Synthetic virions reveal fatty acid-coupled adaptive immunogenicity of SARS-CoV-2 spike glycoprotein

Materials

18:1 DOPC 1,2-dioleoyl-sn-glycero-3-phosphocholesteroline, 18:1 DOPE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine, L-α-phosphatidylglycerol (EggPG), L-α-phosphatidylcholine (EggPC), LissRhod PE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl), 18:1 DGS-NTA(Ni) 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] (nickel salt), 18:1 1,2-dioleoyl-sn-glycero-3-phospho-(1’-myo-inositol) (ammonium salt), 18:1 1,2-di-(9Z-octadecenoyl)-sn-glycero-3-phospho-L-serine (sodium salt), 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamin Atto488-conjugate, 18:1, cholesterol, 18:0 N-stearoyl-D-erythro-sphingosylphosphorylcholine, and extrude set with 50 nm pore size polycarbonate filter membranes were purchased from Avanti Polar Lipids, USA Dulbecco’s Modified Eagle Medium (DMEM) High Glucose, heat-inactivated fetal bovine serum, penicillin-streptomycin (10,000 U/mL), L-Glutamine (200 mM), trypsin-EDTA (0.05%), FluoroBrite DMEM, CellTracker Green CMFDA dye, Hoechst 33342, phosphate-buffered saline were purchased from Thermo Fischer Scientific, Germany. Dexamethasone (BioReagent), Tretinoin (pharmaceutical standard), Acitretin (>98%), Vitamine K₂, Carbenoxolone disodium salt (>98%), palmitic acid (>99%), oleic acid (>99%), heparin, linoleic acid (>99%) arachidonic acid (analytical standard), oxalyl chloride (>99%), 3-picolylamine (>99%) were purchased from Sigma Aldrich, Germany. LinRGD was custom synthesized by PSL GmbH, Germany. Transparent flat-bottom 96-well plates were purchased from TTP, Switzerland. Human Insulin was purchased from Millipore Sigma. Lipidex-1000 resin was purchased from Perkin Elmer, Germany. DMSO for cell culture use was purchased from Omnilab, Germany. Sicastar Ni²⁺-NTA silica beads were purchased from Micromod GmbH, Germany. Human IgG COVID-19 convalescent plasma fractionated purified and lyophilized was purchased from Innovative Research, USA. Anti SARS-CoV-2 S, clone CR3022 (FITC) was purchased from Novus Biologiclas, Germany. Histidine-tagged recombinant SARS-CoV-2 spike RBD (Val16-Arg685(D614G)), human recombinant ACE2 were purchased from Sino Biologicals, Germany. MCF-7 cells, HUVEC cells and endothelial cell growth medium were obtained from ATCC, USA. QCM-D sensor crystals (QS-QSX303) were obtained from Quantum Design GmbH, Germany.

SUV/MiniV preparation

SUVs and MiniVs were produced by manual extrusion through track-etched polycarbonate filter membranes^50,51. For this, lipids dissolved in chloroform stock solutions were mixed at the desired lipid ratio in glass vials and subsequently dried under vacuum for at least 15 min to evaporate the chloroform completely. The obtained lipid film was rehydrated to a final lipid concentration of 6 mM in PBS for at least 5 min and afterwards shaken for at least 5 min at 1000 rpm on a horizontal shaker. This liposome solution was extruded nine times through a 50 nm radius pore size filter.

For immobilization of recombinant S ectodomains, the SUV solution was diluted to a final concentration of 100 µM in PBS, corresponding to 1 µM final DGS-NTA(Ni²⁺). To this, 0.5 µM of recombinant histidine-tagged S was added and incubated for coupling for at least 15 min. For loading of FAs on MiniVs, FAs were dissolved in DMSO to a final concentration of 100 mg/mL. From these stocks, 100 µg/mL dilutions in PBS were prepared.

S proteins expression and purification

Recombinant wild-type (Wuhan) S protein with a mutated furin site was produced as described previously¹. The construct comprises amino acids 1–1213, lacks the native transmembrane domain, which is replaced with a C-terminal thrombin cleavage site followed by a T4-foldon trimerization domain and a hexa-histidine tag. The polybasic cleavage site has been removed (RRAR to A mutation). Briefly, S protein was expressed using the MultiBac baculovirus expression system in Hi5 cells⁵². Medium from transfected cells was harvested 3 days after-transfection by centrifuging the cultures at 1000 × g for 10 min followed by another centrifugation of supernatant media at 5000 × g for 30 min. This final supernatant was then incubated with 10 mL HisPur Ni-NTA Superflow Agarose (Thermo Fisher Scientific) per 4 l of expression culture for 1 h at 4 °C. Subsequently, the resin-bound with SARS-CoV-2 S protein was collected using a gravity flow column (Bio-Rad) and then extensively washed with 30 column volumes (CV) of wash buffer (65 mM NaH₂PO₄, 300 mM NaCl, 20 mM imidazole, pH 7.5). Finally, the protein was eluted using a step gradient of elution buffer (65 mM NaH₂PO₄, 300 mM NaCl, 235 mM imidazole, pH 7.5). After analysing elution fractions using reducing SDS-PAGE, fractions containing SARS-CoV-2 S protein were pooled and concentrated using 50 kDa MWCO Amicon centrifugal filter units (EMD Millipore) and finally buffer-exchanged in phosphate-buffered saline (PBS) pH 7.5. The protein was then subjected to size exclusion chromatography (SEC) using a Superdex 200 increase 10/300 column (GE Healthcare) in PBS pH 7.5. Peak fractions from SEC were then analysed using reducing SDS-PAGE and then fractions containing SARS-CoV-2 S protein were pooled and concentrated using 50 kDa MWCO Amicon centrifugal filter units (EMD Millipore) and finally aliquoted and flash-frozen in liquid nitrogen for storage at −80 °C until further use.

The UK (or ‘Kent’) B1.1.7 variant⁵³ S ectodomain gene sequence was synthesized and inserted into pACEBac1 plasmid (Genscript Inc., New Jersey USA). Expression and purification were carried out as described above for wild-type S.

The S(R403A) mutant was prepared by modifying the wild-type (Wuhan) S expression construct with the point mutation using the QuickChange site-directed mutagenesis kit (Qiagen).

SARS-CoV S encoding gene was synthesized (Genscript Inc, New Jersey USA). The construct comprises amino acids 14 to 1193, preceded by a GP64 secretion signal sequence (amino acids MVSAIVLYVLLAAAAHSAFA) and contains a C-terminal thrombin cleavage site followed by a T4-foldon trimerization domain and a hexa-histidine affinity purification tag. The synthetic gene was inserted into pACEBac1s⁵². Protein was produced and purified as described above for SARS-CoV-2 S.

MERS-CoV S encoding gene was synthesized (Genscript Inc, New Jersey USA) and cloned into pACEBac1⁵². This construct comprises amino acids 18–1294, preceded by the GP64 secretion signal sequence and contains a C-terminal thrombin cleavage site followed by a T4-foldon trimerization domain and a hexa-histidine affinity purification tag. Protein was produced and purified as described above for SARS-CoV-2 S.

ApoS protein lacking free fatty acid was produced from purified SARS-CoV-2 S by Lipidex treatment as described¹. Briefly, purified SARS-CoV-2 S protein was incubated with pre-equilibrated lipidex-1000 resin (Perkin Elmer; cat no. 6008301) in PBS pH 7.5 overnight at 4 °C on a roller shaker. Following this, Lipidex-treated S protein was separated from the resin using a gravity flow column. The integrity of the protein was confirmed by size exclusion chromatography (SEC) using a S200 10/300 increase column (GE Healthcare) and SDS-PAGE.

ADAH11 selection, expression, and purification

Neutralizing nanobody ADAH11 against the RBM of S was selected from a synthetic library using in vitro selection by ribosome display⁵⁴. Following selection, the ADAH11 coding sequence was cloned into pHEN6 plasmid⁵⁵ containing a PelB signal sequence at the N-terminus, and a hexa-histidine and 3X FLAG tag at the C-terminus. ADAH11 was expressed in E. coli TG1 cells in 2x YT medium overnight at 30 °C induced with 1 mM IPTG (Isopropyl ß-D-1-thiogalactopyranoside). Cells were harvested by centrifugation at 3200 × g for 15 min at 4 °C. Cells were then resuspended in 5 mL cold TES buffer (50 mM Tris pH 8.0, 20% Sucrose, 1 mM EDTA, complete protease inhibitor tablet) for each gram of pellet. This resuspension was incubated on a roller shaker for 45 min at 4 °C. Then, 7.5 mL of ice-cold shock buffer (20 mM Tris pH 8.0, 5 mM MgCl₂) was added per gram of pellet and again incubated on a roller shaker for 45 min at 4 °C. Then, the supernatant-containing periplasm was collected by centrifugation at 13,000 × g for 12 min at 4 °C. This supernatant was incubated with 0.5 ml HisPur Ni-NTA Superflow Agarose (Thermo Fisher Scientific) per liter of expression for 1 h. The resin was pre-equilibrated in ADAH11 wash buffer 1 (50 mM Tris, 200 mM NaCl, 10 mM Imidazole pH 8.0). A gravity flow column (Bio-Rad) was used to separate the resin-bound with ADAH11 from the unbound lysate. This resin was then washed with 20 CV of ADAH11 wash buffer 1, followed by 30 CV wash with ADAH11 wash buffer 2 (50 mM Tris pH 8.0, 200 mM NaCl, 20 mM Imidazole pH 8.0). Protein was then eluted using a step gradient of ADAH11 elution buffer (50 mM Tris, 200 mM NaCl, 500 mM Imidazole pH 8.0). Elution fractions were analyzed using reducing SDS-PAGE. Elution fractions containing ADAH11 protein were pooled and dialyzed against PBS pH 7.5. Dialyzed protein was concentrated using 10 kDa MWCO Amicon centrifugal filter units (EMD Millipore) and then injected on a Superdex 200 increase 10/300 size exclusion chromatography (SEC) column (GE Healthcare) in PBS pH 7.5. Peak fractions from SEC were analyzed using reducing SDS-PAGE and fractions containing ADAH11 were pooled and concentrated using 10 kDa MWCO Amicon centrifugal filter units (EMD Millipore). Finally, the protein was aliquoted and flash-frozen in liquid nitrogen for storage at −80 °C until further use.

DLS + zeta potential

Size and zeta potentials of SUVs and MiniV variants were measured with a Malvern Zetasizer Nano ZS system at a total lipid concentration of 100 µM in PBS⁵⁶. Temperature equilibration time was set to 300 s at 25 °C, followed by three repeat measurements for each sample at a scattering angle of 173° using the built-in automatic run-number selection. The material refractive index was set to 1.4233 and solvent properties to η = 0.8882, n = 1.33 and ε = 79.0.

Hoechst staining and nuclei counting

To assess blocking of cell adhesion under linRGD incubation, we imaged adherent cells after seeding and washing on fibronectin-coated well plates following previously developed protocols^57,58. To this end, MCF-7 cells were seeded at a density of 50,000 cells/well in flat-bottom 96-well plates in 100 µL culture medium. Cells were either incubated with 20 µM linRGD or with the addition of 20 µL PBS (mock) for 24 h. Subsequently, Hoechst33342 was added to a final concentration of 10 µM to the cell layers and incubated for 10 min. Cells were then washed twice with 200 µL PBS and fixed for 20 min with 4% paraformaldehyde. Cell nuclei were then imaged in the whole well with a Leica DMi8 inverted fluorescent microscope equipped with a sCMOS camera and 10× HC PL Fluotar (NA 0.32, PH1) objective with DAPI emission/excitation filters. For automated nuclei counting, TIFF images from three wells (i.e., three replicates) were background segmented by global histogram thresholding and automated particle counting (particle analysis) with ImageJ software (NIH). Before particle counting, a watershed algorithm was applied to separated overlapping nuclei and nuclei counting was restricted to particles in the size range between 1 µm² and 100 µm².

Cryo-TEM tomography

For cryo-TEM imaging, MiniVs were diluted to a final particle concentration of 2 × 10¹⁰ particles/mL. The samples were applied to glow-discharged C-Flat 1.2/1.3 4 C grids (Protochips) and plunge-frozen using a Vitrobot Mark IV (FEI, now Thermo Scientific) at the following settings: temperature of 22 °C, the humidity of 100%, 0 s wait time, 2–3 s blot time, blot force −1, and 0 s drain time. The data were acquired on a Titan Krios electron microscope operated at 300 kV equipped with a Falcon 3EC direct detector, a Volta phase plate, and a Cs Corrector (CEOS GmbH). The micrographs were acquired with the software EPU (Thermo Scientific) in linear and counting mode at magnified pixel sizes of 1.39 Å and 0.85 Å, at doses of 40–60 e-/sqÅ, at defocus values ranging from −0.25 to −2.0 µm using the Volta phase plate or the 100 µm objective aperture. Tilt series were acquired with the software TOMO (Thermo Scientific) at tilt angles ranging from −60° to 60° at 3° increments with defocus values ranging from −0.3 to −1.0 µm with individual micrograph movies recorded with 10 frames in counting mode at a magnified pixel size of 0.85 Å, a total dose per micrograph of 3e-/sqÅ, using the Volta phase plate. Micrograph movies were aligned using MotionCor2 in the RELION3 software suite^59,60. Tilt series were binned by a factor of four for further processing in Etomo (IMOD)⁶¹. For each tilt series, the micrographs were aligned using patch tracking. The final tomograms were generated with a filtered back-projection of the aligned micrographs. Tomogram slices were FFT band-pass filtered between 3 and 40 pixel and subsequently a 2 × 2 pixel Gaussian blur filter was applied. Images were contrast corrected by visual inspection.

QCM-D + imaging SLB

For QCM-D measurements, sensor crystals, AT-cut gold electrodes coated with a 50 nm thick layer of silicon oxide were used. SiO₂ surfaces were cleaned as described elsewhere⁶². Briefly, QCM-D sensor crystals were cleaned using an aqueous 2% SDS solution water and activated by UV/ozone for 10 min. A QSense Analyzer equipped with a four-channel system from QuantumDesign was used for measurements. All measurements were performed at 22 °C in an open mode. The resonance frequency and dissipation shifts were recorded at several harmonics simultaneously. Before adding the samples, the frequency and dissipation changes were base-lined by averaging over the last 5 min of the buffer wash (PBS). Formation of supported lipid bilayers (SLBs) was achieved by absorption and rupture of SUVs on cleaned SiO₂ surfaces. SUVs with a lipid composition of 20 mol% L-α-phosphatidylglycerol (EggPG), 77 mol% L-α-phosphatidylcholine (EggPC), 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamin, 2 mol% 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] (18:1 DGS-NTA), and 1 mol% 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamin Atto488-conjugate were used for SLB formation at a lipid concentration of 1.2 mM and a final MgCl₂ concentration of 2 mM. Two hundred microliters of this solution was added to each QCM-D sensor to form SLBs after the sensors were equilibrated with 200 µL PBS for 7 min and incubated for 17 min. SLB formation was confirmed by the characteristic changes in frequency and dissipation as it is described elsewhere⁶³. SLBs were washed with PBS to remove non-ruptured SUVs for 7 min. For immobilization of recombinant human AC2 200 µL PBS containing 300 nM histidine-tagged ACE2 was added to the according sensor crystals (S1, S2, and S4) and incubated for 16 min. Unbounded ACE2 receptors were removed washing with PBS. The crystal sensor S3 not being functionalized with the ACE2-receptor was treated in the same way with the washing buffer (5). After 11 min, the washing buffer was removed and 200 µL of a solution containing 1.5 µM MiniVs (total lipid concentration) was added to sensors S1 and S3. To the crystal of sensor S2, a 200 µL PBS solution containing 1.5 µM (total lipid concentration) naive SUVs was added. Sensor S4 was treated with 200 µL PBS containing soluble recombinant S. The solutions were incubated on the crystal sensors for 55 min and afterwards washed with PBS to remove unbound components (7).

Cell culture and viruses

MCF-7 cells were cultured in Dulbecco’s Modified Eagle Medium supplemented with 4.5 g/l glucose, 1% L-glutamine, 1% penicillin/streptomycin, 0.01 mg/mL recombinant human insulin, and 10% fetal bovine serum. HUVEC cells were cultured in F-12K medium supplemented with 0.1 mg/ml heparin, 10% fetal bovine serum and 30 µg/ml endothelial growth supplement. Cells were routinely cultured at 37 °C and 5% CO₂ atmosphere and passaged at ~80% confluency by detachment with 0.05% trypsin/EDTA treatment. MCF-7 cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 4.5 g/l glucose, 1% L-glutamine, 1% penicillin/streptomycin, 0.01 mg/mL recombinant human insulin, and 10% fetal bovine serum. A549-ACE2 cells²⁴ were kindly provided by M. Cortese and R. Bartenschlager (Heidelberg University) and cultivated in DMEM supplemented with 10% fetal calf serum (Capricorn), 100 U/mL penicillin, 100 µg/mL streptomycin, and 1% non-essential amino acids (all from Gibco) and 0.5 mg/mL G418. Cells were routinely cultured at 37 °C and 5% CO₂ atmosphere and passaged at ~80% confluency based on 0.05% trypsin/EDTA treatment.

The SARS-CoV-2 isolate Bavpat1/2020 was obtained by the European Virology Archive (Ref-SKU: 026V-03883) at passage 2. Virus stocks were generated by passaging the virus two times in VeroE6 cells. Virus stocks were titrated by infection of VeroE6 cells as previously described²⁴.

Confocal microscopy

Confocal microscopy was performed with a laser-scanning microscope LSM 800 (Carl Zeiss AG). Images were acquired with a 20x (Objective Plan‐Apochromat ×20/0.8 M27, Carl Zeiss AG) and a ×63 immersion oil objective (Plan‐Apochromat ×63/1.40 Oil DIC, Carl Zeiss AG). Images were analyzed with ImageJ (NIH) and adjustments to image brightness and contrast, as well as background corrections, were always performed on the whole image and special care was taken not to obscure or eliminate any information from the original image.

Mass spectrometry

For loading of ApoS samples with defined FFA for MS analysis, pure FFA (PA, OA, LA, AA) solutions were prepared at a stock concentration of 100 mg/mL in DMSO. Then, a predilution of 100 µg/mL in PBS was prepared and 0.2 mg ApoS samples were mixed with respective FFA predilutions in a 1:10 molar ratio and incubated for 2 h. Subsequently, the histidine-tagged ApoS protein was pulled down with 1 mg of 300 nm NTA(Ni2 + )-conjugated silica beads through 30 min incubation and subsequent centrifugation at 11,000 × g for 5 min. ApoS was eluted by the addition of 100 mM imidazole and subsequently subjected to MS samples preparation (see below). For mass spectrometry detection of the FA content in S samples, a targeted LC-coupled MS/MS approach with multiple reaction monitoring was applied. For MS analysis, all protein samples were pre-diluted to a final concentration of 400 µg/mL. To extract the FAs from the S protein, S samples were mixed in a 1:4 (v/v) ratio with chloroform for 2 h on a horizontal shaker in a Teflon-sealed glass vial at 25 °C. The top organic phase was then transferred to a new glass vial and the chloroform was evaporated for 15 min in a desiccator. Subsequently, a derivatization approach enabling improved MS detection of FFAs was applied(2). This approach is based on the activation of carboxylic FA head groups with oxalyl chloride (OC) and subsequent derivatization with 3-picolylamine (PA). For this, 200 µL of a 2 M OC solution prepared in dichloromethane (DCM) was added to the samples using a glass syringe and incubated for 5 min at 65 °C. The OC and DCM were then removed in a desiccator and 150 µL of a 3-picolylamine solution (1% (v/v) in acetonitrile) was added to the samples and incubated for 5 min. The residual solvent was again evaporated in a desiccator and the samples were dissolved in 50 µL acetonitrile. To all samples, a deuterated LA internal standard (dissolved in DCM) was added to a final concentration of 20 µg/mL before derivatization. Finally, the resulting samples were diluted 1–500 in 80% ACN aq. (MS grade).

The subsequent LC-MSMS analysis was performed using a Shimadzu Nexera system hyphenated to a Sciex QTRAP 4500 system. The LC system was supplied with water (A) and acetonitrile (B) in LCMS grade from Biosolve. The solvents were supplemented with 0.1% LCMS grade formic acid. 6 µL of the sample solutions were fractionated with a Supelco Titan C18 column 100 × 2.1 mm, 1.9 µ, at 45 °C. Whereas the following gradient was applied: 0 min—40%B; 1.5 min: 40% B; 6.5 min—98%B; 8.0 min—98%B; 8.1 min—40%B and 9.5 min—98% B. The mass spectrometer was controlled using the Analyst 1.7 software and was operated in ESI positive mode. An optimal target compound ionization was achieved by setting the following source parameters: Curtain gas 35, temperature 550 °C, ionization voltage 5500 V, nebulizer gas 65, heater gas 80 and collision gas 9.

The targeted derivatized fatty acids were detected in MRM mode. Fragmentation of the monitored PA derivatized FA yielded a characteristic MSMS fragment with m/z 92. For a linoleic acid additional parent to fragment MSMS transitions were recorded. The following compound dependent MSMS parameters were applied: PA derivatized linoleic acid (LAP-92: 371.3–92.1 Da, Dwell 90 ms, declustering potential (DP) 80, entrance potential (EP) 10, collision energy (CE) 43, cell exit potential (CXP) 11, LAP-109: 371.3–109.1 Da, Dwell 40 ms, DP 80, EP 10, CE 36, CXP 10, LAP-163: 371.3–163.1 Da, Dwell 15 ms, DP 80, EP 10, CE45, CXP 11); PA derivatized deuterated linoleic acid (LADP-92: 375.3–92.1 Da, Dwell 50 ms, DP 80, EP 10, CE 43, CXP 11); PA derivatized palmitic acid (PAP-92: 347.3–92.1 Da, Dwell 50 ms, DP 80, EP 10, CE 43, CXP 11); PA derivatized oleic acid (OAP-92: 373.3–92.1 Da, Dwell 50 ms, DP 80, EP 10, CE 43, CXP 11); PA derivatized arachidonic acid (AAP-92: 395.3–92.1 Da, Dwell 50 ms, DP 80, EP 10, CE 43, CXP 11). Subsequent data analysis was performed using the Analyst and Multiquant 3.0.2 software.

A set of positive and negative control samples was measured under the same conditions to ensure high-quality results. As positive controls derivatized LA standard dilutions in 80% ACN aq. (MS grade) starting from the following concentration levels were used: 5, 10, and 25 µg/mL. Four-fold injection of the 25 µg/mL LAP standard yielded a CV < 2%. A sample preparation related contamination of the samples was excluded using a negative control sample, that was generated by using a water aliquot instead of the protein solution for the sample work-up. Moreover, we ensured the chromatographic separation of LADP (C18H28D4O2) and the ubiquitous as well as isobaric stearic acid (C18H36O2).

Retention assays

For quantification of MiniV-cell-binding, we developed retention assays to measure the amount of MiniVs retained within culture plates after incubation and washing. This assay is based on quantification of the rhodamine fluorescence form the SUV lipid membrane and can therefore not discriminate between attachment and uptake of MiniVs. For retention analysis, SUVs and MiniVs with different recombinant S ectodomains (as indicated in the figure legends), were added to MFC-7 cell cultures in flat-bottom 96-well plates with 100 µL culture medium to a final lipid concentration of 10 µM. Before the addition, cell medium was exchanged from the seeding medium (DMEM supplemented with phenol red, 4.5 g/l glucose, 1% L-glutamine, 1% penicillin/streptomycin, 0.01 mg/mL recombinant human insulin, and 10% fetal bovine serum) to low-serum medium (DMEM supplemented without phenol red, 4.5 g/l glucose, 1% L-glutamine, 1% penicillin/streptomycin, 0.01 mg/mL recombinant human insulin, and 0.5% fetal bovine serum) to reduce the amount of serum-derived FAs far below physiologically relevant concentrations. After incubation of MiniVs with cells for 24 h, rhodamine fluorescence was measured at 9 different positions and 1300 µm distance to well wall in each well using an Infinite M200 TECAN plate reader controlled by TECAN iControl software with an in-built gain optimization and excitation/emission setting adjusted to 555/585 nm. Wells were then washed three times with PBS and subsequently fixed with 100 µL 4% paraformaldehyde. After 10 min fixation, rhodamine fluorescence was again measured in each well with the settings mentioned above. For retention analysis binding could then be deduced from residual fluorescence in each well (for assessment of patient IgG neutralization) or retention values could be calculated by dividing the residuals fluorescence to the initial fluorescence intensity before washing. All measurements were performed in triplicates. As retention show significant variations depending on cell seeding density, cell viability and vesicle preparation, a condition of naïve SUVS (i.e., without protein on the surface) was added to all experimental batches for normalization proposes. For time-resolved measurements of retention, separate wells for each time point were prepared and evaluated sequentially.

For retention assay involving an assessment of FFA influence, stock of pure FFA (PA, OA, LA and AA) were prepared in DMSO at a final concentration of 100 mg/mL. From these stock, predilutions (100 µg/mL) in PBS were prepared. Individual FFAs, as indicated in the figures, were added to the MiniVs of the retention assays at a final concentration of 1 µM. For retention assays involving TMPRSS2 inhibition by camostat mesylate, camostat mesylate was added to the cell cultures at indicated concentration 2 h prior to MiniV addition.

Competition experiments of D614G and B1.1.7S

For assessment of competitive cell-binding between MiniVs presenting WT(D614G) and B1.1.7 S on the surface, two differently fluorescent SUV samples were prepared. SUVs harbouring Atto448 we decorated with WT(D614G) S and SUVs harbouring rhodamine B with B1.1.7 S. 10 µM final lipid concentration of both MiniV types were incubated together with MCF-7 cells and retention assays were performed as described above for the rhodamine and Atto488 (emission excitation set to 488/512) signals separately. Naïve SUV controls were performed for both vesicles types and retention were normalized accordingly.

Based on the differing fluorescence of the two MiniV populations, multiplicity of infection (MOI) measurements were performed by confocal imaging. For this, both MiniV types (D614G and B1.1.7) were incubated with MCF-7 cells in glass-bottom 8-well LabTek chambers for 24 h. After incubations, cells were washed three times with PBS and subsequently imaging in at least 8 z-positions (1 µm stack distance) in both channels. Maximum intensity projects were made by ImageJ and MiniVs of each type were manually counted for 5 individual cell groups.

Retention assays (RGD)

For assessment of RGD-motif-mediated effects in MiniV-cell-binding, retention assay under integrin blocking conditions with linRGD was performed. For this, 100,000 MCF-7 cells/well were seeded in flat-bottom 96-well plates and allowed to form confluent monolayers overnight. Subsequently, the seeding medium was exchanged to low-serum (0.5%) cell culture medium and a final linRGD concentration of 20 µM was added to each well from a 10 mM stock solution. MiniVs (with S-configurations as indicated in the figure legends) were added to a final lipid concentration of 10 µM. Retention assays were performed as detailed above.

To quantify differences in FA-regulated S-integrin engagement, retention was calculated as linRGD-normalized retention. For this, retention of MiniVs was measured with and without incubation of linRGD. Differences in retention were calculated and normalized by:

Where R is the retention value of the respective MiniV S configuration without linRGD, R_linRGD is the retention value of the corresponding MiniV S configuration under the addition of 20 µM linRGD and R_nativeS is the retention value MiniVs presenting native S.

Retention assays (FABP drug assessment)

For assessment of drug-modulated S binding of potential pharmacologic FABP binders, retention assays under drug incubation were performed. For this, 100,000 MCF-7 cells/ well were seeded in flat-bottom 96-well plates and allowed to form confluent monolayers overnight. Subsequently, the seeding medium was exchanged to low-serum (0.5%) cell culture medium and a final drug concentration of 1 µM was added to each well from DMSO stock solutions. MiniVs (with S-configurations as indicated in the figure legends) were added to a final lipid concentration of 10 µM and incubated for 24 h. For normalization purposes, also retention of naïve SUVs under drug treatment was measured to account for any drug-induced changes in cellular phenotypes. Retention assays were performed as detailed above and drug-normalized retention was calculated from:

where R_corr is calculated by

and R is is the retention value of the respective MiniV ApoS configuration under drug incubation, R_apo is the retention value of MiniVs with ApoS without drug incubation, R_SUVs is the retention value of SUVs without drug incubation and R_drugSUVs is the retention value of SUVs under drug incubation. IC50 values of MiniV retention for dexamethasone and vitamin K2 were measured by serial dilutions and IC50 values were calculated using nonlinear regression.

Protein structure visualization and ASA calculation

Previous studies have demonstrated the value of accessible surface area calculations for assessment of IgG epitope characterization in SARS-CoV-2 S and conformational changes upon protein binding^64,65. For visualization of S cryo-TEM structures, the PDB 3D viewer was applied⁶⁶ and protein structure was retrieved from https://www.wwpdb.org/pdb?id=pdb_00007bnn; https://www.wwpdb.org/pdb?id=pdb_00006zb5; https://www.wwpdb.org/pdb?id=pdb_00007a97. ASAs were calculated from molecular surface representations of residue accessible surface area properties in the PBD. A rolling probe radius of 1.4 nm was applied. For ASA calculations hydrogen atoms were taken into consideration. Tracing atoms were not considered for the analysis and no line size attenuation was applied.

Nanoparticle tracking analysis

MiniV particle concentration was determined by NTA with a ZetaView Quatt Video Microscope PMX-420 (Particle Metrix, Inning am Ammersee, Germany). Alignment was performed using 100 nm polystyrene beads diluted 1:250 000 (v:v) in MilliQ water. MiniV particles were diluted in PBS to a final concentration of 50–150 particles/frame. The observation cell was equilibrated with PBS before 1 mL of the sample was injected. One acquisition cycle was performed at 11 positions of the observation cell, in scatter mode using the 488 nm laser and at a temperature of 24 °C. The following settings were used for acquisition: sensitivity 80, shutter 100, frame rate 30, medium quality. Post-acquisition parameters were set as follows: minimal brightness 30, minimal area 10, maximal area 1000, trace length 15. All 11 positions were included in the analysis, using over 500 traced particles in total.

MiniV neutralization assays

For MiniV neutralization assays, 100,000 MCF-7 cells/ well were seeded in flat-bottom 96-well plates and allowed to form confluent monolayers overnight. Subsequently, the seeding medium was exchanged to low-serum (0.5%) cell culture medium and MiniVs (with S-configurations as detailed in the figure legends) were added to a final lipid concentration of 10 µM. Three types of neutralizing immunoglobulins were tested (1) IgG CR3022 (2) ADAH11, and (3) purified IgG from convalescent COVID-19 donors. CR3022 neutralization was assessed by the addition of 132 nM purified IgG. ADAH11 neutralization was titrated in a concentration range of 7.4 nM–1.5 µM and retention assays for ADAH11 neutralization were performed at 1 µM ADAH11. Neutralization of purified donor IgGs was measured at a final concentration of 3.3 µg/mL.

For assessment of neutralization with different FFA profiles, immunoglobulin mediated reduction in retention was calculated by:

where R is the retention value of the respective MiniV ApoS configuration without immunoglobulin, R_{immunoglobulin} is the retention value of the respective MiniV ApoS configuration under immunoglobulin incubation and R_NativeS is the retention value of MiniVs presenting Native S without immunoglobulin incubation. For the assessment of MiniV neutralization as a function of FFAs, individual FFAs were added from DMSO stocks (100 mg/mL) to the culture medium of the retention assay at a final concentration of 1 µM. To mimic the basal, low FFA levels, for COVID-19 donor IgG neutralization, no additional FFAs were added to the culture medium as the 0.5% serum concentration of the retention assay culture medium already provide approximately 0.1 µM LA and AA⁶⁷.

Competition assays

5 × 10⁴ A549-ACE2 cells were seeded in 24-well plates on the day prior to infection. Cells were incubated with 300 µL of SUV or MiniV dilution for 2.5 h. 100 µL of virus suspension containing 5 × 10⁴ infectious virus particles (multiplicity of infection of 1) were added to each well and incubated for 2 h. The medium was removed and cells were washed two times with sterile PBS before the addition of 1 mL of fresh medium. Cells were harvested 18 h post-infection and total RNA extracted using NucleoSpin RNA Plus kit (Macherey-Nagel), following manufacturer’s instructions. cDNA was generated using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) following the manufacturer’s instructions. Expression levels of GAPDH and SARS-CoV-2 Orf7a mRNA were determined by using the iTaq Universal SYBR Green 2x (Bio-Rad). Reactions were performed on an CFX96 (Bio-Rad) using the following program: 95 °C for 3 min and 45 cycles as follows: 95 °C for 10 s, 60 °C for 30 s. GAPDH mRNA level was used for the normalization of input RNA. Relative abundance of each specific mRNA was determined by using the ΔΔCT method as previously described⁶⁸. The following primers were used:

GAPDH-For 5′ – GAAGGTGAAGGTCGGAGTC – 3′

GAPDH-Rev 5′ – GAAGATGGTGATGGGATTTC – 3′

CoV-2 Leader_For 5′ – TCCCAGGTAACAAACCAACCAACT- 3′

CoV-2 Orf7a_Rev 5′ – AAATGGTGAATTGCCCTCGT- 3′.

Live virus neutralization assays

Purified IgG antibodies were serially diluted 2-fold in Opti-MEM, starting with a dilution of 1:20 and mixed with an equal volume of Opt-MEM containing 5 × 10⁴ pfu SARS-CoV-2 (final multiplicity of infection of 1). IgGs/virus mixes were incubated for 1 h at 37 °C and subsequently transferred to 24-wells containing 5 × 10⁴ A549-ACE2 cells seeded the day prior to infection. Cells were infected for 2 h at 37 °C, subsequently washed once with sterile PBS, and cultured for a further 6 h in a fresh medium. Cells were washed and harvested for RNA extraction and ORF7a mRNA expression levels were quantified by qRT-PCR as described above. Values were normalized to infection levels in absence of IgG antibodies. Relative inhibitory concentration of 50% (IC50) values were calculated using nonlinear regression.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.