Stapled Peptides Based on Human Angiotensin-Converting Enzyme 2 (ACE2) Potently Inhibit SARS-CoV-2 Infection In Vitro

SARS-CoV-2 is a novel virus with many unknowns. No vaccine or specific therapy is available yet to prevent and treat this deadly virus.

Here, we took advantage of Helix-1 of ACE2 to design peptide-based therapeutics for COVID-19. However, Helix-1 is expected to lose its helical conformation in solution; therefore, it will also lose the orientation of most, if not all, of the binding site residues to make any proper interaction. To counter this, we adopted the very-well-studied concept of hydrocarbon stapling techniques mentioned above to reinforce the bioactive helical conformation of Helix-1 (Fig. 2). We hypothesize that stapled peptides will be a-helical and will be proteolytically stable. They are expected to bind to the SARS-CoV-2 RBD and inhibit virus entry to cells.
In this report, we present the design of four such stapled peptides based on Helix-1 of ACE2. Three of these peptides showed potent SARS-CoV-2-inhibitory activity in an in vitro pseudovirus and in an authentic virus infectivity assay. These stapled peptides have potential as a therapeutic for COVID-19 by functioning as a decoy of the ACE2 binding site for SARS-CoV-2 and preventing the virus from binding to the ACE2 receptor, thereby preventing virus entry and infection.

RESULTS AND DISCUSSION
Design of double-stapled peptides. Structures recently solved by cryo-EM and X-rays provided intricate details on how SARS-CoV-2 RBD binds to the host receptor ACE2. The major binding sites are located on Helix-1 of ACE2 (Fig. 1). A recent report indicated that a computationally designed short peptide from Helix-1 of ACE2 showed a nanomolar binding affinity to SARS-CoV-2 RBD (6). However, the usefulness of such a linear peptide in isolation as a drug is yet to be proven. In general, when a helix is taken out as a peptide from a protein, it loses its secondary structure. These peptides are also prone to proteolytic degradation. In contrast, if such small peptides are constrained by hydrocarbon stapling, an enhancement of helical content can be achieved together with resistance to proteolysis and improved binding affinity and biological activity (11-21, 23, 24). Therefore, we envisioned that the stapling of Helix-1 would render it a-helical, resistant to proteolysis, and, therefore, a highly potent inhibitor of SARS-CoV-2.
i/i14 and i/i17 hydrocarbon stapling. We used the well-established stapling techniques to synthesize hydrocarbon-stapled peptides (25)(26)(27) from Helix-1 of ACE2, which are expected to maintain high a-helicity, have high binding affinity to SARS-CoV-2 RBD, and have improved antiviral activity and stability from proteolysis. We decided to use double stapling because of the relatively longer helical region that binds to the SARS-CoV-2 RBD. Double stapling has been reported to confer striking protease resistance (20,28,29). Furthermore, Mourtada et al. recently reported 2-to 4fold enhancement of antimicrobial potency by double-stapled peptides over single stapling (23). Double stapling was also reported to achieve improved pharmacokinetic profiles, including oral absorption of peptide-based HIV-1 fusion inhibitors (28). For the i,i14 and i,i17 stapling, we used nonnatural amino acids as illustrated in Fig. 2 to introduce the hydrocarbon staples to Helix-1. The stapling sites were selected based on the binding site information of SARS-CoV-2 RBD with ACE2 (Helix-1) from the cryo-EM and X-ray structure mentioned above so that they do not interfere with the critical binding site residues. We have synthesized (CPC Scientific, Inc.) four stapled peptides, as depicted in Fig. 2. We also synthesized the linear peptide NYBSP-C as a control. Besides, we purchased StRIP16 (29), which is a Rab8a GTPase-binding double-stapled peptide and unrelated to ACE2, to use as a control.
Effect of double stapling on the helical propensity. Due to the extended size of Helix-1, we decided to double staple Helix-1 of ACE2 to preserve the a-helical content of the peptides. To determine the helical propensity of these peptides, we resorted to circular dichroism (CD) spectroscopy for analyzing secondary structures. The data in Fig. 3A indicate that all the stapled peptides had the maxima at 190 nm and two minima at 208 and 222 nm, confirming that all these stapled peptides had a-helical structure. The percent helicity calculation shows remarkable improvement over the linear peptide. The data in Fig. 3B indicate that NYBSP-1 had the highest helical propensity of 94%. Stapled peptides NYBSP-2 and NYBSP-4 had percent helicities of 61 and 80, respectively. NYBSP-3 showed the lowest percent helicity (50%) among the four stapled peptides. The control peptide, NYBSP-C, as expected, showed no indication of a-helical propensity, with only 19% helicity. The control stapled peptide, StRIP16, was reported to have 38% helicity (29).
Binding affinity of a double-stapled peptide by SPR analysis. We used surface plasmon resonance (SPR) to determine the binding affinity of one double-stapled peptide, NYBSP-4, with SARS-CoV-2 RBD. This method is useful in measuring the binding constant (K D ) as well as k on (also known as association constant, k a ) and k off (also known as dissociation constant, k d ). We also used the linear peptide NYBSP-C as a control. Various concentrations of biotinylated SARS-CoV-2 RBD were manually printed onto the chip (immobilization) via biotin-avidin conjugation. Peptides were passed through the chip surface, and the signal changes (in arbitrary units [AU]) of each peptide at varied concentrations were recorded ( Fig. 4A and B). The resulting data were fit to a 1:1 binding model, and the binding affinity K D and kinetic parameters k on and k off of the interaction of SARS-CoV-2 RBD with NYBSP-C and NYBSP-4 were determined (Fig. 4C). The K D value of NYBSP-4 was 2.2 mM, which was very much in line with the antiviral activity we observed against SARS-CoV-2 pseudovirus ( Table 1). As expected, the linear peptide NYBSP-C, used as a control, showed poor binding (73.7 mM), which was about 33.5-fold lower than NYBSP-4. The SPR analysis also confirmed that the double-stapled peptides bind directly to the SARS-CoV-2 RBD.
Proteolytic stability of stapled peptides in human plasma. Proteolytic stability remains a potential concern in developing peptide-based therapy. We wanted to address this problem by introducing double stapling in Helix-1 of the ACE2 receptor. Double stapling is known to enhance protection from proteolysis. However, it was prudent for us to test how the stapled peptides that we designed withstand proteolysis in human plasma. The peptides were incubated with human plasma at 37°C, and the fragments formed from proteolysis were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS) at different time points (0, 10, 30, 60, and 120 min). We used a small-molecule drug, propantheline bromide, as an assay control. The data in Fig. 5 show that stapled peptide NYBSP-4 is highly stable during the time used for the assay. The calculated half-life (T 1/2 ) of NYBSP-4 was .289 min, but surprisingly, the Stapled Peptides as Inhibitors of SARS-CoV-2 ® linear control peptide NYBSP-C also showed similar stability (T 1/2 .289 min). Usually, small peptides are known to get proteolytically cleaved quickly. The unusual stability of the linear peptide in human plasma in our test is yet to be understood.
Validation of the SARS-CoV-2 pseudovirus. We prepared SARS-CoV-2 pseudovirus capable of single-cycle infection by transfecting HEK293T/17 cells with an HIV-1 Envdeleted proviral backbone plasmid, pNL4-3DEnv-NanoLuc, and pSARS-CoV-2-S D19 plasmid (30). The incorporation of the spike protein in the pseudovirus was validated by Western blot analysis using the SARS spike protein antibody (Novus Biologicals) (Fig. 6A). This antibody targets explicitly amino acids 1124 to 1140 of the spike protein S2. We found a specific band at 80 kDa, which identifies the subunit S2, and a second band at about 190 kDa, which corresponds to the full-length S protein (S1 1 S2), as previously reported (31,32), confirming the incorporation of the SARS-CoV-2 S protein in the pseudovirus. Moreover, the correlation of SARS-CoV-2 pseudovirus with the expression of the ACE2 receptor was analyzed by infecting five different cell types including human fibrosarcoma HT1080 cells overexpressing ACE2 (HT1080/ACE2) and the parental HT1080 cell type, human lung carcinoma A549 cells overexpressing ACE2 (A549/ RLU) compared to that for HT1080/ACE2 but about 9-fold higher than what we detected for the parental cell type A549. These results were confirmed by analyzing the expression levels of the ACE2 receptor in the different cell lines by Western blotting (Fig. 6C). As shown in Blot-1, we found that ACE2 expression was undetectable in the parental cell line HT1080, while it was overexpressed in HT1080/ACE2 cells. The lower infection detected in A549/ACE2 cells suggested a lower ACE2 expression in these cells. For this reason, in Blot-2, we loaded a larger amount of proteins (75 mg). As expected, the expression levels of ACE2 were undetectable in the parental cell line A549 and the HeLa cells. The expression was moderate in the A549/ACE2 cells compared to HT1080/ACE2. These data confirmed that SARS-CoV-2 pseudovirus infects the cells through its interaction with the ACE2 receptor.
Anti-SARS-CoV-2 activity and cytotoxicity of double-stapled peptides in cellbased assays. We evaluated the anti-SARS-CoV-2 activity of the double-stapled peptides by infecting the two cell types overexpressing hACE2, the human fibrosarcoma HT1080/ACE2 cells and the human lung carcinoma A549/ACE2 cells, with a SARS-CoV-2 pseudovirus that was pretreated with escalating concentrations of the peptides for 30 min. The linear peptide NYBSP-C, an unrelated double-stapled peptide (StRIP16) (29), and an anti-ACE2 monoclonal antibody (Mab) (AC384 from AdipoGen Life Sciences) were used as controls. The concentration required to inhibit 50% (IC 50 ) of SARS-CoV-2 pseudovirus infection was calculated. The results are shown in Table 1. The doseresponse plots (Fig. 7) indicate that all the double-stapled peptides efficiently inhibited the infection of HT1080/ACE2 cells and A549/ACE2 cells by the SARS-CoV-2 pseudovirus with low-micromolar potency. Three of the four stapled peptides displayed potent antiviral activity in both cell lines. The IC 50 s in HT1080/ACE2 and A549/ACE2 cells for NYBSP-1 were 4.1 6 0.26 and 2.2 6 0.14, for NYBSP-2 were 2.9 6 0.27 and 2.68 6 0.14, and for NYBSP-4 were 1.97 6 0.14 and 2.8 6 0.08 mM, respectively. NYBSP-3 was the least active peptide with an IC 50 of 12.9 6 0.35 mM in HT1080/ACE cells and ;25 mM in A549/ACE2 cells. Most significantly, as expected, neither of the two controls, the linear peptide NYBSP-C and the double-stapled peptide StRIP16, showed any antiviral activity while the IC 50 detected for the anti-ACE2 MAb was ,70 ng/ml (at this concentration, we detected about 65 to 80% viral inhibition). It is interesting that NYBSP-3 had the lowest percent a-helicity among all the stapled peptides designed, and it also showed FIG 5 The plasma stability of double-stapled peptide NYBSP-4 (red) and control linear peptide NYBSP-C (green). Propantheline bromide (a small-molecule antimuscarinic agent) was used as an assay control.
Stapled Peptides as Inhibitors of SARS-CoV-2 ® the least antiviral activity. Moreover, none of the peptides showed activity against the control pseudovirus vesicular stomatitis virus glycoprotein (VSV-G) ( Table 1), suggesting that their inhibitory activity is specific to SARS-CoV-2.
The 50% cytotoxicity (CC 50 ) of the peptides was assessed in parallel with the inhibition assays. We found that the peptides did not induce any apparent cell toxicity at the highest dose tested. More specifically, ,10% toxicity was detected at the higher dose used in this assay (Table 1).
Additionally, the antiviral activity of the peptides was evaluated by infecting Vero

MATERIALS AND METHODS
Peptide synthesis: general procedure of coupling. Fmoc (9-fluorenylmethoxy carbonyl) Rink amide MBHA resin (0.1 mmol) was swelled in dimethylformamide (DMF) (20 ml) for 1 h. The suspension was filtered. Twenty percent piperidine in DMF (20 ml) was added to the resin to remove Fmoc. The suspension was kept at room temperature for 0.5 h, while a stream of nitrogen was bubbled through it. The mixture was filtered, and the resin was washed with DMF (6 Â 20 ml). Fmoc-Asn(Trt)-OH (0.3 mmol) was preactivated with N,N9-diisopropylcarbodiimide (DIC) (0.3 mmol) and hydroxybenzotriazole (HOBt) (0.3 mmol) in 10 ml DMF, and then the mixture was added into the resin. The reaction was carried out in a nitrogen atmosphere. The Kaiser ninhydrin test was used to indicate reaction completion. After the reaction was completed, the suspension was filtered, and the resin was washed with DMF (3 Â 20 ml). After that, all sequences of amino acids were coupled and completed to give the peptidyl resin full protection, which was used to do the cyclization with Grubbs first catalyst.
The peptidyl resin was washed with DCM (3 Â 20 ml). Then, 10 ml 1,2-dichloroethane was added in the reaction vessel, and the solution was bubbled by N 2 . Fifteen minutes later, 30 mg of Grubbs first catalyst was added to the reaction vessel and bubbled by N 2 for 4 h and filtered, and the resin was washed with methanol (MeOH) (2 Â 150 ml), DCM (2 Â 150 ml), and MeOH (2 Â 150 ml). The resin was dried under vacuum overnight to give cyclized peptidyl resin.
E solution (trifluoroacetic acid [TFA]-thioanisole-phenol-ethanedithiol [EDT]-H 2 O = 87.5:5:2.5:2.5:2.5, 350 ml) was added into the peptidyl resin, and the suspension was shaken for 3.5 h followed by filtration. Ether (2,000 ml) was added into the filtrate, and peptide was precipitated. The mixture was centrifuged, and the ether layer was decanted. The peptide was washed three times with ether and dried under vacuum overnight to give the crude peptide.  Stapled Peptides as Inhibitors of SARS-CoV-2 ® All crude peptides were purified by pre-high-performance liquid chromatography (pre-HPLC) to .90% purity. They were acetylated at the C terminus and amidated at the N terminus. The molecular weight of all peptides was confirmed by electrospray mass spectrometry.
Circular dichroism spectroscopy. The secondary structure analysis to calculate percent a-helicity was performed by Creative Biolabs (New York, NY, USA). In brief, peptides were dissolved in 10 mM phosphate-buffered saline (PBS) to make a final concentration of 0.125 mg/ml in PBS. The CD spectra were obtained on a Chirascan spectropolarimeter (Applied Photophysics, Leatherhead, United Kingdom). The measurement parameters were set up as follows: wavelength, 185 to 260 nm; step size, 0.5 nm; temperature, 25°C. The percent helicity was calculated using the software CDNN (Circular Dichroism analysis using Neural Networks) (http://gerald-boehm.de/download/cdnn) (33).
SPR analysis. The binding study of one double-stapled peptide, NYBSP-4, and the linear control peptide, NYBSP-C, was performed by Profacgen, New York, NY. The bare gold-coated (thickness, 47 nm) PlexArray Nanocapture sensor chip (Plexera Bioscience, Seattle, WA, USA) was prewashed with 10Â PBS-Tween (PBST) for 10 min, 1Â PBST for 10 min, and deionized water twice for 10 min before being dried under a stream of nitrogen prior to use. Various concentrations of biotinylated recombinant SARS-CoV-2 spike RBD protein (Bioss Antibodies, Woburn, MA) dissolved in water were manually printed onto the chip with Biodo bioprinting at 40% humidity via biotin-avidin conjugation. Each concentration was printed in replicate, and each spot contained 0.2 ml of the sample solution. The chip was incubated in 80% humidity at 4°C overnight and rinsed with 10Â PBST for 10 min, 1Â PBST for 10 min, and deionized water twice for 10 min. The chip was then blocked with 5% (wt/vol) nonfat milk in water overnight and washed with 10Â PBST for 10 min, 1Â PBST for 10 min, and deionized water twice for 10 min before being dried under a stream of nitrogen prior to use. SPR imaging (SPRi) measurements were performed with PlexArray HT (Plexera Bioscience, Seattle, WA, USA). Collimated light (660 nm) passes through the coupling prism, reflects off the SPR-active gold surface, and is received by the charge-coupled device (CCD) camera. Buffers and the test peptides were injected by a nonpulsatile piston pump into the 30-ml flow cell that was mounted on the coupling prim. Each measurement cycle contained four steps: washing with running buffer containing 10 mM HEPES, pH 7.4, 150 mM NaCl, 3 mM EDTA, and 0.05% Tween 20 at a constant rate of 2 ml/s to obtain a stable baseline, sample injection at 5 ml/s for binding, surface washing with running buffer at 2 ml/s for 300 s, and regeneration with 0.5% (vol/vol) H 3 PO 4 at 2 ml/s for 300 s. All the measurements were performed at 25°C. The signal changes after binding and washing (in AU) are recorded as the assay value.
Kinetics fitting and analysis. Selected protein-grafted regions in the SPR images were analyzed, and the average reflectivity variations of the chosen areas were plotted as a function of time. Real-time binding signals were recorded and analyzed by the Data Analysis Module (DAM; Plexera Bioscience, Seattle, WA, USA). Kinetic analysis was performed using BIAevaluation 4.1 software (Biacore, Inc.). The resulting data were fit to a 1:1 binding model using Biacore evaluation software (GE Healthcare).
Plasma stability study. The plasma stability study was performed by Creative Biolabs (New York, NY, USA). In brief, the pooled frozen human plasma (BioIVT, United Kingdom) was thawed in a water bath at 37°C prior to the experiment. Plasma was centrifuged at 4,000 rpm for 5 min, and the clots, if any, were removed. The pH was adjusted to 7. was added to precipitate protein and mixed thoroughly. Sample plates were centrifuged at 4,000 rpm for 10 min. An aliquot of supernatant (100 ml) was transferred from each well to the plate. The samples were shaken at 800 rpm for about 10 min before submission to LC-MS/MS analysis. The percent remaining of the test peptides after incubation in plasma was calculated using the following equation: % remaining = 100 Â (PAR at appointed incubation time/PAR at time zero), where PAR is the peak area ratio of analyte versus internal standard (IS).
Pseudovirus preparation. To prepare pseudoviruses capable of single-cycle infection, 10 Â 10 6 HEK293T/17 cells were transfected with a proviral backbone plasmid and an envelope expression vector by using FuGENE6 (Promega, Madison, WI) and following the manufacturer's instructions. To obtain the SARS-CoV-2 pseudovirus, the cells were transfected with the HIV-1 Env-deleted proviral backbone plasmid pNL4-3DEnv-NanoLuc DNA and the pSARS-CoV-2-S D19 plasmid (30). For the VSV-G pseudovirus, the cells were transfected with the Env-deleted proviral backbone plasmid pNL4-3.Luc.R-E-DNA and the pHEF-VSV-G envelope expression vector. Pseudovirus-containing supernatants were collected 2 days after transfection and filtered, their titers were determined, and they were stored in aliquots at 280°C. Pseudovirus titers were determined to identify the 50% tissue culture infectious dose (TCID 50 ) by infecting different cell types. For the titers in HT1080/ACE2 cells, 2 Â 10 4 cells were added to 100-ml aliquots of serial 2-fold dilutions of pseudoviruses in a 96-well plate and incubated for 24 h. For the titers in A549/ACE2 cells, 1 Â 10 4 cells were added to 100-ml aliquots of serial 2-fold dilutions of pseudoviruses in a 96-well plate and incubated for 48 h. Following the incubation time, the cells were washed with PBS and lysed with 50 ml of the cell culture lysis reagent (Promega, Madison, WI). For the SARS-CoV-2 titers, 25 ml of the lysates was transferred to a white plate and mixed with the same volume of Nano-Glo luciferase reagent (Promega). For the VSV-G titers, 25 ml of the lysates was transferred to a white plate and mixed with 50 ml of luciferase assay reagent (luciferase assay system, Promega). We immediately measured the luciferase activity with a Tecan Spark multifunctional microplate reader (Tecan, Research Triangle Park, NC). The wells producing relative luminescence unit (RLU) levels 10 times the cell background were scored as positive. We calculated the TCID 50 according to the Spearman-Karber method (37).
Analysis of the incorporation of spike proteins into SARS-CoV-2 pseudovirus. To confirm the incorporation of the spike protein into the SARS-CoV-2 pseudovirus, 2 ml of the pseudovirus-containing supernatant was ultracentrifuged for 2 h at 38,000 rpm on a 20% sucrose cushion to concentrate the viral particles. Viral pellets were lysed and processed for protein analysis. The viral proteins were resolved on a NuPAGE Novex 4 to 12% Bis-Tris gel (Invitrogen, Carlsbad, CA) and immunodetected with a SARS spike protein antibody (NB-100-56578; Novus Biological, Littleton, CO). An anti-rabbit-IgG horseradish peroxidase (HRP)-linked whole antibody (GE Healthcare, Chicago, IL) was used as a secondary antibody. Proteins were visualized using chemiluminescence. Additionally, the correlation of the pseudovirus SARS-CoV-2 with the expression of the ACE2 receptor was analyzed by infecting five different cell types expressing different amounts of the ACE2 receptor with the same volume of the pseudovirus-containing supernatant. Briefly, 50 ml of SARS-CoV-2 diluted with 50 ml serum-free medium was added to wells of a 96-well cell culture plate. Next, the cells were added as follows: HT1080\ACE2 and HT1080 cells were added to the respective wells at the concentration of 2 Â 10 4 cells/well and incubated for 24 h at 37°C; A549\ACE2, A549, and HeLa cells were added to the respective wells at a concentration of 1 Â 10 4 cells/ well and incubated for 48 h at 37°C. Uninfected cells for all cell lines were used as a negative control. Following the incubation time, the cells were washed with PBS and lysed with 50 ml of the cell culture lysis reagent. Twenty-five microliters of the lysates was transferred to a white plate and mixed with the same volume of Nano-Glo luciferase reagent. The luciferase activity was immediately measured with a Tecan Spark.
Evaluation of the ACE2 expression in different cell lines. The expression levels of the ACE2 receptor was evaluated by Western blotting to find a correlation with the infection levels detected in the five cell lines HT1080\ACE2, HT1080, A549\ACE2, A549, and HeLa. Cell pellets were lysed and processed for protein analysis. For Blot-1, we loaded 50 mg of proteins (medium, HeLa, HT1080, and HT1080/ACE2), and for Blot-2, we loaded 75 mg of proteins (medium, HeLa, A549, and A549/ACE2). The proteins were resolved on a NuPAGE Novex 4 to 12% Bis-Tris gel and immunodetected with a human anti-ACE2 MAb (AC384) (AdipoGen Life Sciences, San Diego, CA). The enhanced chemiluminescence (ECL) mouse IgG, HRP-linked whole Ab (from sheep) (Amersham, Little Chalfont, United Kingdom) was used as a secondary antibody. Proteins were also immunodetected with the housekeeping gene b-actin as a loading control. Proteins were visualized using chemiluminescence.
Measurement of antiviral activity. (i) HT1080/ACE2 cells. The antiviral activity of the peptides was evaluated in single-cycle infection assay by infecting HT1080/ACE2 cells with the SARS-CoV-2 pseudovirus, as previously described with minor modifications (31). Briefly, in 96-well culture plates, aliquots of SARS-CoV-2 at about 3,000 TCID 50 /well at a multiplicity of infection (MOI) of 0.1 were preincubated with escalating concentrations of the peptides for 30 min. Next, 2 Â 10 4 cells were added to each well and incubated at 37°C. HT1080/ACE2 cells cultured with medium with or without the SARS-CoV-2 pseudovirus were included as positive and negative controls, respectively. The linear peptide NYBSP-C, StRIP16 (29) (Tocris, Bristol, United Kingdom), which is a SARS-CoV-2-unrelated Rab8a GTPase-binding doublestapled peptide, and an anti-ACE2 MAb (AC384; catalog no. AG-20A-0037PF; AdipoGen Life Sciences, San Diego, CA) were used as controls. In the case of the anti-ACE2 MAb, the cells were preincubated with escalating concentrations of the MAb for 30 min and then infected with the pseudovirus SARS-CoV-2. Following 24 h of incubation, the cells were washed with 100 ml of PBS and lysed with 50 ml of lysis buffer (Promega). Twenty-five microliters of the lysates was transferred to a white plate and mixed with the same volume of Nano-Glo luciferase reagent (Promega). The luciferase activity was measured immediately with the Tecan Spark. The percent inhibition by the peptides and the IC 50 (the half-maximal inhibitory concentration) values were calculated using GraphPad Prism 7.0 software (San Diego, CA). Additionally, to test the specificity of the double-stapled peptides, we evaluated their activity against pseudovirus VSV-G by following the infection protocol described above. Following 24 h of incubation, 25 ml of the lysates was transferred to a white plate and mixed with 50 ml of a luciferase assay reagent. The luciferase activity was immediately measured.
(ii) A549/ACE2 cells. For the evaluation of the antiviral activity in A549/ACE2 cells, aliquots of the pseudovirus SARS-CoV-2 at about 1,500 TCID 50 /well at an MOI of 0.1 were preincubated with escalating concentrations of the double-stapled peptides for 30 min. Next, 1 Â 10 4 cells were added to each well and incubated. A549/ACE2 cells cultured with medium with or without the SARS-CoV-2 pseudovirus were included as positive and negative controls, respectively. The linear peptide NYBSP-C, the doublestapled peptide StRIP16, and the anti-ACE2 MAb were used as controls. In the case of the anti-ACE2 MAb, the cells were preincubated with escalating concentrations of the Ab for 30 min and then infected with the pseudovirus SARS-CoV-2. Following 48 h of incubation, the cells were washed with PBS and lysed with 50 ml of lysis buffer. Twenty-five microliters of the cell lysates was processed as reported above to measure the luciferase activity and calculate the percent inhibition by the peptides and the Stapled Peptides as Inhibitors of SARS-CoV-2 IC 50 . Moreover, the double-stapled peptides were evaluated against pseudovirus VSV-G by following the infection protocol described above. Following 48 h of incubation, 25 ml of the lysates was transferred to a white plate and mixed with 50 ml of a luciferase assay reagent. The luciferase activity was immediately measured.
SARS-CoV-2 MN assay. The standard live virus-based microneutralization (MN) assay was used (38)(39)(40). Briefly, serially 2-fold-diluted and duplicate dilutions of individual peptides were incubated with 120 PFU of SARS-CoV-2 (US_WA-1/2020) at room temperature for 1 h before transferring into designated wells of confluent Vero E6 cells (ATCC, CRL-1586) grown in 96-well cell culture plates. Vero E6 cells cultured with medium with or without the same amount of virus were included as positive and negative controls, respectively. After incubation at 37°C for 3 to 4 days, individual wells were observed under the microscope to determine the virus-induced formation of cytopathic effect (CPE). The efficacy of individual drugs was expressed as the lowest concentration capable of completely preventing virus-induced CPE in 100% of the wells.
Evaluation of cytotoxicity. (i) HT1080/ACE2 cells. The evaluation of the cytotoxicity of the peptides in HT1080/ACE2 cells was performed in parallel with the antiviral activity assay and measured by using the colorimetric CellTiter 96 Aqueous One Solution cell proliferation assay {MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium]} (Promega, Madison, WI) following the manufacturer's instructions. Briefly, aliquots of 100 ml of the peptides at graded concentrations were incubated with 2 Â 10 4 /well HT1080/ACE2 cells and cultured at 37°C. Following 24 h of incubation, the MTS reagent was added to the cells and incubated for 4 h at 37°C. The absorbance was recorded at 490 nm. The percentage of cytotoxicity and the CC 50 (the concentration for 50% cytotoxicity) values were calculated as described above.
(ii) A549/ACE2 cells. For the cytotoxicity assay in A549/ACE2 cells, aliquots of escalating concentrations of the peptides were incubated with 1 Â 10 4 /well A549/ACE2 cells and cultured at 37°C. Following 48 h of incubation, the MTS reagent was added to the cells and incubated for 4 h at 37°C. The absorbance was recorded at 490 nm. The percentage of cytotoxicity and the CC 50 values were calculated as described above.
Conclusions. The current pandemic made it urgent to develop therapeutics against COVID-19. We took advantage of the structural knowledge of the binding site of SARS-CoV-2 RBD and the ACE2 receptor. We envisioned that a stapled peptide based on the helix region of ACE2 that binds with the SARS-CoV-2 RBD might work as a decoy receptor, bind SARS-CoV-2 preferentially, and prevent its entry into the host cells and thereby the subsequent infection by the virus. We reported the successful design of double-stapled peptides with potent antiviral activity in HT1080/ACE2 and human lung carcinoma cells, A549/ACE2. Most significantly, all three active stapled peptides with potent antiviral activity against SARS-CoV-2 pseudovirus showed high helical content (60 to 94%). We also evaluated the antiviral activity of the peptides by infecting Vero E6 cells with the replication-competent authentic SARS-CoV-2 (US_WA-1/2020). NYBSP-1, NYBSP-2, and NYBSP-4 completely (IC 100 ) inhibited virus-induced CPE at lowmicromolar doses. One of the most active stapled peptides, NYBSP-4, demonstrated a binding affinity (K D ) of 2.2 mM with SARS-CoV-2 RBD, which was in line with the antiviral activity of the stapled peptide against SARS-CoV-2 pseudovirus. NYBSP-4 also showed substantial resistance to degradation by proteolytic enzymes in human plasma. The lead stapled peptides are expected to pave the way for further optimization of a clinical candidate.