STING-Mediated Autophagy Is Protective against H2O2-Induced Cell Death

Stimulator of interferon genes (STING)-mediated type-I interferon signaling is a well characterized instigator of the innate immune response following bacterial or viral infections in the periphery. Emerging evidence has recently linked STING to various neuropathological conditions, however, both protective and deleterious effects of the pathway have been reported. Elevated oxidative stress, such as neuroinflammation, is a feature of a number of neuropathologies, therefore, this study investigated the role of the STING pathway in cell death induced by elevated oxidative stress. Here, we report that the H2O2-induced activation of the STING pathway is protective against cell death in wildtype (WT) MEFSV40 cells as compared to STING−/− MEF SV40 cells. This protective effect of STING can be attributed, in part, to an increase in autophagy flux with an increased LC3II/I ratio identified in H2O2-treated WT cells as compared to STING−/− cells. STING−/− cells also exhibited impaired autophagic flux as indicated by p62, LC3-II and LAMP2 accumulation following H2O2 treatment, suggestive of an impairment at the autophagosome-lysosomal fusion step. This indicates a previously unrecognized role for STING in maintaining efficient autophagy flux and protecting against H2O2-induced cell death. This finding supports a multifaceted role for the STING pathway in the underlying cellular mechanisms contributing to the pathogenesis of neurological disorders.


Introduction
Type-I interferons (IFNs) are pleiotropic cytokines that have been implicated in neuropathologies, including Gaucher disease [1], Aicardi-Goutieres syndrome [2] and a model of prion disease [3]. Supporting this, several reports have identified a detrimental role for type-I IFNs in animal models of Alzheimer's disease (AD) [4,5], Parkinson's disease (PD) [6] and traumatic brain injury (TBI) [7], with elevated expression of type-I IFNs found in post-mortem human AD, PD and TBI brains [4,6,7]. Classically, type-I IFNs (IFN-α and IFN-β) mediate their proinflammatory effects through the Janus activated kinases (JAK)-signal transducer and activator of transcription (STAT) pathway. Following binding to their cognate receptor, composed of interferon receptor 1 (IFNAR1) and interferon receptor 2 (IFNAR2) subunits, type-I IFNs signal through this pathway to induce the upregulation of type-I IFN and other proinflammatory cytokine gene expression.
Type-I IFNs can alternatively be activated through the stimulator of interferon genes (STING)-dependent pathway. Foreign materials released by invading pathogens, including DNA or self-derived nucleic acids from dying cells, known as damage-associated molecular patterns (DAMPs), have been shown to activate STING. Through a signaling cascade involving tumor necrosis factor (TNF) receptor-associated factor NF-κB activator (TANK)-binding kinase 1 (TBK1) and interferon regulatory factor-3 (IRF3), this trigger increases type-I IFN production. Aberrant STING-mediated type-I IFN a potential target for therapeutic intervention in neuropathologies with oxidative-stress mediated cellular injury.

The STING Pathway Is Activated Following H 2 O 2 Treatment
H 2 O 2 has been used experimentally as a potent ROS inducer, rendering cells into a heightened oxidative stress state, such as that which occurs in a number of neuropathologies. However, the underlying mechanisms that contribute to oxidative stress-mediated cell death are complex, and the roles of the STING and type-I IFN pathways have not been investigated. To confirm a possible role for STING in H 2 O 2 -induced oxidative stress, mRNA expression was analyzed by qPCR analysis. STING expression was induced by 500 µM H 2 O 2 in WT MEF SV40 cells in a time-dependent manner, with increased STING mRNA levels as early as 30 min (2.06 ± 0.34 fold; n.s. p = 0.2808) after H 2 O 2 treatment as compared to the control group. Interestingly, a second wave of increased STING mRNA expression was also detected at 24 h (1.96 ± 0.72; n.s p = 0.3958) of H 2 O 2 treatment ( Figure 1A). Furthermore, Western blot analysis confirmed increased phosphorylation of the STING (pSTING) protein at 30 min (3.43 ± 0.68 fold; n.s. p = 0.8656) and 6 h (5.058 ± 1.712 fold; * p < 0.05) ( Figure 1B,C) after H 2 O 2 treatment as compared to the untreated control group. However, pSTING expression was unchanged in cells treated for 12-48 h 500 µM H 2 O 2 treatment (data not shown). following H2O2 treatment. Furthermore, the sustained and elevated expression of autophagy markers, including LC3, p62 and LAMP2 levels in H2O2-treated STING −/− cells, suggests inefficient autophagy flux. Collectively, this study has identified a beneficial role for STING in protecting against H2O2-induced cell death and proposes STING as a potential target for therapeutic intervention in neuropathologies with oxidative-stress mediated cellular injury.

The STING Pathway is Activated Following H2O2 Treatment
H2O2 has been used experimentally as a potent ROS inducer, rendering cells into a heightened oxidative stress state, such as that which occurs in a number of neuropathologies. However, the underlying mechanisms that contribute to oxidative stress-mediated cell death are complex, and the roles of the STING and type-I IFN pathways have not been investigated. To confirm a possible role for STING in H2O2-induced oxidative stress, mRNA expression was analyzed by qPCR analysis. STING expression was induced by 500 μM H2O2 in WT MEF SV40 cells in a time-dependent manner, with increased STING mRNA levels as early as 30 min (2.06 ± 0.34 fold; n.s. p = 0.2808) after H2O2 treatment as compared to the control group. Interestingly, a second wave of increased STING mRNA expression was also detected at 24 h (1.96 ± 0.72; n.s p = 0.3958) of H2O2 treatment ( Figure 1A). Furthermore, Western blot analysis confirmed increased phosphorylation of the STING (pSTING) protein at 30 min (3.43 ± 0.68 fold; n.s. p = 0.8656) and 6 h (5.058 ± 1.712 fold; *p = 0.0418) ( Figure 1B,C) after H2O2 treatment as compared to the untreated control group. However, pSTING expression was unchanged in cells treated for 12-48 h 500 μM H2O2 treatment (data not shown). Increased STING expression was detected by qPCR analysis (A) (n = 4 for each time point) in WT MEF SV40 cells exposed to H2O2 treatment as compared to vehicle control. Increased phosphorylation of STING (pSTING) was confirmed at the protein level by western blot analysis (B) with a lack of STING expression identified in STING −/− cells (representative blot of n = 5 blots). Quantification of pSTING expression relative to total STING in (B) is shown in (C). All data are expressed as mean ± SEM, *p < 0.05.
Downstream of STING activation, pTBK1 ( Figure 2A) and TBK1 ( Figure 2B) protein expression was also determined by Western blot analysis, with increased pTBK1 expression identified in WT cells at early time points (15 min-6 h) of H2O2 treatment but not in the STING −/− cells. A significant A B C Figure 1. H 2 O 2 induces an upregulation in stimulator of interferon genes (STING) expression. Increased STING expression was detected by qPCR analysis (A) (n = 4 for each time point) in WT MEF SV40 cells exposed to H 2 O 2 treatment as compared to vehicle control. Increased phosphorylation of STING (pSTING) was confirmed at the protein level by western blot analysis (B) with a lack of STING expression identified in STING −/− cells (representative blot of n = 5 blots). Quantification of pSTING expression relative to total STING in (B) is shown in (C). All data are expressed as mean ± SEM, * p < 0.05.
Downstream of STING activation, pTBK1 ( Figure 2A) and TBK1 ( Figure 2B) protein expression was also determined by Western blot analysis, with increased pTBK1 expression identified in WT cells at early time points (15 min-6 h) of H 2 O 2 treatment but not in the STING −/− cells. A significant increase in pTBK1 levels was detected at 30 min (4.65 ± 1.43 fold; * p < 0.05) and 1 h (4.629 ± 1.33 fold; * p < 0.05) ( Figure 2C)  increase in pTBK1 levels was detected at 30 min (4.65 ± 1.43 fold; *p = 0.0103) and 1 h (4.629 ± 1.33 fold; *p = 0.0110) ( Figure 2C) after H2O2 treatment as compared to their genotype control group, suggesting H2O2-induced pTBK1 expression is STING-mediated. Collectively, these results confirmed our hypothesis that the STING pathway is activated after H2O2 treatment. Figure 2. Activation of the downstream mediator of STING, TBK1, is increased following H2O2 treatment. Western blot analysis confirmed increased phosphorylation of TBK1 (pTBK1) (A) in WT MEF SV40 cells following treatment with 500 μΜ H2O2, with levels unchanged in STING −/− MEF SV40 cells (representative blot of n = 4 blots). Total TBK1 expression (B) was unchanged by H2O2 treatment in both cells (representative blot of n = 4 blots). Quantification of (A) and (B) is shown in (C), demonstrating an increase in pTKB1 expression relative to total TBK1 levels in H2O2-treated WT MEF SV40 cells. Data represent the mean ± SEM, *p < 0.05.

H2O2 Induces Type-I IFN Signalling in a STING-Dependent Manner
To further characterize the activation of the STING pathway following H2O2 insult, mRNA expression of IRF3, IRF7, IFNα and IFNβ were measured by qPCR analysis. Consistent with STING activation, increased IRF3 mRNA levels were identified in WT cells after H2O2 treatment, with a reduced response in STING −/− cells ( Figure 3A). IRF3 was significantly reduced in STING −/− cells at 48 h of H2O2 treatment as compared to the WT cells (WT = 4.24 ± 1.22 vs. STING −/− = 1.079 ± 0.22; **p = 0.0015). In contrast, increased IRF7 mRNA expression was induced by H2O2 in both WT and STING −/− cells with no significant difference between these two genotypes ( Figure 3B). These results suggest H2O2-induced IRF3 expression is STING-mediated while IRF7 can be activated independent of STING after H2O2 treatment.   Data represent mean ± SEM, * p < 0.05, ** p < 0.001, *** p < 0.0001.

STING is Required for Cellular Survival following H2O2 Insult
The role of STING in cellular death induced by H2O2-induced oxidative stress was assessed in a concentration-and time-dependent manner. Firstly, WT and STING −/− cells were exposed for 24 h with H2O2 in a dose-dependent manner (300 μM, 500 μM and 1 mM) and cell viability was assessed by an MTT assay. STING −/− cells were more susceptible to cell death induced by 300μM (% cell viability of WT = 71.04 ± 2.02 vs. STING −/− = 34.58 ± 0.67; *** p < 0.001) and 500μM H2O2 (% cell viability of WT = 49.87 ± 0.95 vs. STING −/− 31.20 ± 1.19; *** p < 0.0001) as compared to WT cells ( Figure 4A). A dose of 1 mM of H2O2 was seen to be extremely toxic to the cells of both genotypes with almost 100% of cells killed after a 24 h incubation (% of cell viability of WT = 0.15 ± 0.01 vs. STING −/− = 0.05 ± 0.001; ns p > 0.9999). This result suggests that STING is protective in vitro following H2O2 treatment. Accordingly, the next experiment was designed to determine the time-dependent toxicity effects of H2O2-induced cell death in WT and STING −/− MEF SV40 cells. Supporting the previous results, it was found that STING confers protection after H2O2-induced cell death with 24 h ( Data represent mean ± SEM, * p < 0.05, ** p < 0.01, *** p < 0.001.

STING Is Required for Cellular Survival Following H 2 O 2 Insult
The role of STING in cellular death induced by H 2 O 2 -induced oxidative stress was assessed in a concentration-and time-dependent manner. Firstly, WT and STING −/− cells were exposed for 24 h with H 2 O 2 in a dose-dependent manner (300 µM, 500 µM and 1 mM) and cell viability was assessed by an MTT assay. STING −/− cells were more susceptible to cell death induced by 300 µM  Data represent mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.0001.

STING Is a Negative Regulator of H2O2-Induced ROS Production
Generation of intracellular ROS following H2O2 treatment is known to contribute to oxidative stress [36][37][38]. To confirm that H2O2 treatment leads to ROS generation and to further understand the modulation of the H2O2-induced oxidative stress by STING in this study, intracellular ROS levels in H2O2 treated WT and STING −/− cells were measured by a 2′, 7′-dichlorofluorescin (DCF) assay. Increased ROS levels were confirmed in both WT and STING −/− cells after H2O2 treatment as compared to the untreated control group. Importantly, STING -/ cells exhibited higher ROS levels as compared to the WT cells with a rapid burst of intracellular ROS production detected at 15 min (WT = 13.55 ± 1.67; n.s. p = 0.9666 and STING −/− = 19.23 ± 0.32; *p = 0.0303) that gradually decreased over H2O2 treatment ( Figure 5). This suggests a role for STING as an inhibitor of H2O2-induced oxidative stress. Further, we confirmed increased ROS generation induced by H2O2 indicative of elevated oxidative stress in this in vitro model of inflammation.

STING is Required to Maintain a Normal Autophagy Flux in Response to H2O2 Treatment
We demonstrated that STING is required to promote cellular survival in response to H2O2 treatment with increased intracellular ROS levels identified in the absence of STING. To investigate A B

STING Is a Negative Regulator of H 2 O 2 -Induced ROS Production
Generation of intracellular ROS following H 2 O 2 treatment is known to contribute to oxidative stress [36][37][38]. To confirm that H 2 O 2 treatment leads to ROS generation and to further understand the modulation of the H 2 O 2 -induced oxidative stress by STING in this study, intracellular ROS levels in H 2 O 2 treated WT and STING −/− cells were measured by a 2 , 7 -dichlorofluorescin (DCF) assay.  Data represent mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.0001.

STING Is a Negative Regulator of H2O2-Induced ROS Production
Generation of intracellular ROS following H2O2 treatment is known to contribute to oxidative stress [36][37][38]. To confirm that H2O2 treatment leads to ROS generation and to further understand the modulation of the H2O2-induced oxidative stress by STING in this study, intracellular ROS levels in H2O2 treated WT and STING −/− cells were measured by a 2′, 7′-dichlorofluorescin (DCF) assay. Increased ROS levels were confirmed in both WT and STING −/− cells after H2O2 treatment as compared to the untreated control group. Importantly, STING -/ cells exhibited higher ROS levels as compared to the WT cells with a rapid burst of intracellular ROS production detected at 15 min (WT = 13.55 ± 1.67; n.s. p = 0.9666 and STING −/− = 19.23 ± 0.32; *p = 0.0303) that gradually decreased over H2O2 treatment ( Figure 5). This suggests a role for STING as an inhibitor of H2O2-induced oxidative stress. Further, we confirmed increased ROS generation induced by H2O2 indicative of elevated oxidative stress in this in vitro model of inflammation.

STING is Required to Maintain a Normal Autophagy Flux in Response to H2O2 Treatment
We demonstrated that STING is required to promote cellular survival in response to H2O2 treatment with increased intracellular ROS levels identified in the absence of STING. To investigate A B Figure 5. STING −/− cells display greater ROS production in response to H 2 O 2 treatment compared with WT cells. WT and STING −/− cells were treated with 500 µM H 2 O 2 for the indicated time points before intracellular ROS levels were determine by DCF assay. Data represent mean ± SEM, * p < 0.05, n = 5.

STING Is Required to Maintain a Normal Autophagy Flux in Response to H 2 O 2 Treatment
We demonstrated that STING is required to promote cellular survival in response to H 2 O 2 treatment with increased intracellular ROS levels identified in the absence of STING. To investigate whether autophagy plays a role in mediating the protective effects of STING after H 2 O 2 treatment, we examined LC3, p62 and LAMP2 expression profiles in WT and STING −/− cells treated with 500µM H 2 O 2 . Conversion of LC3-I to LC3-II is representative of increased autophagy activation [39] while the degradation of p62 and LAMP2 at the late step of the autophagy process serves as a marker for normal autophagic flux. Western blot analysis identified increased LC3-II levels at earlier time points in both WT and STING −/− cells as compared to the control levels, indicating that autophagy is activated in response to H 2 O 2 treatment ( Figure 6A). Interestingly, a prolonged H 2 O 2 treatment (24 h and 48 h) ( Figure 6B)  STING cells, LAMP2 levels were measured. We detected sustained and increased LAMP2 expression in the STING −/− cells challenged with 500 μM H2O2 across all time points ( Figure 6G, H), with peak LAMP2 levels detected at 24 h (LAMP2 = 2.413 ± 0.489 vs. vehicle; *p = 0.0456) while reduced in the WT MEF SV40 cells (LAMP2 = 0.808 ± 0.156 vs. vehicle, n.s p > 0.9999) ( Figure 6I). The accumulation of LAMP2 expression suggests a block in the autophagosome-lysosomal degradation step, indicating impaired autophagy flux in the STING −/− cells in response to H2O2.  The increased LC3-II/LC3-I levels detected at earlier time points in both WT and STING −/− cells indicate a key role for STING in either inducing or blocking autophagic flux. To confirm whether STING is a key mediator of autophagy flux by H2O2, LC3-II levels in the presence or absence of a known autophagy inhibitor, bafilomycin A1 (BafA1), were analyzed ( Figure 7A). If STING is an autophagy flux inducer, co-treatment with BafA1 and H2O2 would lead to higher LC3-II levels as compared to H2O2 alone. In contrast, if STING inhibited autophagy flux, LC3-II levels would remain unchanged [40,41]. WT cells treated for 3 h with H2O2 and BafA1 displayed higher LC3-II/LC3-I levels as compared to that of H2O2 alone (WT LC3-II/LC3-I ratio of H2O2 = 3.953 ± 0.781 vs. BafA1 ± H2O2 = 10.81 ± 2.061; **p = 0.0054) ( Figure 7B), confirming that STING is an inducer rather than an inhibitor of autophagic flux. Similar treatment of BafA1 and H2O2 had no effect on LC3-II/LC3-I levels in the STING −/− (STING −/− LC3-II/LC3-I ratio of H2O2 = 2.393 ± 0.621 vs. BafA1 ± H2O2 = 6.098 ± 1.052; ns p > 0.9999) cells, confirming STING as the mediator of increased autophagy flux induced by H2O2. This result suggests that the increased LC3-II levels observed in H2O2-treated WT cells are due to increases in autophagosome formation rather than a block in autophagy flux. The increased LC3-II/LC3-I levels detected at earlier time points in both WT and STING −/− cells would indicate a key role for STING in either inducing or blocking autophagic flux. Interestingly, the addition of an autophagy inducer, rapamycin (Rap), and H2O2 in STING −/− (STING −/− LC3-II/LC3-I ratio of VEH = 1.610 ± 0.256 vs. Rap ± H2O2 = 8.417 ± 1.479; **p = 0.0060) cells induced further accumulation of LC3-II levels which was similar to that observed in WT (WT LC3-II/LC3-I ratio of VEH = 3.138 ± 0.452 vs. Rap ± H2O2 = 11.63 ± 2.042; ***p = 0.0001) cells, supporting a role for STING in mediating the autophagosome-lysosomal degradation step of autophagy. The reduced cellular viability observed in the H2O2-treated STING −/− cells as compared to the WT cells can be attributed to the reduced protective effects of normal autophagy activity. Taken together, these results confirmed that autophagy is activated in response to H2O2 and identified STING as a critical mediator of efficient autophagy flux in this cellular stress context.

Discussion
A B the WT cells can be attributed to the reduced protective effects of normal autophagy activity. Taken together, these results confirmed that autophagy is activated in response to H 2 O 2 and identified STING as a critical mediator of efficient autophagy flux in this cellular stress context.

Discussion
The activation of the STING pathway and induction of type-I IFN signaling in response to H 2 O 2 -induced oxidative stress in vitro was confirmed in this study. Previously, the detrimental effects of the STING pathway in mediating the neuro-inflammatory response were demonstrated in a CCI-induced TBI model reported by our group [7]. Surprisingly, the results in this study suggest that the STING pathway can play a protective role, as demonstrated by a reduced cellular viability in the STING −/− cells in response to H 2 O 2 treatment. This finding proposes a previously unrecognized role for STING in modulating H 2 O 2 -induced oxidative stress and reveals a multifaceted role for STING in different disease/cellular models and stress-related pathophysiological processes.

STING Is Protective in H 2 O 2 Induced Cell Death
Aberrant neuroinflammatory responses and increased oxidative stress are the major components that underlie many pathologies. However, the underlying molecular mechanisms contributing to oxidative stress involving STING are unknown. In this study, using a cell-based model of oxidative stress in vitro, a role for STING in response to H 2 O 2 -induced ROS generation was confirmed. STING is thought to regulate ROS production and oxidative stress through its anti-viral and anti-microbial activity [42]. It is known that an increased ROS level, generating oxidative stress and subsequent cell death, is a host defense mechanism to limit pathogen replication following infection [43,44]. Emerging evidence also supports a role for STING in inducing both apoptotic [45,46] and necroptotic cell death pathways [47,48]. However, evidence linking STING and the generation of ROS-induced oxidative stress is still lacking. More importantly, the underlying mechanisms that contribute to oxidative stress-induced cell death involving STING in a disease context, including central nervous system (CNS) injury, remains mostly unknown.
Intriguingly, the findings from this study identify a protective role for the STING pathway in H 2 O 2 -induced oxidative stress. It was found that 500 µM H 2 O 2 induced higher production of intracellular ROS levels in the STING −/− cells ( Figure 5) which is associated with reduced pTBK1 (Figure 2A) and type-I (IFNα and IFNβ) levels ( Figure 3C,D) as compared to WT cells. Importantly, the elevation in ROS levels corresponds with reduced cellular viability in the STING −/− cells, suggesting STING ( Figure 4A,B) may exert its protective effect by limiting the generation of detrimental ROS. Additionally, STING might also act as an antioxidant or play a role in regulating antioxidant components, such as superoxide dismutases (SODs), catalases, peroxidases and reductases, all of which eliminate excessive H 2 O 2 -induced ROS generation. However, the precise mechanisms by which STING protects against H 2 O 2 -induced oxidative stress in this study warrant further investigation.

The Type-I IFN and STING Pathways Are Activated after H 2 O 2 Treatment
Furthermore, an increase in STING expression, both at the transcript and protein level, after H 2 O 2 treatment was identified ( Figure 1A,B). Increases in STING-mediated pTBK1 (Figure 2), IRF3 ( Figure 3A), IFNα ( Figure 3C) and IFNβ ( Figure 3D) levels were also identified, confirming STING pathway activation after H 2 O 2 treatment. These results, while not surprising, contradict the previous observations of STING in a TBI model. These discrepancies could be attributed to the complexity of the signaling pathways involved after TBI as compared to the simpler and controlled in vitro culture conditions that limit the activation of specific pathways induced by H 2 O 2 . Additionally, the peripheral-derived MEF cell line used in this study may elicit a different STING response as opposed to the complex interplay between neurons, microglia and astrocytes in the TBI brain. Henceforth, it can be postulated that the STING response may produce different outcomes in different cell types. Type-I IFNs can signal through the STING-TBK1-IRF3 axis induced by the nucleic acids, including cyclic-dinucleotide (CDN) or DNA, released by invading pathogens, resulting in the inhibition of viral and bacterial replication [49]. Recent studies have demonstrated STING-induced type-I IFN signaling through host-derived DAMPs, including mtDNA and self-DNA, in response to cellular injury [50,51]. Additionally, increased intracellular H 2 O 2 -induced ROS levels have been shown to promote the release of DAMPs, including ATP, uric acids, DNA, high mobility group box 1 (HMGB1) protein and mtDNA [52][53][54][55]. Henceforth, it is reasonable to postulate that the increases in STING ( Figure 1) and IFNβ ( Figure 3D) levels that coincide with reduced cellular viability ( Figure 4B) can be attributed to DAMPs released from injured or dying cells after 500 µM H 2 O 2 treatment. However, future studies confirming the exact molecules and underlying mechanisms leading to STING and type-I IFN signaling activation in this model are warranted.
It remains to be determined whether the downstream STING effectors, including increases in pTBK1, IRF3 and type-I IFN levels, contribute to the protective effects seen in the WT cells after 500 µM H 2 O 2 treatment. However, activation of type-I IFN signaling mediated by the NF-κB transcription factor has been shown to promote cellular survival against pro-apoptotic stimuli [55,56] and STING is known to induce type-I IFN production through NF-κB activation [57]. Furthermore, it is interesting to note that increased STING and IFNβ mRNA levels showed a similar trend of activation ( Figures 1A  and 3D) with the first wave of activation detected at 15 min and a second wave of activation detected at 24 h after 500 µM H 2 O 2 treatment. Therefore, given that STING is protective after 500 µM H 2 O 2 treatment, while activation of STING through DAMPs is consistent with a pro-death effect of STING reported in the literature [58,59], it is reasonable to postulate that STING may exert its protective effects through NF-κB-dependent type-I IFN signaling. However, future studies to confirm this are required.

Dual Role for STING in Regulating Autophagy Activity
The role of autophagy has been widely implicated under elevated oxidative stress with increased autophagic marker expression observed in disease pathogenesis [29,60]. However, its precise role and the mechanisms that trigger its induction and the involvement of the STING pathway remain unclear. Both the STING and type-I IFN pathways have emerged as critical players in autophagy activation in other cellular and disease models [61,62]. However, their interaction and regulation mechanisms following oxidative stress-induced cellular injury are unknown. Autophagy is a dynamic and complex cellular degradation process that requires careful analysis to identify and interpret its activity accurately. This study assessed hallmark markers of autophagy, including microtubule-associated protein 1 light chain 3 (LC3), SQSTM1/p62 and lysosomal-associated membrane protein 2 (LAMP2), in the H 2 O 2 -induced oxidative stress model. Conversion of LC3-I to LC3-II is representative of increased autophagy activation [63] while the degradation of p62 and LAMP2 at the late step of the autophagy process serves as a marker for normal autophagic flux. Given the observation of a protective effect in the WT cells following H 2 O 2 treatment, this suggests that the increased LC3-II and p62 levels observed in STING −/− cells do not indicate enhanced autophagy flux but rather impaired autophagy activity that contributes to cellular damage following elevated oxidative stress. Indeed, this study confirmed a protective role for STING following H 2 O 2 -induced oxidative stress in vitro. Evidence in the literature has also associated H 2 O 2 -induced oxidative stress with autophagy activation [58,64]. Here, we show H 2 O 2 -induced autophagy activation, with a trend for increased LC3-II and p62 levels in both H 2 O 2 -treated WT and STING −/− cells. However, STING −/− cells showed significantly increased and higher LAMP2 expression as compared to WT cells in response to H 2 O 2, suggesting a role for STING in regulating normal autophagy flux. It was also confirmed that increased and impaired autophagic activity, as measured by significant increases in LAMP2 protein levels, was evident in STING −/− cells following H 2 O 2 treatment. However, reduced LAMP2 levels in WT cells suggest an adaptation to normal autophagic activity in the presence of STING after H 2 O 2 insult. This normal autophagic activity may serve as a protective mechanism to remove damaged cells and promote a protective environment, thus partially contributing to the reduced cell death observed in WT cells after H 2 O 2 treatment.
The protective effect of STING in promoting cellular survival by maintaining normal autophagy activity after H 2 O 2 treatment differs from its detrimental effects observed in our TBI model. However, it is not surprising as the increases in the normal autophagy response found in this study coincide with elevated intracellular ROS which may act as a defense mechanism to protect cells against ROS-induced oxidative stress, as reported in the literature [65,66]. It is also noteworthy that impaired autophagy activity in the STING −/− cells correlated with reduced type-I IFN and STING-TBK1-IRF3 protein expression following H 2 O 2 treatment. Given observations that the STING and type-I IFN pathways play a role in autophagy activation, these results highlight a critical role for STING in modulating normal autophagic activity rather than just as an autophagy inducer. Interestingly, a recent report demonstrated that cGAS-mediated autophagic activity protects cells from ischemia/reperfusion-induced oxidative stress independent of STING [67], while increased STING activation that is detrimental following TBI was found to be associated with its role in normal autophagy flux [7]. These discrepancies observed underscore our lack of understanding of how STING modulates autophagic activity in different disease models. Evidence in the literature has also associated H 2 O 2 -induced oxidative stress with autophagy activation [26,36]. It is possible, therefore, that STING may act as a molecular switch in promoting either detrimental or beneficial outcomes in TBI and H 2 O 2 -induced oxidative stress, respectively, by modulating autophagy activity. However, the underlying mechanisms leading to oxidative stress during CNS injury involving the STING and type-I IFN pathways remain to be investigated.

MTT Assay
The WT and STING -/-MEF SV40 cells were a gift from Dr Kate McArthur (Walter and Eliza Hill Institute, Melbourne, Australia) and were cultured in Dulbecco's minimal essential media (DMEM; Thermoscientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS) and 5% penicillin/streptomycin (P/S; Thermoscientific, Waltham, MA, USA) at 37 • C/5% CO 2 . Cells were seeded at a density of 1 × 10 5 cells/well in 6-well plates to assess cell viability by the conventional 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction assay [68]. The WT and STING -/-MEF SV40 cells were treated for the indicated times with H 2 O 2 before being incubated with MTT (Sigma, St. Louis, MO, USA) solution (final concentration; 0.5 mg/mL) for 1 h at 37 • C/5% CO 2 . The medium was removed and 150 µL of dimethyl sulfoxide (DMSO) was added into each well, which was then read at 490 nm in a microplate reader. Results were expressed as percentages of the untreated control for that genotype.

DCF Assay
Intracellular ROS levels were quantitated using the 2 , 7 -dichlorofluorescin diacetate (DCFH-DA) assay (Sigma, St. Louis, MO, USA) [69]. WT and STING -/-MEF SV40 cells were seeded at a density of 2 × 10 5 cells/mL in a black 96-well plate with a transparent bottom and allowed to adhere overnight. Following H 2 O 2 treatment (15 min to 48 h; 5 replicates for each genotype and time point), the cells were incubated with DCFH-DA (10 µM) for 30 min in the dark at 37 • C. Fluorescence was measured at excitation 485 nm/emission 535 nm using a microplate fluorescence reader (FlexStation ® 3 Benchtop Multi-Mode Microplate Reader, Molecular Devices, San Jose, CA, USA). The signal intensity was calculated relative to the genotype specific DMSO control group with the relative fluorescence intensity proportional to the intracellular ROS levels.

RNA Extractions and cDNA Synthesis
Cell lysates were homogenized in 1 mL Trizol (Invitrogen, Waltham, MA, USA) and incubated at room temperature for 10 min. 0.2 mL chloroform (Chem Supply, Melbourne, Australia) per 1 mL Trizol was added to the samples and they were centrifuged at 12,000× g for 15 min at 4 • C. The colorless, aqueous phase of each sample containing the RNA was transferred into a fresh 1.7 mL microcentrifuge tube. RNA was precipitated by adding 0.5 mL Propan-2-ol (Chem Supply) per 1 mL Trizol and the samples were centrifuged again at 12,000× g for 10 min at 4 • C. The supernatant from the tubes was discarded, and the RNA pellet was washed with 75% ethanol (Chem Supply, Melbourne, Australia) in diethyl pyrocarbonate (DEPC)-treated water (Sigma, St. Louis, MO, USA), vortexed and centrifuged at 7500× g for 5 min at 4 • C. The RNA pellet was air-dried and redissolved in RNAse-free H 2 O (Invitrogen, Waltham, MA, USA). The concentration and purity of the RNA samples was measured using the NanoDrop 1000 Spectrophotometer (Thermo Scientific, Waltham, MA, USA).

Quantitative Real Time Polymerase Chain Reaction (qPCR) Analysis
cDNA was transcribed from 1 µg RNA using a high capacity cDNA reverse transcription kit (Applied Biosystems, Waltham, MA, USA) as previously described [4]. Genes of interest were detected using Taqman (Applied Biosciences, Waltham, MA, USA) ( Table 1) or SYBR green (GeneWorks, Thebarton, SA, Australia) ( Table 2) primers. Ct values were obtained for each sample and relative transcript levels for each gene were calculated using the 2 −∆∆Ct Method [70].

Western Blot Analysis
Protein concentrations were determined by Bradford assay (Bio-rad, Hercules, CA, USA) with 50 µg of protein lysed in 2× Novex ® Tris-glycine SDS sample buffer (Invitrogen, Waltham, MA, USA) for 10 min at 100 • C before resolution on 8% or 12% acrylamide SDS PAGE gels. Blots were then transferred to polyvinylidene fluoride (PVDF) membranes using a semi-dry transfer apparatus (BioRad, Hercules, CA, USA). Membranes were blocked in 5% skimmed milk in TBS-T for 1 h and incubated with primary antibodies (Table 3) in 2% skim milk in TBS-T at 4 • C overnight. Membranes were washed three times for 10 min each with TBS-T before being incubated with HRP-conjugated secondary antibodies (diluted in 2% skim milk in TBS-T) for 60 min at room temperature. Membranes were washed with TBS-T (3 × 10 min) and signals were detected using an ECL prime ® Western blotting detection kit (Amersham, Chicago, IL, USA) and visualized with the IQ350 imaging machine (GE Healthcare, Chicago, IL, USA). Post-image densitometry was performed using ImageJ software (NIH), whereby signal intensity was calculated in arbitrary units. For densitometry calculations, phosphorylation intensity was measured in arbitrary units and normalized to the β-actin loading control. These values were then calculated as fold change relative to untreated or vehicle control.

Statistical Analysis
Data are expressed as mean ± SEM and were analyzed using Graph Pad Prism 8.5 software. For qPCR and Western blot data, a one-way or two-way analysis of variance (ANOVA) was performed as appropriate, followed by a Bonferroni's post-hoc analysis, with a value of p < 0.05 considered statistically significant.

Conclusions
This study provides evidence to suggest a novel role for STING in mediating the type-I IFN pathway in a model of elevated oxidative stress. This finding proposes a previously unrecognized role for STING in modulating H 2 O 2 -induced oxidative stress and reveals the multifaceted role of STING in different disease/cellular models and elevated oxidative stress-related pathophysiological processes.  Signal transducer activator of transcription STING Stimulator of interferon genes TBI Traumatic brain injury TBK1 Tumour necrosis factor (TNF) receptor-associated factor NF-κB activator (TANK)-binding kinase 1