Selective Inhibition of the Interaction between SARS-CoV-2 Spike S1 and ACE2 by SPIDAR Peptide Induces Anti-Inflammatory Therapeutic Responses

Since December 2019, the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is causing a worldwide pandemic, leading to the highly morbid and potentially fatal coronavirus disease 2019 (COVID-19). Common symptoms of COVID-19 are fever, cough, and shortness of breath (1, 2), and currently, >4 million people have died worldwide from COVID-19. Although anyone is susceptible to COVID-19, the ones over 60 years of age or with preexisting conditions, such as hypertension, obesity, asthma, or diabetes, are more vulnerable to severe symptoms. It appears that COVID-19 is more lethal in men than women (3). It has been suggested that in severely ill patients with COVID-19, cytokine storm may be responsible for lung injury and multiorgan failure (4). Although ∼50% people in the United States have been fully vaccinated and a number of drugs (e.g., remdesivir, dexamethasone, etc.) have been approved for repurposing for COVID-19, it is always important to describe a specific antiviral and anti-inflammatory agent for treating COVID-19.

Angiotensin-converting enzyme 2 (ACE2), being the main effector of the classical renin-angiotensin system, is an important molecule for the regulation of blood pressure and hypertension, as the prototype function of ACE2 is to convert angiotensin II, a vasoconstrictor, to Ang1-7, a vasodilator (5, 6). However, SARS-CoV-2 requires ACE2 to enter into human cells. The spike protein on the surface of SARS-CoV-2 binds to ACE2 (2, 7), and during infection, the spike protein is cleaved into S1 and S2 subunits. The spike S1 subunit harbors the receptor-binding domain (RBD). Because ACE2 is predominant in lung, heart, and kidney (8), SARS-CoV-2 easily infects these organs, causing multiorgan failure in severe cases of COVID-19.

Therefore, inhibition of ACE2 should reduce the SARS-CoV-2 infection and associated inflammation. However, either inhibition or knockdown of ACE2 may not be a valid therapeutic option for COVID-19, as it is a beneficial molecule. Therefore, to target COVID-19 from a therapeutic angle, we took an out-of-the-box approach and designed a hexapeptide corresponding to spike S1–interacting domain of ACE2 receptor (SPIDAR). Interestingly, wild-type (wt)SPIDAR, but not mutated (m)SPIDAR, inhibited the association between RBD-containing spike S1 and ACE2 and suppressed spike S1–induced activation of NF-κB and expression of IL-6 in human lung cells. Accordingly, the wtSPIDAR also inhibited the entry of pseudotyped SARS-CoV-2, but not vesicular stomatitis virus (VSV), into human ACE2 (hACE2)–expressing human embryonic kidney 293 (HEK293) cells. Moreover, after intranasal administration, wtSPIDAR, but not mSPIDAR, reduced lung inflammation, decreased lung neutrophil infiltration, reduced fever, inhibited arrhythmias, and enhanced locomotor activities in an animal model of COVID-19. Therefore, intranasal SPIDAR may be beneficial for COVID-19.

SARS-CoV-2 spike-pseudotyped lentiviral particles, VSV-G–pseudotyped particles, and hACE2-expressing HEK293 cells were purchased from GeneCopoeia (Rockville, MD). Recombinant COVID-19 spike protein S1 was purchased from Abeomics (San Diego, CA). Whereas anti–SARS-CoV-2 spike S1 Ab was bought from BioVision (Milpitas, CA), anti-hACE2 Ab was purchased from R&D Systems (Minneapolis, MN). Anti-human MyD88 Ab was bought from Millipore (Burlington, MA). Human A549 lung carcinoma cell line and F-12K medium were obtained from American Type Culture Collection (Manassas, VA). HBSS, 0.05% trypsin, and antibiotic-antimycotic mixture were bought from Mediatech (Washington, DC). FBS was obtained from Atlas Biologicals (Fort Collins, CO). An ACE2:SARS-CoV-2 Spike Inhibitor Screening Assay Kit was purchased from BPS Bioscience (San Diego, CA). Anti-human IL-6 Ab and human IL-1β and IL-6 ELISA kits were bought from Thermo Fisher Scientific (Waltham, MA).

To study the interaction between hACE2 and SARS-CoV-2 spike S1, we performed in silico structural analysis as described earlier (9–11). Briefly, by using the protein preparation tools from the Schrodinger platform, at first, we evaluated the quality of the crystal structure of hACE2 and SARS-CoV-2 spike S1 followed by addition of hydrogens to the hydrogen bond orientation, charges, missing atoms, and side chains of the different residues of both of the proteins. Finally, the complex structure was subjected to energy minimization in the Optimized Potential for Liquid Simulations 3 force field to make it torsion free. Then, the spike protein was extracted out from the ACE2 to apply the dynamic hydrogen bonding module for finding potential hydrogen bonds between the two structures. We also evaluated other interactions, such as hydrophobic interactions, between the two structures.

The peptides used were: wtSPIDAR: EDLFYQ; and mSPIDAR: EKLFYG. SPIDAR peptides (>98% pure) were synthesized in GenScript (Piscataway, NJ).

The efficacy of wtSPIDAR and mSPIDAR to dissociate the binding of SARS-CoV-2 spike S1 with ACE2 was investigated using the ACE2:SARS-CoV-2 Spike S1 Inhibitor Screening Assay Kit (BPS Bioscience) according to the manufacturer’s instructions, as described by us recently (11). Briefly, a 96-well nickel-treated plate provided by the manufacturer was coated with ACE2 solution. After washing with immunobuffer and treatment with blocking buffer, different concentrations of wtSPIDAR and mSPIDAR were added to each well followed by addition of SARS-CoV-2 Spike (RBD)-Fc. After washing and incubation with blocking buffer, plates were treated with anti-mouse Fc-HRP followed by addition of an HRP substrate. Resultant chemiluminescence was monitored in a Victor X5 multimode microplate reader (PerkinElmer).

Human ACE2-expressing HEK293 cells were plated in 24-well plates at 70–80% confluence followed by infection with lentiviral particles pseudotyped with either SARS-CoV-2 spike S1 or VSV-G at a multiplicity of infection (MOI) of 0.5. Cells were also infected with empty lentiviral (lenti-naked) particles as control. After 48 h of infection, the entry of pseudovirus into HEK293 cells was checked by either assaying luciferase activity in total cell extracts using a TD-20/20 Luminometer (Turner Designs) or monitoring GFP fluorescence in an Olympus BX-41 fluorescence microscope.

Total RNA was isolated from mouse lungs and human A549 lung cells using Ultraspec-II RNA reagent (Biotecx Laboratories, Houston, TX) and an RNeasy Mini Kit (Qiagen, Germantown, MD), respectively. To remove any contaminating genomic DNA, total RNA was digested with DNase. RT-PCR was carried out as described earlier (12, 13) using an RT-PCR kit (Clontech, Mountain View, CA) and following primers of human proinflammatory molecules: TNF-α sense, 5′-CTG AGT CGG TCA CCC TTC TCC AGC T-3′ and antisense, 5′-CCC GAG TGA CAA GCC TGT AGC CCA T-3′; IL-1β sense, 5′-GGA TAT GGA GCA AC A AGT GG-3′ and antisense, 5′-ATG TAC CAG TTG GGG AAC T-3′; IL-6 sense, 5′-TTT TGG AGT TTG AGG TAT ACC TAG-3′ and antisense, 5′-GCT GCG CAG AAT GAG ATG AGT TGT-3′; and GAPDH sense, 5′-GGT GAA GGT CGG AGT CAA CG-3′ and antisense, 5′-GTG AAG ACG CCA GTG GAC TC-3′.

GAPDH gene was used to confirm the synthesis of an equivalent amount of cDNA from different samples. Amplified products were electrophoresed on 1.8% agarose gels and visualized by ethidium bromide staining.

Real-time PCR was performed in the ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA), as described earlier (12, 13).

Cells were plated in six-well plates at a density of 1.8 × 10⁶ cells/well with DMEM/F-12 containing FBS (10%) and antibiotic-antimycotic mixture at 37°C with 5% CO₂. After 24 h, cells were infected with CRISPR scrambled single-guide RNA (sgRNA) all-in-one lentivirus and human MyD88 sgRNA CRISPR all-in-one lentivirus (viral titer 9 × 10⁶ IU/ml) according to the protocol provided by the manufacturer (Applied Biological Materials, Richmond, BC, Canada). Following the infection, cells were incubated at 37°C with 5% CO₂. After overnight incubation, culture medium was replaced with 2 ml complete medium, and cells were incubated for 48 h followed by monitored genomic editing by Western blot.

A549 cells plated at 60–70% confluence in 12-well plates were transfected with 0.25 μg pNF-κB-Luc using Lipofectamine Plus (Invitrogen). After 24 h of transfection, cells were stimulated with different stimuli under serum-free conditions for 4 h. Firefly luciferase activities were analyzed in cell extracts using the Luciferase Assay System Kit (Promega) in a TD-20/20 Luminometer (Turner Designs), as described earlier (14, 15).

Mice were maintained and experiments conducted in accordance with National Institutes of Health guidelines and approved by the Rush University Medical Center Institutional Animal Care and Use Committee. C57/BL6 mice (6–8 wk old; Envigo) of both sexes were intoxicated with recombinant SARS-CoV-2 spike S1 (50 ng/mouse/d) intranasally, as described by us recently (11). Briefly, recombinant spike S1 was dissolved in 2 μl normal saline, mice were held in a supine position, and 1 µl volume was delivered into each nostril using a Pipetman. Control mice received only 2 µl saline.

Starting from 5 d of SARS-CoV-2 spike S1 intoxication, C57/BL6 mice (6–8 wk old; Envigo) of both sexes were treated intranasally with wtSPIDAR or mSPIDAR (100 ng/mouse/d) for 7 d. Briefly, SPIDAR peptides were dissolved in 2 μl normal saline, mice were held in a supine position, and 1 µl volume was delivered into each nostril using a Pipetman.

Electrocardiogram (ECG) recording was performed as described by us recently (11). Briefly, mice were acclimatized to the ECG pulse transducer pad (TN 012/ST; ADInstruments, Colorado Springs, CO) and the experimental housing conditions prior to ECG recording. The ECG pulse transducer pad was placed around the heart of each animal, and ECG recording was carried out for 120 s. For ECG analysis, ECG data were exported from the LabChart Pro, version 8.0 (Power Lab 4/35 model) in raw data format, and the digital signal processing was performed using this software. The recording was conducted for 120 s, the ECG signals were recorded at a sampling range of 20 mV with 4 beats/s sampling rate, and different ECG parameters were calculated as described (11). In the detection setting, typical QRS width was kept at 10 ms and R wave reserved at least 60 ms apart with the alignment sustained at QRS maximum. During analysis, prebaseline was kept at 10 ms with a maximum at 50 ms. We selected for rodent waves to measure ST segment height at 10 ms. Moreover, the recording and analysis settings were kept same for all of the experimental mice included in this study.

After treatment, animals were anesthetized with ketamine/xylazine injectable followed by transcardial perfusion (16). The lungs were collected and processed for histological studies. H&E (Sigma-Aldrich, St. Louis, MO) staining was performed from 4-µm-thick paraffin-embedded sections and used for studying the general lung tissue morphology. The number of epithelial cells and number of infiltrated neutrophils in alveolar spaces and interstitial space were analyzed by ImageJ (National Institutes of Health). At least 10 ×40 fields from each group were chosen for the counting of the epithelial and infiltrated neutrophils. Lung injury score was measured as described by Matute-Bello et al. (17) using a scale (11) in a blinded manner.

Immunostaining was performed as described previously (18, 19). Briefly, paraffin-embedded sections were deparaffinized, rehydrated, and blocked with PBS containing 4% normal horse serum and 2% BSA for 1 h. For double labeling, sections were incubated overnight at 4°C with Abs for ACE2 (1:100) and IL-6 (1:500). Following washings, sections were incubated with Cy2- or Cy5-labeled secondary Abs and imaged under an Olympus BX41 fluorescence microscope. Mean fluorescence intensity of target proteins was measured using ImageJ. Intensity of τ and phospho-τ diaminobenzidine staining was quantified using Fiji (ImageJ2).

Recruitment of NF-κB to the IL-6 promoter in vivo in the lung of mice was examined by in situ chromatin immunoprecipitation (ChIP) analysis as described before (13). Briefly, after formaldehyde fixation, lungs were kept in 4% paraformaldehyde overnight followed by washing with PBS and then homogenization in Tris-EDTA buffer (pH 7.6). The homogenates were kept in 500 μl lysis buffer at 52°C for overnight until tissue fragments were dissolved completely. Then, the genomic DNA was isolated and sonicated, followed by immunoprecipitation with Abs against p65, p50, transcriptional coactivator p300, and RNA polymerase according to standard protocol as described by us (20, 21). Immunoprecipitated DNA was analyzed by PCR and real-time PCR using the following primers: sense, 5′-CCAATCAGCCCCACCCACTCTGGCCCC-3′; and antisense, 5′-GGAATTGACTATCGTTCTTGGTGGGCT-3′.

IL-6 ELISA was performed in lung homogenates and mouse serum as described earlier (22) using an assay kit (eBioscience) according to the manufacturer’s instructions. C-reactive protein (CRP) ELISA was performed in mouse serum using a kit from Abcam.

Statistical analyses were performed using Prism 9.0.0 (GraphPad Software, La Jolla, CA). Results were examined by one-way ANOVA followed by a Tukey multiple-comparison post hoc test. Data were expressed as means ± SEM. Statistical differences between means were calculated by the Student t test (two-tailed) while comparing between two groups. A p value <0.05 was considered statistically significant.

Because the interaction of SARS-CoV-2 with its receptor ACE2 is critical for the infection of host cells and the pathogenesis of COVID-19, we decided to target the interaction between ACE2 and the RBD of SARS-CoV-2 spike S1 (23). Therefore, we applied a rigid-body protein–protein interaction tool to model the interaction between ACE2 and RBD of spike S1 and found that Glu³⁷ to Gln⁴² region of ACE2 is responsible for binding with spike S1 (Fig. 1A). Therefore, we designed a small peptide (Fig. 1B) corresponding to the SPIDAR to perturb the communication between ACE2 and SARS-CoV-2 Spike S1: wtSPIDAR, ³⁷EDLFYQ⁴²; and mSPIDAR: ³⁷EKLFYG⁴².

In these peptides, we have underlined the position of mutations. Next, we investigated whether wtSPIDAR inhibited the binding of SARS-CoV-2 spike S1 with ACE2 receptor. Therefore, we employed a chemiluminescence-based ACE2:SARS-CoV-2 spike S1 binding assay (catalog number 79936; BPS Bioscience). We found that SARS-CoV-2 spike S1 binding to immobilized ACE2 was strongly inhibited by wtSPIDAR (Fig. 1C). As expected, we did not see any inhibition in binding with mSPIDAR (Fig. 1C), indicating the specificity of the effect.

Because wtSPIDAR inhibited the association between ACE2 and SARS-CoV-2 spike S1, next, we examined whether wtSPIDAR inhibited viral entry. Pseudoviruses are appropriate for virus entry assays, as they permit viral entry to be distinguished from other virus life cycle stages. Therefore, we used lentiviral particles pseudotyped with the SARS-CoV-2 spike S1 protein. Because HEK293 cells do not have any detectable ACE2, we employed hACE2-expressing HEK293 cells for entry assay. In pseudovirus luciferase assay, viral entry into cells is correlated to the levels of luciferase signals in the cells. Although lenti-naked infection did not increase luciferase signals in hACE2-expressing HEK293 cells, marked increase in luciferase activity was seen in pseudo–SARS-CoV-2–infected cells (Fig. 1D), indicating the entry of pseudo–SARS-CoV-2 into hACE2-HEK293 cells. As evident from (Fig. 1E, neutralizing Abs against hACE2, but not control IgG, markedly inhibited the admission of pseudo–SARS-CoV-2 into HEK293 cells, confirming that the entry of pseudo–SARS-CoV-2 into hACE2-HEK293 cells is dependent on ACE2. Although wtSPIDAR at a concentration of 0.1 µM did not inhibit pseudo–SARS-CoV-2–induced luciferase activity, marked reduction of luciferase activity was seen at 0.5 and 1.0 µM wtSPIDAR (Fig. 1F), suggesting that wtSPIDAR inhibits the entry of pseudo–SARS-CoV-2 into hACE2-HEK293 cells. This result was specific, as mSPIDAR at the same concentration remained unable to inhibit luciferase activity (Fig. 1F). Recently, we have described that a wild-type peptide corresponding to the ACE2-interacting domain of SARS-CoV-2 (wtAIDS) also inhibits the interaction between spike S1 and ACE2 (11). Therefore, in this study, we wanted to determine which one between wtAIDS and wtSPIDAR is more potent in inhibiting the entry of SARS-CoV-2 pseudovirus in hACE2-HEK293 cells. Similar to wtSPIDAR, wtAIDS also reduced luciferase activity in pseudo–SARS-CoV-2–infected hACE2-HEK293 cells (Fig. 1F). However, wtSPIDAR was more effective than wtAIDS (Fig. 1F) in inhibiting luciferase activity in pseudo–SARS-CoV-2–infected hACE2-HEK293 cells. Moreover, similar to pseudo–SARS-CoV-2, infection with pseudo-VSV also led to marked increase in luciferase activity in hACE2-HEK293 cells (Fig. 1G). However, either SPIDAR or AIDS remained unable to modulate pseudo–VSV-induced luciferase activity in hACE2-HEK293 cells (Fig. 1G), indicating the specificity of both SPIDAR and AIDS peptides.

To further confirm inhibition of viral entry by wtSPIDAR, we monitored GFP fluorescence. Although no GFP expression was seen in the cells infected with lenti-naked viral particles (Fig. 2A, 2B), marked GFP expression was found in cells infected with both SARS-CoV-2 pseudovirus (Fig. 2A) and VSV pseudovirus (Fig. 2B). However, wtSPIDAR, but not mSPIDAR, markedly inhibited GFP expression induced by pseudo–SARS-CoV-2 (Fig. 2A, 2C). In contrast, either wtSPIDAR or mSPIDAR had no effect on VSV pseudovirus-induced GFP expression in hACE2-HEK293 cells (Fig. 2B, 2D). Together, these results suggest that wtSPIDAR is capable of inhibiting the entry of pseudo–SARS-CoV-2, but not pseudo-VSV, into hACE2-HEK293 cells.

Some patients with COVID-19 display a severe symptom of acute respiratory distress syndrome with high mortality. This high severity is dependent on pulmonary inflammation induced by a cytokine storm (4) that is most likely mediated by IL-6 and other proinflammatory cytokines. Therefore, at first, we investigated if SARS-CoV-2 pseudovirus infection leads to the expression of proinflammatory cytokines in A549 lung cells. Marked upregulation of IL-6 (Fig. 3A) and IL-1β (Fig. 3B) mRNA expression in lenti–SARS-CoV-2–infected, but not lenti-naked–infected, cells suggest that SARS-CoV-2 pseudovirus infection is capable of upregulating proinflammatory cytokines in human lung cells. Abrogation of SARS-CoV-2 pseudovirus-induced mRNA expression of IL-6 and IL-1β by neutralizing Abs against SARS-CoV-2 spike S1 indicates that the induction of proinflammatory cytokine gene expression in A549 cells may be due to SARS-CoV-2 spike S1. It could be also due to the fact that the entry of the virus is blocked, and therefore, there is a lack of intracellular pathogen-associated molecular pattern detection. Next, we examined if SPIDAR was capable of suppressing the expression of IL-6 and IL-1β in SARS-CoV-2 pseudovirus-infected A549 lung cells. Inhibition of pseudo–SARS-CoV-2–induced expression of IL-6 (Fig. 3C) and IL-1β (Fig. 3D) by wtSPIDAR, but not mSPIDAR, suggests that wtSPIDAR is capable of knocking down the expression of proinflammatory molecules in SARS-CoV-2 pseudovirus-infected A549 cells. Alternatively, these results also suggest that wtSPIDAR, but not mSPIDAR, inhibits pseudoviral entry, thus preventing pathogen-associated molecular pattern–mediated expression of IL-6 and IL-1β. This result was specific, as either wtSPIDAR or mSPIDAR remained unable to inhibit the mRNA expression of IL-6 and IL-1β in VSV pseudovirus-infected A549 cells.

Recently, we have demonstrated that recombinant SARS-CoV-2 spike S1 is capable of inducing the activation of NF-κB and the expression of IL-6 in human A549 lung cells (11). Although boiled spike S1 does not induce the expression of proinflammatory molecules in A549 cells, anti–spike S1 neutralizing Ab (catalog number A3000-50; BioVision) neutralized the proinflammatory function of recombinant SARS-CoV-2 spike S1 (11). To further confirm that the induction of proinflammatory molecules in A549 cells by recombinant SARS-CoV-2 spike S1 is not due to any contaminant, we examined the effect of spike S1 in MyD88-depleted A549 cells. We performed CRISPR-mediated editing of MyD88 gene in human A549 cells followed by Western blotting with anti-MyD88 Ab (catalog number AB16527; Millipore). As evident from (Fig. 4A and 4B, CRISPR editing markedly reduced the level of MyD88 compared with scrambled control. However, CRISPR knockdown of MyD88 remained unable to inhibit SARS-CoV-2 spike S1–induced expression of TNF-α (Fig. 4C), IL-6 (Fig. 4D), and IL-1β (Fig. 4E) mRNAs in A549 cells. Because most of the known TLRs, except TLR3, require MyD88 as an adapter, these results indicate that the proinflammatory function of SARS-CoV-2 spike S1 is not due to any contamination by TLR ligands.

SARS-CoV-2 spike S1 is known to interact with ACE2 receptor (2). Therefore, to confirm that the proinflammatory effect of SARS-CoV-2 spike S1 in human A549 lung cells is also dependent on ACE2, we examined the effect of anti-hACE2 Ab (R&D Systems) on SARS-CoV-2 spike S1–induced expression of proinflammatory cytokines. As expected, stimulation with SARS-CoV-2 spike S1 led to marked induction of IL-6 (Fig. 5A) and IL-1β (Fig. 5B) mRNAs in A549 cells. However, anti-ACE2 Ab, but not control IgG, at different doses tested markedly suppressed the SARS-CoV-2 spike S1–induced expression of IL-6 (Fig. 5A) and IL-1β (Fig. 5B) mRNAs, indicating an essential role of ACE2 in SARS-CoV-2 spike S1–stimulated expression of proinflammatory cytokines in A549 cells. It has been shown that human A549 cells are poor hosts for SARS-CoV (24). A key feature in the infection of host cells with live SARS-CoV-2 is the involvement of transmembrane protease serine 2 (TMPRSS2). Only ACE2 expression is not enough for SARS-CoV-2 infection, because TMPRSS2 is necessary for the priming of the spike protein of SARS-CoV-2 (25, 26). However, once spike protein is cleaved and processed, it binds with ACE2 and does not require TMPRSS2. Similarly, recombinant spike S1 also does not need the involvement of TMPRSS2. In this study, because we are stimulating A549 cells with recombinant SARS-CoV-2 spike S1, the basal level of ACE2 present in A549 cells is sufficient for biological activity of spike S1.

Because wtSPIDAR inhibited SARS-CoV-2 pseudovirus-induced expression of IL-6 and IL-1β in A549 cells, next, we examined if SPIDAR was capable of suppressing inflammation in human A549 lung cells induced by recombinant SARS-CoV-2 spike S1. To understand the specificity of the effect, cells were also stimulated with proinflammatory cytokine TNF-α, dsRNA mimic polyinosinic-polycytidylic acid (poly-IC), and HIV-1 Tat. Recombinant spike S1 (Fig. 6A), TNF-α (Fig. 6E), poly-IC (Fig. 6I), and HIV-1 Tat (Fig. 6M) induced the activation of NF-κB in A549 cells as monitored by NF-κB–driven reporter (luciferase) activity. However, wtSPIDAR inhibited spike S1–induced activation of NF-κB (Fig. 6A). In contrast, wtSPIDAR remained unable to subdue the activation of NF-κB in A549 cells induced by TNF-α (Fig. 6E), poly-IC (Fig. 6I), and HIV-1 Tat (Fig. 6M). These results were specific, as mSPIDAR had no effect on the activation of NF-κB induced by any of the stimuli. To confirm these results, we also monitored the expression of IL-6 and IL-1β, proinflammatory cytokines that are driven by activated NF-κB. Spike S1 (Fig. 6B–D), TNF-α (Fig. 6F–H), poly-IC (Fig. 6J–L), and HIV-1 Tat (Fig. 6N–P) increased the expression of IL-6 and IL-1β in A549 cells. However, wtSPIDAR inhibited SARS-CoV-2 spike S1–induced mRNA expression of IL-6 and IL-1β in A549 cells (Fig. 6B–D). In contrast, wtSPIDAR could not inhibit the mRNA expression of IL-6 and IL-1β induced by TNF-α (Fig. 6F–H), poly-IC (Fig. 6J–L), and HIV-1 Tat (Fig. 6N–P). Moreover, the inability of mSPIDAR to inhibit the mRNA expression of IL-6 and IL-β induced by any of the stimuli used further indicated the specificity of the anti-inflammatory function of SPIDAR.

Although SARS-CoV-2 does not easily bind to ACE2 and infect normal mice, recently, we have seen that intranasal intoxication of SARS-CoV-2 spike S1 induces fever and important cardiac and respiratory symptoms of COVID-19 in normal C57/BL6 mice (11). To further characterize this SARS-CoV-2 spike S1–mediated mouse model, we performed the following: first, we monitored the status of ACE2 in lungs of C57/BL6 mice before and after spike S1 intoxication. Lungs of normal mice expressed ACE2, which decreased upon SARS-CoV-2 spike S1 intoxication (Supplemental Fig. 1A, 1B). The decrease in ACE2 expression in lung cells could be due to the internalization of ACE2 receptor–spike S1 complex via a clathrin-dependent manner (27) and/or suppression of ACE2 expression by proinflammatory cytokines generated from spike S1 intoxication (A. A. Potdar, S. Dube, T. Naito, G. Botwin, T. Haritunians, D. Li, S. Yang, J. Bilsborough, L. A. Denson, M. Daly, et al., manuscript posted on medRxiv, DOI: 10.1101/2020.04.19.20070995). Anyway, the decrease in ACE2 expression was specific, as the lungs of SARS-CoV-2 spike S1–intoxicated mice, but not normal mice, expressed IL-6 (Supplemental Fig. 1A, 1C). Therefore, normally, mouse lungs express ACE2, which is decreased by spike S1 insult. Second, we examined the efficacy of heat-denatured SARS-CoV-2 spike S1 in inducing fever and locomotor abnormalities in C57/BL6 mice. SARS-CoV-2 spike S1 was boiled for 5 min for denaturation. Failure of boiled recombinant SARS-CoV-2 spike S1 to induce fever (Supplemental Fig. 2A), increase serum lactate dehydrogenase (LDH) (Supplemental Fig. 2B), and cause impairment in locomotor activities (Supplemental Fig. 2C–G) suggests that the induction of these symptoms in mice is due to SARS-CoV-2 spike S1 protein. Third, we studied whether anti-spike S1 Ab could neutralize functions of SARS-CoV-2 spike S1 in C57/BL6 mice. Neutralization of SARS-CoV-2 spike S1–mediated increase in body temperature (Supplemental Fig. 2A), upregulation in serum LDH (Supplemental Fig. 2B), and induction of hypolocomotion (Supplemental Fig. 2C–G) in mice by coadministration of anti–SARS-CoV-2 spike S1 Ab indicate the involvement of SARS-CoV-2 spike S1 protein in causing these problems in mice. According to Mossel et al. (24), mouse ACE2 could be a poor receptor for SARS-CoV-2. However, in this study, we have challenged mice with recombinant SARS-CoV-2 spike S1, not live SARS-CoV-2.

Therefore, next, we examined if wtSPIDAR was capable of controlling these symptoms in mice. Because patients with COVID-19 are and/or will be treated with drugs usually after the diagnosis of the disease, we examined whether wtSPIDAR administered 5 d after initiation of the disease (Fig. 7A) was still capable of protecting mice from COVID-19–related complications. We selected the 5-d window, as all SARS-CoV-2 spike S1–intoxicated mice exhibited a body temperature of ∼100°F on 5 d of intoxication (Fig. 7B). Parallel to that observed in human lung cells, intranasal insult with recombinant SARS-CoV-2 spike S1 (Fig. 7A) led to the activation of NF-κB in vivo in the lung of C57/BL6 mice (Fig. 7C), which was strongly inhibited by intranasal treatment with wtSPIDAR, but not mSPIDAR (Fig. 7C). In situ ChIP analysis showed the recruitment of classical NF-κB subunits (p65 and p50), RNA polymerase II, and transcriptional coactivator p300 to the IL-6 gene promoter in vivo in the lungs of SARS-CoV-2 spike S1–intoxicated mice that was markedly inhibited by intranasal treatment with wtSPIDAR, but not mSPIDAR (Fig. 8). Consistently, wtSPIDAR, but not mSPIDAR, also inhibited the expression of IL-6 mRNA (Fig. 7D) and protein (Fig. 7F) as well as another proinflammatory cytokine (IL-1β) mRNA (Fig. 7E) in the lungs of SARS-CoV-2 spike S1–insulted mice. Similarly, wtSPIDAR treatment also suppressed the serum level of IL-6, an important contributor of the so-called cytokine storm–related illness (28), and CRP, an important biomarker for monitoring COVID-19–associated lethality (29), in SARS-CoV-2 spike S1–insulted mice (Fig. 7G, 7H).

Fever is probably one of the most common symptoms of COVID-19 (2, 30), and intranasal treatment with wtSPIDAR, but not mSPIDAR, also led to the normalization of body temperature in SARS-CoV-2 spike S1–intoxicated mice (Fig. 7I). SARS-CoV-2 spike S1–intoxicated mice also feature widespread neutrophil infiltration into the lungs (11). Therefore, we monitored the status of neutrophil infiltration upon SPIDAR treatment. As evident from (Fig. 9A, treatment with wtSPIDAR, but not mSPIDAR, led to marked inhibition of neutrophil infiltration into the lungs of SARS-CoV-2 spike S1–intoxicated mice. Cell counting as well as assessment of lung injury also showed that wtSPIDAR, but not mSPIDAR, reduced lung neutrophil infiltration (Fig. 9B), normalized lung epithelial cells (Fig. 9C, 9D), and reduced overall lung injury (Fig. 9E) in SARS-CoV-2 spike S1–intoxicated mice.

Because many cardiac features of COVID-19 are modeled in SARS-CoV-2 spike S1–intoxicated mice (11), we examined if wtSPIDAR was capable of improving heart functions in these mice. Noninvasive ECG showed cardiac arrhythmias in SARS-CoV-2 spike S1–intoxicated mice as compared with control untreated mice (Fig. 10A, 10B). However, wtSPIDAR, but not mSPIDAR, normalized electrical activity of the heart, as evident from ECG (Fig. 10A–D). Similarly, wtSPIDAR, but not mSPIDAR, also stabilized heart rate (Fig. 10E), ORS interval (Fig. 10F), QT interval (Fig. 10G), RR interval (Fig. 10H), and heart rate variability (Fig. 10I) in SARS-CoV-2 spike S1–intoxicated mice. As expected, serum LDH level was also markedly higher in SARS-CoV-2 spike S1–intoxicated mice than normal mice (Fig. 10J). However, wtSPIDAR, but not mSPIDAR, reduced and/or normalized serum LDH in spike S1–intoxicated mice (Fig. 10J).

Recently, we have seen that SARS-CoV-2 spike S1 intoxication also causes functional discrepancies in C57/BL6 mice (11). Therefore, we examined whether wtSPIDAR treatment was capable of improving such behavioral deficits. Interestingly, wtSPIDAR, but not mSPIDAR, treatment increased overall locomotor activities as evident by heat map (Fig. 11A), distance traveled (Fig. 11B), velocity (Fig. 11C), cumulative duration (Fig. 11D), center zone frequency (Fig. 11E), and rotarod performance (Fig. 11F). We did not notice any drug-related side effect (e.g., hair loss, appetite loss, weight loss, untoward infection and irritation, etc.) in any mouse upon treatment with intranasal SPIDAR.

Unraveling the mechanism of the disease process of COVID-19 to design an effective therapeutic approach for controlling the disease process is of paramount importance. Although ACE2 plays an important role in vascular diseases (6), SARS-CoV-2 binds to ACE2 for entering into human cells. Therefore, COVID-19 infection becomes deadly in patients with underlying cardiovascular issues. Under such situations, it is expected to use inhibitors of ACE2 (31); however, such inhibitors may not be beneficial for patients with severe COVID-19 in the intensive care unit, as ACE2 inhibition may further aggravate symptoms in patients with COVID-19 with hypertension and cardiovascular issues. Therefore, to dissociate the interaction between ACE2 and SARS-CoV-2 spike S1 without altering the level and function of ACE2, we analyzed the interaction between ACE2 and SARS-CoV-2 spike S1 and designed a small hexapeptide corresponding to the SPIDAR. Resembling the spike S1-binding domain of ACE2, wtSPIDAR dissociated the interaction between ACE2 and SARS-CoV-2 spike S1. Accordingly, in a pseudovirus cell entry assay, wtSPIDAR inhibited the entry of SARS-CoV-2 pseudovirus into hACE2-expressing HEK293 cells. In contrast, wtSPIDAR remained unable to modulate the entry of VSV pseudovirus into hACE2-expressing HEK293 cells, indicating the functional specificity of SPIDAR. Recently, we have demonstrated that another peptide corresponding to AIDS also detaches the association between ACE2 and SARS-CoV-2 spike S1 (11). Although wtAIDS also inhibited the entry of SARS-CoV-2 pseudovirus into hACE2-expressing HEK293 cells, wtSPIDAR seemed to be more effective than wtAIDS in suppressing the viral entry. Although COVID-19 is the first documented coronavirus pandemic in the history (32), there are several human-infecting coronaviruses with different degrees of virulence. However, the first step of coronavirus pathogenesis is the attachment of coronavirus to the host via viral spike glycoprotein. Therefore, in addition to COVID-19, our SPIDAR and AIDS technologies may be effective against other coronaviruses.

Inflammation plays an important role in the pathogenesis of many human disorders, including COVID-19. It has been reported that patients with COVID-19 with severe symptoms suffer from cytokine storm (33). We have seen that spike S1 alone is sufficient to drive the expression of proinflammatory cytokines in human lung cells (11). Therefore, spike S1 may be responsible for the cytokine storm seen in patients with COVID-19. However, consistent with the inhibition of association between SARS-CoV-2 spike S1 and ACE2, wtSPIDAR decreased the activation of NF-κB and the expression of IL-6 and IL-1β in SARS-CoV-2 spike S1–intoxicated A549 lung cells. Inability of wtSPIDAR to inhibit the activation of NF-κB and associated expression of proinflammatory cytokines induced by TNF-α (a prototype proinflammatory cytokine), poly-IC (viral dsRNA mimic), and Tat (transactivator of HIV-1 transcription) delineates the selective nature of wtSPIDAR. This is important because there are many anti-inflammatory therapies (e.g., steroids) available to take care of the cytokine storm in patients with COVID-19. However, it is a double-edged sword because very often, these treatments cause immunosuppression. Therefore, it has been suggested that reducing hyperinflammation by anti-inflammatory drugs in severely affected patients with COVID-19 should be approached with caution. Because wtSPIDAR inhibits cytokines produced only by SARS-CoV-2 spike S1, but not other inflammatory stimuli, wtSPIDAR is not expected to cause immunosuppression.

Another highlight of our finding is that wtSPIDAR corresponds to peptide sequence of SARS-CoV-2 from the binding site with ACE2. Therefore, wtSPIDAR will only inhibit the binding of ACE2 with SARS-CoV-2 without affecting beneficial enzyme activities and basal level of ACE2. Similarly, our recently described wtAIDS should also specifically inhibit the association of ACE2 with SARS-CoV-2 (11). However, wtAIDS binds to ACE2, as this sequence has been designed from the ACE2-binding site of spike S1. In contrast, wtSPIDAR should bind to spike S1 of SARS-CoV-2 as this sequence is derived from the spike S1–binding region of ACE2. Although we have not examined if wtAIDS binds to ACE2, it may have ACE2-inhibitory activity. Under that condition, by suppressing ACE2 activity, wtAIDS may pose risks for patients with preexisting cardiovascular problems because the most important function of ACE2 is to convert a physiological vasoconstrictor (angiotensin II) to a vasodilator (Ang1-7), which is known to function through Mas receptor to transduce antiproliferative/vasodilatory activities (34). Moreover, because myocardial NADPH oxidase is responsible for myocardial oxidative stress and inflammation, by suppressing myocardial NADPH oxidase, Ang1-7 is protective for the heart (35). Therefore, any inhibition and/or dysfunction of ACE2 may not be beneficial for patients with COVID-19 with hypertension, diabetes, and preexisting vascular problems. In contrast, SPIDAR should not cause these problems, as it would bind to SARS-CoV-2 spike S1, but not ACE2. Therefore, SPIDAR will function only in the presence of SARS-CoV-2.

Although vaccination against COVID-19 started and about half the population in the United States has been vaccinated, their distribution globally will take months and possibly years in some parts of the world. On top of that, storage of some of the widely used COVID-19 vaccines at −40°C to −80°C throughout the world is another big issue. Moreover, vaccines may not entirely prevent the spread of COVID-19. We also must remember that despite flu vaccination, ∼40,000 to 50,000 people die each year in the United States from the flu. Therefore, a specific medicine for handling cytokine storm and taking care of respiratory and cardiac complications caused by SARS-CoV-2 infection will be necessary for better management of COVID-19 even in the postvaccine era. Although some mAbs against SARS-CoV-2 spike proteins (e.g., sotrovimab, casirivimab, imdevimab, etc.) and IL-6 receptor (e.g., tocilizumab) have been approved for COVID-19 treatment (36, 37), other than mAbs, until now, no other COVID-19–specific therapy was available. Most of the non-Ab therapies being investigated and tried for COVID-19 are repurposed drugs. For example, remdesivir is being repurposed from HIV to COVID-19 emergency use (38). Similarly, dexamethasone, a corticosteroid that is used to take care of wide range of conditions for its anti-inflammatory and immunosuppressive effects, has been proposed for COVID-19 (39). Aviptadil is a drug for the treatment of erectile dysfunction. It is being tried to alleviate cardiac problems of critically ill patients with COVID-19. Hydroxychloroquine, a prototype antimalarial drug, was also considered for COVID-19 before clinical trials ruled it out (30). Similarly, multiple sclerosis drug IFN-β-1b is also being considered for lowering mortality rate in COVID-19 (40). In contrast, SPIDAR is very selective to inhibit cellular entry of SARS-CoV-2 pseudovirus, but not VSV pseudovirus, and knock down inflammation only associated with SARS-CoV-2 spike S1. Protection of lungs, normalization of heart functions, reduction of fever, decrease in serum markers, and improvement in locomotor activities in SARS-CoV-2 spike S1–intoxicated mice by nasally administered SPIDAR suggest that selective targeting of the ACE2-to-SARS-CoV-2 contact by SPIDAR may be beneficial for COVID-19.

The authors have no financial conflicts of interest.