Innate immunity induces an immune response through pattern recognition receptors (PRRs), which recognize pathogen- and danger-associated molecular patterns (PAMPs and DAMPs) (Matzinger 1994; Bianchi 2007). Cyclic GMP-AMP synthase (cGAS) is a type of PRRs that protects the host from disease by sensing cytoplasmic double-stranded DNA (dsDNA) of various pathogens such as bacteria, viruses, or retroviruses (Tan et al. 2018). Furthermore, cGAS plays a pivotal role in preventing inflammatory diseases and cancer by recognizing misplaced nuclear or mitochondrial dsDNA from the host (Nagata et al. 2010; Kanneganti et al. 2015; Harding et al. 2017; Liu et al. 2018). When various intracellular dsDNA bind to cGAS, it uses GTP and ATP to produce 2’3’-cGAMP (c [G (2’, 5’)pA (3’, 5’) p]), which then activates the downstream factor STING (Ablasser et al. 2013; Zhang et al. 2013; Kato et al. 2017). The subsequent signaling pathway progresses through TANK-binding kinase 1 (TBK1) and Interferon regulatory factor 3 (IRF3) activation, ultimately leading to the induction of the type I interferon and other cytokine components of the innate immune responses (Kato et al. 2017).

While the cGAS/STING pathway plays a pivotal role in the activation of innate immunity by cytosolic DNA, inappropriate activation can lead to diseases such as autoimmune disorders. Acardi-Goutieres syndrome (AGS) is a representative inflammatory neurodevelopmental disorder caused by excessive cGAS activation. This disease is characterized by excessive levels of type I interferons (Ahn and Barber 2014; Gao et al. 2015). Deletion of cGAS in a mouse model with AGS was found to suppress inflammation and autoimmune responses (Stetson et al. 2008; Gray et al. 2015). Additionally, myocardial infarction is a disease that occurs when blood flow is reduced or stopped in heart, causing damage to the cardiac muscle. dsDNA released from the damage can activate an inflammatory response by exposing to dendritic cells, leading to myocardial destruction and an increased risk of early mortality (King et al. 2017; Cao et al. 2018).

cGAS has emerged as a key therapeutic target for the treatment of autoimmune disorders. Inhibition of cGAS can be achieved by direct binding or by regulating upstream or downstream pathways. Efforts have been made to discover small molecules that can inhibit cGAS to treat various diseases caused by abnormal cGAS activation. One promising cGAS inhibitor, RU.521 binds to the active pocket of cGAS, reducing its activity and inhibiting interferon secretion, but its development was discontinued in in vivo experiments due to toxicity and off-target effects (Vincent et al. 2017; Wiser et al. 2020). Suramin was shown to inhibit the generation of 2’3’-cGAMP by dsDNA in vitro and inhibit the binding of cGAS to DNA. Suramin is thought to compete with DNA and RNA as a nucleic acid mimetic. However, there have been no in vivo studies conducted (Wang et al. 2018).

We attempted to find and develop a new cGAS inhibitor that can overcome the limitations of compounds currently under development. To identify promising cGAS inhibitors, we first screened approximately 6,000 compounds using IFN-stimulated response elements (ISRE)-dependent luciferase assays. Among the compounds that showed 90% or greater inhibition of luciferase activity, butaclamol was selected for the study.

Reagents

(+)-Butaclamol hydrochloride was purchased from Sigma Aldrich (St. Louis, MO, USA). G3-YSD, 2’3’-cGAMP, Poly I:C, ODN 2395, ODN-INH-18, and 5’ppp dsRNA were purchased from Invivogen (San Diego, CA, USA). M62812 and CU CPT 4a were purchased from Tocris (Bristol, UK). LPS was purchased from Biological Laboratory Inc (Campbell, CA, USA).

Animals

Animal care and the experimental protocols were carried out in accordance with the guidelines of the Institutional Animal Care and Use Committee (IACUC) of the Catholic University of Korea (permission #, 2022-006, 2014-006). C57BL/6 mice (Purbey et al. 2017) were obtained from Raon Bio (Seoul, Korea). The mice were housed in a room controlled for optimal temperature (23 ± 3℃) and relative humidity (40-60%) under specific pathogen-free condition and were acclimated in specific pathogen-free conditions in an animal facility for at least one week before experimentation.

Cell culture

Human THP-1 monocyte like cells (THP-1, ATCC, Manassas, VA, USA) were maintained in RPMI1640 medium (Gibco, Waltham, MA, USA) containing 10% (v/v) heat-inactivated fetal bovine serum (FBS) (Corning, Steuben, NY, USA), 50 µM of 2-mercaptoethanol (Gibco), 100 units/mL of penicillin (Gibco), 100 µg/mL of streptomycin, and 2 mM of glutamine. THP-1 Lucia ISG cells and THP-1 cGAS knockout (KO) cells were purchased Invivogen. THP-1 Lucia ISG cells were maintained in the same medium as THP-1 with the addition of 100 µg/mL of Normocin, and 100 µg/mL of Zeocin (Invivogen). THP-1 cGAS KO cells were cultured in the same medium as THP-1 Lucia ISG cells with the addition of 10 µg/mL of Blastacidin (Invivogen). Human embryonic kidney cells (HEK 293T cells) were cultured in Dulbecco’s modified eagle medium (DMEM) containing 10% (v/v) FBS, 100 units/mL of penicillin, and 100 μg/mL of streptomycin. Preparation of bone marrow-derived dendritic cells (BMDCs) was performed as previously described (Yeon et al. 2017). Bone marrow cells were isolated from bone marrow of C57BL/6 mice and differentiated into dendritic cells in RPMI1640 medium containing 10% (v/v) FBS, 40 µM of 2-mercaptoethanol, 100 units/mL of penicillin, 100 units/mL of streptomycin, 2 mM of glutamine, and 3% of J558L hybridoma cell culture supernatant for 6 days. Non-adherent cells were collected and used as dendritic cells. Cells were maintained at 37℃ in a 5% CO₂/air environment.

Luciferase assay

G3-YSD was complexed with LyoVec (Invivogen) according to the manufacturer’s instruction. After THP-1 Lucia ISG cells were stimulated with G3-YSD, cell culture supernatants were harvested and luciferase activity was measured using QUANTI-Luc™ Luciferase reagent (Invivogen), and CentroXS3 LB960 luminometer (Berthold Technologies, Bad Wildbad, Germany) according to the manufacturer’s instrument.

Cell viability assay

Cell viability was determined by MTT assay. MTT (5 mg/mL) was added to each well and incubated at 37℃ for 4 h. Cell culture medium was removed, and the cells were dissolved in DMSO. The intensity was measured at 570 nm using a VersaMax Microplate Reader (Molecular devices, San Jose, California, US).

Transfection and overexpression study

The IFNPRD III-I promotor luciferase reporter plasmid, pTRIP-pCMV-GFP-hcGAS plasmid (Addgene, Watertown, MA, USA), and pCMV-Myc-hSTING plasmid (kindly provided by Dr. Andrew Bowie) was used. HEK293T cells were seeded 24 h before transfection. Plasmids and Superfect (Qiagen, Hilden, Germany) were incubated for 10 minutes at a ratio of 1 to 5 and then treated to the cells with media. The compounds were treated 24 h after transfection and harvested with a passive lysis buffer of the Dual luciferase assay kit (Promega, Madison, WI, USA). Dual-Luciferase assay was detected using Centro XS3 LB 960 luminometer (Berthold Technologies) according to manufacturer’s instrument.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed as previously described (Kim et al. 2010). The IFN-β protein levels in the culture supernatants and 2’,3’-cGAMP in the cells were determined by ELISA (R&D Systems, Minneapolis, MN, USA, and Cayman, Ann Arbor, MI, USA, respectively) according to the manufacturer’s instructions. The concentration ranges of the standard curves were 46.875 to 3000 for IFN-β and 9.7 to 10,000 for 2’,3’-cGAMP. Samples were properly diluted to be measured within the standard curve ranges. The intensity was read with a microplate reader (Molecular Devices, San Francisco, CA, USA).

Reverse transcription (RT) and quantitative PCR analysis

PCR was performed as previously described (Joung et al. 2011). Total RNAs were isolated with trizol reagent (Invitrogen, Carlsbad, CA, USA). RNAs were reverse transcribed with ImProm-II™ Reverse Transcriptase (Promega). Synthesized cDNAs were amplified with IQ™ SYBR^® Green Supermix (Bio-Rad, Hercules, CA, USA) and primer using an IQ™5 (Bio-Rad) for quantitative real-time PCR. The specificity of the amplified PCR products was analyzed by a melting curve analysis. Fold-induction of gene expression was calculated after mRNA levels of each target gene were normalized to β-actin levels in corresponding samples. Primer sequences were as follows: hIfn-β, 5’-GCTCTCCTGTTGTGCTTCTCCAC-3’; and 5’-CAATAGTCTCATTCCAGCCAGTGC-3’; mIfn-β, 5’-TCCAAGAAAGGACGAACATTCG-3’ and 5’-TGAGGACATCTCCCACGTCAA-3’; β-actin, 5’-TCATGAAGTGTGACGTTGACATCCGT-3’ and 5’-TTGCGGTGCACGATGGAGGGGCCGGA-3’

Statistical analysis

Statistical analysis was performed using the software GraphPad Prism7 (GraphPad Software, San Diego, CA, USA). All data were expressed as the mean ± SD (n = 3). Datasets were analyzed by one-way ANOVA followed by Turkey’s multiple comparison test. p-values < 0.05 were considered significant.

To investigate whether butaclamol (Fig. 1A) inhibited the cGAS activation, the effect of butaclamol on ISRE-dependent luciferase activity induced by G3-YSD, a cGAS agonist, was determined in THP-1 Lucia ISG cells. G3-YSD is a potent cGAS agonist consisting of Y-shaped dsDNA with a guanosine overhang (Herzner et al. 2015). Treatment of THP-1 Lucia ISG cells with G3-YSD induced ISRE-dependent luciferase expression, while butaclamol treatment at 5 and 10 µM reduced G3-YSD-induced luciferase expression (Fig. 1B). The suppression by butaclamol was comparable to the inhibition by RU.521, a commercial inhibitor of cGAS (Fig. 1B). To examine whether the reduction of luciferase expression by butaclamol was due to cytotoxicity, a MTT assay was performed. Cell viability in butaclamol treatment groups was similar with that in G3-YSD alone group, indicating butaclamol did not induce cytotoxicity and the inhibitory effects of butaclamol was not due to the cytotoxicity (Fig. 1C). These findings demonstrate that butaclamol inhibits the dsDNA-induced cGAS pathway activation. The activation of cGAS culminates the production of 2’3’-cGAMP as a result of cGAS enzymatic activity after recognition of dsDNA by cGAS using GTP and ATP. Therefore, we measured whether butaclomol reduced the production of 2’3’-cGAMP as a result of cGAS inhibition. Butaclamol reduced the levels of 2’3’-cGAMP increased by G3-YSD in THP-1 cells, further confirming the inhibition of cGAS enzymatic activity (Fig. 1D).

Figure 1. Butaclamol inhibits cGAS activation induced by G3-YSD. (A) Structure of butaclamol. (B) After THP-1 Lucia ISG cells were pre-treated with vehicle, butaclamol (1, 5, 10 µM), or RU.521 (10 µM) for 1 h, the cells were further treated with G3-YSD (0.5 µg/mL) for 24 h. Luciferase activity was measured and expressed as a fold induction as compared with vehicle control. (C) Cytotoxicity assay was performed under the same conditions as the (B). (D) After THP-1 cells were pre-treated with vehicle, butaclamol, or RU.521 for 1 h, the cells were stimulated with 1 µg/mL G3-YSD for 2 h. Intracellular 2’,3’-cGAMP were analyzed by ELISA. (E) After HEK293T cells were transfected with an IFN-β PRDIII-I luciferase reporter gene, a renilla reporter gene, and plasmids expressing cGAS (pTRIP-pCMV-GFP-cGAS) or STING (pCMV-STING), cells were pre-treated with butaclamol for 1 h and then treated with 1 µg/mL G3-YSD for 24 h. Cell lysates were analyzed for IFN-β PRDIII-I luciferase and renilla luciferase activities using the dual-luciferase assay. Data are expressed as the mean ± SD (n = 3). ^#Significantly different between vehicle and G3-YSD alone, p < 0.05. *Significantly different from G3-YSD alone, p < 0.05.

To confirm the inhibitory activity of butaclamol, we employed a gain-of-function approach. HEK293T cells were transfected with expression plasmids expressing cGAS or STING together with a reporter luciferase gene containing an IFN-β promoter region and the cells were pre-treated with butaclamol for 1 h and then stimulated with G3-YSD for 24 h. Butaclamol suppressed G3-YSD-induced reporter luciferase expression in cGAS/STING overexpressed HEK293T cells (Fig. 1E), showing the inhibition of cGAS/STING pathway by butaclamol.

2’3’-cGAMP produced upon the activation of cGAS by dsDNA activates STING, the downstream signaling molecule of cGAS, that further activates TBK1 and IKKβ to induce the activation of transcription factors such as IRF3 and NF-κB. We next, investigated whether the inhibitory effects of butaclamol on luciferase reporter expression in Fig. 1 was due to the suppression of STING, which is a downstream component of cGAS. THP-1 Lucia ISG cells were pre-treated with butaclamol, and further treated with 2’3’-cGAMP the ligand of STING. ISRE-dependent luciferase expression was increased by 2’3’-cGAMP in THP-1 Lucia ISG cells, while butaclamol did not inhibit 2’3’-cGAMP-induced luciferase expression (Fig. 2A). H-151, a well-known STING antagonist (Haag et al. 2018), suppressed 2’3’-cGAMP-induced luciferase expression (Fig. 2A). Butaclamol did not affect cell viability of THP-1 Lucia ISG cells (Fig. 2B). The data show that butaclamol does not inhibit the 2’3’-cGAMP-induced activation of STING.

Figure 2. Butaclamol does not inhibit luciferase reporter expression induced by STING agonist. (A) THP-1 Lucia ISG cells were pre-treated with vehicle, butaclamol (1, 5, 10 µM), or H-151 (10 µM) for 1 h, and then further treated with 2’3’-cGAMP (10 µg/mL) for 24 h. (B) Cytotoxicity assay was performed under the same conditions as the (A). (C) THP-1 cGAS knockout (KO) cells were pre-treated with vehicle, butaclamol, or RU.521 for 1 h and stimulated with G3-YSD for 2 h. (D) THP-1 cGAS knockout (KO) cells were treated with the same conditions as the (A). Data are expressed as the mean ± SD (n = 3). ^#Significantly different between vehicle and 2’3’-cGAMP alone, p < 0.05. *Significantly different from G3-YSD or 2’3’-cGAMP alone, p < 0.05.

In cGAS knockout (KO) THP-1 cells, G3-YSD, a cGAS agonist, did not increase the expression of luciferase reporter gene and butaclamol did not reduce the luciferase reporter gene expression (Fig. 2C). In contrast, 2’3’-cGAMP, a STING agonist, increased the expression of luciferase reporter gene in cGAS KO THP-1 cells, while butaclamol did not suppress 2’3’-cGAMP-induced luciferase reporter gene expression (Fig. 2D). These further confirm that butaclamol does not inhibit activation of STING.

The activation of cGAS culminates in the production of type I interferons. Therefore, we examined whether the inhibition of cGAS by butaclamol resulted in the suppression of type I interferon expression. Butaclamol reduced the mRNA levels of IFN-β increased by G3-YSD in THP-1 cells (Fig. 3A). Similarly, butaclamol blocked the increase of IFN-β mRNA levels in primary bone marrow-derived dendritic cells (BMDCs) stimulated by G3-YSD (Fig. 3B). Consistently, butaclamol inhibited the protein levels of IFN-β increased by G3-YSD in BMDCs (Fig. 3C). The inhibitory potency of butaclamol was comparable with RU.521, a cGAS inhibitor (Fig. 3). The results show that butaclamol suppresses the cGAS target gene expression.

Figure 3. Butaclamol suppresses the expression of IFN-β induced by a cGAS agonist. (A) THP-1 cells were pre-treated with vehicle, butaclamol (1, 5, 10 µM), or RU.521 (10 µM) for 1 h, and then treated with G3-YSD (1 µg/mL) for 8 h. Cells were harvested, and mRNA levels of IFN-β were analyzed by quantitative real-time PCR. (B) Primary bone marrow-derived dendritic cells were treated under the same conditions of (A). (C) Primary bone marrow-derived dendritic cells were pre-treated with the same condition as the (B), and then treated with G3-YSD for 24 h. Protein levels of IFN-β was measured by ELISA. Data shown as the mean ± SD (n = 3). ^#Significantly different between vehicle and G3-YSD alone, p < 0.05. *Significantly different from G3-YSD alone, p < 0.05.

There are other pattern-recognition receptors, of which activation leads to the production of type I interferons. Therefore, we next investigated whether butaclamol inhibited other pattern-recognition receptors. Butaclamol did not reduce the mRNA levels of IFN-β increased by poly I:C (a TLR3 agonist), LPS (a TLR4 agonist), and ODN2395 (a TLR9 agonist) in THP-1 cells (Fig. 4A-4C). Inhibitors of each TLR reduced the IFN-β mRNA levels increased by the corresponding TLR agonist (Fig. 4A-4C). RIG-I is a cytosolic sensor to detect viral infection, inducing the IFN-β production. 5’ppp dsRNA, a RIG-I ligand, increased the IFN-β mRNA levels (Fig. 4D). Butaclamol did not reduce 5’ppp dsRNA-induced the IFN-β mRNA (Fig. 4D). These results suggest butaclamol rather specifically inhibits cGAS, but not other pattern-recognition receptors.

Figure 4. Butaclamol does not reduce the IFN-β mRNA levels of induced by agonists of various pattern recognition receptors. THP-1 cells differentiated with phorbol 12-myristate 13-acetate (PMA, 100 nM) for 3 days. Differentiated THP-1 cells were pre-treated with vehicle, butaclamol (1, 5, 10 µM), or various PRR inhibitors for 1 h, and then treated receptor agonist. Cells were harvested and mRNA levels of IFN-β was analyzed by quantitative real-time PCR. (A) CU CPT 4a (27 µM) was pre-treated as a TLR3 inhibitor, and then treated Poly I:C (5 µg/mL) for 2 h. (B) M62812 (3 µg/mL) was pre-treated as a TLR4 inhibitor, and then treated LPS (100 ng/mL) for 8 h. (C) ODN-INH-18 (2.5 µM) was pre-treated as a TLR9 inhibitor, and then treated ODN 2395 (2.5 µM) for 4 h. (D) Cells were treated 5’ppp dsRNA (1 µg/mL) for 8 h. Data are expressed as the mean ± SD (n = 3). ^#Significantly different between vehicle and (A) poly I:C, (B) LPS, or (C) ODN2395 alone, p < 0.05. *Significantly different from Poly I:C, LPS, ODN2395, or 5’ppp dsRNA alone, p < 0.05.

We screened approximately 6,000 library compounds using an ISRE-luciferase assay and discovered small molecule compounds that inhibit the cGAS activation. Butaclamol exhibited an inhibitory potency to reduce ISRE-luciferase expression increased by a cGAS activator, dsDNA, without evident cytotoxicity (Fig. 5). Butaclamol suppressed dsDNA-induced generation of 2’3’-cGAMP, of which production is dependent on cGAS enzymatic activity. Butaclamol demonstrated no off-target effects on other innate immune sensors such as TLRs and RIG-I at concentrations that inhibit cGAS activity, and also reduced the generation of 2’3’-cGAMP, a product of cGAS activated by dsDNA. Additionally, we confirmed the effectiveness of butaclamol on both human and mouse cGAS activity by measuring a decrease in interferon mRNA levels in human monocytes (THP-1 cells) and primary mouse bone marrow-derived dendritic cells.

Figure 5. Schematic model of cGAS inhibition by butaclamol. Butaclamol inhibits the cytosolic DNA sensor cGAS that generates cGAMP for STING activation and production of type I interferons, but not STING. As an inhibitor of cGAS/STING signaling, butaclamol has potential to modulate the innate immune response against DNA viruses, RNA-associated pathogens, and various bacteria. The figure was illustrated using BioRender.com.

The role of cGAS in innate immunity is well known. cGAS is an important protein that detects dsDNA derived from invading pathogens and damaged host cells within cytosolic compartment of immune cells and recognizes dsDNAs to trigger an inflammatory response. cGAS interacts with dsDNA to generate a second messenger molecule called 2’3’-cGAMP, which is involved in activating the immune system and initiating inflammatory responses. Therefore, cGAS plays a crucial role in detecting threats, and activating the immune system to protect cells from danger signals. Despite its important role, excessive activation of cGAS has been implicated in the development of inflammation and autoimmune diseases such as Acardi-Goutieres syndrome (AGS) and lupus. Therefore, there are many trials to develop cGAS pathway inhibitors by directly or indirectly inhibiting cGAS or its downstream signaling factors. Currently, there are no cGAS inhibitors that have entered clinical trials as drugs for treating autoimmune or inflammatory diseases. Our research has demonstrated that a new small molecule compound, butaclamol, can inhibit the activation of cGAS induced by dsDNA. Butaclamol was originally developed as an agent classified as an antidepressant and antipsychotic. It showed significantly lower potency compared to other enantiomers (Voith and Cummings 1976). It works by inhibiting the reuptake of serotonin and norepinephrine in the central nervous system, increasing concentration of neurotransmitters in the brain and improving symptoms of depression and anxiety. Additionally, butaclamol is primarily used to relive anxiety and tension, and may also have some effects in relieving pain such as headache, neuralgia, and myalgia (Hall and Strange 1997). However, it is no longer being developed for reasons that are unknown. Further investigation is needed to determine how butaclamol regulates cGAS activity and its efficacy in disease models. It will also be necessary to compare the effects on dopamine receptor antagonism at therapeutic concentrations. However, our study proposes a potential cGAS inhibitor with cellular activity and suggests a new candidate for developing autoimmune disease therapeutics that are related to cGAS activation in immune cells.

The authors declare that they have no conflict of interest.

This study was supported by grants from the National Research Foundation of Korea (NRF‐2019R1A2C2085739) and the Bio & Medical Technology Development Program of the National Research Foundation (No. 2022M3E5F1017414) funded by the Korean government (Ministry of Science and ICT). The Figure 5 was illustrated with BioRender.com.