Nanobodies: Two Studies Find Tiny Segments of Antibodies for Neutralizing SARS-CoV-2 Virus

• One Study Used Immunization of Llamas to Isolate Nanobodies 
• Another Study Used Phage-Display to Isolate Nanobodies 
• Both Studies Engineered Nanobody Multimers to Increase the Potency of Nanobodies 

In the November 5, 2020 issue of Science, I read about two investigations that independently discovered different nanobodies (Nbs) that neutralized SARS-CoV-2 virus, the causative agent of COVID-19. In both cases, the Nbs tightly bound to the virus’s external Spike protein, blocking entry into host cells. Interestingly, each study found the tightest binding Nbs by using fundamentally different methods. As will be outlined here, Xiang et al. used immunization of a llama, while Schoof et al. screened a yeast surface-displayed library of synthetic Nb sequences.

Before reading these two papers, the Zone was unaware of Nbs and subsequently searched the literature for background information. In doing so, the Zone learned that Nbs are remarkably small molecular relatives of much larger antibodies. Moreover, these markedly downsized antibodies have attracted considerable attention since they were first reported in Nature by Hamers-Casterman et al. in 1993, as indicated by the chart of PubMed articles shown here.

Nanobodies

Structure: The most cited review of Nbs was published in 2013 and titled Nanobodies: Natural Single-Domain Antibodies. It was written by Serge Muyldermans, who was involved in the pioneering discovery of Nbs and continues that work as a professor at Vrije Universiteit Brussel, Belgium. The Zone’s key takeaways from this comprehensive review begin with the discovery of Nbs as structurally unique components of the immune system of the biological family Camelidae, which comprises of camels (one- and two-humped), llamas, and alpacas.

Although all camelids have conventional IgG1 antibodies similar in structure to those of humans, these animals also have IgG2 and analogous IgG3 antibodies, which lack light chains and, upon protease digestion, yield the smallest intact single-domain functional antigen-binding fragment, termed VHH, as depicted here. This ~15-kDa VHH fragment is ~10-times smaller than IgG1, which is what led to VHHs being labeled as Nanobodies® in 2003 by Ablynx, then a new biotechnology company, as discussed in a review by Siontorou. Since then, this term and its singular form have been universally used in place of VHHs and VHH, without inclusion of the registered sign.

Production: According to a 2018 review by Schumacher et al., Nbs (and other immunoglobulin-based recombinant antigen-binding proteins) can be generated by immunizing a camelid with the antigen of interest. Typically, up to six injections of ~0.5 mg of antigen or immobilized antigen (e.g., a conjugate with bovine serum albumin) are performed within several weeks. The mRNA is isolated from lymphocytes and its complementary DNA (cDNA) is synthesized using reverse transcriptase. Next, the specific segment encoding the VHH domain is amplified and potent binders isolated, most commonly by phage display (depicted here), which is capable of screening up to 10¹¹ sequences per library.

Schumacher et al. add that alternative screening strategies include yeast and bacterial display, in which VHH binders can be selected by multiparameter and quantitative flow cytometry, as well as mRNA display and ribosome display, both of which are well suited for the selection of large libraries of up to 10¹⁵ sequences. Once a potent Nb is selected, it can be readily expressed in high yields of up to several g/L in E. coli, S. cerevisiae, etc.

Importantly, Doshi et al. at the University of California, San Diego, described the first adaptation of mRNA display for the rapid, automatable discovery of Nbs against desired targets. They then used this adaptation to discover the first-ever reported Nb against the human full-length glucose transporter, GLUT-1, as a proof-of-concept. The researchers envision their “streamlined method as a bench-top platform technology, in combination with various molecular evolution techniques, for expedited Nb discovery.” This breakthrough development was enabled by use of

modified nucleotide 5’-triphosphates obtained from TriLink. Briefly, in order to synthesize a mutagenic library based on a Nb DNA template, error-prone PCR was performed using the 5’-triphosphates of 6-(2-deoxy-b-D-ribofuranosyl)-3,4-dihydro-8H-pyrimido-[4,5-C>

[1,2>

oxazin-7-one (2'-deoxy-P-nucleoside, dPTP) and of 8-oxo-2’-deoxyguanosine (8-oxo-dGTP) in the total dNTP mix, for promiscuous incorporation.

Applications: The following is largely taken from an excellent 2018 review by Schumacher et al. titled Nanobodies: Chemical Functionalization Strategies and Intracellular Applications. Advantageous properties of Nbs for advanced applications in molecular biology include high thermal and conformational stability, which allow for the retention of binding activity after prolonged incubation at elevated temperatures, high salt concentrations, and under different pH conditions. Moreover, Nbs are able to efficiently refold and fully restore their antigen affinity after thermal denaturation, thus opening novel opportunities for studying the dynamics of protein folding. These robust properties allow Nbs to capture their respective antigens in vitro as well as in vivo.

While Nbs used in imaging have been traditionally expressed as fusions with fluorescent proteins, such as green fluorescent protein (GFP), depicted here, labelling with small organic probes is expanding the scope of utility of Nbs as tools for biological research. According to Schumacher et al., the chemical labelling of Nbs with fluorophores, immobilization on solid supports, or functionalization with recognition motifs, delivery agents, and other chemical groups broadly expands their applicability for imaging, proteomics, and novel therapeutic tools. 

In initial proof-of-principle studies, Nbs were randomly labelled at solvent-exposed lysine residues by N-hydroxysuccinimide (NHS) ester-containing fluorophores to give heterogeneous protein mixtures, as the result of multiple available lysine residues. Increased homogeneity of protein conjugates can be achieved by using cysteine residues, which are less abundant compared to lysine. To bypass heterogeneity due to multiple cysteine residues resulting from disulfide reduction, a single cysteine can be introduced at the C terminus of the Nb, thus ensuring that the conjugation site is most distal from the antigen-binding interface. This approach has been used to covalently attach polyethylene glycol (PEG) linkers to the C terminus of single-cysteine Nbs, prolonging in vivo circulation time.

In addition, there are various chemoenzymatic site-specific labelling methods that use bacterial biotin ligases, transglutaminases, transpeptidase Sortase A, or the lipoic acid ligase LpIA. These enzymes, as well as various strategies for direct cellular delivery of Nbs – a long-standing goal in cell biology and medicine – can be read about in detail in the review by Schumacher et al.

SARS-CoV-2 Neutralizing Nbs via Llama Immunization

SARS-CoV-2 expresses a surface Spike (S) glycoprotein, depicted here, consisting of of S1 and S2 subunits that form a homotrimeric viral spike-like structure to interact with host cells. The interaction is mediated by the outermost S1 receptor-binding domain (RBD), which binds the peptidase domain (PD) of angiotensin-converting enzyme-2 (hACE2) as a host receptor. 

To produce high-quality SARS-CoV-2 neutralizing Nbs, Xiang et al. immunized a llama with the recombinant RBD. The llama serum efficiently neutralized “pseudotyped SARS-CoV-2” (see Xiang et al. Supplementary Materials and FDA FAQs) at the half-maximal neutralization titer (NT₅₀) of ~310,000 orders of magnitude higher than the convalescent sera obtained from recovered COVID-19 patients. To further characterize these initial activities, the single-chain VHH antibodies were separated from the IgGs, which led to confirmation that the single-chain antibodies achieve specific, high-affinity binding to the RBD and possess sub-nM half-maximal inhibitory concentration (IC₅₀ = 509 pM) against the pseudotyped virus.

Thousands of high-affinity VHH Nbs from the RBD-immunized llama serum were then identified using a mass spectrometric-based proteomic strategy. These Nbs included ~350 structures with unique CDR3s (complementarity-determining regions), from which 109 highly diverse Nb sequences were selected to cover various biophysical, structural, and potentially different antiviral properties, based on the presence of unique CDR3s. Of these, 94 Nbs were purified and tested for RBD binding by ELISA, which led to 49 Nbs with high-solubility and high-affinity (ELISA IC₅₀ below 30 nM), promising candidates for the screening and characterization of antiviral activities in several SARS-CoV-2 pseudovirus neutralization assays.

The resultant 3 most potent candidates, Nbs 89, 20, and 21, showed IC₅₀ neutralization values of 0.133 nM, 0.102 nM, and 0.045 nM, respectively, in the SARS-CoV-2 pseudovirus assay and approximately the same values in the SARS-CoV-2 assay. The thermostability of Nbs 89, 20, and 21 was determined to be 65.9, 71.8, and 72.8°C, respectively, and the shelf-life of Nb 21 at room temperature was >6 weeks. 

Importantly, Xiang et al. found up to a ~30 fold improvement for the homotrimeric multimer (as depicted here, n = 1) of Nbs 21 and 20, termed Nb 21₃ (IC₅₀ = 1.3 pM) and Nb 20₃ (IC₅₀ = 4.1 pM), respectively, compared to the corresponding monomeric form, in the SARS-CoV-2 pseudovirus assay. This led Xiang et al. to suggest that “a cocktail consisting of ultrapotent, multivalent constructs that bind simultaneously a variety of epitopes with potentially different neutralization mechanisms will likely efficiently block virus mutational escape.”

SARS-CoV-2 Neutralizing Nbs via Yeast Surface-Display

In contrast to llama immunization, Schoof et al. isolated Nbs that neutralized SARS-CoV-2 by in vitro screening a yeast surface-displayed library (see above depiction) of >2 x 10⁹ synthetic Nb sequences for binders to the a mutant form of SARS-CoV-2 Spike (SpikeS2P) as the antigen. SpikeS2P lacks one of the two proteolytic cleavage sites between the S1 and S2 domains, and introduces two mutations and a trimerization domain to stabilize the pre-fusion conformation. The researchers labeled SpikeS2P with biotin or with fluorescent dyes and selected Nb-displaying yeast over multiple rounds, first by magnetic bead binding and then by fluorescence-activated cell sorting (FACS), a widely applicable methodology depicted here. 

Three rounds of selection yielded 21 unique Nbs that bound SpikeS2P and showed decreased binding (i.e. competitive inhibition) in the presence of a dimeric construct of the ACE2 extracellular domain (ACE2-Fc) target. These Nbs fall into two classes. Class I binds the RBD and directly competes with ACE2-Fc, whereas Class II binds to SpikeS2P but displays no binding to RBD alone. According to Schoof et al. these findings, suggest that Class I Nbs target the RBD to block ACE2 binding, whereas Class II Nbs target other epitopes, as supported by surface plasmon resonance (SPR) experiments that demonstrated that Class I and Class II nanobodies can bind SpikeS2P simultaneously.

Next, Schoof et al. prioritized two Class I Nbs, Nb 6 and Nb11, that combine potent SpikeS2P binding with relatively small differences in faster association rate constant (k_a) between binding to SpikeS2P or RBD. For Class II nanobodies, they prioritized Nb 3 because of its relatively high yield during purification. To define the binding sites for Nb 6 and Nb 11, they were bound to SpikeS2P and the structures were determined by cryogenic electron microscopy (cryo-EM), a now widely employed method, for which the 2017 Chemistry Nobel Prize was awarded.

The structure of Nb 6 bound to closed SpikeS2P enabled Schoof et al. to engineer bivalent and trivalent Nbs predicted to lock all RBDs in the down-state via significantly slowed dissociation rates due to enhanced avidity. To do so, they inserted flexible Gly-Ser linkers (shown here) of either 15 or 20 amino acids to span the 52 Å distance between adjacent Nb 6 monomers bound to down-state RBDs in closed SpikeS2P. These linkers are too short to span the 72 Å distance between Nb 6 molecules bound to open Spike. SPR experiments demonstrated that bivalent Nb 6 with a 15 amino acid linker (Nb 6₂) and trivalent Nb 6 with two 20 amino acid linkers (Nb 6₃) afforded 750-fold and >200,000-fold gains in K_D, respectively, consistent with the expected slower dissociation rates.

Schoof et al. next tested the neutralization activity of monovalent and trivalent versions of their top Class I (Nb 6 and Nb 11) and Class II (Nb 3) Nbs against SARS-CoV-2 pseudotyped lentivirus, using a previously described assay. Compared to Nb 6, trivalent Nb 6₃ showed a 2000-fold enhancement of inhibitory activity, with an IC₅₀ of 1.2nM, whereas trimerization of Nb 11 and Nb 3 resulted in more modest gains of 40- and 10-fold (51 nM and 400 nM), respectively. These neutralization activities were confirmed with a viral plaque assay using live SARS-CoV-2 virus infection of VeroE6 cells, in which Nb 6₃ proved exceptionally potent, neutralizing SARS-CoV-2 with an average IC₅₀ of 160 pM, while Nb 3₃ neutralized SARS-CoV-2 with an average IC₅₀ of 140 nM. 

Schoof et al. further optimized the potency of Nb 6 by selecting a saturation mutagenesis library targeting all three CDRs. Two rounds of selection identified high-affinity mutations that were then incorporated into Nb 6 to generate matured Nb 6 (mNb 6), which binds with a 500-fold increased affinity to SpikeS2P and inhibits both pseudovirus and live SARS-CoV-2 infection with low nanomolar potency, a ~200-fold improvement compared to Nb 6. Based on the cryo-EM structure for bound mNb 6, trivalent mNb 6 with a 20 amino acid linker (mNb 6₃) was engineered, and was found to give gains in potency in both pseudovirus and live SARS-CoV-2 infection assays with IC₅₀ values of 120 pM and 5pM, respectively, leading Schoof et al. to describe mN 6₃ being an “exceptionally potent” SARS-CoV-2 neutralizing molecule. Equally important, this trimer was stable to lyophilization, aerosolization, and heat (1 hour at 50°C).

Concluding Comments

Strategies to prevent SARS-CoV-2 entry into its host cell aim to block the ACE2-RBD interaction. Although high-affinity monoclonal antibodies are leading the way as potential therapeutics, they are expensive to produce by mammalian cell expression and need to be intravenously administered by healthcare professionals, according to Schoof et al. In addition, they say, large doses are needed for prophylactic use, as only a small fraction of systemic antibodies cross the epithelial cell layers lining the airway. By contrast, Nbs can be inexpensively produced in bacteria or yeast. The inherent stability of Nbs enables aerosolized delivery directly into the nasal and lung epithelia, which has been demonstrated by delivery of a trimeric Nb targeting respiratory syncytial virus to decrease measurable viral load in hospitalized infants. 

In both reports featured in this blog, it is evident that multimers of Nbs in the form of bi- and trivalent Nbs can result in remarkable increases in potency in anti-SARS-CoV-2 assays, which bodes well for future translation into clinical effectiveness. While the arguments presented for Nbs being superior to antibodies may ultimately prove to be true, in practice, they need to be juxtaposed to the amazing preliminary protection-efficacies recently reported by both BioNTech/Pfizer and Moderna for Spike-protein mRNA-based vaccines against COVID-19. Given the current dire global situation caused by the COVID-19 pandemic, the more approaches there are for prevention and treatment of the virus, the better. The Zone is mightily impressed by the promise of Nbs, and looks forward to future clinical investigations with these tiny segments of antibodies.

With all these, and other modalities of combating COVId-19, there does indeed seem to be growing glimmers of the light at the end of the tunnel.

What do you think?

As usual, your comments are welcomed.