SH3 domain derivatives targeting Receptor Binding Domain (RBD) of SARS-CoV-2 spike protein

FIELD OF THE INVENTION

The present invention is directed to the field of engineered binding proteins. Particularly, the present invention is directed to SH3 domain derivatives having a specific binding affinity to the receptor binding domain (RBD) of SARS-CoV-2 spike glycoprotein. In this respect, the invention specifically provides SH3 domain derivatives of the ciliary adaptor protein nephrocystin (NPHP1). The present invention is also directed to the prevention and treatment of COVID-19.

BACKGROUND OF THE INVENTION

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a human coronavirus that first emerged in 2019 and has caused a pandemic of acute respiratory disease, namely coronavirus disease 2019 (COVID-19), which threatens human health and public safety. SARS- CoV-2 is a highly transmissible and pathogenic coronavirus. The emergence of novel SARS- CoV-2 variants of concern (VOCs) resistant to neutralizing antibodies emphasizes the need for urgent discovery and development of preventive and therapeutic molecules. Recently, four VOCs B.l.1.7 (alpha), B.1.351 (beta), P.l (gamma), and B.l.617.2 (delta), have emerged in the United Kingdom, South Africa, Brazil, and India, respectively, with numerous substitutions in both the N-terminal and receptor binding (RBD) domains of the trimeric spike envelope glycoprotein. ¹ The fast-spreading, highly contagious delta variant that first emerged in India has since become the dominant strain globally. ² The rapid dominance of delta over pre- existing lineages may be due to its increased infectivity, longer asymptomatic period, and reduced sensitivity to neutralizing antibodies. SARS-Co-V-2 vaccines play a crucial role in limiting viral spread and minimizing severe COVID-19 disease, and recent studies have shown that they are effective also against the delta variant. Although the greatest risk of transmission is among unvaccinated individuals, delta (as well as other VOCs) is able to partially escape neutralizing antibodies elicited by previous SARS-CoV-2 infection or by vaccines. ¹ Long-term follow-up studies have documented waning of the immune response, steady decline of antibody levels and a growing risk of breakthrough infection over time among vaccinated individuals. ³ For SARS-CoV-2 infection, the nasal epithelium of respiratory tract is at first the dominant replication site followed by virus aspiration into the lung. ⁴ High viral load in the respiratory tract correlates with severe disease in COVID-19 patients. ⁵ In animal models, monoclonal antibodies targeted against the spike envelope protein of SARS-CoV-2, have been shown to be more effective for COVID-19 prophylaxis than for treatment, probably mirroring the SARS- CoV-2 virulence, speed of virus replication, and rapid onset of symptoms. ⁵ Circulating IgG antibodies are not able to efficiently enter to mucosal compartments, and antibody levels in the lungs have been shown to be up to 500-fold lower than those in serum after intravenous infusion. ⁶ Both treatment and prophylaxis by monoclonals require exceedingly high administration doses (typically several grams) and intravenous delivery rather than direct administration to the respiratory tract where the virus is mainly found. ⁴ While hard-to-escape antibodies able to neutralize all known SARS-CoV-2 strains as well as related sarbecoviruses are under development, resistance against many therapeutic antibodies that have been authorized for emergency use to date represent a major challenge for antibody-based therapy of COVID-19. ¹ Lately, increasing attention has been shifted to inhibitory targeting of SARS- CoV-2 spike using smaller antibody fragments and antibody mimetics.

Antibodies and antibody fragments are currently the most widely used targeting molecules in biological therapies, prophylaxis as well as in vivo diagnostic applications. However, the intrinsic biochemical properties of the immunoglobulin heavy and light chains of antibodies pose limitations to their engineering, which has led to the emergence of antibody-mimetics as a popular topic in this field of research. ⁷ Antibody-mimetics can be engineered based on a variety of different polypeptide backbones (termed scaffold proteins) that naturally show superior biochemical and biophysical characteristics compared to immunoglobulins. When artificial antigen-binding interfaces are introduced into such scaffold proteins, novel antibody-mimetics can be obtained that show antibody-like antigen recognition combined with the desired properties of the original scaffold.

Numerous different polypeptides backbones derived from mammalian as well as bacterial proteins, or even created via de novo design have been developed into targeting scaffolds ⁷ . In addition to human lipocalin scaffold-based Anticalins and the designed ankyrin repeat proteins (DARPins), which are perhaps the most established scaffolds in therapeutic applications, engineered human SH3 domains originally described by Hiipakka et al. ^{8 9} have proven to be very useful and versatile targeting scaffolds. Extensive work by Covagen on human Fyn SH3 domain-derivatives known as Fynomers have validated the utility of this technology in therapeutic targeting, and led to drug candidates that have reached phase 2 clinical trials.

Recent work by the inventors has shown that among the repertoire of approximately 300 different human SH3 domains, NPHP1 (also known as nephrocystin) SH3 domain provides an exceptionally suitable engineering/targeting scaffold that gives rise to phage libraries with superior functionality when compared with other human SH3 domains subjected to similar loop region sequence diversification (US2018230456A1). This patent application relates to binding proteins, specifically SH3 domain-derivatives, that have been targeted to bind to the receptor binding domain of SARS-CoV-2 spike protein.

SUMMARY OF THE INVENTION

One of the objects of the present invention is to provide a recombinant binding protein having a specific Src homology 3 (SH3) domain based binding affinity to a receptor binding domain (RBD) of SARS-CoV-2 spike glycoprotein, said binding protein comprising a nephrocystin (NPHP1) derived SH3 domain with an RT-loop and a n-src-loop, wherein TAQQVG (SEQ ID NO:39) sequence of the wild type NPHP1 SH3 RT-loop is substituted with an amino acid sequence

(X ₁ )(X ₂ )(X ₃ )(X ₄ )(X ₅ )(X ₆ ), wherein the amino acids (X ₁ ) to (X ₆ ) of the RT loop of said SH3 domain correspond to the following amino acids:

(X ₁ ) is W or F,

(X ₂ ) is S or T,

(X ₃ ) is I, M, A, N, S, T, or Q,

(X ₄ ) is D or S,

(X ₅ ) is any amino acid,

(X ₆ ) is any amino acid; wherein said nephrocystin (NPHP1) derived SH3 domain has an amino acid sequence having at least 80% sequence identity to the amino acid sequence of SEQ ID NO:1 outside the RT-loop and the n-src-loop; and wherein the n-src-loop may comprise any deletion, insertion or amino acid substitution within the NPHP1 SH3 wild type n-src-loop sequence.

Another object of the present invention is to provide a fusion protein comprising the binding protein as defined herein.

Further objects of the invention is to provide polynucleotides, vectors and host cells for the production of the binding proteins.

Another object of the invention is to provide a use of amino acid sequence comprising (W/F)SX(S/D)XX, wherein X is any amino acid, as a RBD of SARS-CoV-2 spike protein binding motif in a recombinant binding protein specific to SARS-CoV-2 spike protein, wherein said recombinant binding protein comprises a SH3 domain.

Another object of the invention is to provide a binding protein or a fusion protein as defined in the present disclosure for use in the prevention or treatment of COVID-19.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1: Analysis of thirtyone individual drSH3 clones obtained by affinity selection of pooled phage libraries displaying loop-randomized NPHP1 SH3-derivatives using recombinant RBD-mFc protein as a target. Individual phage supernatants corresponding to these fifteen clones were tested using two-fold serial dilutions (1:8 fold dilution is shown) in phage-ELISA for binding to RBD-mFc, control protein containing mouse Fc only, or an antibody against an invariant E-tag peptide epitope displayed by each phage between the drSH3 domain and the pill coat protein. The experimental details of the phage-ELISA are described under Materials and methods.

Figure 2. Neutralization of SARS-CoV-2. A) Pseudoviruses carrying the RBD mutations found in the spike of Wuhan-Hu-1 wild-type virus or the B.1.351 (beta) variant of concern were incubated with serially diluted concentrations of the trimeric ARM100 and monomeric ARM92 prior to transduction of ACE2 overexpressing HEK293 cells. Normalized luciferase activity indicating viral internalization was measured at 48 h post transduction. B) Neutralization of B.1.1.7 (alpha/United Kingdom), B.1.351 (beta/South Africa) and B.l.617.2 (delta/India) clinical virus isolates in Vero E6 cells by ARM100. The experimental details of both neutralization assays are described under Materials and methods. Figure 3. Prophylactic efficacy of ARM100 in mice. Two independent experiments (EXP 1 and 2) were performed, in which Balb/c mice (n values are indicated in the figure) were intranasally administered either with a standard dose (2.5 mg/kg) or with a low dose (0.25 mg/kg) of ARM100 1, 4 or 8 h prior to infection with SARS-CoV-2 B.1.351 clinical virus isolate (2x10 ⁵ PFU). Viral RNA in the lungs of Balb/c mice was measured 3 days after the intranasal challenge. RNA extracted from lungs was analysed for SARS-CoV-2 viral load using qRT-PCR for RdRp, E and subE genes as well as mouse actin mRNA, and normalized using subE as previously described. ^45-47 Viral RNA was detectable only in two animals (n=2) of the pre- treated mice in Experiment 2.

DESCRIPTION OF THE EMBODIMENTS

SARS-CoV-2-binding proteins, such as SH3 domains, can be used as components of many different kinds of prophylactic, therapeutic or diagnostic approaches, which may involve for example polypeptides and multimeric macromolecules.

The prophylactic or therapeutic effect of such molecules may be due to blocking of SARS-CoV- 2 receptor binding, inhibiting viral attachment and entry to target cells, or cell-cell transmission.

Likewise, the possible designs and modes of action of in vitro and in vivo diagnostic approaches based on SARS-CoV-2 spike RBD binding proteins are numerous, typically involving modification of a RBD-binding protein with a compound that can be directly or indirectly (e.g. by catalyzing an enzymatic reaction) detected, localized, and/or quantitated.

Of note, the same SARS-CoV2 spike RBD-binding protein-containing molecule may be prophylactic, therapeutic and diagnostic at the same time, i.e. have theragnostic potential.

In a preferred embodiment, SARS-CoV-2 spike RBD-binding SH3 domains can mediate RBD- targeting alone, but they can also be used as components of bi- or multispecific molecules by fusing them with other targeting molecules, such as additional targeted SH3 domains, other targeting scaffolds, antibodies or antibody fragments, which may target another site in SARS- CoV-2 spike protein or the RBD, but may also bind to additional molecules, for example to improve mucosal rentention, enhance stability and pharmacokinetics or aid production and purification. The SARS-CoV-2 spike RBD-binding protein-containing molecules may be produced in vitro but may also be expressed by genetically engineered cells that produce prophylactic or therapeutic molecules in the body (eg. gene therapeutic delivery of a biologic drug).

In a specific embodiment, the present invention provides a recombinant binding protein having a specific Src homology 3 (SH3) domain based binding affinity to a receptor binding domain (RBD) of SARS-CoV-2 spike glycoprotein, said binding protein comprising a nephrocystin (NPHP1) derived SH3 domain with an RT-loop and a n-src-loop, wherein TAQQVG (SEQ ID NO:39) sequence of the wild type NPHP1 SH3 RT-loop is substituted with an amino acid sequence

(X ₁ )(X ₂ )(X ₃ )(X ₄ )(X ₅ )(X ₆ ), wherein the amino acids (X ₁ ) to (X ₆ ) of the RT loop of said SH3 domain correspond to the following amino acids:

(X ₁ ) is W or F,

(X ₂ ) is S or T,

(X ₃ ) is I, M, A, N, S, T, or Q,

(X ₄ ) is D or S,

(X ₅ ) is any amino acid,

(X ₆ ) is any amino acid; wherein said nephrocystin (NPHP1) derived SH3 domain has an amino acid sequence having at least 80% sequence identity to the amino acid sequence of SEQ ID NO:1 outside the RT-loop and the n-src-loop; and wherein the n-src-loop may comprise any deletion, insertion or amino acid substitution within the NPHP1 SH3 wild type n-src-loop sequence.

In a preferred embodiment, the present invention is directed to a recombinant binding protein having a specific Src homology 3 (SH3) domain based binding affinity to a receptor binding domain (RBD) of SARS-CoV-2 spike glycoprotein, said binding protein comprising a nephrocystin (NPHP1) derived SH3 domain with an amino acid sequence

EEYIAVGDF (X ₁ ) ( X ₂ ) ( X ₃ ) ( X ₄ ) ( X ₅ ) ( X ₆ ) DLTFKKGEI LLVIE (X ₇ ) ( X ₈ ) ( X ₉ ) ( X ₁₀ ) ( X _{1 1} ) ( X ₁₂ ) (X ₁₃ ) ( X ₁₄ ) (X ₁₅ ) DGWWI AKDAKGNEGLVPRTYLEPY ( SEQ I D NO : 1 ) , wherein the amino acids (X ₁ ) to (X ₆ ) of the RT loop of said SH3 domain correspond to the following amino acids:

(X ₁ ) is W or F,

(X ₂ ) is S or T,

(X ₃ ) is I, M, A, N, S, T, or Q,

(X ₄ ) is D or S,

(X ₅ ) is any amino acid,

(X ₆ ) is any amino acid; wherein each of the amino acids (X ₇ ) to (X ₁₅ ) of the n-src loop of said SH3 domain may be independently any amino acid or absent preferably so that when amino acids (X ₇ ) (X ₈ ) and (X ₁₅ ) are KKP, respectively, then amino acids (X ₉ ) - (X ₁₄ ) can be absent, or when amino acids (X ₇ ) (Xs) and (X ₁₅ ) are absent, then at least amino acids (X ₉ ) - (X ₁₂ ) are present; and wherein said nephrocystin (NPHP1) derived SH3 domain has an amino acid sequence having at least 80% sequence identity to the amino acid sequence of SEQ ID NO:1 outside the RT-loop and n-src loop.

In another preferred embodiment, the amino acids (X ₉ ) to (X ₁₄ ) of the n-src loop of said SH3 domain correspond to the following amino acids:

(X ₉ ) is T, P, Q, F, R, L, or I,

(X ₁₀ ) is K, L, A, S, T, N, W, or A,

(X ₁₁ ) is S, D, E, N, P, G, A, or E,

(X ₁₂ ) is P, N, T, G, or D,

(X ₁₃ ) is N, Q, A, G, or V, and

(X ₁₄ ) is L, Q, M, F, T, N, S, or R.

In another preferred embodiment the amino acids (X ₁ ) to (X ₆ ) correspond to the following amino acids:

(X ₁ ) is W or F,

(X ₂ ) is S or T,

(X ₃ ) is any amino acid,

(X ₄ ) is D or S,

(X ₅ ) is any amino acid, (X ₆ ) is any amino acid.

In another preferred embodiment, the amino acids (X ₁ ) to (X ₆ ) correspond to the sequence WSISAE (SEQ ID NO:4), WTIDSA (SEQ ID NO:5), WSMSLD (SEQ ID NO:6), WSMDSA (SEQ ID NO:7), WSADRG (SEQ ID NO:8), WSISSA (SEQ ID NO:9), WSMDVE (SEQ ID NO:10), WSNDYG (SEQ ID NO:11), WSNSAG (SEQ ID NO:12), WSSDPL (SEQ ID NO:13), WSNDAD (SEQ ID NO:14), FSTDPA (SEQ ID NO:15), WSQDET (SEQ ID NO:40), WSNSQS (SEQ ID NO:41), WSNSSA (SEQ ID NO:42), WSQDIT (SEQ ID NO:43), WSNDMG (SEQ ID NO:44), WSADSD (SEQ ID NO:45), WSSSSA (SEQ ID NO:46), WSQDKG (SEQ ID NO:47), WSQDKT (SEQ ID NO:48), WSQDAG (SEQ ID NO:49), WSNDPN (SEQ ID NO:50), WSNSPI (SEQ ID NO:51), WSNSPG (SEQ ID NO:52), WSQDST (SEQ ID NO:53), WSQDPY (SEQ ID NO:54), or WSQDNS (SEQ ID NO:55).

In another preferred embodiment, said nephrocystin (NPHP1) derived SH3 domain has an amino acid sequence having at least 85%, 90% or at least 95% sequence identity to the amino acid sequence of SEQ ID NO:1 excluding amino acid positions corresponding to the RT-loop and n-src-loop.

In this invention, the location of the RT loop in the NPHP1 SH3 domain corresponds to amino acid positions 8-17 of SEQ ID NO:1 or 16, preferably 10-15 of SEQ ID NO:1 or 16. The n-src- loop in the NPHP1 SH3 domain is defined to locate between the amino acid positions 28-34 of SEQ ID NO:16, preferably 30-32 of SEQ ID NO:16, or between the amino acid positions 28-40 of SEQ ID NO:l.

In another embodiment, the present invention provides a recombinant binding protein having a specific Src homology 3 (SH3) domain based binding affinity to a RBD of SARS-CoV- 2 spike glycoprotein, said binding protein comprising a nephrocystin (NPHP1) derived SH3 domain with an amino acid sequence

EEYIAVGDFFSTDPADLTFKKGEILLVIE (X ₉ ) (X ₁₀ ) (X _n ) (X ₁₂ ) (X ₁₃ ) (X ₁₄ ) DGWWI AKDAKGNEGLVPRTYLEPY ( SEQ ID NO : 2 ) wherein

X ₉ to X ₁₂ are each any amino acid, X ₁₃ and X ₁₄ are each any amino acid or are absent, wherein the RT loop of said SH3 domain corresponds to amino acid positions 8-17 of SEQ ID NO:2, wherein the n-src-loop corresponds to amino acid positions 27-37 of SEQ ID NO:2, and wherein said nephrocystin (NPHP1) derived SH3 domain has an amino acid sequence having at least 85% sequence identity to the amino acid sequence of SEQ ID NO:2 outside the RT-loop and n-src loop.

In another preferred embodiment, the amino acids (X ₉ ) to (X ₁₄ ) of SEQ ID NO:2 correspond to the sequence RGTSAG (SEQ ID NO:36), TRVPEG (SEQ ID NO:37) or HNPH- (SEQ ID NO:38).

In another preferred embodiment, said binding protein comprises a dimer, trimer, tetramer, or multimer of said nephrocystin (NPHP1) derived SH3 domain.

The present invention also provides a fusion protein comprising a binding protein as defined above. Preferably, the binding protein is fused with another protein such as another SH3 domain, an antibody or a fragment thereof. However, the specific targeting properties of the binding proteins of the present invention also allow for substituting antibodies in the known fusion proteins comprising an antibody or a fragment thereof. The binding protein may also be fused to any pharmaceutically and/or diagnostically active component, which may be a non-polypeptide component such as a label.

Preferably, the recombinant binding protein of the invention or the fusion protein defined herein, has a specific binding affinity to a RBD of SARS-CoV-2 spike protein of 10 ^-5 to 10 ^-12 M, more preferably 10 ^-6 to 10 ^-12 M, 10 ^-7 to 10 ^-12 M or 10 ^-8 to 10 ^-12 M.

The present invention also provides a polynucleotide coding for the recombinant binding proteins described as well as vectors comprising said polynucleotide. For the production of binding proteins, one can use a host cell comprising said polynucleotide and/or a vector comprising said polynucleotide.

In other embodiments, the present invention provides a use of amino acid sequence comprising (W/F)SX(S/D)XX, wherein X is any amino acid, as a RBD of SARS-CoV-2 spike protein binding motif in a recombinant binding protein specific to RBD of SARS-CoV-2 spike protein, wherein said recombinant binding protein comprises a SH3 domain. More preferably, said SH3 domain is from the NPHP1 protein. In another preferred embodiment, the (W/F)SX(S/D)XX sequence is one of the following:

WSQSXX (SEQ ID NO:3), WSISAE (SEQ ID NO:4), WSMSLD (SEQ ID NO:6), WSMDSA (SEQ ID NO:7), WSADRG (SEQ ID NO:8), WSISSA (SEQ ID NO:9), WSMDVE (SEQ ID NO:10), WSNDYG (SEQ ID NO:11), WSNSAG (SEQ ID NO:12), WSSDPL (SEQ ID NO:13), WSNDAD (SEQ ID NO:14, WSQDET (SEQ ID NO:40), WSNSQS (SEQ ID NO:41), WSNSSA (SEQ ID NO:42), WSQDIT (SEQ ID NO:43), WSNDMG (SEQ ID NO:44), WSADSD (SEQ ID NO:45), WSSSSA (SEQ ID NO:46), WSQDKG (SEQ ID NO:47), WSQDKT (SEQ ID NO:48), WSQDAG (SEQ ID NO:49), WSNDPN (SEQ ID NO:50), WSNSPI (SEQ ID NO:51), WSNSPG (SEQ ID NO:52), WSQDST (SEQ ID NO:53), WSQDPY (SEQ ID NO:54), or WSQDNS (SEQ ID NO:55).

In other embodiments, the present invention is directed to a method for detecting the presence of SARS-CoV-2 spike (RBD) protein in a biological sample comprising the step of contacting said biological sample with a recombinant binding protein as defined above or a fusion protein as defined above and detecting the presence of said biomarker by detecting the presence of a complex comprising said binding protein and said biomarker or said fusion protein and said biomarker.

In other embodiments, the present invention is directed to a binding protein or a fusion protein as defined in the present disclosure for use in the prevention or treatment of COVID- 19.

In one embodiment of the present invention, the binding protein or the fusion protein as defined in the present disclosure can be incorporated into pharmaceutical compositions. Such compositions of the invention are prepared for storage by mixing the peptide having the desired degree of purity with optional physiologically acceptable carriers (such as nanocarriers), excipients, preservatives or stabilizers (Remington's Pharmaceutical Sciences, 22nd edition, Allen, Loyd V., Jr, Ed., (2012)), in the form of lyophilized cake or aqueous solutions. Acceptable carriers, excipients, preservatives or stabilizers are non-toxic to recipients at the dosages and concentrations employed, and include buffers such as phosphate, citrate, and other organic acids; antioxidants including ascorbic acid; low molecular weight (less than about 10 residues) polypeptides; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, mannose, or dextrins; chelating agents such as EDTA; sugar alcohols such as mannitol or sorbitol; salt-forming counter-ions such as sodium; and/or non-ionic surfactants such as Tween, Pluronics or polyethylene glycol (PEG).

The expressions "any amino acid" or "any amino acid residue" refers herein to any naturally occurring L-amino acid such as alanine (A), arginine (R), asparagine (N), aspartic acid (D), cysteine (C), glutamine (Q), glutamic acid (E), glycine (G), histidine (H), isoleucine (I), leucine (L), lysine (K), methionine (M), phenylalanine (F), proline (P), serine (S), threonine (T), tryptophan (W), tyrosine (Y), and valine (V).

It must also be noted that, as used in the specification and the appended claims, the singular forms "a," "an" and "the" include plural referents unless otherwise specified.

As used herein, the term "antibody" encompasses naturally occurring and engineered antibodies, as well as full length antibodies, functional fragments, or analogs thereof that are capable of binding e.g., the target immune checkpoint or epitope [e.g. retaining the antigen- binding portion). The antibody may be from any origin including, without limitation, human, humanized, animal or chimeric, and may be of any isotype, and further may be glycosylated or non-glycosylated. The term antibody also includes bispecific or multispecific antibodies so long as the antibody(s) exhibit the binding specificity herein described.

As used herein, the term “fragment” includes native peptides (either degradation products, synthetically synthesized peptides or recombinant peptides) and modified peptides, which may have, for example, modifications rendering the peptides more stable or less immunogenic. Such modifications include, but are not limited to, cyclization, N-terminus modification, C-terminus modification, peptide bond modification, backbone modification and residue modification. The fragment may also comprise further elongations, deletions, substitutions or insertions.

As used herein, the term "polypeptide" refers herein to any chain of amino acid residues, regardless of its length or post-translational modification (e.g., glycosylation or phosphorylation).

Although particular embodiments have been disclosed herein in detail, this has been done by way of example for purposes of illustration only, and is not intended to be limiting with respect to the scope of the appended claims that follow. In particular, it is contemplated by the inventor that various substitutions, alterations, and modifications may be made to the invention without departing from the scope of the invention as defined by the claims. For instance, the choice of protocols and buffers are believed to be a matter of routine for a person of ordinary skill in the art with knowledge of the embodiments described herein.

Having generally described the invention above, the same will be more readily understood by reference to the following Experimental Section, which is provided by way of illustration and is not intended as limiting.

Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and materials are described below. The materials, methods, and examples are illustrative only and not intended to be limiting.

EXPERIMENTAL SECTION

Materials and methods

The following sections detail the techniques and reagents used for the generation of highly diverse phage-displayed drSH3 libraries based on the human NPHP1 SH3 domain [Gene ID: 4867] as a display scaffold, and use of these libraries to discover drSH3 domains targeted against receptor binding domain (RBD) of SARS-CoV-2 spike protein protein.

Construction of double-randomized drSH3 phage display libraries

For display of drSH3 libraries we have generated an optimized pUC119-based M13-phagemid vector pCA11N that allows selection both with ampicillin/carbenicillin as well as with chloramphenicol. Otherwise it contains the generic features of commonly used pill-fusion display vectors, including a PelB signal sequence, and unique Sfil and Notl restriction enzyme sites for cloning of the codon-optimized drSH3 genes fused to the full-length pill gene by the E-tag peptide sequence. The pCA11N phagemid contains a double-stranded DNA (dsDNA) origin of replication (dsDNA ori) and replicates as a double-stranded plasmid when inserted into an E. coli host. A single-stranded DNA (ssDNA) filamentous phage origin of replication (f1 ori) contains all of the DNA sequences necessary for packaging of the viral DNA into phage particles upon superinfection with a helper phage. While wild-type pill is present at five copies per phage particle in total, the pill-fused drSH3 is expected to be displayed in a monovalent format.

To generate highly diverse pCA11N-based libraries of modified NPHP1 SH3 domain sequences we have used standard methods for oligonucleotide-directed mutagenesis and high-efficiency bacterial transformation ¹⁹ . For generation of completely random libraries, unique codons in the template DNA were replaced by NNK degenerate codons (where N = A/G/C/T and K = T/G), which encode all 20 natural amino acids. The TAG amber codon is suppressed by insertion of glutamine in a suppressor E. coli strain such as TGI and XL-1 blue suitable for phage propagation.

The randomization strategies used to generate two different NPHPl-derived drSH3 libraries with RT- and n-src-loop modifications are presented below where X denotes a randomized residue, i.e. any of the 20 natural amino acids. Hyphens have been introduced into these sequences to better align them for easier comparison.

Wild-type human NPHP1 SH3 domain (SEQ ID NO:16):

EEYIAVGDFTAQQVGDLTFKKGEILLVIEKK - PDGWWIAKDAKGNEGLVPRTYLEPY

Library A (SEQ ID NO: 17):

EEYIAVGDFXXXXXXDLTFKKGEILLVIE--XXXXXX-DGWWIAKDAKGNEGLVPRT YLEPY

Library B (SEQ ID NO: 18):

EEYIAVGDFXXXXXXDLTFKKGEILLVIEKKXXXXXXPDGWWIAKDAKGNEGLVPRT YLEPY

In the first step, random mutations were introduced into an ssDNA template. The single- stranded phagemid DNA was purified from CJ236 dut-/ung- E. coli strain, which specifically incorporates uracil instead of thymine in DNA. The uracil-containing ssDNA was used as a template onto which mutagenic oligonucleotides were annealed. The mutagenic oligonucleotides were designed such that they share a minimum of 15 nucleotide complementarity with the template both up- and downstream of the region targeted for mutagenesis to ensure efficient annealing. The oligonucleotides were annealed to the uracil- containing ssDNA template to prime the synthesis of a complementary DNA strand by T7 DNA polymerase. Subsequently, T4 ligase was used to form covalently linked circular dsDNA, containing mismatches in the region targeted for mutagenesis.

Covalently linked circular dsDNA was affinity-purified and transformed by high-efficiency electroporation into a dut+/ung+ SS320 E. coli host, which preferentially replicates the nascent DNA containing the mutagenic oligonucleotide instead of the uracil-containing parental strain. The E. coli SS320 strain has been designed for high-efficiency DNA transformation by mating MC1061 and XLl-blue and selecting on tetracycline and streptomycin medium. ¹³ The strain thus encompasses the high-efficiency transformation qualities of MC1061, and contains the F' episome from XLl-blue, critical for bacteriophage infection and propagation. Once transformed into the SS320 host, the DNA is resolved through DNA repair and replication, and the resulting library is packaged into phage particles.

Affinity panning of the drSH3 libraries

To develop specific binders for the receptor binding domain (RBD) of SARS-CoV-2 spike protein, phage affinity selection process was conducted by panning against recombinant RBD- mFc protein (produced in-house) using a standard solid phase sorting strategy. ¹³ ’ ¹⁴ The immobilized RBD-mFc ein (30 pg/ml in PBS; Maxisorp Immunotubes, Nunc) was incubated in the presence of infectious naive drSH3 phage library (in 2.5% milk-PBS-0.1%Tween20). Non- specific phages were removed by extensive washing (PBS-0.1% PBS-Tween), and the remaining pool of phage were eluted and amplified in E. coli XLl-Blue host according to standard protocols. ¹³ ’ ¹⁴ The amplified pool of phages was collected, and the process was reiterated over three rounds to enrich for a pool of phage-displayed drSH3-domains specific to the RBD target protein Wuhan-Hu-1 strain.

The amplified phage populations obtained after 2-3 rounds of affinity panning were tested for specific binding to RBD using a modification of the enzyme-linked immunosorbent assay (ELISA). In phage-ELISA the target protein and negative control protein (mFc alone) were immobilized on an immunoplate. Affinity-selected phage populations were then incubated with the RBD-mFc of mFc-coated control wells, and non-bound phages were removed by washing. A horseradish peroxidase (HRP) conjugated monoclonal antibody raised against the M13 phage particle was used to detect phage binding to the immobilized proteins. In addition, the phages were allowed to bind to plates coated with polyclonal anti-E-tag antibody (LifeSpan BioSciences) to normalize the quantity of phage displaying drSH3-E-tag-pIII fusion protein on their surface instead of total phage. Testing of phage binding to Fc-only protein again served as a negative control. Upon addition of the HRP substrate, binding of the clones was detected by a spectrophotometric readout.

More specifically, phage-ELISA was performed in 96-well Maxisorp microtiter plates (Nunc) coated overnight at 4°C with 100 μl of target and control proteins (1 pg/ml in PBS). The wells were washed 3 x with PBS-0.05% Tween20 and blocked with 5% skimmed milk powder in PBS (milk-PBS) for 2 h at RT. Appropriate dilutions of drSH3-displaying phage pools were prepared in milk-PBS and incubated with the coated target protein for 1 h at RT followed by washes 5 x with PBS-0.05% Tween20 to remove unbound phage. The detection was performed with HRP-conjugated mouse monoclonal anti-M13 antibody (GE Healthcare), and TMB (3,3' 5,5'-tetramethylbenzidine) substrate. The staining reaction was stopped with 1 M sulfuric acid and absorbance measured at 450 nm using Multiskan Ascent ELISA-reader (Thermo Fisher Scientific).

Analysis of individual anti-RBD drSH3 clones

Individual phagemid clones from panning rounds two to three (P2-P3) were randomly picked for further analysis. Homogenous phage-supernatants were produced from them, and tested in phage-ELISA for binding to RBD-mFc, control mFc protein, or monoclonal anti-E-tag antibody as described above for the phage population supernatants. The drSH3 inserts of the phagemid clones showing the most promising RBD-binding-capacity were sequenced to determine the amino acid sequences of their RT- and n-src loop regions.

Representative RBD-binding drSH3 clones were PCR amplified using the BamHI (sense) and Notl (antisense) restriction site-containing primers matching with their codon optimized NPHP1 SH3 backbone 5'-TTTTGGATCCATGGCCCAGGGCGCGCTG-3' (sense, SEQ ID NO:19) and 5'-TTTTGCGGCCGCTCAGGAATATGGTTCCAGATAG-3' (antisense, SEQ ID NO:20) and inserted in the corresponding cloning sites in the bacterial expression vector pGEX-4T-l (GE Healthcare, 28-9545-49) to be expressed as recombinant GST-drSH3 fusion proteins in the BL21(DE3) E. coli cells, and purified by affinity chromatography using Glutathione Sepharose 4B (GE Healthcare, 17-0756-01 according to the instructions provided by the vendor. The eluted GST-drSH3 fusion proteins were dialyzed overnight against PBS, and stored in aliquots at -70°C.

To evaluate the relative affinities by antigen capture-ELISA, the RBD-binding GST-drSH3 fusion proteins were coated on 96-well Maxisorp microtiter plates (Nunc) over night at 4°C with 100 μl (1.5-fold dilution series starting from 10 pg/ml in PBS), washed 3 x with PBS- 0.05% Tween20 and blocked with 5% skimmed milk powder in PBS (milk-PBS) for 2 h at RT. The wells were incubated for 1 h at RT with serially diluted RBD-His protein in PBS. The wells were then washed three times with PBS-0.05%Tween20, followed by incubation at RT for 60 min with HRP-conjugated anti-His (Thermo Fisher) diluted 1:5000 into PBS-0.1%Tween20. After four washes with PBS-0.1%Tween20 the binding was detected using and TMB (3,3' 5,5'- tetramethylbenzidine) substrate. The staining reaction was stopped with 1 M sulfuric acid and the absorbance was measured at 450 nm using Multiskan Ascent ELISA-reader (Thermo Fisher Scientific).

As described above, one of the strongest binders, ARM92, was further modified and produced as a trimeric GST fusion protein, namely ARM100, which contained three tandem copies of ARM92 interconnected with flexible 15-Gly-Ser linkers.

Neutralization assays using pseudovirus and clinival virus isolates

SARS-CoV-2 pseudovirus-mediated expression of luciferase is directly proportional to the quantity of internalized virus. To assess the inhibition activity, ARM92, ARM100 and appropriate control proteins were serially diluted in complete medium for desired concentrations. 12.5 μl of protein dilutions were mixed with 37.5 μl of luciferase encoding SARS-CoV-2 pseudotyped reporter viruses in 96-well cell culture plates and incubated at 37°C for 30 min. After incubation, 20 000 HEK-ACE2 cells (in 50 μl) were added on the wells and the plates were further incubated at 37°C for 48 h. The amount of internalized pseudovirus in infected cells was quantified by measuring luciferase activity using Renilla-GLO assay (Promega). The relative luciferase units were normalized to those of control samples. Half maximal inhibitory concentrations (IC ₅₀ ) were determined from three parallel experiments.

HEK293T and HEK 293T-ACE2 cells were maintained in DMEM supplemented with 10% fetal bovine serum, 2% L-Glutamine, and 1% penicillin/streptomycin (complete medium). Angiotensin-converting enzyme 2 (ACE2) expressing HEK293T cells (HEK-ACE2) were generated by lentivirus-mediated gene transduction. Briefly, pWPI-puro plasmid containing ACE2 cDNA (AB046569.1) was co-transfected with p8.9NdSB and vesicular stomatitis virus G protein (VSV-G) expressing envelope plasmids into HEK293T cells in complete medium using polyethylenimine. The recombinant lentivirus containing supernatant was collected 48 h post-transfection, filtered and used to infect wild-type HEK293T cells. Transduced cells were selected with puromycin.

Luciferase encoding SARS-CoV-2 pseudotyped reporter virus was generated by transfecting HEK293T cells with p8.9NdSB, pWPI-GFP expressing Renilla luciferase, and pCAGGS, an expression vector containing the SARS-CoV-2 S protein cDNA of the Wuhan-Hu-1 reference strain (NC_045512.2). The last 18 amino acids containing an endoplasmic reticulum (ER)- retention signal of the spike protein was removed to enhance transport to the plasma membrane. Pseudovirus stocks were harvested 48 hours after transfection, filtered and stored at -80°C. D614G (Wuhan-Hu-1 pseudovirus) mutation as well as RBD region mutations K417N, E484K, N501Y (B.1.351; beta/South Africa pseudovirus) mutations in SARS-Cov-2 S protein were generated with standard PCR techniques using synthetic DNA fragments (Integrated DNA Technologies).

A cytopathic effect (CPE)-based microneutralization (MNT) ¹⁵ was performed at the biosafety level 3 laboratory. Briefly, trivalent ARM100 in duplicates was 2-fold serially diluted in EMEM supplemented with penicillin, streptomycin and 2% of heat-inactivated fetal bovine serum. Virus was added to obtain 100 x TCID ₅₀ per well and the 96-well tissue culture plates were incubated for 1 h at +37°C, 5% CO2. African green monkey kidney epithelial (Vero E6) cells were added, following incubation at +37°C, 5% CO ₂ for 4 days. Wells were fixed with 30% formaldehyde and stained with crystal violet. Results were expressed as MNT titers corresponding to the reciprocal of the ARM100 dilution that inhibited 50% of SARS-CoV-2 infection observed by the CPE of inoculated cells.

Wild-type virus Finl-20 virus isolation and propagation were performed in Vero E6 cells. ¹⁵ Variant viruses were isolated and propagated (passages 1-2) in VeroE6-TMPRSS2-H10 cells ¹⁶ , and further propagated in Vero E6 cells (passage 3) for MNT.

Neutralization of SARS-CoV-2 VOCs in mice

We have recently shown that SARS-CoV-2 B.1.351 (beta) variant attains infectibility to female BALB/c mice, and causes pulmonary changes consistent with COVID-19. The B.1.351 (beta) variant used in the animal studies was isolated using transmembrane serine protease 2 (TMPRSS2) -transduced Vero E6 cells from SARS-CoV-2 infected patient nasopharyngeal samples as described in ^{10 ,17} .

Balb/c mice (Envigo) were transported to the University of Helsinki (Finland) biosafety level 3 (BSL-3) facility and acclimatized to individually ventilated biocontainment cages (ISOcage; Scanbur) for seven days with ad libitum water and food (rodent pellets). After the acclimatization period, 9 week old female Balb/c were placed under isoflurane anesthesia and intranasally inoculated with 25 μl per nostril of ARM100 (25 or 2.5 pg/nostril) followed by infection with 20 μl of SARS-CoV-2 B.1.351 clinical isolate (2x10 ⁵ PFU) 1, 4 or 8 h after. Immediately following the inoculation, the isoflurane was switched off and the animals were held in an upright position for a few seconds to allow the liquid to flush downwards in the nasal cavity. Their wellbeing was carefully monitored throughout the experiment for signs of illness (changes in posture or behaviour, rough coat, apathy, ataxia and weight loss), but none of the mice showed any clinical signs. Euthanasia was performed under terminal isoflurane anesthesia with cervical dislocation. Experimental procedures were approved by the Animal Experimental Board of Finland (license number ESA VI/28687/2020). Following euthanasia and dissections, RNA extractions from lung tissues were performed using Trizol (Thermo Scientific) according to the manufacturer's instructions. Isolated RNA was directly subjected to one-step RT-qPCR analysis using TaqMan fast virus 1-step master mix (Thermo Scientific) and AriaMx instrumentation (Agilent) as described previously for RdRp ¹⁰ , as well as for E and subE genes ¹¹ . The RT-qPCR for actin was conducted as previously described. ¹²

Results and conclusions

To select SARS-CoV-2 spike RBD-specific binders two different NPHPl-derived drSH3 phage display libraries (Library A and Library B) were screened using recombinant RBD-mFc fusion protein as an affinity target. Three panning rounds involving affinity selection followed by amplification of the selected phages were performed against RBD-mFc protein immobilized on plastic (immunotubes).

The progress of the panning procedure was monitored by calculating the enrichment ratio, i.e. the number of phages bound to immunotubes coated with the target protein divided by the number of phages bound to uncoated control tubes. In addition, the amplified pools of phages produced during the three rounds of panning were analyzed by phage-ELISA, where the binding of the phage supernatants were tested against 96-well plates coated with RBD-Fc, an irrelevant mFc-fusion protein, or a monoclonal antibody against an epitope tag present in all drSH3-displaying phages.

Based on these criteria the enrichment of RBD-specific clones started to increase already after the first round of panning. Altogether 192 different phage clones were picked from panning round 3, and were amplified and tested individually for RBD-specific binding using phage- ELISA as described above. 98% of clones were specific for RBD at the third round. Clones that bound to RBD-mFc but not to the control mFc-fusion protein were ranked according to their RBD binding profiles, and the residues in their RT- and n-src loop regions were determined.

DNA sequencing of 64 individual phagemid clones revealed a total of 28 unique RBD-binding drSH3 clones (ARM51-ARM62; ARM252; ARM254-259; ARM277; ARM291-ARM292; ARM294-ARM296; ARM315-ARM317). The translated amino acid sequences of these unique clones are shown below, and the randomized RT- and n-src loop regions providing them with RBD-specific affinity are indicated in bold. ARM53 (SEQ ID NO: 21) EEYIAVGDFWSISAEDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM56 (SEQ ID NO:22) EEYIAVGDFWTIDSADLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM57 (SEQ ID NO: 23) EEYIAVGDFWSMSLDDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM59 (SEQ ID NO:24) EEYIAVGDFWSMDSADLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM60 (SEQ ID NO: 25) EEYIAVGDFWSADRGDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM61 (SEQ ID NO: 26) EEYIAVGDFWSISSADLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM62 (SEQ ID NO:27) EEYIAVGDFWSMDVEDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM51 (SEQ ID NO:28) EEYIAVGDFWSNDYGDLTFKKGEILLVIEKKQKSPNLPDGWWIAKDAKGNEGLVPRTYLE PYS ARM52 (SEQ ID NO:29) EEYIAVGDFWSNSAGDLTFKKGEILLVIEKKTLDNGQPDGWWIAKDAKGNEGLVPRTYLE PYS ARM55 (SEQ ID NO: 30) EEYIAVGDFWSSDPLDLTFKKGEILLVIEKKPAETQMPDGWWIAKDAKGNEGLVPRTYLE PYS ARM58 (SEQ ID NO:31) EEYIAVGDFWSNDADDLTFKKGEILLVIEKKRSNNAFPDGWWIAKDAKGNEGLVPRTYLE PYS ARM54 (SEQ ID NO: 32) EEYIAVGDFFSTDPADLTFKKGEILLVIEKKFTPGGQPDGWWIAKDAKGNEGLVPRTYLE PYS ARM252 (SEQ ID NO: 56) EEYIAVGDFWSQDETDLTFKKGEILLVIEKQTTYGDGWWIAKDAKGNEGLVPRTYLEPYS ARM254 (SEQ ID NO:57) EEYIAVGDFWSNSQSDLTFKKGEILLVIEKKILADGRPDGWWIAKDAKGNEGLVPRTYLE PYS ARM255 (SEQ ID NO:58) EEYIAVGDFWSNSSADLTFKKGEILLVIEHSSPLGDGWWIAKDAKGNEGLVPRTYLEPYS ARM256(SEQ ID NO:59) EEYIAVGDFWSQDITDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM257 (SEQ ID NO: 60) EEYIAVGDFWSNDMGDLTFKKGEILLVIEKKLNGDGTPDGWWIAKDAKGNEGLVPRTYLE PYS ARM258 (SEQ ID NO: 61) EEYIAVGDFWSADSDDLTFKKGEILLVIEKEHEGSDGWWIAKDAKGNEGLVPRTYLEPYS ARM259(SEQ ID NO: 62) EEYIAVGDFWSSSSADLTFKKGEILLVIETSSLYGDGWWIAKDAKGNEGLVPRTYLEPYS ARM277 (SEQ ID NO: 63)

EEYIAVGDFWSQDKGDLTFKKGEILLVIEKKRLRNGSPDGWWIAKDAKGNEGLVPRT YLEPYS ARM291 (SEQ ID NO: 64) EEYIAVGDFWSQDKTDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM292 (SEQ ID NO: 65) EEYIAVGDFWSQDAGDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM294 (SEQ ID NO: 66) EEYIAVGDFWSNDPNDLTFKKGEILLVIEKKLQANGNPDGWWIAKDAKGNEGLVPRTYLE PYS ARM295 (SEQ ID NO: 67) EEYIAVGDFWSNSPIDLTFKKGEILLVIEKPEAFNDGWWIAKDAKGNEGLVPRTYLEPYS ARM296 (SEQ ID NO: 68) EEYIAVGDFWSNSPGDLTFKKGEILLVIEKKPWGGVSPDGWWIAKDAKGNEGLVPRTYLE PYS ARM315 (SEQ ID NO: 69) EEYIAVGDFWSQDSTDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS ARM316 (SEQ ID NO:70) EEYIAVGDFWSQDPYDLTFKKGEILLVIERRSAAGDGWWIAKDAKGNEGLVPRTYLEPYS ARM317 (SEQ ID NO:71) EEYIAVGDFWSQDNSDLTFKKGEILLVIEKKPDGWWIAKDAKGNEGLVPRTYLEPYS Ten of these 28 clones were derived from Library B, in which the randomized hexapeptide sequence was inserted in the n-src loop region without deleting any NPHP1 residues, whereas seven clones contained a modified RT loop and wild-type n-Src-loop. These clones can be originated from either Library A or B.

The engineered RT-loops in almost all clones shared clear similarity with each other, mostly conforming to the hexapeptide sequence consensus (W/F)SX(S/D)XX (in which X is any amino acid residue), whereas their engineered n-src loop sequences showed more diversity. Interestingly, in twelve of the SARS-CoV-2 spike glycoprotein targeting clones (ARM53, ARM56, ARM57, ARM59, ARM60, ARM61, ARM62, ARM265, ARM291, ARM292, ARM315, ARM317) only the RT-loop contained a modified sequence, whereas the n-src loop sequence in these clones had escaped randomized during the process of library drSH3 library generation, and contained wild-type human NPHP1 SH3 sequence. Thus, the (W/F)SX(S/D)XX consensus-containing sequences of the engineered RT-loops in the said twelve clones were sufficient to provide them with robust and specific SARS-CoV-2 spike glycoprotein binding capacity and the role of the modified n-src loops in the 16 other clones presumably was limited to assisting in binding to the SARS-CoV-2 spike glycoprotein.

As an affinity maturation initiative, one RBD-targeted clone (ARM54) with Library B-type n- Src-loop sequence was subjected to further sequence diversification to create two new RBD- biased phage libraries (Library C and Library D) in which the RT-loop region was kept constant, but the n-Src-loop region was fully randomized. The resulting new customized libraries C and D were pooled, and used for screening of additional RBD-specific binders, as described above for screening of the generic libraries A and B. Sequencing of the most promising clones from these screens led to the identification of 3 additional RBD-targeted drSH3 domains (ARM92, ARM102 and ARM103) shown below.

ARM92 ( SEQ ID NO : 33 ) EEYIAVGDFFSTDPADLTFKKGEILLVIERGTSAGDGWWIAKDAKGNEGLVPRTYLEPYS ARM102 ( SEQ ID NO : 34 ) EEYIAVGDFFSTDPADLTFKKGEILLVIETRVPEGDGWWIAKDAKGNEGLVPRTYLEPYS ARM103 ( SEQ ID NO : 35 ) EEYIAVGDFFSTDPADLTFKKGEILLVIEHNPHDGWWIAKDAKGNEGLVPRTYLEPYS

Thus, based on this work we identified 31 novel drSH3 clones that targeted the phages displaying them for robust binding against RBD of SARS-CoV-2 but not against irrelevant control proteins similarly expressed as recombinant mFc-fusion proteins (Figure 1). To confirm that these phage-displayed drSH3 domains are useful and retain their RBD- binding capacity also when expressed as recombinant proteins, we cloned 15 of them into the bacterial expression vector pGEX-4T-l to produce them as N-terminally GST-tagged proteins. These GST-drSH3 fusion proteins could be expressed at high levels in E. coli BL21(DE3) cells using standard conditions, and easily purified using glutathione-affinity chromatography with no apparent problems related to protein stability or aggregation. These proteins were next subjected to affinity comparison using a semi-quantitative RBD-capture ELISA. Each protein was immobilized to 96-well plates in varying concentrations and allowed to bind to soluble RBD-His protein at different dilutions. A robust and dose-dependent binding could be observed using an HRP-conjugated monoclonal antibody as a detector, indicating apparent KD values ranging from 60 to 30 nM for the RBD-binding GST-drSH3 proteins. The clone ARM92 was among the strongest binders in this assay, and was chosen for further development. To optimize the ability of ARM92 to bind to the trimeric spike and prevent SARS-CoV-2 internalization, we constructed a GST fusion protein, namely ARM100, which contained three tandem copies of ARM92 interconnected with flexible 15-Gly-Ser linkers.

The capacity of ARM100 as a neutralizing agent was first tested using a pseudovirus model widely used in SARS-CoV-2 research, which is based on luciferase expressing lentiviral vectors that enter ACE2-overexpresing HEK239 target cells in a Spike-dependent manner. ¹⁸ Pseudoviruses carrying the RBD mutations found in the Spike of Wuhan-Hu-1 wild type virus or the B.1.351 (beta) VOC strain were incubated with serially diluted concentrations of the trimeric ARM100 and monomeric ARM92. A uniform and strikingly potent neutralization was observed with ARM100, and subnanomolar half maximal inhibitory concentrations (IC ₅₀ ) were measured with both pseudoviruses (0.15 nM and 0.05 nM, respectively; Figure 2), whereas the corresponding IC ₅₀ values for the monomer were markedly higher 27.5 and 7 nM. Thus, the trimerization resulted in approximately 140- and 180-fold improvement in the neutralization efficiency of the Wuhan-Hu-1 and the B.1.351 (beta) pseudoviruses, respectively.

Subsequently, ARM100 was tested in a more traditional microneutralization test in VeroE6 cells using clinical SARS-CoV-2 VOC isolates. Confirming our pseudovirus model data, a dose- dependent and potent inhibition was observed. Complete (99.99%) neutralization was reached with all virus isolates tested, with the calculated IC ₅₀ values being 1.5, 0.6, and 1.2 nM, for B.l.1.7 (alpha), B.1.351 (beta) and for B.l.617.2 (delta) viruses, respectively. These data indicate that the trimeric ARM100 is a highly potent inhibitor of SARS-CoV-2 infection, and binds to the Spike RBD region in a favorable manner that is insensitive to the various combinations of immune escape mutations found in the relevant VOCs.

To evaluate the prophylactic efficiency of ARM100 in vivo, Balb/c mice were first intranasally administered with ARM100 in two different concentrations followed by viral challenge with SARS-CoV-2 B.1.351 beta variant (2x10 ⁵ PFU). Viral RNA in the mice lungs was measured 3 days after by quantitative real-time PCR (RT-qPCR). For the prophylactic groups, no viral RNA was detected in 100% of the mice when a standard dose 25 pg (n=9) or a low dose 2.5 pg (n=4) of ARM100 (average of 2.5 and 0.25 mg/kg, respectively) was delivered per nostril 1 h before viral infection as evidenced by RT-qPCR, whereas in control mice (n=10) the infection was evident (Figure 3). In parallel, the effective prophylactic time was evaluated in three different time points using the standard dose (2.5 mg/kg) of ARM100 per nostril 1, 4 or 8 h before viral challenge. Mice were also successfully protected in each group, as viral RNA was undetectable in 4/4 (1 h), 4/4 (4 h) and 3/4 (8 h) of the ARM100 pre-treated and infected mice. In ARM100 pretreated mice (Experiment 1), no histological abnormality was recognized. The mice showed no evidence of airway and/or pulmonary infection and no viral antigen was observed in the trachea or in the lungs, whereas in the control mice the infection was evident in nose and airways (trachea, bronchi and bronchioles) and in a few alveoli adjacent to the affected bronchioles. The changes were consistent with a mild bronchiolitis, with minimal alveolar changes. Together, intranasal administration of ARM100 provided comprehensive protection against infection by SARS-CoV-2 South African B.1.351 beta variant infection under the investigated conditions.

REFERENCES

Patent literature

US2018230456A1

Non-patent literature

1. Singh, J. et al. Structure-Function Analyses of New SARS-CoV-2 Variants B.l.1.7, B.1.351 and B.l.1.28.1: Clinical, Diagnostic, Therapeutic and Public Health Implications. Viruses 13 (2021).

2. Singh, J., Rahman, S.A., Ehtesham, N.Z., Hira, S. & Hasnain, S.E. SARS-CoV-2 variants of concern are emerging in India. Nat Med 27, 1131-1133 (2021).

3. Naaber, P. et al. Dynamics of antibody response to BNT162b2 vaccine after six months: a longitudinal prospective study. Lancet Reg Health Eur, 100208 (2021).

4. Hou, YJ. et al. SARS-CoV-2 Reverse Genetics Reveals a Variable Infection Gradient in the Respiratory Tract. Cell 182, 429-446 e414 (2020).

5. Baum, A. et al. REGN-COV2 antibodies prevent and treat SARS-CoV-2 infection in rhesus macaques and hamsters. Science 370, 1110-1115 (2020).

6. Iwasaki, A. Exploiting Mucosal Immunity for Antiviral Vaccines. Annu Rev Immunol 34, 575-608 (2016).

7. Simeon, R. & Chen, Z. In vitro-engineered non-antibody protein therapeutics. Protein & cell 9, 3-14 (2018).

8. Hiipakka, M., Poikonen, K. & Saksela, K. SH3 domains with high affinity and engineered ligand specificities targeted to HIV-1 Nef. Journal of molecular biology 293, 1097-1106 (1999).

9. Hiipakka, M. & Saksela, K. Versatile retargeting of SH3 domain binding by modification of non-conserved loop residues. FEBS letters 581, 1735-1741 (2007).

10. Virtanen, J. et al. Kinetics of Neutralizing Antibodies of COVID-19 Patients Tested Using Clinical D614G, B.l.1.7, and B 1.351 Isolates in Microneutralization Assays. Viruses 13 (2021).

11. Dagotto, G. et al. Comparison of Subgenomic and Total RNA in SARS-CoV-2 Challenged Rhesus Macaques./ Virol (2021).

12. Zivcec, M., Safronetz, D., Haddock, E., Feldmann, H. & Ebihara, H. Validation of assays to monitor immune responses in the Syrian golden hamster (Mesocricetus auratus)./ Immunol Methods 368, 24-35 (2011).

13. Sidhu, S.S., Lowman, H.B., Cunningham, B.C. & Wells, J.A. Phage display for selection of novel binding peptides. Methods Enzymol 328, 333-363 (2000).

14. Viti, F., Nilsson, F., Demartis, S., Huber, A. & Neri, D. Design and use of phage display libraries for the selection of antibodies and enzymes. Methods Enzymol 326, 480-505 (2000).

15. Haveri, A. et al. Serological and molecular findings during SARS-CoV-2 infection: the first case study in Finland, January to February 2020. Euro Surveill 25 (2020).

16. Rusanen, J. et al. A Generic, Scalable, and Rapid Time-Resolved Forster Resonance Energy Transfer-Based Assay for Antigen Detection-SARS-CoV-2 as a Proof of Concept. mBio 12 (2021).

17. Cantuti-Castelvetri, L. et al. Neuropilin-1 facilitates SARS-CoV-2 cell entry and infectivity. Science 370, 856-860 (2020).

18. Crawford, K.H.D. et al. Protocol and Reagents for Pseudotyping Lentiviral Particles with SARS-CoV-2 Spike Protein for Neutralization Assays. Viruses 12 (2020).

19. Tonikian, R., Zhang, Y., Boone, C. et al. Identifying specificity profiles for peptide recognition modules from phage-displayed peptide libraries. Nat Protoc 2, 1368- 1386 (2007).