A diamidobenzimidazole STING agonist protects against SARS-CoV-2 infection

Coronaviruses are a family of RNA viruses that cause acute and chronic diseases of the upper and lower respiratory tract in humans and other animals. SARS-CoV-2 is a recently emerged coronavirus that has led to a global pandemic causing a severe respiratory disease known as COVID-19 with significant morbidity and mortality worldwide. The development of antiviral therapeutics are urgently needed while vaccine programs roll out worldwide. Here we describe a diamidobenzimidazole compound, diABZI-4, that activates STING and is highly effective in limiting SARS-CoV-2 replication in cells and animals. diABZI-4 inhibited SARS-CoV-2 replication in lung epithelial cells. Administration of diABZI-4 intranasally before or even after virus infection conferred complete protection from severe respiratory disease in K18-ACE2-transgenic mice infected with SARS-CoV-2. Intranasal delivery of diABZI-4 induced a rapid short-lived activation of STING, leading to transient proinflammatory cytokine production and lymphocyte activation in the lung associated with inhibition of viral replication. Our study supports the use of diABZI-4 as a host-directed therapy which mobilizes antiviral defenses for the treatment and prevention of COVID-19.


INTRODUCTION
As of January 2021, the recently emerged severe-acute respiratory syndrome coronavirus 2 has led to over 2 million deaths and over 100 million infections globally (1). SARS-CoV-2 is a member of the Coronaviridae family of viruses. Respiratory infections with SARS-CoV-2 can result in asymptomatic, mild or severe forms of a disease known as COVID- 19. More severe cases of COVID-19 result in death due to acute respiratory distress syndrome and damage to the alveolar lumen (2). Currently, there are few treatment options for COVID-19 patients. The antiviral RNA-dependent polymerase inhibitor remdesivir reduces length of hospitalization and deaths from COVID-19 (3). In addition, the steroid dexamethasone has also been approved for use in severe COVID-19 (4). To date numerous efficacious vaccines have been developed and rolled out (5,6). Despite these advances additional antiviral therapeutics will be required for the treatment of future endemic infections. An ongoing global effort is now underway to identify and develop new antiviral and anti-inflammatory therapeutics to reduce COVID-19 related hospitalizations and deaths.
The ER-resident adaptor protein Stimulator of Interferon Genes is a key signaling molecule that is activated following cytosolic DNA detection. Cyclic GMP-AMP synthase is an innate immune sensor of cytosolic DNA. Upon DNA binding cGAS converts ATP and GTP into the cyclic dinucleotide cGAMP, which in turn binds and activates STING (7). Conformational changes in STING lead to C-terminal phosphorylation of STING, dimerization, oligomerization and subsequent activation of autophagy, NF-κB, IRF3 and transcription of proinflammatory cytokines and type I IFNs (8)(9)(10). Activation of STING can elicit a potent anti-tumor response and the use of STING agonists in oncology alone or in combination with checkpoint blockade is an emerging therapeutic area (11)(12)(13). A recent study has identified a new class of STING agonists with systemic in vivo activity. Diamidobenzimidazole based compounds are potent, specific activators of STING and possess superior stability, tissue penetrance and potency over traditional cyclic dinucleotide STING agonists (14). While the therapeutic applications of STING agonists have been reported for use in oncology, the antiviral potential of STING agonists remains underexplored. Given the potent type I IFN CORONAVIRUS A diamidobenzimidazole STING agonist protects against SARS-CoV-2 infection (Page numbers not final at time of first release) 2 response induced by diABZI compounds, we hypothesized that pharmacological activation of STING may elicit protection from SARS-CoV-2 infection. RESULTS

diABZI-4 protects against herpes simplex encephalitis
To evaluate the potential antiviral effects of systemically administered diABZI-4, we first tested the efficacy of diABZI-4 in a mouse model of herpes simplex encephalitis. cGAS-and STING-deficient mice are highly susceptible to acute HSV-1 infection (15). Indeed, both cGAS and STING-deficient mice displayed increased weight loss, hydrocephalus and decreased survival following corneal infection with a neuroinvasive strain of HSV-1 ( Fig. S2A-F). Thus, we hypothesized that diABZI-4 may compensate for cGAS deficiency and confer protection from HSE through activation of STING. Treatment of cGAS −/− mice with a single dose of diABZI-4 delivered via retro-orbital injection resulted in complete protection from HSE ( Fig. 1J-M). cGAS −/− mice receiving diABZI-4 were protected from HSV-1-induced weight loss (Fig. 1J), lethality (Fig. 1K), hydrocephalus (Fig. 1L) and symptoms of neurological disease (Fig. 1M). In addition, cGAS −/− mice receiving diABZI-4 had a significant reduction in HSV-1 titers in brain tissue (Fig. 1N).

diABZI-4 protects against SARS-CoV-2 infection
Given that diABZI-4 strongly inhibited SARS-CoV-2 replication in lung epithelial cells we next determined if diABZI-4 could also prevent SARS-CoV-2 infection in vivo. SARS-CoV-2 cannot bind to murine ACE2; thus, a SARS-CoV-2 infection cannot be established in conventional laboratory mouse strains (16,19). K18-hACE2 transgenic mice (K18-ACE2), express human ACE2 under the control of the epithelial cell cytokeratin-18 (K18) promoter (20). Following intranasal inoculation with SARS-CoV-2, K18-ACE2 mice began to lose weight 4 days post-infection and die 7-8 days after infection (21,22). Thus, we utilized the K18-ACE2 mouse model to determine the effects of diABZI-4 on SARS-CoV-2 infection in vivo. K18-ACE2 mice inoculated intranasally with SARS-CoV-2 resulted in weight loss and lethality 8 days post-infection. However, K18-ACE2 mice administered a single intranasal dose of diABZI-4 were protected from SARS-CoV-2 induced weight loss (Fig. 3A) and lethality (Fig, 3B). Remarkably, di-ABZI-4 also protected mice from SARS-CoV-2 induced weight loss (Fig. 3C) and lethality ( Fig. 3D) when administered 12 hours post-infection. Administration of the same dose of di-ABZI-4 via intraperitoneal injection failed to protect K18-ACE2 mice from SARS-CoV-2 infection (Fig. 3E-F) indicating that direct delivery of diABZI-4 to mucosal surfaces was required for its protective effects. diABZI-4 and diABZI-3 conferred comparable protection from SARS-CoV-2-induced weight loss (Fig. S3A) and lethality (Fig. S3B). A concentration of 0.25 mg/kg was required for complete protection from SARS-CoV-2-induced weight loss (Fig. S3C) and lethality (Fig. S3D). Given these striking effects we next assessed the effect of diABZI-4 on viral loads in the lung. Pre-treatment of K18-ACE2 mice with diABZI-4 resulted in a decrease in the expression of numerous SARS-CoV-2 genes including N-protein (Fig. 3G), Nsp14 (Fig. 3H) and ORF1 (Fig. 3I) in lung tissue collected 48 hours post-infection. Pre-treatment with diABZI-4 also resulted in a decrease in the genome copy number of SARS-CoV-2 when compared to lung tissue from K18-ACE2 mice treated with vehicle control (Fig. 3J). NanoString analysis also demonstrated a significant decrease in SARS-CoV-2 RNA transcripts which correlated with a reduction in expression of a number of inflammatory genes in lung tissue from diABZI-4-treated K18-ACE2 mice indicative of enhanced, rapid viral clearance (Fig. 3K, Fig. S4A). Analysis of hematoxylin and eosin (H&E)-stained lung sections from K18-ACE2 mice infected with SARS-CoV-2 demonstrated severe lung inflammation 5 days post-infection. Vehicle-treated SARS-CoV-2 infected K18-ACE2 mice displayed immune cell infiltrates in large areas of the lung with focal accumulation in the adjacent alveolar spaces and severe thickening of the alveolar wall, whereas mice pre-treated with diABZI-4 were protected from these effects (Fig. 3K). Interestingly, diABZI-4 also protected against influenza A (IAV)-induced lethality ( Fig. S3E) and IAV replication in lung tissue (Fig. S3F).

diABZI-4 promotes myeloid and lymphocyte activation in the lung
We next sought to investigate the mechanism by which diABZI-4 mediates its protective effects. Intranasal administration of diABZI-4 resulted in the potent oligomerization of STING in the lung (Fig. 4A). diABZI-4 also induced the subsequent expression of a large number of genes ( Fig. S4B) including interferon-stimulated genes such as Cxcl10 (Fig.  4B), Ifit1 (Fig. 4C), Isg15 (Fig. 4D), Mx1 (Fig. 4E) and Stat1 (Fig. 4F) in a STING-dependent manner. Previous reports have demonstrated that STING activation elicits a type I interferon response that stimulates interferon receptor signaling in tumor-resident dendritic cells and leads to antitumor CD8 + T cell and NK cell responses (23)(24)(25). In addition, CD8 + T cells are essential for inhibition of tumor growth by diABZI based STING agonists (14). Thus, we next profiled immune cells in the lung following intranasal delivery of diABZI-4 ( Fig. 4G-L, Fig. S5A-C). Except for neutrophils (Fig. 4H,  Fig. S5A) and B cells (Fig. 4I, Fig. S5B), no significant changes in overall cell numbers were observed following intranasal treatment with diABZI-4 ( Fig. 4H-J, Fig. S5A-B)). Interestingly, diABZI-4 treatment promoted the activation of myeloid cells, γδ T cells and NK cells as evidenced by a significant increase in CD69 + cells ( Fig. 4J-K, Fig. S6A-B). di-ABZI-4 did not alter the frequency of central memory and effector T cells (Fig. 4L, Fig. S6C). Despite the antiviral, proinflammatory environment elicited by diABZI-4 in the lung, pathology evaluation of lung tissue 1, 2 and 5 days after intranasal delivery of diABZI-4 showed no gross pathological changes to the lung parenchyma. These data indicate that a single dose strategy is sufficient to eliminate viral infection without any pathological damage to the lung tissue.

diABZI-4 protects against SARS-CoV-2 infection through IFN-dependent and -independent mechanisms
Type I IFNs play an important role in restricting viral infection. Thus, we next assessed if induction of type I IFNs mediated the protective effects of diABZI-4 against SARS-CoV-2 infection. Intranasal treatment with anti-IFNAR inhibited diABZI-4 induced ISG expression ( Fig. S7A-B). Anti-IFNAR neutralizing antibody delivered intranasally partially but not completely blocked the protective effects of diABZI-4 in K18-ACE2 mice infected with SARS-CoV-2 ( Fig. 4O-P). Furthermore, a single dose intranasal dose of diABZI-4 conferred a much greater level of protection against SARS-CoV-2 infection when compared to a single dose of intranasal universal IFN or poly(I:C) (Fig. S7C-D). Universal IFN is a human IFN-alpha A/D hybrid which exhibits activity across multiple species. These data indicate that both IFNdependent and -independent downstream of STING mediate the protective effects of diABZI-4 in controlling SARS-CoV-2 infection in vivo.

DISCUSSION
We have identified a host-directed therapy that is efficacious for the treatment of SARS-CoV-2 infection. Pharmacological activation of STING in the lung during SARS-CoV-2 infection elicits a rapid short-lived antiviral response via type I IFNs, NF-κB driven cytokine production and lymphocyte activation resulting in inhibition of viral replication and prevention of severe respiratory disease. Use of diABZI-4 over other immunotherapies such as recombinant IFN offers several significant advantages including cost, enhanced stability, room temperature storage and potential for efficacy at low dose treatments.
SARS-CoV-2 is a respiratory pathogen which can infect ACE2 positive cells in the upper and lower respiratory tract. Type II alveolar cells are readily infected by SARS-CoV-2 and represent approximately 15% of alveolar cells. Type II alveolar cells are essential to produce surfactant proteins and support epithelial barrier integrity, innate immune responses and airway regeneration following lung insult (26). iAT2-ALI 3D cultures mimic the human airway and mirror host transcriptional responses to SARS-CoV-2 (27). diABZI-4 showed strong inhibition of SARS-CoV-2 replication in iAT2 cells and inhibited SARS-CoV-2-induced cell death of SP2 positive alveolar cells. Thus, the use of diABZI-4 in the human lung is predicted to be effective against SARS-CoV-2.
In vivo, diABZI-4 was efficacious against SARS-CoV-2 prior to and after infection. diABZI-4 administered 12 hours after SARS-CoV-2 infection provided comparable protection to diABZI-4 given as a pre-treatment. The infection of hACE2transgenic mice with SARS-CoV-2 results in rapid weight loss and morbidity over a period of 6-8 days. Thus, the relative time between infection and peak viral load is extremely short which provides limitations for the use of diABZI-4 to be administered therapeutically. Indeed, the use of other immunotherapies such as intranasal IFN or poly(I:C) are also limited by the rapid kinetics of SARS-CoV-2 infection. Therapeutic utility of multiple doses of universal IFN or poly(I:C) in a hamster model of SARS-CoV-2 was also limited to 24 hours after infection (28). While type I IFNs play an essential role in initiating the adaptive immune response to SARS-CoV-2, they alone are not sufficient to control SARS-CoV-2 infection (29,30). However, other reports in human studies have demonstrated an essential role for type I IFNs in preventing severe COVID-19 (31). In support of this some clinical trials have reported positive results for use of IFN in the early stages of COVID-19 (32). STING activation triggers type I IFN production which mediates activation of CD8+ T-cell responses. However, in addition to type I IFN responses NF-κB and non-canonical autophagy are also induced downstream of STING (10). Indeed, recent studies have identified IFNindependent roles for STING in the control of DNA virus infection and anti-tumor immunity (33)(34)(35). In addition to SARS-CoV-2, diABZI-4 also conferred protection from IAV infection. Thus, the host-directed immune response activated by diAZBI-4 may have broad treatment applications to other respiratory pathogens. Like other immunotherapies, future work will be required to determine the appropriate dose and means of delivery for the use of diABZI-4 in humans. Our study provides molecular and cellular characterization of di-ABZI-4 in the prevention of SARS-CoV-2 replication, treatment of COVID-19 and highlights its potential use as a treatment for COVID-19 and its potential for the treatment of future pandemics caused by respiratory pathogens in humans.

Study design
The aim of this study was to investigate the antiviral properties of a diABZI-based STING agonist. We investigated the effect of diABZI-4 on human coronavirus infections in cultured lung epithelial cells. Using a human ACE2 transgenic mouse model we determined if the immune responses triggered by STING could be leveraged as an antiviral treatment for SARS-CoV-2 infection. Sample sizes used in each experiment are detailed in the figure legends.

Biosafety
All study protocols were approved reviewed and approved by Environmental Health and Safety and Institutional review board at University of Massachusetts Medical School prior to study initiation. All experiments with SARS-CoV-2 were performed in a biosafety level 3 laboratory by personnel equipped with powered air-purifying respirators.

Mice
All animal experiments were approved by the Institutional Animal Care Use Committees at the University of Massachusetts Medical School. Animal were kept in a specific pathogen free (SPF) environment. Hemizygous K18-hACE2 C57BL/6J mice (strain: 2B6.Cg-Tg(K18-ACE2)2Prlmn/J) were obtained from The Jackson Laboratory. cGAS KO mice and STING KO mice were used as previously described (36). Animals were housed in groups and fed standard chow diets. Sample sizes used are in line with other similar published studies.
Culture of SARS-CoV-2 T175 flaks of Vero E6 cells were infected with the USA-WA1/2020 (NR-52281; BEI Resources) at an MOI of 0.1 for 48 hours. Supernatants were centrifuged at 450 g for 10 min and aliquoted and stored at -80°C. Virus titer was determined by TCID50 assay in Vero E6 cells.
Culture of HCoV-OC43 T175 flaks of Vero E6 cells were infected with Hcov-0C43 (NR-52281; BEI Resources) at an MOI of 0.1 for 48 hours. Supernatants were centrifuged at 450 g for 10 min and aliquoted and stored at -80°C. Virus titer was determined by TCID50 assay in Vero E6 cells.

SARS-CoV-2 infection
8-12-week-old male and female mice were anesthetized with intraperitoneal injection of ketamine (100 mg kg −1 body weight) and xylazine (10 mg kg −1 body weight). Mice were then infected intranasally with 2.5 × 10 4 PFU of SARS-CoV-2. Mice were monitored daily for weight loss and survival.

PR8-influenza A infections
8-12-week-old male and female mice, were anesthetized with isoflurane and intranasally infected with 400 PFU of Influenza A/PR/8/34 (H1N1). Mice were monitored daily for weight loss and survival.
HSV-1 infections 8-12-week-old male and female mice were anesthetized with intraperitoneal injection of ketamine (100 mg kg −1 body weight) and xylazine (10 mg kg −1 body weight). Corneas were scratched in a 10 × 10 crosshatch pattern and mice were either inoculated with 2 × 10 5 PFU of HSV-1 in 5 μl of PBS or mock infected with 5 μl of PBS. Mice were monitored daily for weight loss and assessed for ocular hair loss, eye swelling, hydrocephalus and symptoms related to neurological disease as previously described.

HSV-1 plaque assay
Vero cells (5x10 5 ) were plated in 6-well plates in DMEM containing 10% FCS and 1% penicillin streptomycin. The following day, brains were extracted from mice and homogenized in 1ml of medium. Brains were centrifuged for 10 min at 8000 g. Supernatant was collected and serially diluted 2fold in DMEM. 500ul of each dilution was added to the Vero cells for 1 hour with gentle shaking. After 1 hour, medium was removed and 2 ml DMEM containing 15 μg/ml purified human IgG (Sigma I4506). Cells were incubated at 37°C for 2 days. Medium was then removed, and cells were fixed in 100% ice cold methanol for 3 min. Plaques were stained with 10% crystal violet for 20 min with gentle shaking, washed and counted.

SARS-CoV-2 RNA analysis
Two days post infection, mice were euthanized in isoflurane. Lung tissue was placed in a bead homogenizer tube with 1 ml of MEM + 2% FBS. After homogenization, 100 μl of this mixture was placed in 300 μl Trizol LS (Invitrogen) and RNA was extracted with the Direct-zol RNA miniprep kit (Zymo) per the manufacturer's instructions. Quantification of SARS-CoV-2 RNA levels was performed using the QuantiFast Pathogen RT-PCR kit (Qiagen) and the US Centers for Disease Control and Prevention real-time RT-PCR primer/probe sets for 2019-nCoV_N2 (IDT).

SARS-CoV-2 viral titer
SARS-CoV-2 and HCoV-OC43 were titered in Vero E6 cells using median tissue culture infectious dose (TCID50). As previously described. Briefly, supernatants from SARS-CoV-2 infected ACE2-A549 cells were collected and 10 replicates per sample were 10-fold serially diluted in Vero E6 cells for 6 days and assessed for the presence of cytopathic effect (CPE). TCID50 was calculated using the Reed and Muench formula.
cDNA synthesis and real time PCR Total RNA was extracted from whole lung tissue or cells. 1 μg of RNA was reverse transcribed using the iScript cDNA synthesis kit (Bio-Rad). 5 ng of cDNA was then subjected to qPCR analysis using iQ SYBR Green Supermix reagent (Bio-Rad). Gene expression levels were normalized to TATAbinding protein (TBP) or HPRT. Relative mRNA expression was calculated by a change in cycling threshold method as 2 - . Specificity of RT-qPCR amplification was assessed by melting curve analysis. The sequences of primers used in this study are listed in Table 1.
NanoString Total RNA was isolated from whole colon using Aurum TM Total RNA Mini kit (Bio-Rad) and quantitated by a Nanodrop ND-1000 spectrophotometer (ThermoFisher Scientific). 50 ng of RNA was then hybridized with a custom probe set and data was analyzed using the NanoString nSolver analysis system (NanoString technology). Gene expression data was normalized to internal positive and negative controls. Heat map was generated using R-software.
Cell culture Human ACE2-A549 and Vero E6 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin. Human peripheral blood monocyte cell line were cultured in RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin. For isolation of BMDMs, tibias and femurs were removed from wild type mice and bone marrow was flushed with complete DMEM medium. Cells were plated in medium containing 20% (v/v) conditioned medium of L929 mouse fibroblasts cultured for 7 days at 37°C in a humidified atmosphere of 5% CO2. Medium was replaced every 3 days.
Immunofluorescence iAT2-ALI cells were fixed in 4% paraformaldehyde for 20 min at RT. After two PBS washes, cells were incubated with blocking buffer containing 5% donkey serum (Lampire Biological Laboratories, Cat# 7332100) in PBST buffer (0.2% Triton-x-100 in 1X PBS) for 45 min at RT. After blocking, cells were incubated overnight at 4°C in blocking buffer containing primary antibodies against SARS-CoV-2 Spike (Novus Biological, Cat# NBP2-90980G), pro-surfactant protein C (Abcam, Cat # ab90716) and NKX2.1 (Thermo Fisher scientific, Cat # MA5-13961). Next day, cells were washed in PBST buffer 3 times for 5 min each. Cells were incubated with Alexa Fluor conjugated secondary antibodies for 2 hours at RT. After 3 PBST washes, nuclei were stained with Hoechst 33342 (Thermo Fischer Scientific, Cat# H3570) for 7 min and stored in PBS. Images were captured using Nikon Eclipse Ti microscope. ELISA Conditioned media or serum was collected as indicated and mouse or human IFNβ, TNF-α, CXCL10 and IL-6 were quantified by sandwich ELISA (R&D Systems).
Histology and immunohistochemistry Tissue blocks were sectioned at 5 μm thick. For paraffinembedded tissue, lungs were intratracheally inflated with 10% buffered-formalin and dissected from mice. Tissues were fixed in 4% paraformaldehyde overnight before being processed and embedded in paraffin. Five micrometer thin sections were stained by H&E or PAS in an automated stainer (Leica Autostainer XL). Histomorphology of each H&E slide was evaluated by Applied Pathology Systems at low and highpower field on an Olympus BX40 microscope, and the images were captured with Olympus cellSens Entry software at X4 magnifications. Grading of histology scores were performed by Applied Pathology Systems. Tissue sections were stained with H&E for evaluation of inflammation. IHC staining of PR8-IAV nucleoprotein-1 (NP-1) was performed with anti-influenza A virus nucleoprotein antibody (AA5H; ab20343, Abcam) and detected using horseradish peroxidase-conjugated secondary antibodies after heat-induced antigen retrieval as previously described. Diaminobenzidine was used for detection. Images were captured with a Nikon Ds-Ri2 microscope.
CD14 + monocyte isolation PBMC were isolated from whole blood of consenting donors. Blood was diluted 1:1 in sterile PBS and layered over 15 mls of Lymphoprep. Blood was spun at 450g with no brake. The interphase was transferred to a fresh tube using a Pasteur pipette and washed twice in PBS. Red blood cells were lysed in red blood cell lysis buffer for 10 min at room temperature. Cells were washed once more in PBS and counted. CD14 + monocytes were isolated using human CD14 magnetic microbeads (Miltenyi), washed twice in ice-cold macs buffer and plated in RPMI medium.

Ethics
All animal studies were performed in compliance with the federal regulations set forth in the Animal Welfare Act (AWA), the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and the guidelines of the UMass Medical School Institutional Animal Use and Care Committee. All protocols used in this study were approved by the Institutional Animal Care and Use Committee at the UMass Medical School (protocols A-1633).