To construct the expression plasmids for SARS-CoV-2 spike protein, a gene encoding residues 1−1208 of the spike ectodomain with a mutation at the furin cleavage site (residues 682-685) from RRAR to GSAS, proline substitutions at residues 986 and 987, followed by the T4 fibritin trimerization domain, a HRV3C protease cleavage site, a twin Strep Tag and an 8XHisTag, was synthesized and optimized for mammalian expression (Wrapp et al., 2020). An optimized coding sequence was cloned into the mammalian expression vector pHLsec.

Expression plasmids were constructed using synthetic fragments coding for human codon-optimized Spike glycoprotein sequences from CoV-229E (GenBank accession number NC_002645.1; amino acids 1–1113), CoV-HKU1 (GenBank accession number NC_006577.2; amino acids 1-1300), CoV-NL63 (GenBank accession number NC_005831.2; amino acids 1–1289), CoV-OC43 (GenBank accession number NC_006213.1; amino acids 1–1297), CoV-MERS (GenBank accession number AFS88936.1; amino acids 1-1291) (Zhao et al., 2013), CoV-SARS1 (GenBank accession number AY27874; amino acids 11-1195) (Simmons et al., 2004) and CoV-SARS2 (GenBank accession number MN908947; amino acids 1-1208). Fragments were cloned in pHLsec vectors downstream of the chicken β-actin/rabbit β-globin hybrid promoter and followed by a T4 fibritin trimerization domain, an HRV 3C cleavage site, a His-8 tag and a Twin-Strep-tag at the C terminus as previously reported by Wrapp et al. (2020).

Mutations coding for stabilizing proline residues and to eliminate putative furin cleavage sites were inserted in each sequence as follows: For CoV-229E, TI > PP (aa 871-872); for CoV-HKU1, RRKR > GSAS (aa 756-759) and AL > PP (aa 1071-1072); for CoV-NL63, RRSR > GSAS (aa 754-757) and SI > PP (aa 1052-1053); for CoV-OC43, AL > PP (aa 1070-1071); for CoV-MERS, RSVG > ASVG (aa 748), RSAR > GSAS (aa 884-887) and VL > PP 1060-1061; for CoV-SARS1, KV > PP (aa 968-969); for CoV-SARS2, RRAR > GSAS (aa 682-685) and KV > PP (aa 986-987). All sequences were verified by DNA sequencing.

DNA plasmids encoding the Strep-Tag-tagged spike proteins were transfected into HEK293T cells cultured in FreeStyle 293 Expression Medium (12338018, ThermoFisher) by PEI-mediated transfection (MW: 25,000; branched: Sigma-Aldrich 408727) and incubated at 37 °C for 7 days. Supernatants were then collected and cleared by centrifugation followed by filtration. CoV Spike protein trimers were affinity-purified using the Strep-Tactin®XT purification system (IBA Lifesciences) according to the instructions of the manufacturer. In the case of CoV-229E and CoV-NL63, the spike proteins were further purified by SEC (Superose 6 increase 30/100 GL column, GE Life Sciences; elution buffer: Tris 20mM, NaCl 150mM, pH 7) to remove aggregates. The purity of the proteins was assessed by reducing (10% β-mercaptoethanol (β-ME)) and non-reducing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (~3 μg of protein). Purified proteins were concentrated in PBS, quantified by spectrophotometry, sterilized by filtration (Spin-X tube filter; 8160; Costar) and kept at −80 °C until use.

The native SARS-CoV-2 nucleoprotein gene was cloned into a pET28a(+) vector (Novagen) downstream of the coding sequence for an N-terminal hexa-histidine tag and 3C-protease cleavage site (a gift from Fred Antson). Expression was carried out using 2-L cultures of Rosettagami2(DE3)pLysS bacteria (Novagen) in terrific broth medium containing 40 mg/L kanamycin, at 15°C for 40 hr following induction with Isopropyl β-D-1-thiogalactopyranoside (1mM final concentration, Meridian Bioscience). Upon centrifugation (10,000 x g; 20 minutes, 4°C), pellets were resuspended in 60 mL H2O containing 10 mg/ml lysozyme (Sigma). After adding 70 mL buffer S (200 mM Tris pH 8.0, 2.5 M NaCl, 60 mM imidazole, 4 mM MgSO4, 0.2% Triton X-100) the suspension was sonicated (40% amplitude, 10 s on-10 s off cycles, 20 min, 4°C). Turbonuclease (3,000 units, Sigma) and RNaseA (500 units, QIAGEN) were added, and the solution was clarified (20,000 x g, 30 min, 4°C) before purification over a 5 mL HisTrap column (Cytiva), using a 20 mM to 1 M imidazole gradient in 25 mM Tris pH 8.0, 1.5 M NaCl. Nucleoprotein-containing fractions were further purified over a Superdex 200 Increase 10/300 GL column (Cytiva) using a 25 mM Tris pH 8.0, 1 M NaCl running buffer, followed by buffer exchange into phosphate-buffered saline (PBS, Sigma) using PD-10 columns (Cytiva), and heparin-affinity chromatography using a 5 mL HiTrap heparin HT column (Cytiva) and a 0.15 - 1 M NaCl gradient in 40 mM sodium phosphate pH 7.4.

Nickel charged agarose beads (nickel-nitrilotriacetic acid [Ni-NTA]; QIAGEN) were washed 3 times in PBS and then incubated overnight, rotating at 4°C, with His-tagged RBD. Twenty micrograms of protein were added for every 50 μL of beads used in a final incubation volume, twice the bead volume. Beads incubated in the absence of RBD antigen were used as a beads-only, mock control. The beads were then washed 3 times with PBS and precleared for 2h at RT with a pooled SARS-CoV-2 negative plasma at a dilution of 1 in 100 in an incubation volume 2 times the bead volume. Beads were then washed 3 times in PBS and incubated with the human plasma samples of interest at a dilution of 1:50 in PBS+, PBS containing an additional 150 mM NaCl and 20 mM imidazole, for 2h at 4°C (50 μL beads per 200 μL sample). The remaining depleted samples were collected, filter sterilized, and tested for complete depletion by RBD direct ELISA.

Constructs are as described in Huo et al. (2020) and production was as described in Zhou et al. (2020).

To generate human monoclonal antibodies from peripheral blood B cells, CD22+ B cells were isolated from PBMCs using CD22 Microbeads (130-046-401; Miltenyi Biotec). Pre-enriched B cells were stained with anti-IgM-APC, IgA-FITC and IgD-FITC. Double negative memory B cells (IgM-,IgA-/D-cells) were sorted by FACS and plated on 384-well plates at a density of 4 B cells per well. Cells were stimulated to proliferate and produce IgG by culturing with irradiated 3T3-msCD40L feeder cells (12535; NID AIDS Reagent Program), 100 U/ml IL-2 (200-02; Peprotech) and 50 ng/ml IL-21 (200-21; Peprotech) for 13-14 days. Supernatants were harvested from each well and screened for SARS-CoV-2 binding specificity by ELISA. Lysis buffer was added to positive wells containing SARS-CoV-2-specific B cells and immediately stored at −80°C for future use in Ig gene amplification and cloning.

To isolate Spike and RBD-specific B cells, PBMCs were sequentially stained with LIVE/DEAD Fixable Aqua dye (Invitrogen) followed by recombinant trimeric spike-twin-Strep or RBD-biotin. Cells were then stained with antibody cocktail consisting of CD3-FITC, CD14-FITC, CD56-FITC, CD16-FITC, IgM-FITC, IgA-FITC, IgD-FITC, IgG-BV786, CD19-BUV395 and Strep-MAB-DY549 (iba) or streptavidin-APC (Biolegend) to probe the Strep tag of spike or biotin of RBD. Spike or RBD-specific single B cells were gated as CD19+, IgG+, CD3-, CD14-, CD56-, CD16-, IgM-, IgA-, IgD-, Spike+ or RBD+ and sorted into each well of 96-well PCR plates containing RNase inhibitor (N2611; Promega). Plates were centrifuged briefly and frozen on dry ice before storage at −80°C for future use in Ig gene amplification and cloning.

Genes encoding Ig VH, Ig Vκ and Vλ from positive wells were recovered using RT-PCR (210210; QIAGEN). Nested PCR (203205; QIAGEN) was then performed to amplify genes encoding γ-chain, λ-chain and κ-chain with ‘cocktails’ of primers specific for human IgG. PCR products of genes encoding heavy and light chains were joined with the expression vector for human IgG1 or immunoglobulin κ-chain or λ-chain (gifts from H. Wardemann) by Gibson assembly. For the expression of antibodies, plasmids encoding heavy and light chains were co-transfected into the 293T cell line by the polyethylenimine method (408727; Sigma), and antibody-containing supernatants were harvested for further characterization.

Heavy chain expression plasmids of specific antibodies were used as templates to amplify the first fragment, heavy chain vector include the variable region and CH1 until Kabat amino acid number 233. The second fragment of thrombin cleavage site and twin-Strep-tag with overlapping ends to the first fragment were amplified. The two fragments were ligated by Gibson assembly to make the Fab heavy chain expression plasmid.

Heavy chain and light chain expression plasmids of specific antibodies were used as a template to amplify variable region gene of heavy and light chain respectively. First, heavy chain gene products having the AgeI–SalII restriction enzyme sites were cloned into a scFv vector which is a modified human IgG expression vector which has a linker between the H chain and L chain genes followed by a thrombin cleavage site and twin-Strep-tags. Light chain gene products having NheI-NotI restriction enzyme site were cloned into scFv vector containing the heavy chain gene insert to produce scFv expression plasmids.

Protein production was done in HEK293T cells by transient transfection with polyethylenimine in FreeStyle 293 medium. For Fab antibody production, Fab heavy chain expression plasmids were co-transfected with the corresponding light chain. For scFv antibody production, scFv expression plasmid of specific antibody was used for transfection. After 5 days of culture at 37°C and 5% CO2, culture supernatant was harvested and filtered using a 0.22 mm polyethersulfone (PES) filter. Fab and scFv antibody were purified by Strep-Tactin affinity chromatography (IBA lifescience) according to the Strep-Tactin XT manual.

MAXISORP immunoplates (442404; NUNC) were coated with 0.125 μg of StrepMAB-Classic (2-1507-001;iba) at 4°C overnight and blocked with 2% skimmed milk in PBS (for plasma) or 2% BSA in PBS (for mAbs) for 1 h, plates were incubated with 50 μL of 10 μg/mL double strep-tag recombinant spike of SARS-CoV-2, SARS-CoV, MERS-CoV, OC43-CoV, HKU1-CoV, 229E-CoV and NL43-CoV. After one hour, 50 μL of serially diluted plasma or mAbs was added, followed by ALP-conjugated anti-human IgG (A9544; Sigma) at 1:10,000 dilution. The reaction was developed by the addition of PNPP substrate and stopped with NaOH. The absorbance was measured at 405nm. To determine the binding to SARS-CoV-2 RBD, SARS-CoV-2 NP, SARS-CoV-2 spike S1 (40591-V08H; Sino Biological Inc) and SARS-CoV-2 spike S2 (40590-V08B; Sino Biological Inc), immunoplates were coated with 0.125 μg of Tetra-His antibody (34670; QIAGEN) followed by 5 μg/mL of His-tag recombinant SARS-CoV-2 RBD, SARS-CoV-2 NP, SARS-CoV-2 spike S1 and SARS-CoV-2 spike S2. The plasma endpoint titers (EPTs) were defined as reciprocal plasma dilutions that corresponded to two times the average OD values obtained with mock. EC50 of mAbs were evaluated using non-linear regression (curve-fit), GraphPad Prism 8 software.

To determine the binding affinity of antibody to SARS-CoV-2 virus, virus was captured onto plates coated with mouse anti-SARS-CoV-2 spike (mAb31 with murine Fc) and then incubated with serial dilutions of SARS-CoV-2-specific human mAbs (full length IgG or Fab) followed by ALP-conjugated anti-human IgG (A8542, Sigma). The reaction was developed with PNPP substrate and stopped with NaOH. The absorbance was measured at 405 nm.

Results are expressed as the percentage of total binding, with 100% binding determined from the Ab concentration that gave maximum absorbance. GraphPad PRISM software was used to perform nonlinear regression curve-fitting analyses of binding data to estimate dissociation constants (Kd). Percent occupancy at IC50 was determined using the following formula: Percent occupancy = BMax [Ab]/(Kd+[Ab]), where the BMax is percent maximal binding, [Ab] is the concentration of Ab required to reach 50% FRNT and Kd is the concentration of Ab required to reach half-maximal binding.

The neutralization potential of Ab was measured using a Focus Reduction Neutralization Test (FRNT), where the reduction in the number of the infected foci is compared to a no antibody negative control well. Briefly, serially diluted Ab was mixed with authentic SARS-CoV-2/human/AUS/VIC01/2020 (Caly et al., 2020) and incubated for 1 hr at 37°C. The mixtures were then transferred to Vero cell monolayers and incubated for 2 hr followed by the addition of 1.5% semi-solid carboxymethyl cellulose (CMC) overlay medium to each well to limit virus diffusion. A focus forming assay was then performed by staining Vero cells with human anti-NP mAb (mAb206) followed by peroxidase-conjugated goat anti-human IgG (A0170; Sigma). Finally, the foci (infected cells) were visualized by adding TrueBlue Peroxidase Substrate. The percentage of focus reduction was calculated and IC50 was determined using the probit program from the SPSS package.

MAbs were screened for binding to MDCK-SIAT1 cells expressing the N-terminal domain (NTD) of SARS-CoV-2 spike glycoprotein (MDCK-NTD). MDCK-NTD was created by stably transfecting MDCK-SIAT1 cells (ECACC 05071502) (Matrosovich et al., 2003) with cDNA encoding the SARS-CoV-2 NTD (amino acids VNLT…TLKS) fused to the transmembrane domain of haemagglutinin H7 (A/HongKong/125/2017) (EPI977395) at the C terminus for surface expression using a second-generation lentiviral vector system. NTD expressing cells were FACS sorted using the FD7C mAb (Huang et al., 2020). In brief, MDCK-NTD cells were seeded at 3 × 104 per well in flat-bottomed 96-well plates (TPP) in high glucose DMEM containing 10% fetal bovine serum (FBS) at 37°C overnight. The medium was then removed and washed with 2% FBS in PBS (PBS/2% FBS) twice. 10 μg/ml of mAbs supernatants from transfected 293T cells were added (50 μl per well) and incubated at room temperature for 1 h. A second antibody Goat anti-human IgG Fc specific-FITC (F9512, Sigma-Aldrich) diluted 1:300 in PBS/2% FBS was then added (50 μl per well) and incubated for another 1 h at room temperature. After washing twice with PBS, the wells were fixed with 1% formaldehyde in PBS. The binding antibodies were detected by fluorescence intensities using a Clariostar plate reader (BMG, Labtech).

For the ACE2 competition ELISA, 250 ng of ACE2 protein was immobilized to a MAXIXORP immunoplate and the plates were blocked with 2% BSA in PBS. In the meantime, serially diluted Ab was mixed with recombinant RBD-mFc (40592-V05H; Sino Biological) and incubated for 1 h at 37°C. The mixtures were then transferred to the ACE2 coated plates and incubated for 1 h followed by goat anti-mouse IgG Fc-AP (Invitrogen #A16093) at 1:2000 dilution. The reaction was developed by the addition of PNPP substrate and stopped with NaOH. The absorbance was measured at 405 nm. The ACE2/RBD binding inhibition rate was calculated by comparing to antibody-free control well. IC50 were determined using the probit program from the SPSS package.

The stable cell line generation vector pNeoSec was used for cloning of the SARS-Cov2 Spike ectodomain comprising amino acids 27-1208 with mutations of the furin cleavage site (RRAR > GSAS at residues 682-685) and the PP (KV > PP at residues 986-987). At the N terminus, there is a twin StrepII tag and at the C terminus fused with a T4 fibritin trimerisation domain, an HRV 3C cleavage site and a His-8 tag. The human embryonic kidney (HEK) Expi293F cells (Thermo Fisher Scientific) were transfected with the construct together with a phiC31 integrase expression plasmid as described earlier (Zhao et al., 2014). The polyclonal G418 resistant (1 mg/ml) cell population were used for protein production. Expi293F cells were grown in adhesion in roller bottles with the high glucose DMEM (Sigma) with 2% FBS for 6 days at 30°C. The soluble spike protein was captured from the dialysed conditional media with prepacked 5 mL Columns of HisTrap excel (GE Healthcare Life Sciences). The protein was eluted in 300 mM imidazole containing phosphate-buffered saline (PBS) after a 20 mM imidazole PBS wishing step. The protein was further purified with a 16/600 Superdex 200 size exclusion chromatography with an acidic buffer (20 mM Acetate, 150 mM NaCl, pH 4.6) for the low pH Spike incubations, or a neutral buffer (2 mM Tris, 150 mM NaCl, pH 7.5).

Stable HEK293S cell line expressing His-tagged RBD was cultured in DMEM (high glucose, Sigma) supplemented with 10% FBS (Invitrogen), 1 mM glutamine and 1x non-essential amino acids at 37 °C. Cells were transferred to roller bottles (Greiner) and cultured in DMEM supplemented with 2% FBS, 1 mM glutamine and 1x non-essential amino acids at 30 °C for 10 days for protein expression. For protein purification, the dialyzed media was passed through a 5 mL HisTrap Nickel column (GE Healthcare). The column was washed with buffer 20 mM Tris pH 7.4, 200 mM NaCl, 30 mM imidazole and RBD was eluted using buffer 20 mM Tris pH 7.4, 200 mM NaCl, 300 mM imidazole. A volume of 30 μL endoglycosidase H1 (~1 mg ml−1) was added to ~30 mg RBD and incubated at room temperature for 2 h. Then the sample was further purified with a Superdex 75 HiLoad 16/600 gel filtration column (GE Healthcare) using 10 mM HEPES pH 7.4, 150 mM NaCl. Purified RBD was concentrated using a 10-kDa ultra centrifugal filter (Amicon) to 10.6 mg ml−1 and stored at −80°C.

Fab fragments were digested from purified IgGs with papain using a Pierce Fab Preparation Kit (Thermo Fisher), following the manufacturer’s protocol.

Thermal stability was assessed using Thermofluor (DSF). Briefly, 3 μg of the Ab preparation was used in a 50 μl reaction containing 10 mM HEPES pH 7.5, 100 mM NaCl, 3X SYPROorange (Thermo Fisher). Samples were heated from 25-97°C in a RT-PCR machine (Agilent MX3005p) and the fluorescence monitored at 25°C after every 1°C of heating. Melting temperatures (Tm) were calculated by fitting of a 5-parameter sigmoid curve using the JTSA software (P. Bond, https://paulsbond.co.uk/jtsa). Polydispersity was assessed by DLS using 10 μg of the Ab preparation in an UNCLE instrument (Unchained Labs). Freeze thaw experiments on 4 of the mAbs were performed with material at 1 mg/ml by flash-freezing using LN2, thawing and centrifuging an aliquot (10 minutes at 20000 g) before measuring the absorbance at 280nm of the soluble fraction.

Purified RBD was combined separately with Strep-tagged Fab150, Fab58, scFv269 and Fab316 in a 1:1 molar ratio, with final concentrations of 13.2, 9.4, 12.7 and 13.0 mg ml-1, separately. RBD was combined with Fab45 and Strep-tagged Fab88, Fab75 and Fab253, and Fab 75 and Strep-tagged chimeric Fab 253H55L in a 1:1:1 molar ratio all with a final concentration of 7 mg ml−1, separately. Glycosylated RBD was combined with Fab S309 (Pinto et al., 2020) and Fab384 in a 1:1:1 molar ratio with a final concentration of 8 mg ml−1. These complexes were separately incubated at room temperature for 30 min. Initial screening of crystals was set up in Crystalquick 96-well X plates (Greiner Bio-One) with a Cartesian Robot using the nanoliter sitting-drop vapor-diffusion method, with 100 nL of protein plus 100 nL of reservoir in each drop, as previously described (Walter et al., 2003). Good crystals of RBD-150 complex were formed in Molecular Dimensions Morpheus condition C2, containing 0.09 M NPS (nitrate, phosphate and sulfate), 0.1 M MES/imidazole pH 6.5, 10% (w/v) PEG 8000 and 20% (v/v) ethylene glycol and crystals also formed in Hampton Research PEGRx condition D11, containing 0.1 M imidazole pH 7.0 and 12% (w/v) PEG 20000. Some good crystals of RBD-158 were obtained from Index condition C01, containing 3.5 M NaCOOH pH 7.0, while some crystals were formed in Proplex condition C1, containing 0.15 M (NH4)2SO4, 0.1 M Tris pH 8.0 and 15% (w/v) PEG 4000 and further optimized in 0.15 M (NH4)2SO4, 0.1 M Tris pH 7.6 and 14.6% (w/v) PEG 4000. Crystals of RBD-scFv269 complexed were obtained from Index condition F01, containing 0.2 M Proline, 0.1 M HEPES pH 7.5 and 10% (w/v) PEG 3350. Good crystals for the RBD-316 complex were obtained from Index condition G10, containing 0.2 M MgCl2, 0.1 M bis-Tris pH 5.5 and 25% (w/v) PEG 3350. Crystals of RBD-45-88 complex were obtained from PEGRx condition G12, containing 10% (v/v) 2-Propanol, 0.1 M Sodium acetate trihydrate pH 4.0, 22% (w/v) PEG 6000. Crystals of RBD-75-253 complex were obtained from PEGRx condition D8, containing 0.1 M BIS-TRIS pH 6.5, 16% (w/v) PEG 10000. Crystals of RBD-75-253H55L were obtained from Index condition F5, containing 0.1 M ammonium acetate, 0.1 M bis-Tris pH 5.5 and 17% (w/v) PEG 10000. For the RBD-S309-384 ternary complex, good crystals were obtained from Morpheus condition H1, containing 0.1 M amino acids (Glu, Ala, Gly, Lys, Ser), 0.1 M MES/imidazole/ pH 6.5, 10% (w/v) PEG 20000 and 20% (w/v) PEG MME 550.

Crystals were soaked in a solution containing 25% glycerol and 75% reservoir solution for a few seconds and then mounted in loops and frozen in liquid nitrogen prior to data collection. Diffraction data were collected at 100 K at beamline I03 of Diamond Light Source, UK. Diffraction images of 0.1° rotation were recorded on an Eiger2 XE 16M detector with exposure time ranging from 0.004 to 0.01 s per frame, beam size 80 × 20 μm and 100% beam transmission. Data were indexed, integrated and scaled with the automated data processing program Xia2-dials or Xia2-3dii (Winter, 2010; Winter et al., 2018). For RBD-158 crystal form 2, RBD-316 and the ternary complexes of RBD88-45, RBD-253H55L and RBD-384-S309 datasets of 360° were collected from a single frozen crystal each, and 720° of data from 2 crystals for RBD-150, RBD-scFv269, RBD-158 crystal form 1 and RBD-253-75.

The structures were determined by molecular replacement with PHASER (Liebschner et al., 2019) using search models of the RBD, VhVl and ChCl domains of a closely related Fab in sequence for each complex. Sequence corrections to the target Fabs from the search models and model rebuilding were done with COOT (Emsley and Cowtan, 2004). All the structures were refined with PHENIX (Liebschner et al., 2019) resulting in good R-factors and stereochemistry for most of the structures except for RBD-88-45 and RBD-53-75 in each of which there is presence of translational NCS with vectors (−0.003 0.502 0.489) and (0.044, 0, 0.5) and can only be refined to Rwork/Rfree of 0.250/0.285 and 0.242/0.284 to 2.53 Å and 2.50 Å, respectively. The ChCl domains of Fab 88 in the RBD-88-45 complex are disordered. Data collection and structure refinement statistics are given in Table S5.

For all Fab or IgG-Spike complexes, a 3 μL aliquot of S ~0.6 μm (determined by OD) with Fab (1:6 molar ratio) was prepared, aspirated and almost immediately applied to a freshly glow-discharged Cu support Cflat 2/1-200 mesh holey carbon-coated grid (high intensity, 20 s, Plasma Cleaner PDC-002-CE, Harrick Plasma). Excess liquid was removed by blotting for 5-5.5 s with a force of −1 using vitrobot filter paper (grade 595, Ted Pella Inc.) at 4.5°C, 100% reported humidity before plunge freezing into liquid ethane using a Vitrobot Mark IV (Thermo Fisher).

For sample-specific details, refer to Table S6.

Movies were collected in compressed tiff format on a Titan Krios G2 (Thermo Fisher) operating at 300 kV with a K3 detector (Gatan) in super resolution counting mode using a custom version of EPU 2.5 (Thermo Fisher). A defocus range of 0.8-2.6 μm was applied with a nominal magnification of x105,000, corresponding to a calibrated pixel size of 0.83 Å/pixel and with a total dose of 43-47 e/ Å2, see Table S6.

Two-times binned movies were then motion corrected and aligned on the fly using Relion(3.1) scheduler (Zivanov et al., 2018) with a 5 × 5 patch based alignment. CTF-estimation of full-frame non-weighted micrographs was performed with the GCTF (1.06) (Zhang, 2016) module in cryoSPARC(v2.14.1-live) (Punjani et al., 2017).

Data for 88, 150, 158 were collected using a Titan Krios G2 (Thermo Fischer) operating at 300 kV with a K2 camera and a GIF Quantum energy filter (Gatan) with a 30 eV slit. For 159 (IgG), 384 and 316, data were collected as for 88, 150 and 158, except using a 20 keV slit. Rapid multi-shot data acquisition was set up using custom scripts with SerialEM (version 3.8.0 beta) (Mastronarde, 2005) at a nominal magnification of 165 kX, corresponding to a calibrated pixel size of 0.82 Å per pixel. A defocus range of −0.8 μm to −2.6 μm was used with a total dose of ~45-57 e-2 applied across 40 frames. Motion and CTF correction of raw movies was performed on the fly using cryoSPARC live patch-motion and patch-CTF correction (Punjani et al., 2017).

Poor-quality images were discarded after manual inspection of CTF and motion estimations. Particles were then blob picked in cryoSPARC (Punjani et al., 2017) and initially extracted with four times binning. After inspection of 2D classes, classes of interest were selected to generate templates for complete particle picking. Binned particles were then subjected to one to three rounds of reference free 2D classification followed by 3D classification with an ab-initio derived model before further refinement and unbinning.

For both 150 and 158, two data separate data collections were set up on the same grid, and refined particle sets from each collection were separated by exposure groups before being combined. For 150, a total of 77,265 exposure-group split particles were initially combined (51,554 from 4726 movies and 25,711 from 2079 movies), re-classified into five classes, and the two best classes (42,655 particles) subjected to further non-uniform refinement, with obvious density for Fab bound to one RBD in an ‘up’ conformation. Notably, discarded classes included a high proportion of undecorated S (28,463 particles, 4.4 Å reported resolution at GSFSC = 0.143, −43 Å2 B-factor).

Classification using heterogeneous refinement in cryoSPARC was found to be generally poor, and, instead, 3D variability analysis was employed to try to better resolve full spike-Fab structures. Local refinements were also performed with masks focused around the Fab/RBD region (not reported here), but maps were still insufficient to clearly build a model at the RBD/Fab interface and far inferior to the crystallographic maps. 3D variability analysis was found to be essential for isolating the RBD up and RBD down conformations for 159-IgG. Results from this are presented for 159-IgG and 384. Briefly, data were separated into eight clusters using the 3D variability analysis module with a 6 Å resolution filter and a mask around the RBD/Fab region. Masks were generated by initially rigid body fitting a model of the spike and a Fab into a refined map in Chimera before selecting an area of the model including the RBD and fab and using the ‘color zone’ module to crop out this desired part of the map. The resulting map was smoothed with a Gaussian filter (Pettersen et al., 2004), converted into a mask format using Relion3.1 ‘Mask Create’ before import into cryoSPARC. Resolution estimates quoted in the Table S6 were taken from Gold standard-FSC (FSC = 0.143) reported in the local resolution module in cryoSPARC (Punjani et al., 2017).

Competition assay of anti-RBD antibodies was performed on a Fortebio Octet RED96e machine with Fortebio Anti-HIS (HIS2) Biosensors. 2 μg ml−1 of His-tagged RBD dissolved in the running buffer (10 mM HEPES, pH 7.4 and 150 mM NaCl) was used as the ligand and was first immobilized onto the biosensors. The biosensors were then washed in the running buffer to remove unbound RBD. Each biosensor was dipped into different saturating antibodies (Ab1) to saturate the bound RBD, except one biosensor was into the running buffer in this step, acting as the reference. The concentration of saturating antibodies used was 15 μg ml−1. Higher concentrations were applied if 15 μg ml−1 was not enough to obtain saturating. Then all biosensors were washed with the running buffer again and dipped into wells containing the same competing antibody (Ab2). The concentration of competing antibodies used was 5 μg ml−1. The y axis values of signals of different saturating antibodies in this step were divided by the value of the reference channel to get ratio results of different Ab1-Ab2 pairs. Ratio result close to 0 indicated total competition while 1 indicated no competition. In total, 50 IgGs and 4 Fabs (Fabs 40, EY6A [Zhou et al., 2020], FD5D (unpublished) and S309 [Pinto et al., 2020]) were used as the saturating antibodies and 80 IgGs as the competing antibodies.

Competition values were prepared for cluster analysis and binning by capping all competition values between 0 and 1. Competition values between antibodies i and j were averaged with the competition value for j and i when both were available. Cluster4x (Ginn, 2020) was used to cluster antibodies into three distinct groups using single value decomposition on the matrix of competition values.

A surface of the receptor-binding domain was generated in PyMOL (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC) from chain E of PDB code 6YLA. A mesh was generated and iteratively contracted and restrained to the surface of the RBD to provide a smoother surface on which to direct antibody refinement, reducing intricate surface features which could lead to unrealistic exploration of local minima.

In order to provide an objective position for those antibodies of known structure (FD5D (unpublished), EY6A (Zhou et al., 2020), S309 (Pinto et al., 2020) and mAb 40), to reflect the occluded region, all non-hydrogen antibody atoms were found within 20 Å of any RBD atom, and likewise all RBD atoms within 20 Å of an antibody atom. From each group, the atoms with the lowest sum-of-square-lengths from all other members were identified and the midpoint of these two atoms was locked to the nearest vertex on the mesh. Solvent molecules were ignored, but in the case of S309, the glycan cofactor was included in the set of antibody atoms.

On an evaluation of the target function, either all unique pairs of antibodies were considered (all-pairs), or only unique pairs where one of the antibodies was fixed (fixed-pairs), depending on the stage of the minimization protocol. Competition levels were estimated for each pair of antibodies as described by f(x) in Equation 1

where r is the working radius of the antibody, set to 11 Å, accounting for the approximate antibody radius. The distance between the pair of antibodies at a given evaluation of the function is given by d in Angstroms. The target function was the sum of squared differences between the competition estimation and the competition value from SPR data.

Minimization was carried out globally by 1000 macrocycles of Monte Carlo-esque sampling using LBFGS refinement. A random starting position for each antibody was generated by randomly assigning a starting vertex on the RBD mesh and the target function minimized for 20 cycles considering data points for pairs with at least one fixed antibody, followed by 40 cycles for all data points. Between each cycle, antibody positions were locked onto the nearest mesh vertex. Depending on the starting positions of antibodies, results were a mixture of well-refined and poorly refined solutions. Results were ordered in ascending target function scores. Positions of antibodies for each result was passed into cluster4x as dummy C-alpha positions (Ginn, 2020). A clear self-consistent solution was enriched in lower target function scores and separated using cluster4x for further analysis. The average position for each antibody was chosen as the sampled position which had the lowest average square distance to very other sampled position, and the RMSD calculated from all contributing antibody positions.

Tissues were weighed and homogenized with zirconia beads in a MagNA Lyser instrument (Roche Life Science) in 1000 μL of DMEM supplemented to contain 2% heat-inactivated FBS. Tissue homogenates were clarified by centrifugation at 10,000 rpm for 5 min and stored at −80°C. RNA was extracted using the MagMax mirVana Total RNA isolation kit (Thermo Scientific) on a Kingfisher Flex extraction robot (Thermo Scientific). RNA was reverse transcribed and amplified using the TaqMan RNA-to-CT 1-Step Kit (ThermoFisher). Reverse transcription was carried out at 48°C for 15 min followed by 2 min at 95°C. Amplification was accomplished over 50 cycles as follows: 95°C for 15 s and 60°C for 1 min. Copies of SARS-CoV-2 N gene RNA in samples were determined using a previously published assay (PubMed ID 32553273). Briefly, a TaqMan assay was designed to target a highly conserved region of the N gene (Forward primer: ATGCTGCAATCGTGCTACAA; Reverse primer: GACTGCCGCCTCTGCTC; Probe: /56-FAM/TCAAGGAAC/ZEN/AACATTGCCAA/3IABkFQ/). This region was included in an RNA standard to allow for copy number determination. The reaction mixture contained final concentrations of primers and probe of 500 and 100 nM, respectively.

Vero-furin cells (Mukherjee et al., 2016) were seeded at a density of 2.5 × 105 cells per well in flat-bottom 12-well tissue culture plates. The following day, medium was removed and replaced with 200 μL of 10-fold serial dilutions of the material to be titrated, diluted in DMEM+2% FBS. After incubation for 1 h at 37°C, 1 mL of methylcellulose overlay was added. Plates were incubated for 72 h, then fixed with 4% paraformaldehyde (final concentration) in phosphate-buffered saline for 20 min. Plates were stained with 0.05% (w/v) crystal violet in 20% methanol and washed twice with distilled, deionized water prior to plaque enumeration.

Octet RED 96e (ForteBio) was used to determine the binding affinities of antibodies with RBD or spike. Anti-RBD IgGs were immobilized onto AR2G biosensors (ForteBio) while RBD was used as the analyte with serial dilutions. For IgG159, spike was immobilised onto AR2G biosensors with IgG159 acting as the analyte with serial dilutions. Kd values were calculated using Data Analysis HT 11.1 (ForteBio) with a 1:1 global fitting model.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.