NeutrobodyPlex-monitoring SARS-CoV-2 neutralizing immune responses using nanobodies

Abstract

In light of the COVID-19 pandemic, there is an ongoing need for diagnostic tools to monitor the immune status of large patient cohorts and the effectiveness of vaccination campaigns. Here, we present 11 unique nanobodies (Nbs) specific for the SARS-CoV-2 spike receptor-binding domain (RBD), of which 8 Nbs potently inhibit the interaction of RBD with angiotensin-converting enzyme 2 (ACE2) as the major viral docking site. Following detailed epitope mapping and structural analysis, we select two inhibitory Nbs, one of which binds an epitope inside and one of which binds an epitope outside the RBD:ACE2 interface. Based on these, we generate a biparatopic nanobody (bipNb) with viral neutralization efficacy in the picomolar range. Using bipNb as a surrogate, we establish a competitive multiplex binding assay ("NeutrobodyPlex") for detailed analysis of the presence and performance of neutralizing RBD-binding antibodies in serum of convalescent or vaccinated patients. We demonstrate that NeutrobodyPlex enables high-throughput screening and detailed analysis of neutralizing immune responses in infected or vaccinated individuals, to monitor immune status or to guide vaccine design.

Keywords: SARS-CoV-2; immune response; nanobodies; neutralizing antibodies; serological assay.

Figure 1. Generation of nanobodies blocking the SARS‐CoV‐2 RBD:ACE2 interface

Nanobodies (Nbs) are genetically engineered from heavy‐chain only antibodies of alpacas. The interaction between the SARS‐CoV‐2 homotrimeric spike protein and angiotensin‐converting enzyme (ACE) 2 can be blocked by receptor‐binding domain (RBD)‐specific Nbs in the monovalent or biparatopic format.

Figure 2. Biochemical characterization of RBD‐specific Nbs

1. Amino acid sequences of the complementarity determining region (CDR) 3 from unique Nbs selected after two rounds of biopanning are listed (upper panel). Phylogenetic tree based on a ClustalW alignment of the Nb sequences is shown (lower panel).

2. Recombinant expression and purification of Nbs using immobilized metal affinity chromatography (IMAC) and size exclusion chromatography (SEC). Coomassie staining of 2 µg of purified Nbs is shown.

3. For biolayer interferometry (BLI)‐based affinity measurements, biotinylated RBD was immobilized on streptavidin biosensors. Kinetic measurements were performed by using four concentrations of purified Nbs ranging from 15.6 nM to 2 µM. As an example, the sensogram of NM1230 at indicated concentrations is shown (upper panel). The table summarizes affinities (K _D), association (k _on), and dissociation constants (k _off) determined for individual Nbs (lower panel).

Figure EV1. ACE2‐competing Nbs identified by multiplex binding assay

Results from multiplex ACE2 competition assay are shown for the three spike‐derived antigens: RBD, S1‐domain (S1) and homotrimeric spike (Spike). Nbs were diluted from 2.1 µM to 0.12 nM in the presence ACE2 and antigen‐bound ACE2 was measured. MFI signals were normalized to the maximum detectable signal per antigen given by the ACE2‐only control. IC₅₀ values were calculated from a four‐parametric sigmoidal model. Data are presented as mean ± s.d. of three technical replicates.

Figure EV2. Epitope mapping of Nbs by HDX mass spectrometry

Surface structure model of RBD showing the ACE2 interface and the HDX‐MS epitope mapping results of RBD epitopes protected upon Nb binding are highlighted in different colors indicating the strength of protection. Accordingly, amino acid residues, which are part of the Nb recognized epitopes, are highlighted in the RBD sequence.

1. Numbered amino acid residues of RBD (pdb code: 6M17 (Yan et al, 2020)) involved in the RBD:ACE2 interaction site (Lan et al, 2020; Yan et al, 2020) are shown in red.

2. NM1226 (Nb‐Set1)

3. NM1228 (Nb‐Set1)

4. NM1230 (Nb‐Set2)

5. NM1221 (Nb‐Set2)

6. NM1222 (Nb‐Set2)

7. NM1224 (Nb‐Set4)

8. NM1223 (Nb‐Set3)

Figure 3. NM1226 and NM1230 binding epitopes on RBD

1. A

   Overview of the binding epitopes of NM1226 (light green) on RBD (dark green) and NM1230 (magenta) on RBD (light pink), respectively.

2. B, C

   The binding interfaces of NM1226 (B) and NM1230 (C) with the RBD are shown as surface representation (upper panels) and in a close‐up view as balls and sticks with direct hydrogen bonds and salt bridges indicated as dotted lines (lower panels). All residues involved in contact formation at a distance cutoff of < 4 Å are listed and labeled. A close‐up view of both interfaces reports a detailed interaction map of both contact areas.

Figure EV3. Mechanism of ACE2 attachment inhibition by NM1226 and NM1230

1. A, B

   (A) Top and (B) side view of the SARS‐CoV‐2 spike protein (gray) in complex with the ACE2 receptor (yellow, orange, dark orange) (pdb code: 7KMS) superposed on the RBD:Nb complexes NM1226 (light green) and NM1230 (magenta). A close‐up view shown in black boxes elucidates how both Nbs block ACE2 receptor attachment.

Figure 4. Biparatopic NM1267 competes with ACE2 and neutralizes SARS‐CoV‐2 infection

1. Coomassie staining of 1 µg of purified biparatopic Nb NM1267.

2. Table summarizing the affinity (K _D), association (k _on), and dissociation constants (k _off) determined by BLI and IC₅₀ values of the multiplex ACE2 competition assay and virus neutralization assay obtained for NM1267.

3. Sensogram of affinity measurements via BLI using four concentrations (6.25–50 nM) of NM1267.

4. Results from multiplex ACE2 competition assay are shown for the three spike‐derived antigens: RBD, S1‐domain (S1), and homotrimeric spike (Spike). Color‐coded beads coated with the respective antigens were co‐incubated with biotinylated ACE2 and dilution series of NM1267 (8 pM to 126 nM) followed by measuring residual binding of ACE2. MFI signals were normalized to the maximum detectable signal per antigen given by the ACE2‐only control. IC₅₀ values were calculated from a four‐parametric sigmoidal model. Data are presented as mean ± s.d. of three technical replicates.

5. Neutralization potency of NM1267 was analyzed in Caco‐2 cells using the SARS‐CoV‐2‐mNG infectious clones. Infection rate normalized to virus‐only infection control is illustrated as percent of infection (% Infection). IC₅₀ value was calculated from a four‐parametric sigmoidal model, and data are presented as mean ± s.e.m. of three biological replicates (n = 3).

Figure 5. NeutrobodyPlex: multiplex competitive binding assay to monitor a neutralizing immune response in patients

1. Schematic illustration of the NeutrobodyPlex. The displacement of serum‐derived neutralizing IgGs binding to SARS‐CoV‐2 antigens upon addition of bipNb is measured. In presence of neutralizing IgGs, the fluorescent signal from anti‐human‐IgG‐PE is inversely proportional to the applied bipNb concentration.

2. For the NeutrobodyPlex, antigen‐coated beads comprising RBD, S1, or spike were co‐incubated with serum samples from five patients and a dilution series of NM1267 (1 µM to 6 pM) (n = 1). Mean fluorescent intensities (MFI) derived from antigen‐bound IgGs in the presence of bipNb normalized to the MFI values of IgGs in the serum‐only samples, illustrated as MFI (% control), are shown.

3. For two patient‐derived serum samples (#289, #265), differences in the NM1267‐mediated displacement of IgGs binding the three spike‐derived antigens (RBD, S1, spike) are shown.

Figure 6. Specificity of the NeutrobodyPlex in determining individual neutralizing immune responses

Serum samples of 18 convalescent SARS‐CoV‐2‐infected individuals were analyzed using the NeutrobodyPlex and the virus neutralization assay.

1. For the NeutrobodyPlex on RBD, two concentrations of NM1267 (1 µM and 1 nM) were applied (n = 1). Light colored squares (high MFI (%control)) are indicative for IgGs outcompeting NM1267 from the RBD:ACE2 interface; dark colored squares (low MFI (%control)) show a continuous displacement of IgGs from serum samples in the presence of NM1267.

2. For the virus neutralization assay, serial dilutions of the serum samples (1:40 – 1:5,120) were applied (n = 1). Dark colored squares are indicative for a low infection level (low % Infection); light colored squares show a high infection rate (high % Infection).

3. The mean percent of infection (% Infection) derived from all individual serum dilutions obtained by the virus neutralization assay was calculated and plotted against the respective MFI (% control) obtained from the NeutrobodyPlex on RBD in the presence of NM1267 (1 nM). The Pearson correlation determined a negative correlation with r = −0.77 and R ² = 0.59 (P < 0.05).

Figure EV4. NeutrobodyPlex enables a differentiated analysis on neutralizing IgGs

Serum samples of 112 convalescent SARS‐CoV‐2‐infected individuals were analyzed using the NeutrobodyPlex with antigen‐coated beads comprising RBD, S1, spike, S2‐domain (S2), and nucleocapsid (N) and two concentrations of NM1267 (n = 1). Light colored squares (high MFI (%control)) are indicative for IgGs outcompeting NM1267 from the RBD:ACE2 interface; dark colored squares (low MFI (% control)) show a continuous displacement of IgGs from serum samples in the presence of NM1267.

1. NeutrobodyPlex NM1267 1 µM.

2. NeutrobodyPlex NM1267 1 nM.

Figure 7. NeutrobodyPlex enables a differentiated analysis on neutralizing IgGs compared with total SARS‐CoV‐2 binding IgGs in individuals

To investigate neutralizing capacities in relation to total levels of SARS‐CoV‐2‐specific IgGs in individuals, serum samples of 112 convalescent SARS‐CoV‐2 infected individuals were analyzed using the NeutrobodyPlex (n = 1). Individuals displaying strong neutralizing IgGs but showed an overall low level of SARS‐CoV‐2‐specific antibodies are highlighted with light‐blue background.

1. A, B

   Total IgGs derived from spike‐binding IgGs were plotted against normalized MFI values from IgGs binding to RBD (MFI RBD (%control)) in the presence of both concentrations of NM1267 (1 µM; 1 nM).

2. C, D

   Total IgGs derived from N‐binding IgGs were plotted against normalized MFI values from IgGs binding to RBD (MFI RBD (%control)) in the presence of both concentrations of NM1267 (1 µM; 1 nM).