Targeting SARS-CoV-2 and host cell receptor interactions

Abstract

Despite the availability of vaccines and therapeutics, continual genetic alterations render the severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) a persistent threat, particularly for the immunocompromised and elderly. Through interactions of its spike (S) protein with different receptors and coreceptors on host cell surfaces, the virus enters the cell either via fusion with the plasma membrane or through endocytosis. Angiotensin-converting enzyme 2 (ACE2) has been identified as a key receptor utilized by SARS-CoV-2 and related human coronaviruses to mediate cell entry in the lung airways. Auxiliary SARS-CoV-2 entry receptors such as ASGPR1, Kremen protein 1, integrins have also been reported. In this review, therapeutic approaches to block SARS-CoV-2 and host cell receptor interactions are discussed.

Abbreviations

3CL

3C-like;

ACE2

Angiotensin-converting enzyme 2

ADAM17

A disintegrin and metalloprotease 17

ASGR1

asialoglycoprotein receptor 1

CCR5

C–C chemokine receptor type 5

CD

cluster of differentiation

CLEC4G

C-Type Lectin Domain Family 4 Member G

COVID-19

Coronavirus disease 2019

DSMB

Data and Safety Monitoring Board;

FDA

Food & Drug Administration

gp

glycoprotein

KREMEN1

Kringle Containing Transmembrane Protein 1

mAb

monoclonal antibody

MOA

mechanism of action

NRP-1

Neuropilin-1

NTCP

sodium taurocholate co-transporting polypeptide;

RBD

receptor binding domain

S

Spike;

SARS-CoV-1, -2

severe acute respiratory syndrome coronavirus 1, 2

Siglec-9

Sialic acid-binding Immunoglobulin-like lectin-9

SRB1

scavenger receptor class B member 1

VOCs

variants of concern

WHO

World Health Organization

1. Introduction

As of November 25, 2022, WHO has reported that the 2019 novel coronavirus disease (COVID-19) caused by the Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has infected 636,440,663 people in 228 countries and territories and caused 6,606,624 deaths (WHO, 2022a). Fast-track development and emergency use authorization (EUA) of several vaccines of novel platforms (Khan et al., 2021) greatly reduced global morbidity and mortality rates. Globally, 12.96 billion vaccine doses have since been administered with 5.45 billion persons having received at least one dose and almost 5 billion people fully vaccinated (WHO, 2022b). For people who become infected, vaccination protected against severe illness, hospitalization, and death (Fiolet et al., 2022; Watson et al., 2022). People who recovered from COVID-19 infections still benefited from subsequent vaccinations. Studies have shown that vaccines prevented 40–65% of symptomatic cases (Cerqueira-Silva et al., 2022). Nevertheless, persistent viral immune escape generated waves of SARS-CoV-2 variants (Akkiz et al., 2022) that are more virulent (Delta strain) or transmissible (Omicron strain), lowering the effectiveness of these vaccines (Accorsi et al., 2022; Andrews et al., 2022; Thompson et al., 2022; Zhou et al., 2022). Implementation of repeat booster vaccination programs, particular within relatively short time-frames have raised questions on long-term immunological implications of repeated vaccines (El-Menyar et al., 2022; Fiolet et al., 2022) and prompted efforts to develop vaccines with broader and long-lasting protective effects (Li, 2022; Hussain et al., 2021; Li et al., 2021).

2. Direct antivirals available for COVID-19 treatment

Repurposing drugs successfully accelerated the development of direct anti-virals currently used in the treatment of SARS-CoV-2 infections. They comprise two SARS-CoV-2 polymerase nucleoside inhibitors, Veklury (Remdesivir, originally developed for Ebola virus; Santoro and Carafoli, 2021) and Lagevrio (Molnupiravir; originally developed for HCV and influenza; Tian et al., 2022), as well as one SARS-CoV-2 3CL protease inhibitor, Paxlovid (Nirmatrelvir, derived from a peptidic inhibitor against SARS-CoV-1 co-administered in combination with Ritonavir; Owen et al., 2021). These drugs are typically given as stand-alone drugs and for short dosing durations of 3–5 days for treatment of non-hospitalised patients with mild-to-moderate COVID-19, with the exception of Remdesivir which is also administered to hospitalised COVID-19 patients. Of note, treatments with these drugs come with some limitations. Remdesivir is given intravenously and must be administered only in a hospital or in a healthcare setting. Ritonavir-boosted Nirmatrelvir is not allowed for Covid-19 patients with severe renal or hepatic impairment, hence limiting the treatable patient pool. CDC (USA) has recommended that Molnupiravir be used as a second-line treatment for non-hospitalised patients only when Nirmatrelvir and Remdesivir are not available, feasible to use, or clinically appropriate and not for pregnant patients due to safety concerns (NIH, 2022a).

Furthermore, COVID-19 rebounds have been reported after cessation of Nirmatrelvir and Molnupiravir treatment, especially in patients with underlying medical conditions (Parums, 2022; Wang, 2022). Rates of COVID-19 rebound increased for both drugs with the time after treatments. The mechanisms of rebound COVID-19 remain uncertain but may be due to inadequate viral clearance after treatment, development of drug resistant virus strains, or insufficient drug dosing. Given the propensity of SARS-CoV-2 and other pathogenic CoVs to undergo frequent genetic mutations, it is highly plausible that future SARS-CoV-2 variants or pandemic CoV strains will be less susceptible to these drugs. Thus, there is a need to develop new therapeutics that can be deployed under these conditions. Notably, an oral non-covalent, non-peptidic SARS-CoV-2 3CL protease inhibitor, Xocova (Ensitrelvir), has obtained emergency approval from the Japan's Ministry of Health, Labor and Welfare in November 2022, for the treatment of mild-to-moderate COVID-19 disease (Shionogi, 2022). Unlike Nirmatrelvir, Ensitrelvir was discovered via virtual screening followed by structure-based drug design (Unoh et al., 2022).

3. CoV entry receptors and pathways

Members of Sarbecoviruses from the betacoronavirus genus such as SARS-CoV-1 (Li et al., 2003), −2 (Lan et al., 2020; Shang et al., 2020; Wang et al., 2020), some civet and bat CoVs (Xiong et al., 2022; Millet et al., 2021; Li et al., 2020a; Zhou et al., 2020) as well as the alpha CoV, HCoV-NL63 (Hofmann et al., 2005), utilize the ACE2 as its obligate receptor to enter cells (Fig. 1 ; Saadat et al., 2021). Two entry pathways through the cell surface or endosomes have been described. Both entry pathways involve sequential engagement of two subunits of S, the distal S1 subunit and the S2 subunit which is anchored on the viral lipid membrane. S1 subunit binds ACE2 receptor via its receptor binding domain (RBD) and triggers conformational changes that exposes an internal “S2′ site” that is susceptible to host proteolytic cleavage. Cleavage of S2′ site induces extensive irreversible conformational changes (“activation”) that exposes the fusion peptide in the S2 subunit, leading to fusion with the cell membrane. Depending on the entry pathway, S2’ site cleavage may occur at the plasma membrane by the serine protease TMPRSS2 or within the cell by lysosomal proteases such as Cathepsin L (reviewed in Almaghaslah et al., 2020; Hu et al., 2021).

Fig. 1.

The role of human host proteases on SARS-CoV-2 entry. Virus entry through (A) endosomal pathway and (B) TMPRSS2 and furin (Saadat et al., 2021).

Blocking TMPRSS2 activity has been shown to prevent spread and pathogenicity of SARS-CoV-2 (Meng et al., 2022; Hoffmann et al., 2020a). Unlike SARS-CoV-1, SARS-CoV-2 S protein contains a multi-basic sequence at its S1/S2 junction which is cleaved by intracellular furin during S maturation (Coutard et al., 2020), converting it into a fusion competent form (“priming”) and enhances SARS-CoV-2 cell entry compared to SARS-CoV-1 (Mykytyn et al., 2021). Cleavage of this site has been shown to be essential for S-protein-mediated cell-cell fusion and entry into human lung cells (Johnson et al., 2021; Hoffmann et al., 2020b). S1/S2 priming occurs at the trans-Golgi network of infected cells during virus assembly and secretion and is a prerequisite for subsequent TMPRSS2-mediated activation at the S2′ site, but not for S2′ activation by cathepsin L in TMPRSS2-negative cells (Baggen et al., 2021; Tang et al., 2021). In intestinal organoids, SARS-CoV-2 entry does not require the endosomal Cathepsin B/L proteases, but specifically depends on TMPRSS2 (Beumer et al., 2021). Furthermore, it has been demonstrated that suboptimal S1/S2 cleavage and usage of TMPRSS2 drives Omicron strain towards use of the endocytic pathway for cell entry and a shift in cellular tropism away from TMPRSS2-expressing cells (Meng et al., 2022).

It is highly likely that future pandemic CoVs would similarly utilize these pathways to mediate cell entry. Additional co-receptors for SARS-CoV-2 attachment (e.g. heparan sulfate, CD209, CD299, SRB1, Siglec-9) or entry (e.g. ASGR1, KREMEN1, NRP-1, CD147, CLEC4G) have been reported (reviewed in Everest et al., 2022; Baggen et al., 2021; Saadat et al., 2021 ). These receptors may cooperate with ACE2 to facilitate SARS-CoV-2 cell entry and also expand the tropism of SARS-CoV-2 beyond ACE2-expressing cells (from the lungs, small intestine, testis, kidneys, heart (Li et al., 2020b) and gastro-intestinal tract (Zhang et al., 2020)).

4. Rationale for targeting SARS-CoV-2 host cell receptor interactions

Virus-host cell receptor interaction is requisite step for virus cell entry and the viral life cycle. Preventing this process effectively blocks virus infection and spread in affected tissues. Inhibitors that block virus-receptor binding, in combination with drugs of different MOA (such as viral polymerase or 3CL protease inhibitors) can augment therapeutic effects and prevent emergence of drug-resistant virus. Combination drug therapies that simultaneously inhibit multiple stages in the virus replication cycle have been highly successfully in the treatment of HIV-1 (Menendez-Arias and Delgado, 2022) and HCV (Sacks, 2017). In addition, marketed drugs that block viral entry are exemplified by small molecule inhibitors, Selzentry (Maraviroc; Yost et al., 2009), Rukobia (Fostemsavir; Chahine, 2021) which prevent HIV-1 gp120 envelope protein interaction with its coreceptor, CCR5 or its attachment receptor, CD4, respectively. As well, two peptidic virus drugs that have been developed are Fuzeon (Enfuvirtide; Lalezari et al., 2003), a 36-amino acid peptide that binds to HIV-1 gp41 and blocks HIV-1 cell-fusion and Hepcludex (Bulevirtide), a 47-aa peptide of the HBVpreS1 antigen which blocks HBV/HDV interaction with its liver-specific receptor, NTCP. Hepcludex received conditional marketing authorization in the EU for treatment of chronic HDV infection and is undergoing phase III trials in the US (Yardeni et al., 2021).

5. Approved therapeutics targeting SARS-CoV-2 entry

Since the start of the SARS-CoV-2 pandemic, convalescent plasma therapy along with several single and combination mAb therapeutics together make up the largest class of antivirals that have received EUA for treatment of SARS-CoV-2 patients (Choudhuri et al., 2022; Kumari et al., 2022; Hurt and Wheatley, 2021). Most SARS-CoV-2-neutralizing Abs derived from COVID-19 convalescent people either bind the surface-exposed Spike RBD or N-terminal domain (NTD). Within the RBD, residues aa 438–506 (the receptor binding motif) form contacts with the human ACE2 receptor (Lan et al., 2020). Thus far, all authorized mAbs bind to the RBD. Despite their initial success, clinical benefits of mAbs declined over time as evolving Spike protein mutations led to the emergence of immune escape SARS-CoV-2 variants (Akkız, 2022; Cao et al., 2022).

Amongst the antibodies approved for use under EUA in the US (Table 1 ; Kelley et al., 2022), Bamlanivimab/Etesevimab, REGEN-COV (Casirivimab and Imdevimab) and Xevudy (Sotrovimab), are no longer recommended by FDA for use as they do not work against SARS-CoV-2 Omicron variant (BA.1, FDA, 2022a). Concurrent global circulation of SARS-CoV-2 Omicron subvariants further challenges the effectiveness of using neutralizing antibodies for treatment. According to NIH, COVID-19 treatment guideline updates in October and November 2022, the last-line therapeutic mAb, bebtelovimab, is not likely to work against Omicron subvariants BQ.1 or BQ1.1, whilst the prophylactic antibody cocktail, Evusheld (Tixagevimab/Cilgavimab), is likely to be inefficacious against multiple Omicron variants (BA.1, BA.1.1, BA.2, BA.5, BQ.1, BQ.1.1, BA.4.6, BF.7, BA.2.75.2; FDA, 2022c, d; NIH, 2022a). Like Molnupiravir, Bebtelovimab is recommended as an alternative option, but only when the preferred therapies (Nirmatrelvir and Remdesivir) are not available, feasible to use, or clinically appropriate (NIH, 2022a). Beyond therapeutic mAbs, CoV entry inhibitors of other modalities have yet to progress beyond clinical trials.

Table 1.

COVID-19 mAb products authorized for emergency use in USA, for mild-to-moderate COVID-19, at high risk for progression to severe COVID-19, including hospitalization or death.

Monoclonal antibody	Other name	Company	EUA (FDA)	Monotherapy or cocktail	Route of administration	Dosing	Efficacy	Reference
Casirivimab	REGEN-COV	Regeneron and Roche	Nov-2020	Cocktail	IV infusion	1200 mg	markedly reduced activity against Omicron variant	FDA (2022a)
Imdevimab					SC injection
Bamlanivimab	–	AbCellera and Lilly	Feb-2021	Cocktail	IV infusion	2100 mg	markedly reduced activity against Omicron variant	FDA (2022a)
Etesevimab
Sotrovimab	Xevudy	Vir and GlaxoSmithKline	May-2021	Monotherapy	IV infusion	500 mg	markedly reduced activity against Omicron variant	FDA (2022a)
					IM pending
Tixagevimab	Evusheld	AstraZeneca	Dec-2021	Cocktail	IM injection	600 mg	likely resistant Omicron subvariants: BA.2.75.2, BA.4.6, BF.7, BQ.1, and BQ.1.1	FDA (2022a, c); NIH (2022a)
Cilgavimab
bebtelovimab	–	AbCellera and Lilly	Feb-2022	Monotherapy	IM injection	175 mg	likely resistant Omicron subvariants: BQ.1 and BQ.1.1	FDA (2022a, b).

6. SARS-CoV-2 entry inhibitors in clinical and preclinical testing

6.1. Therapeutic mAb

As of November 2022, over 30 Abs are listed as having completed or still undergoing clinical trials (Antibodysociety.org, accessed November 2022). Nevertheless, due to the dynamic antigenic shifts in SARS-COV-2 strains and subvariants, and the dominance of the Omicron strain from November 2021 (Chavda et al., 2022), some setbacks have been encountered. Most trials are designed to check antibody efficacy in patients with mild or moderate COVID-19 and at high risk of disease progression or hospitalization, with a few early trials designed to assess mAbs efficacy in hospitalised COVID-19 patients.

Unfortunately, a number of the therapeutic antibodies in late stage clinical trials (Table 2 ) work well against earlier SARS-COV-2 variants, but are less capable of fully neutralizing Omicron strains and its subvariants (BA.2 and its sub-lineages; Bruel et al., 2022; Yamasoba et al., 2022). As well, the number of patients infected with non-Omicron (sub)variants have dwindled. Finally, milder symptoms resulting from Omicron infections (Menni et al., 2022) have resulted in fewer patients being available for trial enrolment. For instance, the independent Data and Safety Monitoring Board (DSMB) terminated the phase II/III trials for mAb, ADG20, after determining that continued participation by study subjects would not yield any additional beneficial safety information as it does not work well against Omicron BA.2 strain and its sublineages. Likewise, DSMB also recommended that phase III study for mAb, SAB-185, be stopped due to low Omicron-related COVID-19 hospitalization and death rates that made the study design statistically unworkable (SAB Therapeutics, 2022).

Table 2.

Status of therapeutic mAbs in clinical testing and their efficacy against SARS-COV-2 Omicron.

Drug Name	Ab details	Global Status	Company	Delivery Route	Omicron Inhibitory Activity	Clinical trials ID/Status
Adintrevimab (ADG20)	half-life extended mAb	Phase II/III	Invivyd, Inc. (Adagio Therapeutics), USA	Injectable, I.M.	Not active against Omicron BA.2 strain and its sublineages	NCT04805671: terminated; Active, not recruiting; NCT04859517 (ph II/III): terminated.
C-135-LS and C144-LS (BMS 986414 and BMS-986413)	from infected patient	Phase III	Rockefeller University/Bristol-Myers Squibb, USA	Injectable, S.C.	Not active against Omicron	NCT04518410 (ACTIV-2): Active, not recruiting
COR-101 (BI 767551, STE90-C11)	fully-human monoclonal antibody	Phase III	Corat Therapeutics, Germany/Boehringer Ingelheim, Germany	Injectable, I.V.	Not active against Omicron	NCT04894474 (prophylactic treatment): Withdrawn (project terminated. NCT04822701: Terminated (not due to safety reasons).
COVID-19 hyperimmune globulin	from convalescent plasma	Phase III	Grifols, Spain	Injectable, I.V.	unknown	NCT04546581 (INSIGHT 013): did not meet its primary endpoints with statistically significant results.
MAD-0004J08	Fc-engineered mAb identified from COVID-19 convalescent plasma	Phase III	AchilleS Vaccines, Italy	Injectable, I.M.	reduced potency against Omicron & subvariants BA.1, BA.1.1 and BA.2	NCT04952805: Active, not recruiting
SAB-185 (CSL-451)	human Immunoglobulin polyclonal antibodies generated from cows using SAB's proprietary DiversitAb platform	Phase III	SAB Biotherapeutics, USA/CSL Limited, Australia	Injectable, I.V.	Active against Omicron	NCT04518410: Review board recommended that the study be stopped.
SCTA-01 (HB27)	recombinant humanized anti-SARS-CoV-2 spike protein monoclonal antibody,	Phase III	Sinocelltech, China	Injectable, I.V.	Not active against Omicron	NCT04644185 (Hospitalised Patients With Severe COVID-19): recruiting NCT04683328 (Patients Admitted to High Dependence or Intensive Care): Not yet recruiting
XAV-19	mixture of swine glyco-humanized antibodies	Phase III	Xenothera, France	Injectable, I.V.	Active against Omicron	NCT04928430: recruiting
FBR-002	equine F(ab’)2 polyclonal antibody	Phase IIa	Fab'entech, France	Injectable, I.V.	Active against Omicron	NCT05279352 (Patients Hospitalised With COVID-19 in Need of Supplemental Oxygen and at Risk of Severe Outcome): recruiting
Plutavimab (STI-9199, STI-2099)	next-generation gene-encoded neutralizing antibody (encoded into a gene for delivery utilizing SmartPharm's non-viral nanoparticle platform).	Phase II	Sorrento Therapeutics, USA	Inhaled, transnasal; Injectable, I.V., I.M.	Active against Omicron and BA.1.1	NCT05372783: Active, not recruiting; NCT04900428: Active, not recruiting;
ABBV 2B04 and ABBV-47D11	mAbs	Phase I	AbbVie, USA	Injectable, I.V.	reduced potency against Omicron	NCT04644120: recruitment completed
CT-P63	variant-customized cocktail mAb (No. 32 antibody) targeted against RBD	Phase I	Celltrion, South Korea	Injectable; I.V.	Active against Omicron	NCT05017168: recruitment completed
DMAbs AZD5396 and AZD8076 with Hylenex® Recombinant	DNA-based anti SARS-CoV-2 mAb, using its Cellectra delivery system	Phase I	AstraZeneca, UK, Inovio, USA	Injectable, I.M.	n.a.	NCT05293249: recruiting
DXP-604 (BGB-DXP 604)	human Ab	Phase I	BeiGene, China, Singlomics Biopharmaceuticals, China	Injectable; I.V.	reduced potency against Omicron	NCT04669262: recruitment completed
HFB-30132A (ABL 901)	Human Ab, binds to S with sub-nM affinity	Phase I	HiFiBiO Therapeutics, France, ABL Bio, S. Korea, Pharmsynthez Russian Federation	Injectable; I.V.	n.a.	NCT04590430: recruitment completed
IBIO-123	cocktail of 3 neutralizing IgG monoclonal antibodies	Phase I/II	Immune Biosolutions, Canada	Inhaled	Active against Omicron	NCT05298813:recruiting
JS-026	recombinant fully human mAb targets S1 subunit	Phase I	Shanghai Junshi Biosciences China	Injectable; I.V.	n.a.	NCT05167279: Active, not recruiting
LY-CovMab (BA CovMab)	fully human mAb specifically binds to RBD with high affinity	Phase I	Luye Pharma Group, Hong Kong, S.A.R., China	Injectable; I.V.	n.a.	NCT04973735: Active, not recruiting
mAbs against COVID-19 omicron variant, B.1.1.529	multiple next generation Ab specifically targeting omicron (B.1.1.529) variant	Phase I	Regeneron, USA	Injectable	Active against Omicron BA.1 and BA.2	n.a.
MTX-COVAB (COVAB 36)	mAbs	Phase I	Memo Therapeutics, Switzerland	Injectable; I.V., inhaled	n.a.	NCT05351437: recruitment completed
NP-028	human polyclonal hyperimmune with Abs (COVID-HIG)	Phase I	Emergent BioSolutions, USA	Injectable; I.M., IV, S.C.	n.a.	NCT04661839: completed
SARS-CoV-2 equine antiserum Ig	equine antiserum Ig (purified F(ab)2 fragment)	Phase I	Biological E, India	Injectable; I.V.	n.a.	n.a.
SPK001; SPKM 001	long-acting Ab	Phase I	SpikImm, France	Injectable, I.M.	Active against Omicron BA.1 and BA.2	n.a.

I.V., intravenous; I.M., intramuscular; S.C., subcutaneous; n.a., information not available.

These conditions have prompted the development of new therapeutic antibodies with broad-spectrum activity against multiple SARS-CoV-2 variants, especially against the Omicron subvariants. Abs that recognize RBD epitopes that are conserved among SARS-CoV-2 variants and other Sarbecoviruses are anticipated to be more effective (Cao et al., 2022; Cameroni et al., 2022) and could be relevant for future CoV outbreaks. A good example is NVD200, a combination of two monoclonal antibodies with in vitro neutralizing activity against Omicron variants BA.1, BA.2, BA.4, BA.5, and BA.2.75, pre-Omicron variants of concern (VOCs) and the more divergent SARS-CoV-1 (Invivyd, 2022). Ueno et al. (2022) also reported that a cocktail of antibodies, EV053273 and EV053286, isolated from PBMCs of convalescent COVID-19 patients had neutralizing activity to a broad spectrum of SARS-CoV-2 viruses including the Omicron variants. In addition, Fang et al. (2022) reported that a mAb, SW186, isolated from Spike immunized mice, had similar broad spectrum neutralizing activities and binds to a conserved epitope on the outer surface of RBD. Nevertheless, in both reports, lower potencies for Omicron versus earlier variants was observed. Interestingly, Yuan et al. (2022) generated a human-derived IgG-like bispecific antibody (bsAb), Bi-Nab_35B5-47D10, that simultaneously bound to the two distinct epitopes on RBD and S2. This Ab showed broad inhibition of SARS-CoV-2 wildtype and all VOCs, including Delta (B.1.617.2), Omicron BA.1, and BA.2. Readers are encouraged to refer to excellent reviews by Strohl et al. (2022), Zhang et al. (2022) and Bhattacharya et al. (2022) for deeper insights into the Ig isotope and epitope recognition sites of the different neutralizing antibodies with EUA or under clinical development.

To overcome developability issues encountered with conventional Abs, DNA plasmids encoding antibody cocktails have been employed as delivery vehicles and are undergoing phase I trials (Table 2; Parzych et al., 2022). Similarly, bovine-derived antibodies and camelid-derived nanobodies against S protein have been proposed for use in countering SARS-CoV-2 and its emerging variants and mutants (Saied et al., 2022). It should also be mentioned that a humanized mAb (Crizanlizumab) against P-selectin glycoprotein ligand 1 (PSGL-1) and blocks PSGL-1 interaction with SARS-CoV-2, is currently in phase IV clinical trials. This study under the ACTIV-4A program conducted by National Heart, Lung, and Blood Institute (NHLBI) is recruiting patients to compare the effectiveness of anti-thrombotic and additional strategies for the prevention of adverse outcomes in COVID-19 positive inpatients (NCT04505774). Researchers have also turned to anti-ACE2 approaches to combat COVID-19. These preclinical studies include mAbs P2G3 and P5C3 (Aerium Therapeutics, Switzerland), single domain VHH nanobodies (TB202-3; Twist Biosciences, USA) and ACE2 antibody fusion protein (FYB-207, Formycon, Germany and SCG Cell Therapy, Singapore).

6.2. Recombinant ACE2 proteins and peptides

ACE2 is an integral part of the renin–angiotensin system (RAS) that regulates the body's blood pressure and renal function (Datta et al., 2020). ACE2 can be shed after cleavage by the metalloprotease, ADAM17 and TMPRSS2. The resulting soluble ACE2 (sACE2) retains its catalytic activity, as well as its binding ability to Spike RBD. Hence, recombinant proteins or peptides, like sACE2 that block S protein-ACE2 interaction, may be beneficial in the treatment of COVID-19 and some are being tested in the clinic (Table 3 ). The most advanced candidate is a tri-specific DARPin (Ensovibep), which binds to Spike trimers with high potency, potentially making it less subject to escape mutations (Rothenberger et al., 2022). Ensovibep successfully completed phase II trials for COVID-19 outpatients (NCT04828161), having met the primary endpoint of viral load reduction and reduction in hospitalization and death. Nevertheless, in a subsequent phase III study with hospitalised COVID-19 patients, it did not show clinical benefit. Based on its successful phase II data, and together with positive preclinical data, an application for FDA EUA has been filed. Alternate trimeric forms of DARPin fused with a T4 foldon (FSR16m and FSR22) have been reported and inhibit various SARS-CoV-2 variants, including BA.1.1 in vitro (Chonira et al., 2022).

Table 3.

Recombinant proteins and peptides in clinical and preclinical testing against SARS-COV-2.

Drug Name	Details	Global Status	Company	Delivery Route	Omicron Inhibitory Activity	Clinical trials ID
Ensovibep (MP-0420; SKO-136)	Tri-specific DARPin (designed ankyrin repeat proteins), binds spike protein trimer; for both prophylactic and therapeutic uses.	Phase III	Molecular Partners, Switzerland, Novartis, Switzerland	Injectable, I.V.	active against Omicron BA.1 and BA.2.	NCT04501978 (ACTIV-3): recommendation to stop enrollment due to lack of clinical benefit
APN01 (alunacedase alfa)	soluble recombinant ACE2 extracellular domain	Phase II	Apeiron Biologics, Austria	inhaled, I.V.	n.a.	NCT04335136: completed
Neumifil (mCBM40)	recombinant protein based on CBM40 domain derived from bacterial sialidases	Phase II (influenza)	Pneumagen, UK	Inhaled, transnasal spray	broad-spectrum antiviral (influenza, RSV, COVID-19)	NCT05507567 (Human Influenza virus Challenge Model): recruiting
HLX71	Recombinant ACE2-Fc fusion protein	Phase I	Hengenix Biotech Inc	Injectable, I.V.	n.a.	NCT04583228: recruitment completed
SI–F019	SARS-CoV-2 neutralizing protein, mimics ACE2 and act as a neutralizing decoy	Phase I	Biokin Pharmaceutical, China	Injectable, I.V.	n.a.	NCT04851444: recruiting
ACE2 mutant	ACE2 mutant	Phase I/Ib	SignalChem Lifesciences, Canada; Innophore, Austria	Injectable	n.a.	n.a.
CG-SpikeDown	peptide-based product binding to RBD	Preclinical	Caregen	Inhaled	n.a.	NCT05234320: not yet recruiting
2MA-1	modified soluble human ACE2	Preclinical	Masker MedTech, Switzerland	Inhaled	active against Omicron	n.a.
TB202-3 (TB202-63)	COVID-19 VHH nanobodies, ACE2 inhibitor	Preclinical	Twist Bioscience	Injectable	n.a.	n.a.

I.V., intravenous; n.a., information not available.

Engineered soluble recombinant ACE2 ectodomains (both active and inactive forms or with and without the collectrin domain) with optimized binding affinity and neutralization capacity against SARS-CoV-1 and -2 RBDs, have also been used as decoys to block SARS-CoV-2 entry. As they will likely bind Spike proteins from multiple SARS-CoV-2 variants, they are expected to possess broad-spectrum inhibitory activity. However, prolonged dosing of a mutant sACE2 may run the risk of generating Abs that recognize it as “foreign”, leading to decreased efficacy with time. This possibility would need to be verified in the clinic.

A soluble recombinant ACE2 extracellular domain (APN01) has successfully completed phase II clinical trials for treatment of severe COVID-19 disease. APN01 was originally developed to treat ARDS after the SARS outbreak in 2003 (Shoemaker et al., 2022). By mimicking ACE2, it functions dually to neutralize SARS-CoV-2 as well as reduce inflammation and organ injury through cleavage of angiotensin II. A patient with severe COVID-19 disease has been reported to be successfully treated by APN01 (Zoufaly et al., 2020). Data from phase II trial demonstrated that APN01 treatment led to statistically significant improvement in mechanical ventilator-free days in alive patients and reduction in viral compared to placebo (ABeriron Biologics, 2021).

Additional recombinant mutant ACE2 decoys, ACE2-Fc fusions and a peptide-based RBD binder are also in early stage clinical trials (Table 3), whilst inhibitory ACE2 peptides are in the exploratory phase. Han et al. (2006) linked two noncontinuous ACE2 peptides aa22-44 and aa351-357 via a glycine residue to inhibit SARS-CoV-1 pseudotyped virus with IC₅₀ of 0.1 μM. Using both ACE2 and de novo in silico design, Cao et al. (2020) generated picomolar inhibitory peptides that neutralized SARS-CoV-2 infection. Cryo-EM analyses showed that these peptides bind stoichiometrically to the three RBDs within the spike trimer. Moreover, a 36aa peptide (EK1) derived from the HR2 sequence in OC43 S2 subunit harboured broad CoV anti-viral fusion activity with IC₅₀ values between 0.19 and 0.62 μM (Xia et al., 2019). Crystallography confirmed that EK1 binds to the conserved HR1 domain of both human alpha and beta coronavirus spike protein by forming a stable six-helix bundle structure.

Wines et al. (2022) took further steps to modify the Fc portion of the ACE2 decoy to enable the construct to oligomerise and induce different downstream immunological responses. Trimeric WT ACE2 constructs with superior binding affinity and broad-spectrum neutralization efficacy have also been developed (Guo et al., 2021; Xiao et al., 2021). Readers may find more examples as well as details of computational ACE2 design approaches in earlier reviews by Arimori et al. (2022) and Feng et al. (2021).

Instead of using recombinant ACE2 protein, Sims et al. (2022) employed an AAV vector to deliver an affinity-matured ACE2 decoy protein via intranasal administration in a SARS-CoV-2 challenge mouse model. Long-lasting, passive protection was achieved suggesting such an approach could be used to treat COVID-19 patients. On the other hand, high plasma level of sACE2s is often observed in COVID-19 patients with severe illness and increases with age (Akin et al., 2022; Wissing et al., 2022), raising questions on whether sACE2 therapy will be beneficial to these patients.

Finally, Neumifil is a first-in-class protein with glycan targeted carbohydrate binding modules (CBMs), based on the CBD Family 40 (CBM40) domain derived from bacterial sialidases. It is bacterially expressed as a single hexavalent protein and developed using Pneumagen's GlycoTarge™ platform technology. Neumifil has been shown to have broad-spectrum antiviral against influenza, RSV, COVID-19 though binding of sialic acid receptors (to inhibit influenza virus entry). It has been demonstrated to bind to ACE2 and spike protein of SARS-CoV-2 (Fell et al., 2022) and is undergoing a phase II proof-of-concept Influenza virus human challenge study.

6.3. Small molecules and natural products

Whilst neutralizing antibodies continue to be the mainstream approach to block Spike-receptor interaction, increasingly, screening for small molecule inhibitors (SMIs) of the protein–protein interaction (PPI) between the S protein and hACE2 is gaining interest due to the challenges faced in the ever-changing SARS-CoV-2 Spike mutations. Small molecules have the advantage of being more likely to be administered orally, have broad-spectrum activity and be less immunogenic. Precedence for SMI in blocking virus-receptor interactions are exemplified by HIV-1 and HBV/HDV entry inhibitors (See Section 4). Moreover, more than 40 SMIs that target PPIs are currently in preclinical development with two drugs, Venetoclax and Lifitegrast, approved for use for treatment of cancer and dry eye disease, respectively (Bojadzic et al., 2021).

Researchers have employed both in vitro binding and cell-based assays for screening and hit validation in order to identify inhibitors of Spike-receptor interaction. These include binding assays that utilize recombinant Spike and ACE2 proteins such as ELISA, SPR, differential scanning fluorimetry and cell-based assays that monitor entry of pseudo-typed luciferase reporter viruses bearing Spike protein (Mediouni et al., 2022; Chen et al., 2020, Chen et al., 2021), or NanoBit-RBD/ACE2 PPI assays (Yu et al., 2021). As well, in silico modelling and screening of potential binders in the RBD-Spike interaction site, followed by the use of machine learning and deep learning algorithms, have been extensively explored to identify novel binders and to facilitate rational design.

Majority of these findings are from large-scale compound screenings with clinical-stage or FDA-approved small molecules. These efforts have cumulated in a number of small molecules and natural products that can interfere with Spike-ACE2 binding. These discoveries have been comprehensively covered in past reviews (Liu et al., 2021; Ma et al., 2021; Xiang et al., 2021). Nevertheless, some of the compounds from these screens await further validations with binding or biophysical assays, or virus neutralization assays, to confirm their mechanism of actions, as well as their broad-spectrum anti-CoV activity. On this note, Han et al. (2021) demonstrated that imatinib and quinacrine dihydrochloride (QNHC) bound to ACE2 by SPR, after using a lung organoid model derived from human IPS cells to conduct a screen with the FDA approved drug library. Furthermore, both MPA and QNHC treatment decrease the expression levels of furin.

Besides interfering with virus-receptor interactions, repurposed drugs that block spike-mediated cell fusion (Mediouni et al., 2022; Chen et al., 2020; Riva et al., 2020) or enzymatic activities of ACE2 (MLN-4760), TMPRSS2 (Camostat mesylate and Nafamostat), cathepsins (Eicoplanin, calpain inhibitor II, XII), furin (reviewed in Nepali et al., 2022) have been described. Indeed, a few repurposed drugs have entered the clinic for treatment of COVID-19 (Table 4 ) and they include Niclosamide (oral or inhaled formulation; phase III) which acts as a protonophore to interfere with viral entry and egress (Singh et al., 2022), Nafamostat (Zhuravel, 2021; phase II/III), Camostat (phase I/II), Amiodarone and Verapamil (inhibit endosomal processing (Stadler et al., 2008; phase II/III), Umifenovir (Arbidol; targets S-protein and prevents viral fusion; Nojomi et al., 2020; phase III/IV).

Table 4.

Broad-spectrum antiviral inhibitors tested in in clinical testing against SARS-COV-2.

Drug Name	Details	Global Status	Company	Delivery Route	Inhibitory Activity	Clinical trials ID
niclosamide (ANA 001)	reformulated form	Phase II/III	NeuroBo Pharmaceuticals, USA	Oral	broad-spectrum antiviral	NCT04603924: recruiting
niclosamide (UNI-91103)	reformulated optimized salt form	Phase III	Union Therapeutics, Denmark	Inhaled; transnasal	broad-spectrum antiviral	NCT04932915: terminated (Failure of recruiting patients)
niclosamide (CP–COV03)	orally ingested niclosamide formulated using organic-inorganic hybrid drug delivery system (DDS) technology	Phase II	Hyundai Bioscience, South Korea; CNpharm, South Korea	Oral	broad-spectrum antiviral	n.a.
Neumifil (first-in-class glycan targeted carbohydrate binding modules)	recombinant protein based on CBM40 domain derived from bacterial sialidases	Phase II (influenza)	Pneumagen, UK	Inhaled, transnasal spray	broad-spectrum antiviral (influenza, RSV, COVID-19)	NCT05507567 (Human Influenza virus Challenge Model): recruiting
Umifenovir (Arbidol)	targets S-protein and prevents viral fusion	phase III/IV	Shahid Beheshti University of Medical Sciences, Iran	Oral	broad-spectrum antiviral (HCV, influenza, COVID-19)	NCT04350684: completed
Amiodarone, Verapamil	inhibits endosomal processing by blocking ion channels	phase II/III	Nicolaus Copernicus University, Poland	Oral	broad-spectrum antiviral	NCT04351763: completed
Nafamostat Mesilate	inhibits TMPRSS2	phase II	Chong Kun Dang Pharmaceutical, South Korea	Oral	broad-spectrum antiviral	NCT04623021: completed
		phase II/III	Gyeongsang National University Hospital, South Korea			NCT04418128: completed
		phase II/III	University Hospital Padova, Italy			NCT04352400: recruiting
Camostat Mesilate	inhibits TMPRSS2	phase II	Yale University, USA	Oral	broad-spectrum antiviral	NCT04353284: completed
		phase II	University of Aarhus, Denmark			NCT04321096: completed

n.a., information not available.

Unfortunately, the trial with Umifenovir revealed that it was ineffective in shortening the duration of SARS-CoV-2 in severe patients and did not improve the prognosis in non-ICU patients and mortality (Darazam, 2021). Neither Amiodarone nor Verapamil too, were found to significantly accelerate short-term clinical improvement in hospitalised COVID-19 patients (Navarese et al., 2022). Although overall results reported for Nafamostat were not highly encouraging, one group observed a shorter median time to clinical improvement in a small group of high-risk COVID-19 patients requiring oxygen treatment (Zhuravel, 2021; NCT04623021). Camostat-treated outpatients had more rapid resolution of COVID-19 symptoms and amelioration of the loss of taste and smell but was not associated with SARS-COV-2 viral load reduction (Chupp et al., 2022). For hospitalised patients, it did not impact clinical outcome (Gunst et al., 2021). These findings suggest that virus loads in the patients and additional immunological responses activated after prolonged virus infection can affect drug efficacy. Nevertheless, as in vitro data suggests that cell entry of the Omicron variant is less dependent on TMPRSS2, therapies to target this enzyme may be less impactful.

Whilst developing repurposed drugs for COVID-19 treatment can have significant time and cost advantages, they may have unwanted pharmacological effects if the protein targeted in COVID-19 disease is different from that in its original indication. It is also important to ascertain if the active repurposed drug has valid patent coverage as its unauthorized usage could encounter freedom-to-operate or patent infringement issues. These considerations make the case for continued efforts to find novel COVID-19 SMIs to inhibit virus entry. An interesting approach has been described whereby ACE2 protein S-nitrosylation by amino-adamantane nitrates resulted in interference with SARS-CoV-2 entry in vitro (Oh et al., 2022). These compounds were previously shown to block the ion channel in the envelope of multiple viruses. The authors further confirmed the ACE2 protein S-nitrosylation by this class of compounds using co-IP and mutational studies.

7. Conclusion

SARS-CoV-2, as well as other Sarbecoviruses, use ACE2 as the obligate receptor. Interfering with virus Spike protein-ACE2 binding prevents virus entry and infection. Therapies with neutralizing Abs have become insufficient to combat COVID-19 disease due to continual virus Spike mutations and co-circulations of several dominant virus variants. Protein-based inhibitors, natural products and SMIs that target virus-receptor interaction and virus entry can complement SARS-CoV-2 direct antivirals and Ab-based therapeutics. SMIs are also more likely to have broad-spectrum activity and be orally administered. For identifying such SMIs, researchers have mostly focused on repurposing drugs, either employing HTS or in silico docking. This endeavour can have time and cost savings and has accelerated testing of drugs in the clinic. Thus far, no repurposed anti-viral drug has been approved for COVID-19 treatment. The use of diverse compound libraries to identify virus entry inhibitors can potentially generate novel scaffolds for development of drug candidates with better safety profiles and binding characteristics. These chemical entities could also be collated into a “chemical tool-box” and readily deployed for testing against newly emerged CoVs in future outbreaks.

SARS-CoV-2 entry inhibitors may be applied in pre-or post-exposure prophylaxis and also in treating the early phase of virus infection. Reducing virus replication will prevent progression to severe disease, shorten recovery time and reduce hospitalizations. In this regard, therapeutic antibodies have been useful in managing patients with mild to moderate COVID-19, but who are at risk of hospitalization due to comorbidities or other factors. On the other hand, entry inhibitors are expected to be less efficacious for patients who are hospitalised with severe COVID-19 disease, as the hyper-activated inflammatory response is mainly responsible for driving the disease rather than virus replication.

Almost 10 years ago, the Antiviral Research Journal collated a series of invited papers marking the 10th anniversary of the outbreak of SARS-CoV-2 in 2003 and the MERS epidemic in 2012. In the introduction to the series, Hilgenfeld and Peiris (2013) warned that “… further introductions of highly pathogenic coronaviruses into the human population are not merely probable, but inevitable.” This prophesy became true in late 2019, with the outbreak of SARS-CoV-2. Also, as recounted by both authors, after 2005–2006 (following outbreak resolution), research funding on SARS-CoV-1 in many countries became unavailable, as life returned to normal. This stymied R&D efforts to better prepare for the next outbreaks (MERS and SARS-CoV-2), not only in the discovery of new antiviral therapies and vaccines, but also in preparing for a better public health response. These inadequacies led to chaotic and disparate responses in the initial phases of the SARS-CoV-2 pandemic and exerted heavy tolls on lives and economies.

Going forward, some governments have established initiatives for future pandemic preparedness, including funding the development of anti-virals and vaccines for pathogens of high importance or pandemic potential. For example, NIAID recently awarded $577 million to establish nine Antiviral Drug Discovery (AViDD) Centers for Pathogens of Pandemic Concern (NIH, 2022b) and the Singapore's Ministry of Health (MOH) has launched a research programme (PREPARE) to safeguard Singapore against future pandemics (MOH, Singapore, 2022).

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.

Data availability

No data was used for the research described in the article.