cGAS-STING PATHWAY LIMITS SARS-CoV-2 REPLICATION IN ACE2+ AIRWAY CELL LINES

Background: We previously screened 10 human lung and upper airway cell lines expressing variable levels of endogenous ACE2/TMPRSS2. We found that H522 human lung adenocarcinoma cells supported SARS-CoV-2 replication independent of ACE2, whereas the ACE2 positive cell lines were not permissive to infection. Type I/III interferons (IFNs) potently restrict SARS-CoV-2 replication through the actions of hundreds of interferon-stimulated genes (ISGs) that are upregulated upon IFN signaling. Here we report that a number of ACE2 positive airway cell lines are unable to support SARS-CoV-2 replication due to basal activation of the cGAS-STING DNA sensing pathway and subsequent upregulation of IFNs and ISGs which restrict SARS-CoV-2 replication. Method(s): SARS-CoV-2 WT strain 2019-nCoV/USA-WA1/2020 viral replication was detected through analysis of cell associated RNA. RNA sequencing was used to study the basal level of genes in the type-I IFN pathway in the 10 cell lines, which was further validated by western blotting and qRT-PCR. A panel of 5 cell lines, with varying expression levels of ACE2 and TMPRSS2, were pre-treated with Ruxolitinib, a JAK1/2 inhibitor. A siRNA-mediated screen was used to determine the molecular basis of basally high expression of ISGs in cell lines. CRISPR knockout of IFN-alpha receptor and cGAS-STING pathway components was conducted in parallel Results: Here we show that higher basal levels of IFN pathway activity underlie the inability of ACE2+ cell lines to support virus replication. Importantly, this IFN-induced block can be overcome by chemical inhibition and genetic disruption of the IFN signaling pathway or by ACE2 overexpression, suggesting that one or more saturable ISGs underlie the lack of permissivity of these cells. Ruxolitinib treatment increased SARS-CoV-2 RNA levels by nearly 3 logs in OE21 and SCC25. Furthermore, the baseline activation of the STING-cGAS pathway accounts for the high ISG levels and genetic disruption of the cGAS-STING pathway enhances levels by nearly 2 and 3 logs of virus replication in the two separate ACE2+ cell line models respectively. Conclusion(s): Our findings demonstrate that cGAS-STING-dependent activation of IFN-mediated innate immunity underlies the inability of ACE2+ airway cell lines to support SARS-CoV-2 replication. Our study highlights that in addition to ACE2, basal activation of cGAS-STING pathway, IFNs and ISGs may play a key role in defining SARS-CoV-2 cellular tropism and may explain the complex SARS-CoV- 2 pathogenesis in vivo.

TARGETING THE HOST-VIRUS INTERFACE TO BLOCK SARS-CoV-2 ASSEMBLY IN AIRWAY CELLS

Background: The rapid emergence of the SARS-CoV-2 Omicron variant that evades many therapies illustrates the need for antiviral treatments with high genetic barriers to resistance. The small molecule PAV-104, identified through a moderate-throughput screen involving cell-free protein synthesis, was recently shown to target a subset of host protein assembly machinery in a manner specific to viral assembly with minimal host toxicity. The chemotype shows broad activity against respiratory viral pathogens, including Orthomyxoviridae, Paramyxoviridae, Adenoviridae, Herpesviridae, and Picornaviridae, with low susceptibility to evolutionary escape. Here, we investigated the capacity of PAV-104 to inhibit SARS-CoV-2 replication in human airway epithelial cells (AECs). Method(s): Dose-dependent cytotoxicity of PAV-104 in Calu-3 cells was determined by MTT assay. Calu-3 cells were infected with SARS-CoV-2 isolate USA-WA1/2020 (MOI=0.01). Primary AECs were isolated from healthy donor lung transplant tissue, cultured at air liquid interface (ALI), and infected with SARS-CoV-2 Gamma, Delta, and Omicron variants (MOI=0.1). SARS-CoV-2 replication was assessed by RT-PCR quantitation of the N gene, immunofluorescence assay (IFA) of nucleocapsid (N) protein, and titration of supernatant (TCID50). Transient co-expression of four SARS-CoV-2 structural proteins (N, M, S, E) to produce virus-like particles (VLPs) was used to study the effect of PAV-104 on viral assembly. Drug resin affinity chromatography was performed to study the interaction between PAV-104 and N. Glycerol gradient sedimentation was used to assess N oligomerization. Total RNA-seq and the REACTOME database were used to evaluate PAV-104 effects on the host transcriptome. Result(s): PAV-104 reached 50% cytotoxicity in Calu-3 cells at 3732 nM (Fig.1A). 50 nM PAV-104 inhibited >99% of SARS-CoV-2 infection in Calu-3 cells (p< 0.01) and in primary AECs (p< 0.01) (Fig.1B-E). PAV-104 specifically inhibited SARS-CoV-2 post entry, and suppressed production of SARS-CoV-2 VLPs without affecting viral protein synthesis. PAV-104 interacted with SARS-CoV-2 N and interfered with N oligomerization. Transcriptome analysis revealed that PAV-104 treatment reversed SARS-CoV-2 induction of the interferon and maturation of nucleoprotein signaling pathways. Conclusion(s): PAV-104 is a pan-respiratory virus small molecule inhibitor with promising activity against SARS-CoV-2 in human airway epithelial cells that should be explored in animal models and clinical studies.

Differential innate immune response and replication of SARS-CoV-2 in human diabetic corneal epithelial cells

Purpose : Diabetes predisposes an individual to severe COVID-19. Diabetic cornea is also known to have impaired wound healing, increasing the chances of infection. Earlier, we reported the ability of SARS-CoV-2 to infect conjunctival cells, and the presence of viral RNA and proteins was also detected in the corneas of COVID-19 donors. In this study, we evaluated the effect of diabetes on corneal innate immune response during SARS-CoV-2 infection and sought to determine the underlying mechanisms. Methods : Human primary corneolimbal epithelial cells (HCECs) were isolated from the corneas of three diabetic and three non-diabetic donors. In vitro studies were performed by infecting HCECs with SARS-CoV-2-USA-WA1/2020 strain at MOI 0.5. Viral replication was assessed by viral genome copy number. RNAseq analysis was performed to determine genes/pathways altered by diabetic vs non-diabetic HCECs. qPCR was used to assess the expression of innate inflammatory and antiviral genes. Western blot was performed to detect the protein expression of antiviral signaling molecules. Results : The primary HCECs were found permissive to SARS-CoV-2 infection, as evidenced by increased viral replication which peaked at day 3 p.i. along with an induction of pSTAT1. Interestingly, HCECs from diabetic cornea had higher viral RNA on all three days post-infection. SARS-CoV-2 infected HCECs exhibited induced expression of inflammatory genes and their levels were relatively higher in diabetic cells. RNA-seq analysis revealed significant differences in diabetic vs. non-diabetic SARS-CoV-2 infected cells with alteration in genes regulating viral response, inflammation, and injury. The most affected down-regulated genes are related to lipid metabolism, ferroptosis, and oxidative stress. Conclusions : Our study demonstrates increased SARS-CoV-2 replication and differential innate antiviral and inflammatory response in HCECs from diabetic corneas. These results indicate that diabetes is a potential risk for enhanced infectivity of SARS-CoV-2 for the ocular surface.

A Novel Host-Targeted Viral Assembly Inhibitor Blocks SARS-CoV-2 in Airway Epithelial Cells

Introduction: The rapid emergence of the SARS-CoV-2 Omicron variant that evades many monoclonal antibody therapies illustrates the need for anti-viral treatments with low susceptibility to evolutionary escape. The small molecule PAV-104, identified through a moderate-throughput screen involving cell-free protein synthesis, was recently shown to target a subset of host protein assembly machinery in a manner specific to viral assembly. This compound has minimal host toxicity, including once daily oral dosing in rats that achieves >200-fold of the 90% effective concentration (EC90) in blood. The chemotype shows broad activity against respiratory viral pathogens, including Orthomyxoviridae, Paramyxoviridae, Adenoviridae, Herpesviridae, and Picornaviridae, with low suceptability to evolutionary escape. We hypothesized that PAV-104 would be active against SARSCoV- 2 variants in human airway epithelial cells. Methods: Airway epithelial cells were differentiated from lung transplant tissue at air-liquid interface (ALI) for four weeks prior to challenge with Alpha (Pango lineage designation B.1.1.7), Beta (B.1.351), Gamma (P.1), and Delta (B.1.617.2) SARS-CoV-2 variants. Viral replication was determined by quantitative PCR measurement of the SARS-CoV-2 nucleocapsid (N) gene. Dose-dependent virus inhibition and cytotoxicity of PAV-104 in the Calu-3 airway epithelial cell line was determined by PCR and MTT assay. Student's t-tests were used to evaluate statistical significance. Results: Alpha, Beta, Gamma, and Delta variants of SARS-CoV-2 showed comparable infectivity in human primary airway epithelial cells at ALI (N=3 donors), 47- to 550-fold higher than the parent (USA-WA1/2020) strain. PAV-104 reached 50% cytotoxicity in Calu-3 cells at 240 nM (Fig. 1A). Dose-response studies in Calu-3 cells demonstrated PAV-104 has a 6 nM 50% inhibitory concentration (IC50) for blocking replication of SARS-CoV-2 (USA-WA1/2020) (Fig.1B). In primary cells at ALI from 3 donors tested, there was >99% inhibition of infection by SARS-CoV-2 Gamma variant (N=3, MOI 0.1, P <0.01) with 100 nM PAV-104 (Fig. 1C). Addition of 100 nM PAV-104 2-hours post-infection, but not pre-infection, resulted in >99% suppression of viral replication, indicating a post-entry drug mechanism. PAV-104 bound a small subset of the known allosteric modulator 14-3-3, itself implicated in the interactome of SARS-CoV-2. Conclusion: PAV-104 is a host-targeted, orally bioavailable, pan-viral small molecule inhibitor with promising activity against SARS-CoV-2 variants in human primary airway epithelial cells. (Figure Presented).

Induced Pluripotent Stem Cell Derived Airway to Assess Cystic Fibrosis Intrinsic Epithelial Response to SARS-CoV-2

Rationale There is a lack of knowledge of how CFTR-deficient airway epithelium intrinsically responds to SARS-CoV-2. Though prior work has demonstrated altered CF airway expression of viral entry factors, it is unknown whether these alterations are protective and whether they reflect host genetic variation or secondary response of chronic inflammation. We address this gap by infecting induced pluripotent stem cell (iPSC)-derived airways from CF patients and syngeneic CFTR-corrected controls with SARS-CoV-2 and assessing differential susceptibility to infection and inflammatory and anti-viral response. MethodsCF (F508del homozygous) and syngeneic CFTR-corrected (CRISPR-Cas9) iPSC- were differentiated into airway epithelium cultured at airliquid interface (ALI) by a directed differentiation protocol that generates a pure population of major and rare airway cell-types. After 21 days in ALI culture, the iPSC-airway were infected with either mock or SARS-CoV-2 (isolate USA-WA1/2020) with MOI of 4, and harvested at 0, 1, 3 days post infection (dpi) for RT-PCR and immune-stainingResultsBoth CF and CFTR-corrected iPSC-airway express viral entry factors of ACE2 and TMPRSS2, and are permissive to SARS-CoV-2 infection. CF iPSC-airway exhibited significantly increase in SARS-CoV-2 nucleocapsid protein (N) transcript at 1 dpi, accompanied by increases in IFN2, RSAD2, and CXCL10 at 3 dpi, compared to its CFTR-corrected counter-part. There are no baseline significant differences in ACE2, TMPRSS2, TP63, NGFR, MUC5B, MUC5AC, SCGB1A1, FOXJ1, FOXI1 expression between CF and CFTR-corrected iPSC-airway before SARS-CoV-2 infection. ConclusionsOur preliminary studies indicate increased early SARS-CoV-2 infection in CFTR-deficient epithelium with accompanied subsequent rise in anti-viral and inflammatory response compared to its genetically controlled CFTR-corrected counterpart. Future studies are aimed at assessing differential CF epithelial kinetics of SARS-CoV-2 viral entry and replication, morphological changes, global transcriptomic response, and how treatment with CFTRmodulator would alter the epithelial response. Ultimately, we aim to establish a reductionist, physiologically relevant model system that is coupled with gene-editing technology to study intrinsic CF epithelial response to SARS-CoV-2, which would generate insights to aid practice guidelines for CF patients, and open future directions to evaluate gene-specific mechanisms of airway response to pathogens. (Figure Presented).

Correlation of the VITROS® anti-SARS-CoV-2 Igg quantitative assay with SARS-CoV-2 plaque reduction neutralization test

Background: The VITROS IgG Quant assay∗ is for the quantitative detection of SARS-CoV-2 IgG antibodies with calibration traceable to the first World Health Organization International Standard for Anti- SARS-CoV-2 antibody. Results are reported in both qualitative (reactive/ non-reactive) and quantitative values (Binding Antibody Units [BAU]/ml). PRNT is considered the gold standard method for determining neutralizing antibody (nAb) titers. Aims: This study was designed to assess the correlation of the VITROS Immunodiagnostic Products Anti-SARS-CoV-2 IgG Quantitative Assay (VITROS IgG Quant) to a plaque reduction neutralization test developed at Colorado State University (CSU PRNT). Methods: VITROS IgG Quant: The VITROS IgG Quant assay is a fully automated, high-throughput method run on the VITROS family of immunoassay analyzers. First, antibodies to SARS-CoV-2 present in the sample bind with the S1 subunit of the Spike protein coated on wells. After washing, horseradish peroxidase (HRP)- labelled murine monoclonal anti-human IgG antibodies are added. Following a final wash, bound HRP conjugates are detected using the VITROS signal reagent. The amount of conjugate directly correlates to the amount of SARS-CoV-2 IgG antibody present and is reported in BAU/ml. CSU PRNT: Samples were heat-inactivated for 30 min at 56°C, and serial two-fold dilutions were prepared in a 96-well plate. Viral stock (strain hCoV-19/USA/WA1/2020) containing ∼200 pfu per 0.1 ml was added to each well containing serum dilutions. Following incubation at 37°C in 5% CO2, 6-well plates containing recently confluent Vero cells were inoculated with the virus-serum mixtures. After a second incubation at 37°C, 2 ml of overlay was added to each well. After 24 h incubation at 37°C, a second overlay containing neutral red was dispensed into each well and the number of plaques was counted 48-72 h after initial inoculation. The highest dilution of serum that inhibited plaque formation by 50% (PRNT50) was determined based upon the titre of the samples and the number of plaques present at each dilution. Samples with PRNT50 titers less than or equal to 1:20 are considered negative for nAbs. One hundred forty-nine samples were blind-tested with both the VITROS IgG Quant assay and the CSU PRNT, 74 known positive for SARS-CoV-2 antibody and 75 negatives. The correlation of the VITROS IgG Quant values to CSU PRNT50 was determined. Results: The VITROS IgG Quant results ranged from <2 to 2009 BAU/ml. The CSU PRNT50 results ranged from <1:20 to >1:2560. Pearson correlation coefficient was calculated to be 0.867, demonstrating a good correlation between the VITROS IgG Quant results and the CSU PRNT50 titers. Summary/Conclusions: The VITROS Anti-SARS-CoV-2 IgG Quant assay demonstrates a strong correlation to PRNT50 for the measurement of SARS-CoV-2 nAb titers.

GUNNERA PERPENSA ELLAGITANNINS SYNERGISTICALLY INHIBIT MULTIPLE SARS-CoV-2 VARIANTS

Background: Antivirals are urgently needed to supplement SARS-CoV-2 vaccines and target SARS-CoV-2 variants of concern, particularly in resource-limited regions. Active derivatives from the medicinal plant Gunnera perpensa, already in use as a general antiviral in humans by traditional health practitioners in the Eastern Cape Province of South Africa, warrant further evaluation against SARS-CoV-2. Methods: Active constituents of Gunnera perpensa were identified using hyphenated analytical techniques and for ability to inhibit binding of recombinant SARS-CoV-2 spike with host ACE2 protein as assessed by AlphaScreen. Inhibition was tested against parental (USA-WA1/2020), beta (B.1.351), and delta (B.1.617.2) spike proteins using AlphaScreen and spike-expressing VSVΔG-GFP pseudoviruses. Infection of Vero cells was monitored by high-content imaging of GFP or nucleocapsid-positive Vero-E6 cells in pseudovirus and virus assays, respectively, at 2 days post-infection (dpi). Viral cytopathic effect (CPE) ± GC-376 or remdesivir was also monitored using resazurin viability dye at 4 dpi. All assays were described previously (PMID: 34543092). Synergism was assessed by the Bliss Independence model, and group differences were analyzed by two-sided, paired t-test. Results: Crude extracts from the leaves of Gunnera perpensa were confirmed to inhibit parental spike/ACE2 interactions with an IC50 of 37 ± 23 ng/mL. Bioassay-guided fractionation identified two ellagitannins, punicalin and punicalagin, which inhibited parental, beta, and delta spike/ACE2 binding with IC50s of 2.7 ± 0.6-5.8 ± 4.8 and 6.0 ± 4.5-19 ± 23 nM, respectively. Both compounds inhibited all spike variants in pseudovirus at low to mid micromolar concentrations (see Table). Notably, in CPE-based viral assays, a 1:1 molar mixture of punicalin and punicalagin significantly enhanced antiviral activity (EC50 = 2.9 μM vs. 11.6 and 46.8 μM for single compounds, p < 0.05), on par with activities of preclinical candidate GC-376 (1.3 μM) and remdesivir (2.8 μM;see Table). When combined in a 1:1 molar mixture, punicalin further significantly enhanced activity of GC-376 (EC50 = 0.6 μM, p < 0.05) and remdesivir (EC50 = 1.1 μM, p < 0.05). Conclusion: Punicalin and punicalagin inhibit entry and replication of SARS-CoV-2 variants in vitro and synergize when applied in combination and/or with GC-376 or remdesivir. Ellagitannins and medicinal plant extracts are promising new leads for SARS-CoV-2 antivirals in resource-limited regions.

DESIGN of NOVEL and HIGHLY SELECTIVE SARS-CoV-2 MAIN PROTEASE INHIBITORS

Background: SARS-CoV-2 has caused a global pandemic, yet despite vaccine availability, it continues to inflict morbidity and mortality worldwide. The viral main protease (Mpro) is highly conserved across multiple coronaviruses and has a unique viral substrate specificity. Thus, highly selective Mpro inhibitors are expected to be safe, effective, and elude drug resistance for future coronaviruses. Methods: We used a conformationally restricted peptidomimetic to mimic the bioactive conformation of the Mpro-substrate complex to identify potent, selective Mpro inhibitors. We evaluated protease inhibition in biochemical assays, and cellular efficacy in Vero-E6 cells challenged with live virus representing parental (USA-WA1/2020), beta (B.1.351), and delta (B.1.617.2) variants by monitoring infection at day 2 post-infection measuring nucleocapsid-positive cells by high content imaging, and cytopathic effect (CPE) at day 4 post-infection using resazurin viability dye. Results were compared to reference compounds. Group differences were analyzed by two-sided, paired t-test. Results: AP-8-013 required a 2-hour incubation to achieve maximal dose-dependent Mpro inhibition with an IC50 = 230 ± 18 nM, reflecting its highly constrained conformation, compared to the more flexible Cpd 22 (AP-8-001;IC50 = 11 ± 0.7 nM) or GC-376 (IC50 = 18 ± 1.5 μM). Importantly, AP-8-013 showed exquisite selectivity for Mpro with no inhibition at key mammalian cysteine proteases, cathepsin B and L, or the serine protease thrombin, while Cpd 22 (Cat B IC50 = 24 ± 7.5 nM, Cat L IC50 = 1.8 ± 0.3 nM) or GC-376 (Cat B IC50 = 37 ± 1.5 nM, Cat L IC50 = < 1 nM) showed poor selectivity towards mammalian cysteine proteases. AP-8-013 was active in CPE cell-based assays with comparable potency to reference compounds, with EC50 = 4.7 μM compared to Cmp 22 (EC50 = 1.4 μM) or GC-376 (EC50 = 1.1 μM). Using intact SARS-CoV-2 infection-based assays, AP-8-013 significantly inhibited parental virus as well as beta and delta VOC (EC50s = 2.7, 2.5, and 6.0 μM, respectively). Finally, a 3:1 molar mixture of AP-8-013 and remdesivir significantly enhanced antiviral activity in CPE assays (EC50 = 1.3 μM;p < 0.05) when compared against either compound alone (EC50s = 4.7 and 3.3 μM, respectively). Conclusion: We have identified a novel drug-like Mpro inhibitor lead series which is highly selective over cysteine and serine proteases that can inhibit multiple SARS-CoV-2 VOC and increase the antiviral activity of remdesivir.

EVIDENCE of BRAIN HYPOXIA in NEUROPATHOLOGY of SARS-CoV-2-INFECTED NONHUMAN PRIMATES

Background: Neurological manifestations are a major complication of sudden acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and likely contribute to symptoms of "long COVID". Elucidating the mechanisms that underlie neuropathogenesis in infection is critical for identifying or developing viable therapeutic strategies. While neurological injury in infection is varied, cerebrovascular disease is seen at a high frequency among patients over 50 years of age. Additionally, microhemorrhages and hypoxic-ischemic injury are often described in brain autopsy series of human subjects who died from COVID-19. Here, we report neuropathology in aged SARS-CoV-2 infected non-human primates (NHPs) is consistent with that observed in aged human subjects and provide insight into the underlying cause. Methods: Four adult Rhesus macaques and four African green monkeys were inoculated with the 2019-nCoV/USA-WA1/2020strain of SARS-CoV-2 via a multi-route mucosal or aerosol challenge. Two of each species were included as age-matched controls. Frontal, parietal, occipital, and temporal lobes, basal ganglia, cerebellum, and brainstem were interrogated through histopathological and immunohistochemical techniques to identify and characterize the observed pathology. Results: Like humans, pathology was variable but included wide-spread inflammation with nodular lesions, neuronal injury, and microhemorrhages. Neuronal degeneration and apoptosis were confirmed with FluoroJade C and cleaved caspase 3 IHC, which showed foci of positivity, particularly among cerebellar Purkinje cells. This was seen even among infected animals that did not develop severe respiratory disease but was not seen in age-matched controls. Significant upregulation of the alpha subunit of hypoxia inducible factor 1 (HIF1-α), indicative of tissue hypoxia, was observed in brain of all infected animals, regardless of disease severity. Sparse virus was detected in brain endothelial cells but did not associate with the severity of CNS injury. Conclusion: SARS-CoV-2 infected NHPs are a viable animal model for advancing our current understanding of infection-associated neuropathogenesis. Upregulation of HIF1-α in brain of infected animals suggests cerebral hypoxia may underlie or contribute to neuroinflammation and neuronal injury/death and may provide some insight into neurological manifestations observed among asymptomatic patients or those only suffering mild disease.

MOLECULAR DETERMINANTS of SARS-CoV-2 CELLULAR TROPISM in VITRO

Background: Lung cell lines to model SARS-CoV-2 replication in vitro are greatly limited hampering the rigorous study of SARS-CoV-2-host interactions. We analyzed a panel of 10 airway cell lines with various levels of ACE2 expression to identify models of SARS-CoV-2 infection. We found that none of the ACE2 expressing cell lines supported replication, whereas the H522 human lung adenocarcinoma cells were naturally permissive to SARS-CoV-2 infection despite detectable expression of ACE2. We confirmed that SARS-CoV-2 replication is indeed completely independent of ACE2 in H522s but dependent on heparan sulfates and the E484D substitution within the Spike. Further, we show that many of the ACE2 positive non-permissive cell lines express high basal levels of interferon-stimulated genes, which can be overcome by inhibition of the JAK/STAT pathway or by ACE2 overexpression. Together, our findings highlight ACE2-independent pathways can control the cellular tropism of SARS-CoV-2. Methods: Conventional molecular virology assays have been conducted to study the permissiveness of a panel of 10 cell lines expressing various levels of ACE2. ACE2 independence of SARS-CoV-2 replication was validated by antibody blocking, Fc-ACE2 decoy peptide and CRISPR-based approaches in H522 cells. RNA sequencing was used to study the basal level of genes in the type-I IFN pathway in the panel of 10 cell lines, which was further validated by western blotting and qRT-PCR. A panel of 5 cell lines, with varying expression levels of ACE2 and TMPRSS2, were pre-treated with Ruxolitinib, a JAK inhibitor, and infected with SARS-CoV-2 strain 2019-nCoV/USA-WA1/2020 and spike variants. Viral replication was detected through analysis of cell associated RNA Results: H522 human lung adenocarcinoma supports SARS-CoV-2 replication in a completely ACE2-independent manner. Transcriptomic analysis revealed basal high level of expression of interferon response pathway genes in some ACE2-positive cells recalcitrant to SARS-CoV-2 infection. Infection of OE21 and SCC25 cells required blocking of the IFN response pathway or ACE2 overexpression to allow SARS-CoV-2 infection. Conclusion: These findings suggest that SARS-CoV-2 replication can proceed in complete absence of ACE2 and that the innate immunity is a key determinant of SARS-CoV-2 cellular tropism. These findings may explain the complex SARS-CoV-2 pathogenesis in vivo as it shows that factors independent of ACE2 can define cellular tropism.

RESTRICTED INFECTION of MACROPHAGES by SARS-CoV-2 INDUCES PROINFLAMMATORY RESPONSES

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been associated with immune hyperactivation and high levels of proinflammatory cytokines. Extensive lung infiltration by CD169+ inflammatory monocytes and presence of activated CD169+ alveolar macrophages suggest monocyte/macrophages are key drivers of severe morbidity and mortality. In this study, we determined whether CD169 mediated ACE2-independent SARS-CoV-2 entry and restricted viral genome replication in macrophages triggers pro-inflammatory cytokine expression. Methods: Monocyte-derived macrophages (MDMs) and PMA-differentiated THP-1 macrophages engineered to constitutively express CD169, ACE2, or CD169 and ACE2 were infected with USA-WA1/2020/SARS-CoV-2 isolate with or without Remdesivir pre-treatment. To identify mechanism of innate immune activation, nucleic acid sensing pathways were selectively depleted in CD169+ macrophages. Extent of viral genomic (gRNA) and sub-genomic (sgRNA) expression and induction of pro-inflammatory cytokines was determined by qRT-PCR and single molecule RNA FISH analysis. Viral protein expression and infectious virus particle production was determined by immunofluorescence analysis and TCID50. Results: While productive virus infection (viral protein expression and infectious virus particle release) was only observed in ACE2+ macrophages, SARS-CoV-2 N or S expression and infectious virus production was not observed in CD169+ macrophages. Co-expression of ACE2 and CD169 significantly enhanced infectious virus production and spread. Interestingly, smFISH and RT-qPCR analysis revealed CD169+ cells express cytosolic negative-strand gRNA and positive strand sgRNA. Importantly, CD169-mediated SARS-CoV-2 infection of macrophages and expression of viral mRNAs led to induction of pro-inflammatory cytokines, IL-6, TNFα, and IL-1β, despite lack of viral protein expression in CD169+ macrophages. Pre-treatment with Remdesivir blocked de novo expression of viral mRNAs and induction of inflammatory cytokines in CD169-dependent infection of macrophages. Furthermore, knockdown of cytosolic RLRs (RIG-I and MDA-5) or MAVS significantly attenuated inflammatory cytokine expression in CD169+ macrophages, confirming that nucleic acid sensing of restricted cytosolic viral mRNA expression in macrophages triggers innate immune activation. Conclusion: These results suggest that restricted SARS-CoV-2 infection of CD169+ macrophages contributes to COVID-19-associated hyperinflammatory cytokine response.

EFFECT of ASTODRIMER SODIUM AGAINST SARS-CoV-2 VARIANTS (Α, Β, Γ, Δ, Κ) in VITRO

Background: The dominance of SARS-CoV-2 Variants of Concern (VOC) and Interest (VOI) has challenged the efficacy of public health strategies to control the current pandemic. Astodrimer sodium is a broad-spectrum antiviral dendrimer that has been formulated as a topical nasal spray to help reduce exposure to infectious viral load in the nasal cavity. Astodrimer sodium showed antiviral and virucidal activity against early pandemic isolates of SARS-CoV-2 in vitro and after nasal administration in vivo. The current studies assessed the spectrum of activity of astodrimer sodium against emerging variants of SARS-CoV-2 and other pandemic viruses. Methods: Assays utilized hACE2+ and hTMPRSS2+ HEK-293T cells, Calu-3 and Vero E6 cells. Time of addition studies involved adding astodrimer sodium 1 hour prior to, at the time of, or 1-hour post-infection. Coronavirus spike receptor binding domain (RBD) or S1 binding studies were analysed by ELISA or confocal microscopy. Virucidal studies involved exposing 105 SARS-CoV-2 PFU to 10mg/mL astodrimer sodium for 0.5, 1, 5, 15 and 30 mins. Results: Astodrimer sodium demonstrated potent antiviral and virucidal activity against SARS-CoV-2 VOC α, β, δ and γ, and VOI κ in Vero E6 and Calu-3 cells. Astodrimer sodium reduced infectious viral load of all variants by >99.9% vs virus control. The pan-SARS-CoV-2 activity of astodrimer sodium occurred despite multiple mutations and deletions in the viral spike protein of each variant. The attachment of SARS-CoV-2 early pandemic virus isolates, Wuhan-Hu-1 and USA-WA-1/2020, and SARS-CoV-1 Spike binding to ACE2, as well as attachment of Middle Eastern respiratory syndrome (MERS) coronavirus spike protein to its cellular receptor, was inhibited by astodrimer sodium. Astodrimer sodium did not prevent attachment of the SARS-CoV-2 VOC α and β spike S1, or γ RBD spike protein, to the ACE2 receptor in vitro. Conclusion: Astodrimer sodium mimics negatively charged glycosaminoglycans and provides a potent antiviral and virucidal barrier to viral attachment and entry. The potent broad-spectrum anti-pandemic coronavirus and virucidal efficacy of astodrimer sodium against whole virus is likely due to blocking multiple electrostatic interactions of the spike protein that are not negated by minor or major changes to the isolated RBD of SARS-CoV-2 VOC α, β and γ alone. Astodrimer sodium has the potential to block the binding of pan-SARS-CoV-2, thus reducing the potential for the development of COVID-19.

GALECTIN-9 INDUCES SARS-CoV-2 REPLICATION and INFLAMMATION in AIRWAY EPITHELIAL CELLS

Background: Galectin-9 (Gal-9) is a β-galactoside-binding lectin involved in immune regulation and viral immunopathogenesis. Multiple recent reports demonstrate that plasma levels of Gal-9 are elevated in the setting of severe COVID-19 disease. However, a causal role of Gal-9 in SARS-CoV-2 pathology remains to be elucidated. Here, we determined the impact of Gal-9 on SARS-CoV-2 replication and pro-inflammatory signaling in immortalized and primary human airway epithelial cells (AECs). Methods: Dose-dependent cytotoxicity of recombinant human Gal-9 in the Calu-3 AEC line was determined by MTT assay. Calu-3 cells were infected with SARS-CoV-2 isolate USA-WA1/2020 (MOI=0.01). Primary AECs were isolated from healthy donor lung transplant tissue, cultured at air liquid interface (ALI), and infected with SARS-CoV-2 lineage P.1 (MOI=0.1). SARS-CoV-2 replication was assessed by RT-PCR quantitation of the nucleocapsid (N) gene, immunofluorescence assay (IFA) of N protein, and titration of supernatant (TCID50). Viral entry was measured using luciferase activity of VSV-SARS-CoV-2 S-ΔG-Luciferase reporter pseudovirus. ACE2 and TMPRSS2 cell-surface expression were measured by flow cytometry. Pro-inflammatory factors (IL-6, IL-8, and TNFα) were detected by RT-PCR. Total RNA-seq was used to evaluate Gal-9 effects on the host transcriptome. Groups were compared by Student's t-test, and differential expression analyses were performed using DESeq2. Results: Gal-9 reached 50% cytotoxicity in Calu-3 cells at 597 nM. Gal-9 significantly increased SARS-CoV-2 expression (8.1 to 25.5 fold;p<0.0001) and infectious virus release (1.9 to 17.8 fold;p<0.038) in a dose-dependent manner in Calu-3 cells. Pseudovirus entry into Calu-3 cells was enhanced by Gal-9 (2.4 to 5.6 fold;p<0.0016), and the enhanced entry was inhibited by anti-ACE2 antibody (p<0.0027). Cell surface ACE2 and TMPRSS2 expression were unaffected by Gal-9. Gal-9 treatment accelerated virus-induced expression of IL-6, IL-8, and TNFα (p<0.018) in Calu-3 cells. Gal-9 increased SARS-CoV-2 production (p=0.03) and pro-inflammatory factor expression (p<0.05) in primary AECs (N=5 donors). RNA-seq data revealed that Gal-9 significantly induced IL-17, EIF2, IL-8 and IL-6 signaling pathways in the setting of SARS-CoV-2 infection. Conclusion: Gal-9 facilitates SARS-CoV-2 entry, replication, and virus-induced pro-inflammatory signaling in AECs ex vivo. Our data suggest that pharmacologic manipulation of Gal-9 should be explored as a SARS-CoV-2 therapeutic strategy.

SARS-COV-2 INDUCES EPITHYELIAL-NEURONAL CROSSTALK STIMULATING VASOACTIVE INTESTINAL PEPTIDE RELEASE AS A POTENTIAL MECHANISM OF COVID-19-ASSOCIATED DIARRHEA

Background: Up to 36.6% of COVID-19 patients have diarrheal symptoms and 48.1% test positive for SARS-CoV-2 via stool test. The mechanism of SARS-CoV-2-associated diarrhea remains poorly understood. We hypothesize that crosstalk between enterocytes and the enteric nervous system (ENS) plays a critical role in the pathogenesis of COVID-19-associated diarrhea. We studied the effects of SARS-CoV-2 on induction of endoplasmic reticulum (ER) stress and release of Damage Associated Molecular Patterns (DAMPs), which act on enteric neurons and stimulate the production of neurotransmitters. The influence of ER stress and enteric neuron-derived vasoactive intestinal peptide (VIP) on the expression of electrolyte transporter Na+/H+ exchanger 3 (NHE3) was also examined. Methods: SARS-CoV-2 (2019-nCoV/USA-WA1/2020) was propagated in Vero-E6 cells. Caco-2, a human colon epithelial cell line, expresses the essential SARS-CoV-2 entry receptor ACE2 and was thus used for infection (MOI, ~0.01). We used Western blotting to assess the expression of ER stress (phospho-PERK and Xbp1s) and DAMP (HMGB1) markers at 48 hours post-infection. Primary mouse enteric neurons were co-cultured with Caco-2 cells, pre-treated for 24 hours with 2 μM tunicamycin to induce ER stress. Supernatants from enteric neurons were used to assess the expression of VIP by ELISA. Primary enteric neurons were treated with HMGB1 or ATP (another form of DAMPs), and the expression of c-FOS, a marker of neuronal activity, was determined by Western blotting and immunofluorescence staining. Results: We found that SARS-CoV-2 infection of Caco-2 cells led to increased expression of phospho-PERK and Xbp1s. Compared to uninfected control, infected Caco-2 cells secreted HMGB1 into culture media, indicating epithelial production of DAMPs in response to SARS-CoV-2 infection. Tunicamycin was used to induce ER-stress and secretion of HMGB1 by Caco-2, mimicking SARS-CoV-2 infection. Importantly, enteric neurons co-cultured with tunicamycin-treated Caco-2 cells secreted significantly higher levels of VIP. Treating Caco-2 cells with tunicamycin or VIP on the basolateral side led to decreased surface NHE3 expression, suggesting a potential impairment of intestinal electrolyte/fluid absorption. More-over, HMGB1 and ATP both increased the expression of phospho-c-FOS in cultured enteric neurons, indicating DAMP-induced neuronal activation. Conclusions: Our findings demon-strate that enterocytes infected by SARS-CoV-2 release DAMPs with the capacity to induce VIP secretion by the enteric neurons, which in turn acts on enterocytes and inhibits apical localization of NHE3. These findings establish basic mechanisms relevant to diarrheal disease in COVID-19 patients and identify potential targets for the treatment of SARS-CoV-2 infection of the gastrointestinal tract.