nanoparticles can assist drugs in effectively crossing the blood–brain barrier and targeting delivery into the brain. In theory, it is a promising idea that drug-loaded nanoparticles are encapsulated with cell membranes such as neutrophil and blood cell membranes, which not only can improve their targeting and precise delivery to the lesion site but also can escape immune defense to prevent clearance by the body. Although nanoparticles are an increasingly important research field in the treatment of CNS diseases, there are limited reports on the clinical application of nanoparticles in CNS diseases. Thus, further studies are necessary to evaluate their side effects and toxicity concerning CNS diseases.

Table 1.

The cGAS regulators in CNS diseases

cGAS modulators	Regulatory effects	Disease types	Targeted delivery	Effects	References
RU.521	Inhibitor	Cerebral venous sinus thrombosis	None	Beneficial	Ding et al. (2022)
		Cerebral ischemia–reperfusion injury	None	Beneficial	Li et al. (2023)
		Ischemic stroke	None	Beneficial	Liu et al. (2023a); Zhao et al. (2024)
		Neonatal hypoxic-ischemic encephalopathy	None	Beneficial	Gamdzyk et al. (2020)
		Huntington’s disease	None	Beneficial	Jauhari et al. (2020)
		Amyotrophic lateral sclerosis	None	Beneficial	Yu et al. (2020); Tan et al. (2022)
		Parkinson’s disease	None	Beneficial	Ma et al. (2023); Zhou et al. (2025)
		Alzheimer’s disease	None	Beneficial	Gao et al. (2023a)
PF-06928215	Inhibitor	Alzheimer’s disease	None	Detrimental	Wang et al. (2020)

Table 2.

The STING regulators in CNS diseases

	STING modulators	Regulatory effects	Disease types	Targeted delivery	Effects	References
Brain tumors	c-di-GMP	Agonist	Gliomas	None	Beneficial	Ohkuri et al. (2014)
		Agonist	Glioblastoma multiforme	Nanoparticles	Beneficial	Bielecki et al. (2021)
		Agonist	Glioblastoma	nanocapsule	Beneficial	Zhou et al. (2022)
	IACS8779	Agonist	Glioblastoma	None	Beneficial	Boudreau et al. (2021)
	8803	Agonist	Glioblastoma	None	Beneficial	Najem et al. (2024)
	DMXAA	Agonist	Neuroblastoma	None	Beneficial	Layer et al. (2017)
	SR717	Agonist	Glioma	Nanoparticles	Beneficial	Wang et al. (2022a)
	2′3′-cGAMP	Agonist	Neuroblastoma	Nanoparticles	Beneficial	Wang-Bishop et al. (2020)
	CMA	Agonist	T cell-derived malignancies	None	Uncertain	Gulen et al. (2017)
Stroke	C-176	Inhibitor	Ischemic stroke	None	Beneficial	Peng et al. (2020); Kong et al. (2022)
	CMA	Agonist	Subarachnoid hemorrhage	None	Detrimental	Peng et al. (2020)
		Agonist	Ischemic stroke	None	Beneficial	Kundu et al. (2022)
	H151	Inhibitor	Ischemic stroke	None	Beneficial	Liu et al. (2024c); Wu et al. (2024)
	2′3′-cGAMP	Agonist	Cerebral venous sinus thrombosis	None	Detrimental	Ding et al. (2022)
		Agonist	ischemic stroke	None	Detrimental	Liu et al. (2023a)
		Agonist	Neonatal hypoxic-ischemic encephalopathy	None	Detrimental	Gamdzyk et al. (2020)
	DMXAA	Agonist	Ischaemic stroke	None	Beneficial	Kundu et al. (2022)
Degenerative diseases	c-di-AM(PS)2	Agonist	Alzheimer’s disease	None	Beneficial	Wang et al. (2020)
	Astin C	Inhibitor	Alzheimer’s disease	None	Detrimental	Wang et al. (2020)
	C-176	Inhibitor	Parkinson’s disease	None	Beneficial	Wang et al. (2023)
	H-151	Inhibitor	Amyotrophic lateral sclerosis	None	Beneficial	Yu et al. (2020); Tan et al. (2022)
		Inhibitor	Ataxia- telangiectasia	None	Beneficial	Aguado et al. (2021)
		Inhibitor	Alzheimer’s disease	None	Beneficial	Hou et al. (2021); Gao et al. (2023a)
		Inhibitor	Sporadic amyotrophic lateral sclerosis	None	Beneficial	Zamiri et al. (2023)
		Inhibitor	Parkinson’s disease	None	Beneficial	Hinkle et al. (2022); Zhou et al. (2025)
	DMXAA	Agonist	Parkinson’s disease	None	Detrimental	Zhao et al. (2021); Hinkle et al. (2022)
	palmitic acid	Agonist	Human Neurodegenerative Diseases	None	Detrimental	Ferecskó et al. (2023)
Intracranial infection	DMXAA	Agonist	Herpes simplex encephalitis	None	Beneficial	Ceron et al. (2019)
	2′3′-cGAMP	Agonist	Herpes simplex encephalitis	None	Beneficial	Reinert et al. (2021)
	2′3′-c-di-AM (PS)2(Rp, Rp)	Agonist	Herpes simplex encephalitis	None	Beneficial	Wang et al. (2022b)
	c-di-GMP	Agonist	Microbial infection	None	TRIF required	Wang et al. (2016)
Traumatic brain injury	DMXAA	Agonist	Traumatic brain injury	None	Detrimental	Wangler et al. (2022)
	2′3′-cGAMP	Agonist	Traumatic brain injury	None	Detrimental	Shi et al. (2023)
	cGAMP	Agonist	Traumatic brain injury	None	Detrimental	Cao et al. (2023)
	ADU- S100	Agonist	Traumatic brain injury	None	Detrimental	Zhang et al. (2022)
	C-176	Inhibitor	Traumatic brain injury	None	Beneficial	Zhang et al. (2022); Cao et al. (2023); Shi et al. (2023); Fryer et al. (2024)

Fig. 3.

The chemical structures of cGAS and STING mediators applied in CNS diseases

Multiple alleles have been reported in the coding gene of STING (TMEM173), while noticeable diversity has been detected in TMEM173 genotypes and STING protein expression among different ethnic groups (Fryer et al. 2021). Furthermore, significant differences between mouse and human STING mean that STING agonists exhibiting significant pharmacological effects in mice may not necessarily be validated in humans (Shih et al. 2018). For example, DMXAA (ASA404), one of the earliest reported non-cyclic dinucleotide small molecule STING agonists with preclinical efficacy, binds selectively to mouse STING rather than human STING (Cherney et al. 2022). Similar effects have also been reported in brain tumors (Bähr et al. 2017). These reports highlight the challenges associated with interspecies differences in STING and indicate that future studies should focus on human STING. Clinical trials on STING agonists and inhibitors should receive more attention. In conclusion, although the study of STING signaling remains a challenging area of research, there is enormous potential and significant prospects in disease treatments by targeting the STING signaling pathway.

Author Contributions

MS, JR, and ZZ wrote the original draft; ZY, CW and LL searched and organized the literature; JT and GM drew the figures; GM and MC reviewed and edited the manuscript and figures.

Funding

This work was supported by the National Nature Science Foundation of China (Grant Number NO. 82160243), Foundation of Jiangxi health and Family Planning Commission (Grant Number No. 220210596 and No.202210572) and the Natural Science Foundation of Jiangxi Provincial Department of Science and Technology (Grant Number 20242BAB25476).

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare that they have no competing interests.