Ethics declaration and biosafety statements

The animal experiment was approved by the Institutional Animal Welfare Committee of the Institut de Recerca i Tecnologia Agroalimentàries (CEEA-IRTA, registration number CEEA 365/2023) and by the Ethical Commission of Animal Experimentation of the Autonomous Government of Catalonia (registration number CEA-OH/12069/1) and conducted by certified staff. Experiments with SARS-CoV-2 were performed at the Biosafety Level-3 (BSL-3) facilities of the Biocontainment Unit of IRTA-CReSA (Barcelona, Spain) under the approval of the biosafety committee (registration number CBS 116/2023). Animal studies were carried out in compliance with the ARRIVE guidelines and in accordance with relevant guidelines and regulations.

Convalescent sera were collected from adult volunteers after written informed consent at the Medical University of Graz and were approved by the ethics committee of the Medical University of Graz (EK number: 34-203 ex 21/22). All research involving human samples was performed in accordance with the Declaration of Helsinki and the Research Organisation Act, the Federal Act on Hospitals and Sanatoria and the Clinical Trials Regulation.

Production of ACE2-Fc in glycoengineered plants

Wild-type and mutant ACE2-Fc proteins encompassing ACE2 residues 18-728 were produced as described previously for other ACE2-Fc variants^21,45. Briefly, recombinant ACE2-Fc fusion proteins were transiently expressed in leaves of glycoengineered N. benthamiana ΔXT/FT plants⁴⁶ by agrobacterium-mediated transfection. Batches of 30 g leaf wet weight were extracted and then purified by affinity chromatography using a protein A column, resulting in the isolation of up to 20 mg ACE2-Fc per batch. In contrast to previous studies^21,45, affinity-purified ACE2-Fc was not subjected to a final polishing step by size exclusion chromatography.

Enzyme-linked immunosorbent assays

Enzyme-linked immunosorbent assays were carried out according to standard protocols. Briefly, 96-well plates (Nunc MaxiSorp; Thermo Fisher Scientific, Waltham, MA, USA) were coated overnight at 4 °C with 200 ng per well of in-house produced Wuhan RBD⁴⁷ in 100 µl PBS. Plates were then washed three times with PBS containing 0.05% Tween 20 (PBST). From an initial concentration of 1 µg/ml, twofold serial dilutions of the ACE2-Fc samples were prepared in PBST containing 1% bovine serum albumin (dilution buffer), and 50 µl then added per well. After incubation for 1 h at 37 °C, the plates were washed three times with PBST prior to addition of 50 µl of 0.03 µg/ml peroxidase-conjugated goat antibodies to human Fc (Sigma-Aldrich, St. Louis, MO, USA) in dilution buffer prior to incubation for 1 h at 37 °C. After washing of the plates as above, bound peroxidase activity was detected by addition of 100 µl of ELISA substrate solution (0.1 mg/ml tetramethylbenzidine (Sigma-Aldrich) in 100 mM citric acid/sodium phosphate buffer (pH 5.0) containing 0.006% H₂O₂). After 5–15 min, the reaction was stopped by addition of 100 µl 0.18 M H₂SO₄ prior to measurement of the optical density at 450 nm (reference wavelength: 620 nm) using a Spark multi-channel spectrophotometer (Tecan, Grödig, Austria).

Differential scanning calorimetry

Differential scanning calorimetry experiments were performed as described¹⁷ using a MicroCal PEAQ-DSC Automated system (Malvern Panalytical, Malvern, UK), using 5–10 µM protein solutions in PBS. The heating was performed from 20 to 100 °C at a rate of 1 °C/min. The protein solution was then cooled in situ and an identical thermal scan was run to obtain the baseline for subtraction from the first scan. All measurements were performed in triplicates. Fitting was done with Origin 7.0 for DSC software using the non-2-state transition model.

ACE2 activity assays

The enzymatic activity of wild-type and mutant ACE2-Fc was determined using 50 µM 7-methoxycoumarin-4-yl-acetyl-Ala-Pro-Lys-2,4-dinitrophenyl (Bachem, Bubendorf, Switzerland) as substrate. Assays were performed in 50 mM MES buffer (pH 6.5) containing 300 mM NaCl, 10 µM ZnCl₂ and 0.01% Brij-35⁴⁸ in a final volume of 200 µl. After addition of 10 ng ACE2-Fc (appropriately diluted in assay buffer containing 0.1% bovine serum albumin) and incubation for 5 min at ambient temperature, reactions were stopped by addition of 1 ml 50 mM EDTA prior to analysis by spectrofluorimetry (excitation wavelength: 320 nm,emission wavelength: 405 nm). A standard curve was generated with 1–5 µM completely hydrolysed substrate. One unit of enzymatic activity corresponds to 1 µmol of substrate hydrolysed per min.

Cells

African green monkey kidney epithelial cells (VeroE6) were obtained from Biomedica (Vienna, Austria) and grown in Minimal Essential Medium (MEM) supplemented with Earle’s Salts and L-Glutamine (Thermo Fisher Scientific), 1% Penicillin–Streptomycin stock solution (PenStrep) and 2% fetal bovine serum (Thermo Fisher Scientific). Primary human air liquid interface cultures (MucilAir™) were purchased from Epithelix (Plan-les-Ouates, Switzerland) and cultured according to the manufacturer’s instructions.

Viruses

All experimental procedures with SARS-CoV-2 were performed in a BSL-3 laboratory. SARS-CoV-2 Wuhan-Hu-1 (wh19: Human 2019-nCoV Isolate, Ref-SKU 026V-03883), as disclosed by the Center for Disease Control and Prevention, was purchased from EVAg. The Omicron isolate 2021 (om21), lineage B.1.1.529 was also purchased from EVAg (SKU: 010V-04425). The Delta isolate (de21) and the Omicron isolate 2023 (om23; lineage EG5.1) were isolated at the Medical University of Graz in July 2021 and September 2023, respectively. Virus variants were propagated in VeroE6 cells at 37 °C and 5% CO₂ for 72–96 h. Prior to harvesting, cells were lysed by a freeze–thaw cycle, followed by a centrifugation step (10 min, 3000 × g) to remove cellular debris. Supernatants were filtered with 0.2 µm syringe filters (Thermo Fisher Scientific) and virus stocks were stored at − 80 °C until use. To determine viral titers, confluent VeroE6 cultures were infected with serial tenfold dilutions of virus stocks. After 96 h of incubation at 37 °C and 5% CO₂ cells were fixed with 4% neutral-buffered formalin and cytopathic effects were visualized by crystal violet staining. TCID₅₀ values were calculated by the Reed-Muench method⁴⁹. All virus stocks used in this study were sequenced as described elsewhere⁵⁰. Sequences are provided in the supplementary section (Supplementary Table 1). The virus strain used for the in vivo experiment was hCoV-19/France/GE1973/2020, clade G, D614G (S) supplied by the National Reference Centre for Respiratory Viruses hosted by Institut Pasteur (Paris, France).

Virus neutralization assays

The viral main protease inhibitor PF-07321332 (PF-332) was purchased from Selleckchem (Catalog No. S9866). Convalescent sera from 4 donors were collected at the Medical University of Graz between August 2021 and July 2022. Sera were heat-inactivated and pooled for neutralization assays. Neutralizing activity was assessed by cytopathic effect assays, as described previously⁵¹. Briefly, VeroE6 cells were seeded in 96-well plates at a density of 1 × 10⁴ per well. Antiviral compounds were serially diluted (fourfold) and incubated with SARS-CoV-2 for 15 min at 37 °C. Subsequently, decoy-virus mixtures were added to VeroE6 cells in triplicates and incubated for 96 h at 37 °C. Cells were fixed with 4% neutral-buffered formalin and cytopathic effects were assessed by crystal violet staining. The stained cells were dissolved in 10% acetic acid prior to measuring the optical density of the samples at 595 nm in a plate reader. Data were normalized to a no-decoy control (0% inhibition) and a no-virus control (100% inhibition). IC₅₀ values were calculated by nonlinear regression analysis with variable slopes in GraphPad Prism version 9.

Primary human ALI cultures (MucilAir, Epithelix) were washed once with PBS prior to infection and the basal medium was replenished with MucilAir culture medium containing 20 µg/ml ACE2-Fc. 5 × 10³ pfu SARS-CoV-2 (de21) were incubated with ACE2-Fc (20 µg/ml) and subsequently added to the apical side of the ALI culture. After 1 h incubation at 37 °C, the virus inoculum was removed, and the apical side was washed three times with PBS to remove residual extracellular virus. Virus release was assessed 24 h post-infection (hpi) by washing the apical side with 200 µl PBS (apical wash). Viral RNA was then isolated from the apical wash using the QiaAmp Viral RNA Minikit (Qiagen, Germany), according to the manufacturer’s protocol. SARS-CoV-2 replication was quantified via RT-qPCR using the QuantiTect Multiplex RT-qPCR Kit (Qiagen) with a Rotor Gene Q cycler (Qiagen). The reactions were performed in a total volume of 25 µL at 50 °C for 30 min followed by 95 °C for 15 min and 45 cycles of 95 °C for 3 s and 55 °C for 30 s. Forward primer: 2019-nCoV_N1-F 5′-GACCCCAAAATCAGCGAAAT-3′; reverse primer: 2019-nCoV_N1-R 5′-TCTGGTTACTGCCAGTTGAATCTG-3′; probe: 2019-nCoV_N1-P 5′-FAM-ACCCCGCATTACGTTTGGTGGACC-BHQ1-3′.

Transwell membranes were cut in half for intracellular virus quantification and immunohistochemical analysis. Intracellular virus RNA was isolated using the RNeasy Mini Kit (Qiagen). Viral RNA was quantified by two-step RT-qPCR using LunaScript RT SuperMix (New England Biolabs, Frankfurt, Germany) and Luna Universal qPCR Master Mix (New England Biolabs) together with the N1 primer pair mentioned above. HPRT1 was used as housekeeping gene (primers: 5′-TCAGGCAGTATAATCCAAAGATGGT-3′; 5′-AGTCTGGCTTATATCCAACACTTCG-3′).

In vivo experiment

A total of thirty-six 6- to 8-week-old male and female golden Syrian hamsters (Charles River Laboratories, Wilmington, MA, USA) were divided into 6 groups (6 animals/group). Animals were inoculated with SARS-CoV-2 as described previously⁵² and then treated on 4 consecutive days with either 2.5 mg/ml wild-type ACE2-Fc, 2.5 mg/ml mutACE2-Fc or vehicle (PBS) alone. Non-infected animals were treated on 4 consecutive days with 2.5 mg/ml ACE2-Fc plus 60 µM hexahydroxystilbene⁵³ in PBS. Briefly, animals were intranasally inoculated with 10⁴ TCID₅₀ per animal (100 µL total, 50 µL/nostril) of SARS-CoV-2 (hCoV-19/France/GE1973/2020, clade G, D614G (S)). This was considered day 0 post-infection (0 dpi). Non-infected animals received the same volume of PBS. At 1, 2, 3, and 4 dpi animals were treated intranasally as described above under inhalation anaesthetics (4% isoflurane). Clinical signs, body weight and oropharyngeal swabs were recorded and collected every day after challenge, until the end of the experiment (5 dpi). At 5 dpi all animals were euthanised via intraperitoneal administration of sodium pentobarbital (not exceeding a dose of 60 mg/kg). Prior to this, the animals were anesthetized using inhalation anesthesia (4% isoflurane). Lungs, spleens, hearts, kidneys and livers were then collected for pathological analysis. Further, lung samples were extracted to quantify viral loads and examine the expression levels of inflammatory marker genes. To quantify virus titers, lung tissues were homogenized using bead disruption and centrifuged to remove cellular debris. Supernatants were serially diluted (fivefold) and endpoint titrations were performed on confluent VeroE6 cells. TCID₅₀ values were calculated by the Reed-Muench method. Total RNA from lungs was isolated with the RNeasy Mini kit (Qiagen). Complementary DNA (cDNA) was synthesized from 1 µg of total RNA by reverse transcription (RT) using LunaScript RT SuperMix (New England Biolabs). Primer sequences used for quantitative real-time polymerase chain reaction (qPCR) are listed in Table 1.

Immunohistochemical analysis

Lung, liver, heart, spleen, and kidney samples from hamsters were collected on day 5 after viral challenge, fixed by immersion in 10% buffered formalin and embedded into paraffin blocks. Transwell membranes from ALI cultures were fixed in 4% neutral-buffered formalin and embedded into paraffin blocks. Formalin-fixed paraffin-embedded tissue specimens and ALI cultures were sectioned at 3 µm. SARS-CoV-2 nucleocapsid protein was detected by immunohistochemical staining using the rabbit monoclonal antibody 40143-R019 (Sino Biological, Eschborn, Germany) at 1:500 dilution. For visualization of the immunolabelled cells, the EnVision® + detection system (Agilent, Santa Clara, CA, USA) with 3,3’-diaminobenzidine as peroxidase substrate was used.

Evaluation of histopathological changes

Hematoxylin/eosin stained sections were examined by light microscopy and lung damage was semi-quantitatively scored based on the level of inflammation (0—no; 1—mild, 2 -moderate; or 3—severe) as described previously⁵².

Statistical analysis

Statistical data analysis was performed in GraphPad Prism software version 9. Statistical significances were analyzed by one-way analysis of variances (ANOVA) followed by Bonferroni post hoc test to correct for multiple comparisons or by unpaired t-test. P-values < 0.05 were considered statistically significant.