Engineered multivalent self-assembled binder protein against SARS-CoV-2 RBD.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a global pandemic since December 2019, and with it, a push for innovations in rapid testing and neutralizing antibody treatments in an effort to solve the spread and fatality of the disease. One such solution to both of these prevailing issues is targeting the interaction of SARS-CoV-2 spike receptor binding domain (RBD) with the human angiotensin-converting enzyme 2 (ACE2) receptor protein. Structural studies have shown that the N-terminal alpha-helix comprised of the first 23 residues of ACE2 plays an important role in this interaction. Where it is typical to design a binding domain to fit a target, we have engineered a protein that relies on multivalency rather than the sensitivity of a monomeric ligand to provide avidity to its target by fusing the N-terminal helix of ACE2 to the coiled-coil domain of the cartilage oligomeric matrix protein. The resulting ACE-MAP is able to bind to the SARS-CoV-2 RBD with improved binding affinity, is expressible in E. coli, and is thermally stable and relatively small (62 kDa). These properties suggest ACE-MAP and the MAP scaffold to be a promising route towards developing future diagnostics and therapeutics to SARS-CoV-2.

Introduction

The COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has, since December 2019, caused over 6 million deaths with over 500 million confirmed cases worldwide [1]. Great strides have been made through innovations in rapid testing and neutralizing antibody treatments in an effort to control the spread and fatality of the disease [2], [3], [4]. However, the lack of immediate widespread testing at the beginning of the pandemic has proven fatal [5]. The need for widely available therapies is also clear. If an infected person reaches the stage at which hospitalization is necessary, the COVID-19 patient faces a 21% death rate, more than five times greater than that of influenza [6]. Hospitalization has also been linked to higher viral titers in patients [7]. In terms of surveillance of the virus, testing of SARS-CoV-2 has relied on specialized instruments in addition to costly reagents and supplies for carrying out the reactions [8]. Lateral flow assays (LFAs) and enzyme-linked immunosorbent assays (ELISA), represent a point-of-care test for a simple, inexpensive, and fast diagnosis that predominantly relies on protein-protein interactions (PPIs) [2], [9].

During infection by SARS-CoV-2, the spike (S) protein on the virus surface recognizes the peptidase domain (PD) of the angiotensin-converting enzyme 2 (ACE2) of the host [10]. Structural studies reveal that the N-terminal alpha-helix (residues 1–23) of human ACE2 receptor is critical to binding the S1 receptor binding domain (RBD) of SARS-CoV-2 [11], [12], [13]. Recently engineered recombinant ACE2 exhibits an increased avidity to SARS-CoV-2 compared to the wild-type [11], [14], [15]. Recombinant ACE2 has also been proven to block early infection [16]. However, recent studies have also shown that isolation of this alpha-helix as an antibody or protein domain mimic (PDM) provides weak protein-protein interaction with the SARS-CoV-2 RBD [13], [17]. With this in mind, we have been inspired by nature’s approach to design an antibody against a specific target to create a unique platform for engineering a stable, high-affinity PDM.

In the lifetime of an antibody’s development, a disulfide bonded heavy and light chains self-assemble into a multivalent structure [18]. The multivalency of the protein provides a jump start in the race to increased affinity before undergoing its own evolutionary process, affinity maturation, to become a highly sensitive protein binder to a specific target [19]. Multivalency has proven to be an effective strategy to increase sensitivity and specificity of PPIs including multivalent displays of the SARS-CoV-2 RBDs and nanobodies against SARS-CoV-2 [20], [21], [22], [23], [24], [25]. Whereas previously designed PDMs against the SARS-CoV-2 have used a variety of design techniques to mature affinity [11], [15], [17], [26], we engineer high affinity via a multivalent protein design derived from a self-assembling protein domain ( Fig. 1a). Recently, multivalent protein fusions have been developed bearing the coiled-coil domain of the cartilage oligomeric matrix protein (C) – a protein our lab has studied extensively [27], [28] to create functional nanomaterials [29], [30], [31], [32], [33], [34]. Here we describe the results of an ACE2 mimicking multivalent assembled protein (ACE-MAP) where the N-terminal alpha-helix (ACEBINDER) is joined with a computationally designed kinked linker to C (Fig. 1b, Table A1) leading to high binding affinity of SARS-CoV-2 RBD for future applications as a biosensor or therapeutic.

Fig. 1

a) Schematic for the MAP design strategy used to generate a scaffold for the ACE2 protein linker using SARS-CoV-2 RBD as a target. b) Cartoon representation of computationally designed ACE-MAP. C, kink, linker and binder protein are shown in grey, orange, green and blue color respectively. The residues involved in binding to SARS-CoV-2 RBD, shown in red, are shown in stick representation and dark blue color. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Results and discussion

Computational design

Initial structure of C protein was taken from PDBID 3V2P and the 23-residue “binder protein” was taken from residues 21–44 of chain B in PDB 6M17, which was shown to have micromolar affinity to SARS-CoV-2 RBD [13]. In order to fuse the C and ACEBINDER (derived from the N-terminal α-helix of ACE2), a series of alpha-helix forming rigid linkers consisting of [EAAAK]n, (n = 2,3,4) motifs [35] were designed computationally using Rosetta [36] (data not shown). Initial monomeric structure of ACE-MAP was made in PyMOL [37] (Schrodinger). Using the symmetry information taken from PDBID 3V2P and Rosetta’s symmetric modeling protocol [38], a pentameric structure was produced and relaxed with the FastRelax protocol and REF2015 score function [39]. The first five residues in the linker region were manually designed to provide a kink at the junction of C domain and linker, which provided an opening for binding to multiple S-RBDs (Fig. 1b). To further improve the chance of ACE-MAP•S-RBD interaction, different lengths of the linker sequence were modeled to ensure that the residues involved in ACE2 binding (Q24, T27, K31, E35, D38, Y41) [40] were on the outer surface of ACE-MAP. A linker sequence with 20 residues was chosen that provides adequate length and bend for an exposed binding domain where increasing n = 2 repeats of [EAAAK] exhibited the best Rosetta score per protein length (Table A2).

Binding affinity to SARS-CoV-2 RBD by enzyme-linked immunoassay (ELISA)

The binding affinity of ACE-MAP was tested against SARS-CoV-2 RBD ( Fig. 2). ELISA protocol established by the Krammer Lab [41] and total binding saturation kinetics were used to determine the maximum saturation (Bmax) and binding affinities (Kd) values. ACE-MAP exhibited a Kd of 620 pM. As a negative control, no detectable binding was observed for the parent protein, C (Fig. A1). ELISA assay was performed for ACEBINDER using ACEBINDER to seed the plate and SARS-COV-2 RBD as a primary antibody to avoid the impact of a tag on the ACEBINDER affinity. Most notably, when applied without the MAP scaffold, ACEBINDER did not show any detectable binding activity (Fig. A2), confirming previously reported inactivity of similarly derived binder sequences from ACE2 [13], [26]. When compared to full length ACE2, which possessed a Kd of 11.7 nM, ACE-MAP was 19-fold better at binding SARS-CoV-2 RBD. This demonstrates that the presence of the fusion ACEBINDER to the multivalent scaffold is crucial for affinity to the target. Compared to previously designed binders to SARS-CoV-2 RBD ranging from 970 nM [26] to 100 pM [15], ACE-MAP was within the range and uniquely without further modification.

Fig. 2

Binding of ACE-MAP (blue) vs full length ACE2 (red) vs COMPcc (grey) as a function of SARS-CoV-2 RBD concentration measured by ELISA. Error bars represent the standard error of three independent trials. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Structure and thermostability

Structural studies of ACE-MAP and ACEBINDER were performed via circular dichroism spectroscopy. Wavelength scans performed at 25 ºC revealed a double-minima of − 17,000 ± 700 deg•cm2•dmol−1 208 nm and − 16,000 ± 600 deg•cm2•dmol−1 at 222 nm, indicative of helical conformation ( Fig. 3a).

Fig. 3

a) Wavelength scan of ACE-MAP performed at 25 ºC. Dark band – average MRE. Light shadow – ± standard deviation from average MRE inset: Representative ACE-MAP sample data of relative fraction folded using MRE at 222 nm wavelength from 25 ºC to 85 ºC b) Western blot from 12 % SDS-PAGE of ACE-MAP after BS3. Image Analysis reveals intensity is approximately 62.2 kDa molecular weight corresponding to a pentamer self-assembly.

Analysis via CONTIN illustrated secondary structure of 50.2 ± 2.0 % helical content (Table A3). Relative to the parent C, which was reported to possess 70% helicity [42], a loss in structure was observed due to the addition of the linker and ACEBINDER. Notably, the presence of the multivalent scaffold doubled the helical content when compared to ACEBINDER which revealed only 35% helicity (Fig. A3 , Table A4). To determine the stability of ACE-MAP, a temperature scan was carried out from 25 ºC to 85 ºC (Fig. 3a). While the parent C demonstrated a melting temperature of at 60 ºC [43], ACE-MAP revealed a slight increase in stability with a Tm of 64.18 ± 0.87 ºC relative to C. In contrast, ACEBINDER revealed a lower Tm of 55.2 ºC (Fig. A3).

To assess the extent of n-oligomerization due to the coiled-coil domain, BS3 crosslinking of ACE-MAP was employed. After running the sample on a 12% SDS-PAGE, the gel was subjected to western blot analysis and imaged (Fig. 3b). Corresponding analysis software was used to quantify the band position and purity revealing a single protein band at approximately 62 kDa indicating n-oligomerization of n = 5 (pentamer) only as compared to ACE-MAP without BS3 crosslinking (Figs. A4-A6).

The fusion of ACEBINDER by a kinked linker has resulted in a similarly thermostable protein relative to its C counterpart [42]. In comparison, C with all cysteine residues (C48 and C54) mutated to serines (denoted as CSS) reduces the melting temperature of CSS to 45 ºC [27]. Thus, despite the reduced helical secondary structure, likely due to the kinked region of ACE-MAP reducing the fraction of coiled-coil structure contribution, ACE-MAP maintains a Tm higher than CSS and a Tm more characteristic to wild-type C [42], [44]. Interchain disulfide-bond formation is a property only characteristic of a parallel n = 5 oligomer revealing that similarly ACE-MAP is likely to exist in the pentameric coiled-coil conformation [45], [46], [47]. Also of importance, is that the strong alpha-helical structure of C may lend itself as a scaffold that stabilizes the ACEBINDER region to become functional against the SARS-CoV-2 RBD. The secondary structural data shown here elucidates this picture where ACEBINDER possess just half the helical content of ACE-MAP. Previously, optimization of the domain’s helicity only creates a modest improvement in binding affinity [17]. In addition to the pentamerization after BS3 crosslinking, we deduce that the multivalency created by ACE-MAP through oligomerization of the C domain is critical for the increased binding affinity in comparison to ACEBINDER [13], [17] and ACE2 (Section 2.2 ). The increase in affinity and stability at high temperatures bodes well for future studies into applications of ACE-MAP as a stable biosensor or therapeutic.

Neutralization

ACE-MAP was tested for neutralization of SARS-CoV-2 B.1.1.7 (Delta variant) using a lentiviral pseudotyped assay ( Fig. 4, Tables A4–A7). Inhibitor vs. normalized response kinetics were used to determine the IC50 values. While higher concentrations of ACE-MAP and ACEBINDER exhibited increased infectivity, the high concentration of protein produced noise in the data. Thus, due to protein concentration limits and effects of protein oversaturation in the assay, ACE-MAP IC50 was projected from concentrations inhibiting > 50% infectivity of SARS-CoV-2 pseudotyped virus, which might produce some error in exact IC50 values. From the assay, while ACE-MAP demonstrated stronger inhibition of SARS-CoV-2 than ACEBINDER, it was less potent than ACE2. Direct comparison of the IC50 values revealed that ACE-MAP possessed a 16.7-fold improvement in neutralization activity to ACEBINDER with IC50 values of 1.05 μM and 18 μM, respectively (Fig. 4). In comparison, full-length ACE2 exhibited a much stronger ability to inhibit infectivity with an IC50 value of 1.8 nM, a 573-fold improvement over ACE-MAP.

Fig. 4

SARS-COV-2 pseudotyped lentiviral neutralization fitted with Inhibitor vs. Normalized Response Kinetics using Prism 7 (Graphpad) of ACE-MAP (blue) and ACEBINDER (green) and ACE2 (red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The large improvement of ACE-MAP to its parent protein, ACEBINDER, supports ACE-MAP as a scaffold towards improved protein-protein targeting. The relative response points towards relationship between the structural function of the RBD as compared to a virion. While ACE-MAP shows strong affinity towards the RBD in vitro, the weaker neutralization activity when compared to ACE2 indicates that the larger virion may provide additional challenges such as competing intermolecular interactions, inhomogenous spatial organization, environment-dependent thermodynamic stability [48] and glycosylation [49]. As has often been introduced in the development of effective PDM IC50s, improvements upon the original protein engineered design will be needed to generate an effective agent in vivo [50]. Despite this, ACE-MAP shows a strong improvement over its parent ACEBINDER in neutralization of SARS-CoV-2.

Described in this work is a coiled-coil self-assembled fusion protein capable of binding to SARS-CoV-2 RBD at picomolar affinity. Its characterization has revealed insights into the utility of multivalent alpha-helical binders for this and future coronavirus variants. Several proteins and antibodies have been recently generated to bind to SARS-CoV-2 RBD [3], [4], [15], [24], [26], [51], [52]. These proteins tend to use either mammalian expression systems and/or be > 140 kDa. The protein binders that have been computationally designed to increase the avidity to SARS-CoV-2 based on ACE2 possess Kd values ranging from 970 nM [26] to 100 pM [15].

E. coli expression systems have the unparalleled advantage of low cost, rapid growth, and good productivity; however, it is a host that is out of reach for many recombinant proteins > 60 kDa, and especially antibodies, that require post-translational modifications [53], [54], [55]. In our studies, ACE-MAP yields 74 ± 9 μg/mL (n = 3) from 400 mL cultures measured by BCA assay following purification and concentration. E. coli expression systems tend to yield functional Fab fragments in the range of 400–800 μg (in 400 mL culture) [56] using shake flasks indicating that our yield should be improved for a comparable advantage. Thus, expression of our construct needs to further be optimized to provide improved yields whereas E. coli expression systems may often be enhanced with tailored expression time, bioreactor design, temperature, and gene vectors [56], [57]. While ACE-MAP exhibits a strong affinity to SARS-CoV-2 RBD like antibodies, it possesses the advantage of soluble expression in E. coli, and a significantly smaller size – 12 kDa as a monomer (62 kDa as a pentamer).

Other small proteins or peptides being produced against SARS-CoV-2 RBD suffer from poor binding affinities to the SARS-CoV-2 RBD. Pentelute and coworkers have reported that a series of peptide binders designed against SARS-CoV-2 RBD range from 80 to 970 nM [26]. They have also previously reported that the same sequence used here, ACEBINDER, shows little to no binding affinity to SARS-CoV-2 [13] where we additionally demonstrate ACEBINDER to have low neutralization activity with an IC50 value of 18.0 μM (Fig. 4). Karoyan et al. have investigated and optimized the ACEBINDER sequence (to note they have included the first 27 residues) by retaining important residues and stabilizing its alpha helicity through iterative residue scanning [17]. As a result they have shown agreement that the native ACEBINDER sequence does not produce binding, while their optimized peptides produce up to approximately 3-fold improvement in binding affinity by Biolayer Interferometry compared to ACE2 [17]. Conversely, Baker and coworkers have reported several ‘minibinders’ that are capable of tight IC50 affinities; an increase in affinity is achieved by several rounds of mutagenesis using a ACE2 scaffolded design, followed by tethering another copy as dimer and trimer [15]. In a follow-up to these minibinders, Baker and coworkers report multivalent heterotrimers to develop escape resistance while maintaining impressive IC50 affinities [25]. Our approach fundamentally differs from such prior work as the tethering or multivalency is part of the design from the outset. ACE-MAP joins only a handful of other recombinant PDMs that bind to SARS-COV-2 RBD [11], [15], [17], [25], [26]. Moreover, the conservation of the ACE2 binding sequence in ACE-MAP may prove resilient against mutant escape variants and future coronaviruses.

Conclusions

We have biosynthesized ACE-MAP, a small binder protein against SARS-CoV-2 RBD using a fusion of ACEBINDER to C via a computationally designed kinked linker. ACE-MAP has increased thermostability compared to C [27] and increased binding affinity through multivalency compared to the ACEBINDER region as previously reported [13]. ACE-MAP furthermore demonstrates the utility of generating a PDM utilizing multivalency as the source of high affinity. The increased efficacy of the protein and thermostability at room temperature indicates the utility of simple multivalent fusions. The high avidity of ACE-MAP to SARS-CoV-2 RBD by utilizing its target receptor, ACE2, reveals that ACE-MAP may be useful in a variety of biomedical applications such as the development of biosensors and therapeutics especially among the growing concern for SARS-CoV-2 mutant escape.

Methods and materials

Materials

Chemically competent AFIQ E. coli cells were gifted from David Tirrell at California Institute of Technology. ACE-MAP/pQE30 plasmid was cloned and purchased from Eurofins. Bacto-tryptone, sodium chloride, yeast extract, tryptic soy agar, ampicillin, chloramphenicol, sodium phosphate dibasic anhydrous (Na2HPO4), sodium hydroxide (NaOH), dextrose monohydrate (D-glucose), magnesium sulfate, calcium chloride (CaCl2), manganese chloride tetrahydrate (MnCl20.4H2O), cobaltous chloride hexahydrate (CoCl20.6H2O), isopropyl β-D-1-thiogalactopyranoside (IPTG), Pierce bicinchoninic acid (BCA) assay kit, Pierce snakeskin dialysis tubing 3.5 K MWCO, sodium dodecyl sulfate, Pierce C18 tips with 10 µL bed, bissulfosuccinimidyl suberrate (BS3), ascorbic acid, Immulon 4 HBX ninety-six well plates, Nunc ninety-six well plates, Dulbecco’s Modified Eagle medium (DMEM), Nunc EasYFlask Cell Culture Flasks, Quant-iT PicoGreen dsDNA Assay Kit, IL-6 Mouse ELISA Kit, Pierce High Capacity Endotoxin Removal Spin Columns, and ELISA wash buffer (30X) were acquired from Thermo Fisher Scientific. The twenty naturally occurring amino acids, thiamine hydrochloride (vitamin B), dimethylsulfoxide (DMSO), and 3,3′,5,5′-tetramethylbenzidine (TMB) were purchased from Sigma Aldrich. Hydrochloric acid (HCl), Coomassie® Brilliant Blue G-250, and milk powder (non-fat, skimmed) were purchased from VWR. HiTrap Q HP 5 mL columns for protein purification were purchased from GE Healthcare Life Sciences. Macrosep and Microsep Advance Centrifugal Devices 3 K molecular weight cutoff (MWCO) and 0.2 µm syringe filters were purchased from PALL. Acrylamide/bis solution (30 %) 29:1, Mini Trans-Blot filer paper, Trans-Blot Transfer Medium (nitrocellulose membrane), and natural polypeptide sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) standard were purchased from Bio-Rad, and Dulbecco’s phosphate buffered saline were purchased from ATCC.

Methods

ACE-MAP expression and purification

ACE-MAP protein was expressed in phenylalanine auxotrophic AFIQ E. coli cells. pQE30/ACE-MAP plasmid was transformed via heat shock in chemically competent AFIQ cells. Transformed cells were grown for 14–16 h at 37ºC on tryptic soy agar plates containing 200 µg/mL ampicillin and 35 µg/mL chloramphenicol. A single colony was inoculated in 16 mL supplemented M9 minimal medium (0.5 M Na2HPO4, 0.22 M KH2PO4, 0.08 M NaCl, and 0.18 M NH4Cl) containing all 20 natural amino acids (100 µg/mL), ampicillin (200 µg/mL), chloramphenicol (35 µg/mL), vitamin B (35 µg/mL), D-glucose (100 µg/mL), magnesium sulfate (1 mM), calcium chloride (0.1 mM), and trace metals (0.02% v/v) and incubated at 37ºC and 350 rpm for 16 h. Following, 8 mL of the starter culture was added to 200 mL of supplemented M9 medium and incubated at 37ºC and 350 rpm until the optical density at 600 nm (OD600) reached 0.7. Protein expression was induced with 200 µg/mL IPTG and incubated at 37ºC and 350 rpm for 3 h. After the expression, cells were harvested by centrifugation at 5000 × g at 4 ºC for 20 min in an Avanti J-25 centrifuge (Beckman Coulter) and stored at − 20ºC until purification. Expression of ACE-MAP was confirmed via 12% SDS-PAGE (Fig. A4).

Cell pellets were thawed and resuspended in Buffer A (50 mM Na2HPO4, 250 mM NaCl, 6 M urea, pH 8.0). Cells were lysed via Q500 probe sonicator (QSonica) at 65% amplitude, pulse on for 5 s and off for 30 s for a total of 2 min. The lysed cells were centrifuged at 11,000 × g for 45 min at 4 ºC to remove cell debris (Beckman Coulter). The supernatant was removed and purified using a syringe-pump driven IMAC Q Sepharose high performance 5 mL column (HiTrap Q HP 5, GE Health Sciences) charged with CoCl2. Protein was eluted from the column using a gradient (0–100 %) of Buffer B (50 mM Na2HPO4, 250 mM NaCl, 6 M urea, 500 mM imidazole, pH 8.0) (Fig. A5). Elutions with pure protein were removed and dialyzed using a 3.5 kDa MWCO tubing at 4 ºC. Dialysis was performed using a step-wise decrease of urea from (three buckets from 3 M to 0.75 M urea) succeeded by six buckets with 0 M urea. The protein elutions were then concentrated to 1.5 mL using 3 kDa MWCO Macrosep and Microsep Advance centrifugal devices (Pall Corporation) at 2000 × g. Following, 500 µL volumes were injected into a Fast Purification Liquid Chromatography (FPLC, AKTA pure, GE Healthcare) using a Superdex 75 10/300 GL Size Exclusion Chromatography (SEC) column (GE Healthcare). Protein was eluted using phosphate buffered saline (PBS) pH 7.4. Pure fractions were determined using a 12% SDS-PAGE (Fig. A6) and protein concentration was determined using a bicinchoninic acid (BCA) assay with a standard curve based on bovine serum albumin concentrations.

SARS-CoV-2 RBD expression and purification

The plasmid used for protein expression and purification of SARS-CoV-2 RBD was constructed by insertion of a secretion signal, the coding sequence of a 5 A tag, RBD, and a 6xHis tag into an expression vector pVRC8400 (kindly provided by the Vaccine Research Center, National institute of Health). The gene construct was codon optimized for mammalian cell expression and synthesized by GenScript. The plasmid was transiently transfected into HEK293S cells for 5 days. Cell supernatants were filtered through 0.22-µm filters, loaded onto Ni-nitrilotriacetic acid (NTA) beads, and proteins were eluted with 600 mM imidazole. The elution was then dialyzed in PBS, flash frozen and stored at − 80 ºC.

Circular dichroism

Secondary structure of ACE-MAP and ACEBINDER was measured using the Jasco J-815 CD spectrometer with a PTC-423S single position Peltier temperature control system. Wavelength scans of ACE-MAP (10 µM) were performed from 195 to 250 nm at 1 nm step sizes. ACEBINDER was first reconstituted in DMSO prior to dilution in PBS. SEC was used to remove remaining DMSO prior to CD. Temperature scans were performed from 25 ºC to 85 ºC at 1 ºC step sizes. The mean residue ellipticity (MRE) and melting temperature (Tm) – using 222 nm for ACE-MAP and 214 nm for ACEBINDER – were calculated as described in previous studies [27]. The secondary structure content (α-helicity, β-content, and unordered structure) was predicted with CONTIN/LL software [58], [59], [60].

Chemical crosslinking

Addition of 3 mM bis(sulfosuccinimidyl) suberate (BS3) to a 10 μM concentration of ACE-MAP in PBS pH 7.4 was performed for chemical crosslinking to assess oligomerization. The reaction was allowed to incubate at room temperature and 300 rpm for 3 h in the dark on an Eppendorf Thermomixer C. The reaction was then quenched using 25 mM Tris HCl at pH 7.5 and sampled into a 12 % SDS-PAGE and oligomerization was confirmed using Western Blot analysis at a final concentration of 5 μM next to 5 µL of BioRad Precision Plus Protein Standard. To detect and analyze the oligomer bands, Amersham Imager 680 and corresponding analysis software (Cytiva Life Sciences) were used to detect and analyze relative intensities of oligomer bands.

Enzyme Linked Immunoassay

The ELISA protocol was adapted from previously established protocols [61]. A ninety-six well plate was coated overnight at 4 ℃ with 50 µL per well of a 2 μg/mL solution of SARS-CoV-2 RBD for ACE-MAP assay and with ACEBINDER peptide for ACEBINDER assay. ACEBINDER was first reconstituted in DMSO prior to dilution in PBS. SEC was used to remove remaining DMSO prior to ELISA. The next morning, the coating solution was removed and 100 µL per well of 3 % non-fat milk prepared in PBS with 0.1 % Tween 20 (TPBS) was added to the plate at room temperature (RT) for 2 h as blocking solution. The blocking solution was removed and 100 µL of serial dilutions of ACE-MAP, C, and ACE2 proteins were added to plates coated with SARS-CoV-2 RBD for 2 h at RT. Serial dilutions of SARS-CoV-2 RBD were substituted as the primary antibody for plates coated with ACEBINDER. The plates were washed three times at 200 µL volumes using 0.1 % TPBS. Approximately 100 µL of a 1:3000 anti-Histag horseradish peroxidase (HRP) conjugated secondary antibody (Sino Biological) was prepared in 0.1 % TPBS and added to each well for 1 h. Plates were then washed three times with 200 µL of 1x ELISA wash buffer (Thermo Scientific) and then air dried in a hood. While drying, TMB solution was prepared as described by Sigma Aldrich protocol. 1 mg/mL TMB was prepared in DMSO and then added to 0.05 mM citrate-phosphate buffer with 0.01 % hydrogen peroxide (0.01 %). Once completely dry, 100 µL of 3,3′,5,5′-tetramethylbenzidine (TMB) solution was added to the plates for 10 min. To quench the reaction, 50 µL of 3 M HCl was added to the plates. Absorbance at OD450 was immediately read using a microplate reader (BioTek Synergy H1).

Neutralization assay

ACE-MAP, ACE2, and ACEBINDER were tested against SARS-CoV-2 using the previously established pseudotyped lentiviral neutralization assay [62] which is employed with ACE 0.293 T cells. Cells were plated in a 96 well tissue culture dish a 1 × 104 cells/well. ACE-MAP and ACEBINDER was serially diluted and incubated with pseudotyped virus for 30 min before being added to the target cells. Cells were infected with pseudotyped SARS-Cov-2 and normalized for activity at MOI= 0.2. The next day, culture medium was removed and Nano-Glo luciferase substrate (Promega) was added before reading on an Envision 2103 microplate luminometer (PerkinElmer). Luciferase activity was normalized to the untreated cells and IC50 values were be calculated by fitting a sigmoidal nonlinear regression fit using inhibitor concentration vs. normalized response – Variable slope equation against concentration of protein using GraphPad Prism.

Funding sources

This work was supported by the award numbers IIP-2031364, IIP- 1644681 and award number W911NF-19-1-0150.

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.