TYPE Original Research PUBLISHED 04 December 2023 DOI 10.3389/fimmu.2023.1291972 Novel Spike-stabilized trimers OPEN ACCESS with improved production EDITED BY William Murphy, University of California, Davis, United States protect K18-hACE2 mice and REVIEWED BY David Nemazee, golden Syrian hamsters from the The Scripps Research Institute, United States Arta Monjazeb, highly pathogenic SARS-CoV-2 UC Davis Health, United States *CORRESPONDENCE Beta variant Jorge Carrillo jcarrillo@irsicaixa.es Joaquim Segalés Carlos Ávila-Nieto 1‡, Júlia Vergara-Alert 2,3‡, joaquim.segales@irta.cat Pep Amengual-Rigo 4‡, Erola Ainsua-Enrich 1, Marco Brustolin 2,3†, Victor Guallar victor.guallar@bsc.es Marı́a Luisa Rodrı́guez de la Concepción 1, † PRESENT ADDRESSES Núria Pedreño-Lopez 1, Jordi Rodon 2,3†, Victor Urrea 1, Marco Brustolin, Edwards Pradenas 1, Silvia Marfil 1, Ester Ballana 1,5,6, Department of Biomedical Sciences, Institute of Tropical Medicine, Antwerp, Eva Riveira-Muñoz 1, Mònica Pérez 2,3, Núria Roca 2,3, Belgium Jordi Rodon, Ferran Tarrés-Freixas 1†, Julieta Carabelli 1, Guillermo Cantero 2,3, Institute of Virology, Charité- Anna Pons-Grı́fols 1, Carla Rovirosa 1, Carmen Aguilar-Gurrieri 1, Universitätsmedizin Berlin, Berlin, Germany Raquel Ortiz 1, Ana Barajas 1, Benjamin Trinité 1, Rosalba Lepore 4, Ferran Tarrés-Freixas, Unitat Mixta d’Investigació IRTA-UAB en Jordana Muñoz-Basagoiti 1†, Daniel Perez-Zsolt 1, Sanitat Animal, Centre de Recerca en Nuria Izquierdo-Useros 1,5,6, Alfonso Valencia 4,7, Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Julià Blanco 1,5,6,8, Bonaventura Clotet 1,5,6,8,9, Victor Guallar 4,7*, Barcelona, Spain; IRTA Programa de Sanitat Animal, Centre de Recerca en Sanitat Joaquim Segalés 2,10* and Jorge Carrillo 1,5,6* Animal (CReSA), Campus de la Universitat 1 IrsiCaixa AIDS Research Institute, Badalona, Spain, 2 Unitat Mixta d’Investigació IRTA-UAB en Sanitat Autònoma de Barcelona (UAB), Barcelona, Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Spain Barcelona (UAB), Barcelona, Spain, 3 IRTA Programa de Sanitat Animal, Centre de Recerca en Sanitat Jordana Muñoz-Basagoiti, Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), Barcelona, Spain, 4 Life Unitat Mixta d’Investigació IRTA-UAB en Sciences Department, Barcelona Supercomputing Center (BSC), Barcelona, Spain, 5 Germans Trias i Sanitat Animal, Centre de Recerca en Pujol Research Institute (IGTP), Badalona, Spain, 6 Centro de Investigación Biomédica en Red (CIBER) Sanitat Animal (CReSA), Campus de la de Enfermedades Infecciosas, Instituto de Salud Carlos III (ISCIII), Madrid, Spain, 7 Catalan Institution Universitat Autònoma de Barcelona (UAB), for Research and Advanced Studies, Barcelona, Spain, 8 Centre for Health and Social Care Research Barcelona, Spain; IRTA Programa de Sanitat (CESS), Faculty of Medicine, University of Vic – Central University of Catalonia (UVic – UCC), Animal, Centre de Recerca en Sanitat Vic, Spain, 9 Fundació Lluita contra les Infeccions, Hospital Germans Trias i Pujol, Badalona, Spain, Animal (CReSA), Campus de la Universitat 10 Departament de Sanitat i Anatomia Animals, Facultat de Veterinària, UAB, Cerdanyola del Autònoma de Barcelona (UAB), Barcelona, Vallès, Spain Spain ‡ These authors have contributed equally to this work RECEIVED 10 September 2023 Most COVID-19 vaccines are based on the SARS-CoV-2 Spike glycoprotein (S) or ACCEPTED 20 November 2023 PUBLISHED 04 December 2023 their subunits. However, S shows some structural instability that limits its immunogenicity and production, hampering the development of recombinant S-based vaccines. The introduction of the K986P and V987P (S-2P) mutations increases the production and immunogenicity of the recombinant S trimer, suggesting that these two parameters are related. Nevertheless, S-2P still shows some molecular instability and it is produced with low yield. Here we described a novel set of mutations identified by molecular modeling and located in the S2 region of the S-2P that increase its production up to five-fold. Besides their immunogenicity, the efficacy of two representative S-2P-based mutants, S- Frontiers in Immunology 01 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 29 and S-21, protecting from a heterologous SARS-CoV-2 Beta variant challenge was assayed in K18-hACE2 mice (an animal model of severe SARS-CoV-2 disease) and golden Syrian hamsters (GSH) (a moderate disease model). S-21 induced higher level of WH1 and Delta variants neutralizing antibodies than S-2P in K18-hACE2 mice three days after challenge. Viral load in nasal turbinate and oropharyngeal samples were reduced in S-21 and S-29 vaccinated mice. Despite that, only the S-29 protein protected 100% of K18-hACE2 mice from severe disease. When GSH were analyzed, all immunized animals were protected from disease development irrespectively of the immunogen they received. Therefore, the higher yield of S-29, as well as its improved immunogenicity and efficacy protecting from the highly pathogenic SARS-CoV-2 Beta variant, pinpoint the S- 29 mutant as an alternative to the S-2P protein for future SARS-CoV-2 vaccine development. KEYWORDS COVID-19, SARS-CoV-2, vaccine, neutralizing antibodies, humoral response, Spike glycoprotein 1 Introduction the human immunodeficiency virus (HIV) (13). Several studies have shown that it is possible to stabilize these proteins in their Vaccines have been extensively used to control infectious prefusion state and improve their production and immunogenicity diseases. While smallpox is the only pathogen that has been (13). In this sense, the introduction of two proline mutations in the eradicated in human, mass immunization programs have reduced S2 (S-2P) has been proposed as a common strategy for the the spread of other infectious diseases, including tetanus, polio, and stabilization of this glycoprotein from several coronaviruses, measles (1, 2). In January 2020, the severe acute respiratory including the SARS-CoV-2 (11, 14). In fact, some of the most syndrome coronavirus 2 (SARS-CoV-2) was identified as the used SARS-CoV-2 vaccines, such as BNT162b2, mRNA-1273 or causal agent of the coronavirus disease 2019 (COVID-19), and Ad26.COV2.S are based on the S-2P strategy (15–17). many laboratories rapidly started programs to develop COVID-19 Importantly, the S-2P protein still retains some structural vaccines (3, 4). To date, several COVID-19 vaccines are available instability and generates a poor yield when the protein is and have contributed to the reduction of COVID-19 impact on produced as recombinant protein (around 0.5 mg/L) (11, 18). public health (5, 6). However, COVID-19 is still present in the Several studies have addressed these limitations by introducing world and new SARS-CoV-2 variants continue emerging with high additional stabilizing mutations. In this sense, the incorporation transmissibility and/or resistance to the immune responses elicited of four additional prolines (S-6P) improved the stability of the S after infection and/or vaccination (7). trimer and increased its yield by ten-fold (18). In another Within the SARS-CoV-2 proteome, the Spike (S) mediates virus approach, the incorporation of the mutations D614N, A892P, attachment by binding to the angiotensin-converting enzyme 2 A942P, and V987P stabilized the S protein in a close-prefusion receptor (ACE2) on the surface of target cells. After being primed by state and increased its yield by 6-fold (19). However, whereas the host proteases, S promotes viral entry and cell infection (8, 9). addition of disulfide bridges between different domains of the S Therefore, most SARS-CoV-2 vaccines are based on this protein glycoprotein reduced the motility of the RBD, it failed improving since it is the main target of neutralizing antibodies (NAbs) recombinant trimer production (20, 21). Alternatively, pre-fusion [particularly, the receptor binding domain (RBD)] (3, 4, 10). The stabilizing mutations have also been identified by high- S is a trimer, and each monomer has two subunits: the S1 throughput methods. Thus, the addition of D994Q mutation to extracellular, and the S2 membrane anchor subunits. While the the S-2P backbone increased its production as soluble S1 binds to ACE2 via the RBDs, the S2 domain participates in the recombinant protein by three-fold (22). However, it remains membrane fusion process, which involves drastic conformational poorly understood how all these mutations modify the S changes (8, 11, 12). Thus, the S glycoprotein shows a certain degree immunogenicity compared to S-2P. of structural instability that might hamper its production as Here, we describe a set of novel mutations that increase recombinant protein and modulate its immunogenicity. This SARS-CoV-2 S yields by five-fold, while maintaining the feature is shared with functional homologous surface proteins immunogenicity and protection efficacy against the development from other viruses, including the S of Middle East respiratory of SARS-CoV-2-induced disease in K18-hACE2 transgenic mice syndrome coronavirus (MERS-CoV), the Fusion protein of the and golden Syrian hamsters (GSH) previously observed with the respiratory syncytial virus (RSV), or the Envelope glycoprotein of S-2P prototype. Frontiers in Immunology 02 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 2 Materials and methods Coomassie G-250 staining (ThermoFisher Scientific). Purified proteins were stored at -80°C until use. 2.1 Recombinant trimeric S glycoprotein design and modeling 2.3 Recombinant S proteins quantification Unsolved secondary structures of the trimer in closed (PDB: and RBD exposure 6VXX) and open (PDB: 6VYB) conformations (8) were reconstructed using SwissModel (23). Then, all possible single S variants production was determined in duplicate by an in- mutations in both conformations were modeled using FoldX (24). house developed ELISA. Nunc MaxiSorp ELISA plates were coated For selecting potential variants, two different approaches were used. overnight at 4°C with the anti-6xHis antibody HIS.H8 First, we computed the Gibbs free energy change (DDGopen) (ThermoFisher Scientific) at 2 µg/mL in PBS. Then, plates were between the WT and the mutant using the open state as a blocked with PBS/1% of bovine serum albumin (BSA, Miltenyi reference. Negative values indicate introduction of stabilization. Biotech) for two hours at room temperature. The recombinant Second, comparison of the Gibbs free energy changes upon SARS-CoV-2 Spike S1+S2-His protein (Sino Biological) was used as mutation between the closed (DDGclosed) and open (DDGopen) standard at 1µg/mL followed by 3-fold dilutions. S variants samples conformations (DDG) revealed a set of mutations predicted to (supernatant and purified proteins) were serially diluted so that they strengthen the open conformation in combination with 2P could be quantified. Standards and samples were diluted in blocking (positive values indicate stabilization of the open state). For both buffer. After blocking, 50 µL of each sample and standard were approaches, all single mutations predicting beneficial energies (or added to plates and incubated overnight at 4°C. Next day, the just slightly neutral/worst values) were addressed by inspecting the SARS-CoV-2 Spike S2 IgG rabbit antibody (Sino Biological) was three-dimensional models. In this regard, the final selection was diluted (1/1000) in blocking buffer and added to plates for 2 hours based on: i) selection of mutations predicted to increase the stability at room temperature. The HRP-donkey anti-rabbit IgG (H+L) (1/ of the open-conformation using FoldX; ii) selection of mutations 10.000) (Jackson ImmunoResearch) was used as detection antibody. predicted to increase the stability of the open-conformation over the Plates were revealed with o-Phenylenediamine (OPD) (Sigma closed one using FoldX, iii) selection of mutations creating well- Aldrich) and stopped using 4N of H2SO4 (Sigma-Aldrich). Signal defined intermolecular interactions between the RBD domains was analyzed as the optical density (OD) at 492 nm with noise (including hydrophobic bonds, p-p interactions and cation-p correction at 620 nm. S mutants quantification was done according interactions, ionic bonds, hydrophobic contacts or cavity filling to standard curve using Graphpad. mutations) that would exert a positive impact in the open state or a The RBD exposition was evaluated using an in-house ELISA. negative one on the closing motion of the trimer. Nunc Maxisorp ELISA plates were coated with the HIS.H8 anti- 6xHis tag monoclonal antibody (ThermoFisher Scientific) at 2 µg/ mL in PBS and incubated overnight at 4°C. Then, plates were 2.2 Recombinant protein production washed and blocked for two hours using PBS/1% of BSA (Miltenyi and purification Biotech) at room temperature. After that, each sample was added in triplicate and incubated overnight (4°C) at 0.1 µg/ml. The next day, Recombinant trimeric S glycoproteins based on the Wuhan plates were washed and incubated with a purified ACE2-human IgG WH-1 sequence were designed as previously described by Wrapp Fc-fusion protein at 0.1 µg/mL in blocking buffer for 2 hours at (11). Briefly, the C-terminal end of the S, without the room temperature. Next, plates were incubated for 1 hour at room transmembrane and cytoplasmatic domains, was fused to a T4 temperature with HRP conjugated- (Fab)2 goat anti-human IgG (Fc foldon trimerization domain in tandem with an 8xhis tag and a specific) (1/10000) (Jackson ImmunoResearch). Plates were strep tag II. The furin cleavage site was removed by mutating it to revealed with OPD (Sigma Aldrich) and stopped using 2N of GSAS. All constructs include the K986P and V987P mutations. H2SO4 (Sigma Aldrich). Signal was analyzed as the OD at 492 DNA constructs were supplied by GeneART (ThermoFisher nm with noise correction at 620 nm. ACE2 binding was determined scientific) as pcDNA3.4-based plasmids. Proteins were produced as the signal obtained with the ACE2-human Fc fusion protein, by transient transfection using the Expi293 expression system normalized according to protein concentration. Data are shown as (ThermoFisher Scientific), following the manufacturer fold change compared to S-2P. instructions. Five days after transfection, cell culture supernatants were harvested and clarified by centrifugation (3000xg for 20 minutes) or using Sartoclear Dynamics® Lab V (Sartorius). 2.4 Viral stock preparation Supernatants were then filtered at 0.2 mm using Nalgene Rapid Flow sterile single using vacuum filter units (ThermoFisher In vivo challenge experiments were performed using Cat24 Scientific) and were purified using Ni-Sepharose Excel histidine- SARS-CoV-2 B.1.531 (Beta) variant isolate (EPI_ISL_1663571) tagged protein purification resin (Cytiva), concentrated, and buffer (25, 26). Cat24 was isolated from a nasopharyngeal swab from a exchanged to phosphate buffer saline by ultrafiltration (Merck COVID-19 affected patient, as previously described (25, 26) and Millipore). Integrity and purity of purified proteins were analyzed subsequently grown and titrated in Vero E6 cells (ATCC CRL- by sodium dodecyl sulfate polyacrylamide gel electrophoresis and 1586). Vero E6 cells were cultured in Dulbecco’s modified Eagle Frontiers in Immunology 03 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 medium (Invitrogen) supplemented with 10% fetal bovine serum analysis. Furthermore, blood samples were collected before each (FBS, Invitrogen), 100 U/mL penicillin, and 100 mg/mL immunization, viral challenge, and under euthanasia. Blood was left streptomycin, (all from Invitrogen). at room temperature for two hours for clotting, and serum was collected after centrifugation (10 minutes at 5000xg) and stored at -80°C until use. 2.5 In vivo immunization and challenge experiments 2.6 Quantification of anti-S and anti-RBD K18-hACE2 transgenic mice (B6.Cg-Tg(K18-ACE2)2Prlmn/J; antibodies by ELISA stock #034860; Jackson Laboratories) were maintained by breeding K18-hACE2 hemizygous mice with C57BL/6J mice, An in-house ELISA was developed to evaluate IgG antibodies following the instructions of Jackson Laboratory (https://www.jax. elicited against the S and RBD glycoproteins in serum samples org/strain/034860). Offspring genotyping was performed obtained as described before. Nunc MaxiSorp ELISA plates were according to the protocol 38170: Probe Assay - Tg (K18-ACE2) coated overnight at 4°C. Half plate was coated with 50 ng/well of 2Prlmn QPCR version 2.0 (https://www.jax.org/Protocol? antigen diluted in PBS (S or RBD, Sino Biological) and the other stockNumber=034860&protocolID=38170). GSH were purchased half plate was incubated only with PBS. Next day, the whole plate from Envigo and maintained by brother/sister mating. Both K18- was blocked for two hours at room temperature using blocking hACE2 transgenic mice and GSH colonies were established at the buffer [PBS with 1% of bovine serum albumin (BSA, Miltenyi Centre for Comparative Medicine and Bioimage (CMCiB). Animal Biotech)]. After that, 50 µL of the appropriate standard or diluted studies were evaluated and approved in advance by the Committee on samples were added to each half plate in duplicates and incubated the Ethics of Animal Experimentation of the IGTP and count with overnight at 4°C. All samples were prepared in blocking buffer. For the authorization of Generalitat de Catalunya (Code: 10965 the mouse standard curve, we used anti-6xHis antibody His.H8 and 11094). (ThermoFisher Scientific) starting at 1 µg/mL followed by 3-fold Ninety-one K18-hACE2 mice (balanced female-male ratio, 7-9 dilutions. For GSH standard, a positive GSH serum was used. GSH weeks old) were distributed in five experimental groups: S-2P standard was prepared as seven 1/3 dilution of a stating 1/100 (n=21), S-21(n=22), S-29 (n=22), unvaccinated and challenged dilution. To reduce inter-assay variability, plates were run in controls (n=16), and uninfected negative controls (n=10). In the parallel, and each plate contained samples from all experimental case of GSHs, a total of forty-nine animals were used (balanced groups. The following day, plates were incubated with detection female-male ratio, 5-7 weeks old) and distributed in five antibodies for 1 hour at room temperature. HRP conjugated (Fab)2 experimental groups: S-2P (n=11), S-21 (n=11), S-29 (n=11), Goat anti-mouse IgG (Fc specific (1/20,000 dilution), or Goat anti- unvaccinated and challenged controls (n=11), and uninfected hamster IgG (H-L) (1/20,000 dilution) (all from Jackson negative controls (n=5). Both mice and hamsters from S-2P, S-21, ImmunoResearch) were used as secondary antibodies in the and S-29 groups were immunized with 15 mg of recombinant mouse and GSH IgG ELISA, respectively. Finally, plates were protein with AddaVax™ (Invivogen) as adjuvant in the hock revealed with o-phenylenediamine dihydrochloride (OPD) (Sigma (27). Three weeks later, immunized animals were boosted with a Aldrich) and stopped using 2N of H2SO4 (Sigma Aldrich). Optical second dose of the same formulation. Control animals were primed density (OD) was measured at 492 nm with noise correction at 620 and boosted with PBS and AddaVax™. Two weeks after boosting, nm. The background OD obtained from the half antigen-free plate mice were challenged with 1000 TCID50 of SARS-CoV-2 (Cat24 was subtracted from the half antigen-coated plate to obtain the isolate) and followed up for 14 days. GSHs were challenged three specific signal for each sample. Data are showed as arbitrary units weeks after boosting with 10000 TCID50 of SARS-CoV-2 (Cat24 (AU/ml) according to the standard used. isolate) and followed up for 7 days. After infection, body weight and clinical signs were monitored daily until the end of the experiment. Six mice for each experimental group, except the uninfected 2.7 Neutralizing activity of serum samples controls, were euthanized on days 3 and 6. The remaining mice were followed up until day 14 post-infection. Three and four Sera neutralizing activity was evaluated as described by hamsters from each experimental group, except the uninfected Pradenas et al. (28). HIV reporter pseudoviruses expressing control group, were euthanized on days 2 and 4, respectively. The SARS-CoV-2 S glycoprotein and carrying the luciferase gene were remaining GSHs were euthanized on day 7 post infection. In both produced in Expi293F cells (ThermoFisher Scientific) by co- challenge experiments, uninfected control group was euthanized at transfection of the pNL4-3. Luc.R-. E- plasmid (NIH AIDS the end of the experiment. In addition, any animal showing weight Reagent Program (29)) and plasmids coding for the following loss higher than 20%, a drastic lack of motility, or a significant SARS-CoV-2 S glycoproteins lacking the last 19 amino acid in C- reduction of their response to stimuli were euthanized according to terminal: Wuhan (WH1), Beta, Delta, or Omicron variants. VSV- the humane endpoints defined in the supervision protocol. G-pseudotyped pseudoviruses were used as negative controls. Biological samples were collected after euthanasia, including Transfections were performed with ExpiFectamine293 Reagent kit oropharyngeal swab, nasal turbinate, lung, and brain (only in the (ThermoFisher Scientific). Forty-eight hours later, cell supernatants case of mice) to determine viral loads and perform histopathological were harvested, filtered at 0.45 µm and frozen at -80°C until use. Frontiers in Immunology 04 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 Pseudovirus titration was performed using HEK293T cells gene expression was measured in duplicate for each sample using overexpressing WT human ACE-2 (HEK293T/hACE2) (Integral TaqMan gene expression assay (ThermoFisher Scientific) as Molecular, USA). amplification control. Serum samples were inactivated at 56°C for 60 minutes before use. Once inactivated, serum samples were serially diluted 1/3 in cell culture medium (RPMI-1640, 10% fetal bovine sera) (range 1/60–1/ 2.9 Pathology and immunohistochemistry 14,580) and incubated with 200 TCID50 of SARS-CoV-2-derived pseudoviruses for 1 hour at 37°C. Then, 1x104 HEK293T/hACE2 Nasal turbinate and lung from mice and GSHs, and additionally cells treated with DEAE-Dextran (Sigma-Aldrich) were added. brain from mice, were collected after euthanasia and fixed After 48 hours, plates were read using BriteLite Plus Luciferase by immersion in 10% buffered formalin and embedded into reagent (PerkinElmer, USA) in an EnSight Multimode Plate Reader. paraffin. Then, tissue slides were stained with hematoxylin/ Neutralizing activity was calculated using a four-parameter logistic eosin and examined by optical microscopy to be analyzed equation in Prism 8.4.3 (GraphPad Software, USA) and visualized histopathologically. Samples were scored semi-quantitatively as normalized ID 50 (reciprocal dilution inhibiting 50% of based on the level of inflammation (0-No lesion; 1-Mild, 2- the infection). Moderate, or 3-Severe lesion) as described in (31, 32). The levels of SARS-CoV-2 Nucleoprotein in tissue slides were determined by immunohistochemistry. A rabbit monoclonal 2.8 Viral load quantification in antibody 40143-R019 (Sino Biological) at 1:15,000 dilution, and oropharyngeal swab and tissue samples the EnVision®+ System linked to horseradish peroxidase (HRP, Agilent-Dako) and 3,3’-diaminobenzidine (DAB) were used. A After euthanasia, oropharyngeal swab and samples from nasal semi-quantitative score was used to measure the amount of viral turbinate, lung, and brain were collected in 1.5 mL tubes with antigen in the analyzed tissues (0-No antigen detection, 1-low, 2- DMEM media supplemented with penicillin (100 U/mL) and moderate and 3- high amount of antigen) according to previous streptomycin (100 µg/mL). Tissues were homogenized twice at 25 classifications (31, 32). Hz for 30 seconds using a Tissue Lyser II, and a 1.5 mm Tungsten bead (QIAGEN). After that, samples were centrifuged for 2 minutes at 2000xg and supernatants were collected and processed using the 2.10 Statistical analysis Viral RNA/Pathogen Nucleic Acid Isolation kit and a KingFisher instrument (ThermoFisher Scientific), or an IndiMag pathogen kit Anti-S and anti-RBD IgG data, as well as neutralizing activity (Indical Bioscience) on a Biosprint 96 workstation (QIAGEN), differences among groups at each time point were analyzed using following manufacturer’s instructions. Kruskal-Wallis and Conover’s post-hoc tests with multiple PCR amplification was based on the 2019-Novel Coronavirus comparison correction by using false discovery rate (FDR). Real-Time RT-PCR Diagnostic Panel guidelines, following the Differences among animals within a particular group along time protocol developed by the American Center for Disease Control were analyzed using the Friedman test and Conover’s post-hoc tests and Prevention (https://www.fda.gov/media/134922/download). for paired data and corrected for multiple comparison by FDR. Briefly, 20 mL PCR reaction was set up containing 5 mL of RNA, Krustal Wallis and Dunn’s post-hoc test were used in weight 1.5 mL of N2 primers and probe (2019-nCov CDC EUA Kit, variation in SARS-CoV-2 challenged animals. Severe disease Integrated DNA Technologies) and 10 ml of GoTaq 1-Step RT- incidence was represented by Kaplan Meier plots and Mantel-Cox qPCR (Promega, Madison, WI, USA). Thermal cycling was test was implemented to calculated statistical differences against performed at 50°C for 15min for reverse transcription, followed uninfected group. To analyze SARS-CoV-2 gRNA and sgRNA data, by 95°C for 2 min and then 45 cycles of 95°C for 10 sec, 56°C for 15 a Peto & Peto Left-censored k-sample test corrected by FDR was sec and 72°C for 30 sec in the Applied Biosystems 7500 or performed. Asymptotic Generalized Pearson Chi-Squared test with QuantStudio5 Real-Time PCR instruments (ThermoFisher FDR correction was applied to histopathology analysis. P values are Scientific). For absolute quantification, a standard curve was built indicated as follows: * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. using 1/5 serial dilutions of a SARS-CoV2 plasmid (2019- Statistical analyses were conducted using the R software nCoV_N_Positive Control, 200 copies/mL, Integrated DNA environment (version 4.1). Technologies) and run in parallel in all PCR determinations. Triplicates were performed to determine viral load of each sample, which was extrapolated from the standard curve (in 3 Results copies/mL) and corrected by the corresponding dilution factor. Alternatively, results are shown as Ct or 2-DCt. 3.1 Strategy for S glycoprotein stabilization SARS-CoV-2 subgenomic RNA was quantified as previously described (30) with the following primers (Forward; 5- To increase the S glycoprotein stability and immunogenicity, we CGATCTCTTGTAGATCTGTTCTC-3′; Reverse, 5′-ATATTG followed two different approaches: 1) introduction of point CAGCAGTACGCACACAA-3′) and probe (5′- FAM-ACACTAG mutations into the S sequence to increase its stabilization (using CCATCCTTACTGCGCTTCG-TAMRA-3′). Mouse or GSH gapdh the open state as a reference structure), and 2) increase of RBD Frontiers in Immunology 05 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 exposure by forcing an open conformation. In this regard, we built a hydrophobic pockets) between different chains of the S trimer computational pipeline involving the three-dimensional modeling were prioritized (Figure 1A). of all possible single mutations in both scenarios (see the Methods S mutants, based on the two prolines (K986P/V987P) stabilized section for more details). Moreover, all single mutations that Wuhan WH-1 sequence, (Figure 1B) were then produced, and showed a preference for any of these two conditions were visually yields evaluated (Figure 1C). Based on their production levels, these inspected. Mutations that clearly generated well-defined glycoproteins were classified into three different groups (Figures 1B, interactions (e.g. hydrogen bonds, ionic interactions for filling C). Group 1 included those constructs (i.e., S-29 and S-22) that were A B C D E F G FIGURE 1 Selection of mutations that stabilized S glycoproteins. (A) S stabilizing mutations (red variants, left plot), or amino acids changes that increased RBD exposition (blue variants, right plot) were selected based on energetic filters and visual inspection. Positive energy values indicate stabilization of the open structure versus the closed one. Mutations with neutral (or slightly opposite) energetic trend were included. (B) List of S constructs that incorporate the selected mutations identified in (A). (C) Yields of recombinant S mutants in a five-day cell culture supernatant. Mean plus standard deviation of three experiments are shown. (D) RBD exposure index in selected recombinant proteins. Data are shown as ratio between RBD binding and total protein. Mean plus standard deviation of three experiments are shown. (E) Presence of a cluster of three Glu residues (one Glu1092 from each chain) that are facing each other in close proximity, with no positive residues nearby. Location of Thr912 is underlined as well. Also notice that one of the three Thr has been mutated to Arg clearly showing a salt bridge interaction with the glutamic acid. (F) A detail of the proximity of the 1113 residue, already mutated to Arg, to the Glu1092 cluster. Thr912 is also showed. Structure models were based on 6VXX PDB structure. (G) Detail of the RBD opening process and location of some key residues. The red and green ribbons indicate the difference between the open and closed states, underlying the position of the two consecutive aspartic acid residues, Asp428 and Asp427, at the tip of the RBD domain. In orange ribbons the location of the S758E mutation is shown. Notice that the inserted glutamic residue collides with the neighbor helix (pink ribbons). Models generated from the 6VXX (closed) and 6VYB (open) PDB structures. Frontiers in Immunology 06 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 produced at the highest levels (five-fold compared to the S-2P CoV-2 Beta variant. Blood and tissue samples were collected after protein). Group 2 contained S-21, S-24, S-26, S-27, S-30, and S-31, viral challenge on days 3 (n=6), 6 (n=6) and 14 (n=10 for S-21, S-29 whose production was intermediate (two-fold higher than S-2P). and uninfected controls, and n=8 for S-2P) to analyze tissue damage Last, Group 3 included those S mutants with a protein yield lower and viral replication (Figure 2A). All mice that developed severe than S-2P (S-20, S-23, S-25, S-28, S-32, S-33, S-34, S-35, S-36, S-37, disease (one mouse in the S-2P and S-21 groups, and all mice from and S-38). Remarkably, all constructs designed to increase RBD the positive control group) were euthanized before day 14 following exposure were in Group 3, suggesting that those mutations humane endpoints and were analyzed separately. drastically impacted the S stability and/or its production. Anti-RBD (Figure 2B) and anti-S (Supplementary Figure 1A) However, most constructs with an improved production, also IgG humoral responses were evaluated prior to each immunization showed a better RBD exposure (Figures 1C, D). and viral challenge, and in euthanized animals after infection on Variants S-22 and S-29, with higher expression yield, days 3, 6, and 14, or due to humane endpoints. Regardless of the introduced a positive charge per chain in a local area where the immunogen used, all vaccinated animals developed similar anti- Glu1092 of each chain might cause destabilization. Figure 1E shows RBD (Figure 2B) and anti-S IgG levels (Supplementary Figure 1A), the presence of this cluster of Glutamic acid residues facing each which increased after each immunization and after viral challenge other and how the T912R mutation in S-29 might introduce (p<0.01, Conover´s post-hoc test). Since we did not identify significant stabilization. Similarly, the Q1113R mutant in S-22 significant differences in the humoral responses among vaccinated placed an arginine next to Glu1092 (Figure 1F). Analogous groups after challenge (Supplementary Figures 1B, C), we pooled observations can be extracted of most mutants introducing a net these mice in a single “post-challenge” group to simplify the charge. We also observed that most mutants increasing RBD analysis. Of note, unvaccinated but challenged positive controls exposure, such as S-21, S-24 and S-29, incorporated the S758E elicited low levels of anti-RBD and anti-S IgG antibodies (Figure 2B mutation. This mutation is in the vicinity of the tip of the closed and Supplementary Figure 1A) that were detected from day 6 after RBD domain, where two consecutive Aspartic acid residues, Asp427 viral challenge (Supplementary Figures 1B, C). and Asp428 are located (Figure 1G). After modeling the possible Sera neutralizing activity against SARS-CoV-2 WH1, Beta, positioning of Glu758 (with an initial significant clash with a helix Delta, and Omicron variants was detected in all three vaccinated backbone), we speculate that it would be displaced toward the tip of groups (Figures 2C–F). Interestingly, despite having similar levels of the RBD domain and destabilize the closed conformation by anti-RBD IgGs (Figure 2B), and a slightly higher levels of anti-S IgG electrostatic repulsion. antibodies (Supplementary Figure 1A), S-2P vaccinated mice Interestingly, S-22 and S-29 constructs share the conservative showed lower sera neutralizing activity against WH1 and Delta mutation K947R located in the middle of the heptad repeat 1 (HR1) variants on day 3 than those immunized with S-21 (Figures 2C, E) helix, which could enhance the thermal stability of the protein (33). (WH1 p<0.05; Delta p=0.052, Conover´s post-hoc test). Sera neutralizing activity against Delta and Beta variants increased over time in the S-2P vaccinated group after viral challenge (Beta 3.2 S-21 and S-29 vaccination protects p<0.05; Delta p=0.052, Conover´s post-hoc test), suggesting that K18-hACE2 mice from SARS-CoV-2- infection boosted the humoral response in these animals. In line, induced disease unvaccinated mice developed low sera neutralizing activity against the SARS-CoV-2 Beta variant (Figure 2D) with some cross- To investigate the impact of S mutations on its immunogenicity reactivity with WH1 but limited cross-neutralizing activity against and capability to protect from SARS-CoV-2-induced disease, we other SARS-CoV-2 variants (Figures 2C, E, F) after viral challenge. selected two representative S mutants from group 1 (S-29) and 2 (S- No boost effect on sera neutralizing activity was detected in S-21- 21). Then, we performed an immunization study using K18-hACE2 and S-29-immunized mice after challenge, suggesting that the transgenic mice that were subsequently challenged with SARS- humoral response reached a plateau in these groups. CoV-2 B.1.351 (Beta) variant (Figure 2A). We used this To determine whether S-2P-, S-21-, and S-29-vaccinated mice experimental design for the following reasons: 1) K18-hACE2 were protected against SARS-CoV-2-induced severe disease, we transgenic mice develop a severe form of the disease that leads to measured body weight evolution (Figure 2G), clinical sings, and death (34) unless animals are vaccine-protected; 2) the SARS-CoV- survival rate after viral challenge (Figure 2H). A progressive weight 2 Beta variant is partially resistant to antibodies elicited by natural loss was observed in all unvaccinated but challenged mice starting infection or vaccination with immunogens based on the original on day 2 post-challenge. These mice developed a severe disease on strain (Wuhan, WH1) (35); and 3) the SARS-CoV-2 Beta variant is days 5-9 post-infection and were euthanized following humane one of the most pathogenic SARS-CoV-2 variants tested in K18- endpoints. Conversely, all vaccinated mice (except one S-2P- and hACE2 transgenic mice (34). Thus, we established five experimental one S-21-immunized mice), were disease-free (Figure 2H) and did groups: S-2P (n=21), S-21 (n=22), S-29 (n=22), infected positive not experience weight loss. All mice belonging to S-29 group were controls (n=16), and uninfected negative controls (n=10). Mice protected from severe disease development (Figures 2G, H). from S-2P, S-21, and S-29 groups were immunized twice, three The presence of SARS-CoV-2 in oropharyngeal swabs and weeks apart. Animals from both control groups received antigen- tissue samples from nasal turbinate, lung, and brain was analyzed free doses. Two weeks after the boost, all animals (except the by RT-qPCR. All vaccinated mice had significantly lower levels of negative controls) were intranasally challenged with the SARS- genomic viral RNA (gRNA) in lung on day 3 post-inoculation Frontiers in Immunology 07 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 A B C D E F G H FIGURE 2 Prophylactic activity of S-21 and S-29 immunization and vaccine-induced humoral response elicited in K18-hACE2 transgenic mice challenged with SARS-CoV-2 B.1.351 (Beta) variant. K18-hACE2 transgenic mice were immunized following a prime/boost strategy with S-21, S-29, or S-2P, and challenged with SARS-CoV-2 Beta. The vaccine-induced humoral responses, weight changes, and survival of mice were evaluated after immunization and/or viral challenge. (A) Overview of immunization strategy and infection timeline. Biological samples were collected at indicated time points. (B) Kinetics of anti-RBD antibodies in serum samples. Red triangles: S-2P group (n=21). Yellow squares: S-21 (n=22). Purple diamond: S- 29 (n=22). White circles: unvaccinated-challenge mice (n=16). Grey circles: unvaccinated-uninfected mice (n=10). Groups in each time point were analyzed using Kruskal-Wallis and Conover’s post-hoc tests with multiple comparison correction by FDR. Differences among animals within a particular group along time were analyzed using Friedman and Conover’s post-hoc test for paired data with FDR correction. Sera neutralizing activity against: (C) SARS-CoV-2 WH-1 variant, (D) B.1.351 (Beta) variant, (E) B.1.617.2 (Delta), and (F) B.1.1.529 (Omicron) variants after viral challenge. Neutralization data were analyzed as indicated in “b”. (G) Percentage of weight variation in SARS-CoV-2 B.1.351 infected K18-ACE2 mice over time. Statistical analysis was performed against the unvaccinated and challenged group using Kruskal Wallis with Dunn´s post-hoc test. (H) Kaplan-Meier plot showing the percentage of SARS-CoV-2-infected animals that were disease-free at the end of the experiment. Statistical analysis was performed against unvaccinated group using Mantel-Cox test. * p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001. Mean plus standard errors of the means (SEM) are shown. compared to the positive control group (p<0.05, Peto & Peto Left- over time in all immunized mice regardless of the analyzed sample, censored k-sample test) (Figure 3A). Most notably gRNA was whereas the opposite outcome was observed in mice belonging to scarcely detected in brain of vaccinated mice compared to the challenged control group, and in those vaccinated mice that unvaccinated animals (Figure 3A). Interestingly, S-21- and S-29- developed severe disease (Figure 3A). Similar results were observed vaccinated mice showed lower viral load in nasal turbinate than S- when subgenomic viral RNA (sgRNA) was analyzed using the same 2P and control groups on day 3, and S-2P vaccinated mice on day 6 biological samples (Supplementary Figure 1D). (p<0.05, Peto & Peto Left-censored k-sample test) (Figure 3A). The To confirm active viral replication, nucleoprotein (NP) levels lack of differences with the control group on day 6 could be were analyzed by immunohistochemistry. NP was hardly detected explained due to the small number of unvaccinated mice that in lung and brain samples from S-2P, S-21 and S-29 groups reached this timepoint, since the majority had been euthanized (Figure 3B). These data are in accordance with the viral loads on day 5 post-infection (Figure 2H). Similarly, S-21 and S-29 groups detected in these samples. Despite that, some tissue damage was still exhibited lower viral loads in oropharyngeal swabs than detected in the lungs of all vaccinated groups. Remarkably, limited unvaccinated mice (S-21 p<0.05; S-29 p=0.066; Peto & Peto Left- tissue damage was found in the brain of vaccinated mice, except in censored k-sample test) (Figure 3A). Generally, gRNA decreased those animals euthanized due to humane endpoints (Figure 3C). Frontiers in Immunology 08 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 A B C FIGURE 3 Viral load and histopathology analysis of biological samples from SARS-CoV-2 B.1.351 infected K18-hACE2 transgenic mice after vaccination. SARS- CoV-2 viral loads were analyzed in oropharyngeal swabs, and samples from nasal turbinate, lung, and brain of infected K18-hACE2 mice. Virus distribution and tissue damage were analyzed by immunohistochemistry and histopathology, respectively. (A) Levels of SARS-CoV-2 gRNA (expressed as logarithmic of copies/mL) in oropharyngeal swabs, nasal turbinate, lung, and brain during infection. Dotted line indicates limit of detection (100 copies/mL). Differences among groups were analyzed using Peto & Peto left-censored k-sample test with FDR correction. (B) Detection of SARS-CoV-2 Nucleocapsid protein in brain and lung by immunohistochemistry. Staining score: (0) no antigen, (1) low, (2) moderate, and (3) high viral antigen. (C) Histopathological analysis of brain and lung by hematoxylin and eosin staining. Lesion score: (0) no lesion, (1) mild lesion, (2) moderate lesion, and (3) severe lesion. Differences among groups were analyzed using Asymptotic Generalized Pearson Chi-Squared test with FDR correction. * p<0.05, ** p<0.01. Overall, S-2P, S-21, and S-29 trimers displayed an equivalent In accordance with K18-hACE2 transgenic mice data, the three immunogenicity in K18-hACE2 transgenic mice and protected vaccinated groups (S-2P, S-21 and S-29) developed similar levels of these animals from developing severe disease after SARS-CoV-2 anti-RBD and anti-S binding IgG (Figure 4B and Supplementary Beta variant challenge. Interestingly, S-21- and S-29-immunized Figures 2A, B). Interestingly, the second immunization did not boost animals had lower viral loads in nasal turbinate than S-2P and vaccine-induced anti-S IgGs (Supplementary Figures 2A, B), but infected controls on days 3 and 6 after challenge. Viral loads were slightly increased anti-RBD IgG antibodies by 2 to 3-fold, (p>0.05) also reduced in oropharyngeal swabs of these mice on day 3 (Figure 4B). An increase in anti-S IgG levels was observed after viral compared to infected control groups. challenge in both vaccinated and unvaccinated but challenged mice (Supplementary Figure 2A). Nonetheless, when anti-RBD IgG responses were analyzed, that boosting effect was less evident and 3.3 S-21 and S-29 trimer vaccination only detected in the S-2P and in unvaccinated and challenged groups protects golden Syrian hamsters from (Figure 2B and Supplementary Figure 2B). These results suggest that COVID-19 development viral challenge elicited a rapid humoral response in naïve animals, boosting anti-S IgG responses, but had little effect in vaccinated GSH. To confirm the results obtained in K18-hACE2 mice, a second Despite that, immunized GSH showed higher levels of anti-S and immunization and challenge experiment was performed using GSH anti-RBD antibodies than challenged controls (p<0.001 for S-2P and with the same immunogens. Unlike K18-hACE2 mice, GSH S-29 group, and p<0.01 for S-21 group; Friedman test) (Figure 4B develop a moderate form of SARS-CoV-2-induced disease, from and Supplementary Figure 2A). which they spontaneously recover by day 14 after challenge (31, 36). Sera neutralizing activity against SARS-CoV-2 WH1, Beta and GSH were immunized following a similar prime/boost strategy to Delta, and to a lesser extent, Omicron variants was detected in the previously used for K18-hACE2 transgenic mice. Animals were vaccinated animals at all post-challenge timepoints (Figures 4C–F). intranasally challenged with the SARS-CoV-2 Beta variant and No differences were identified among immunized groups. followed up until day 7 post-challenge (Figure 4A), since it has Remarkably, and contrarily to K18-hACE2 transgenic mice data, been described that GSH start recovering weight from this day sera neutralizing activity against all four SARS-CoV-2 variants were (31, 36). observed in some challenged positive control animals by day 4 after Frontiers in Immunology 09 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 A B C D E F G H FIGURE 4 Vaccine-induced humoral responses and prophylactic activity of S-21 and S-29 in immunized GSH after challenge with the SARS-CoV-2 B.1.351 (Beta) variant. GSH were immunized twice with S-21, S-29 or S-2P, and subsequently challenged with SARS-CoV-2 B.1.351 variants. The humoral response, weight changes, and survival of mice were evaluated after immunization and/or viral challenge. (A) Outline of immunization schedule and infection timeline. Biological samples were collected at the indicated time points. (B) Kinetics of anti-RBD antibodies in serum samples. Red triangles: S-2P group (n= 11). Yellow squares: S-21 (n=11). Purple diamond: S-29 (n=11). White circles: unvaccinated-challenged mice (n=11). Grey circles: unvaccinated-uninfected mice (n=5). Groups in each time point were analyzed using Kruskal-Wallis and Conover’s post-hoc tests with multiple comparison correction by FDR. Differences among animals within a particular group along time were analyzed using the Friedman and Conover’s post-hoc tests for paired data with FDR correction. Sera neutralizing activity after viral challenge against: (C) SARS-CoV-2 WH-1, (D) B.1.351 (Beta), (E) B.1.617.2 (Delta), and (F) B.1.1.529 (Omicron) variants. Neutralization data were analyzed as indicated in (B). (G) Percentage of weight variation in SARS-CoV-2 B.1.351-infected GSH over time. (H) Kaplan-Meier plot showing the frequency of disease-free SARS-CoV-2-infected animals at the end of the experiment. Statistical analysis was performed against the unvaccinated group using Kruskal Wallis and Dunn´s post-hoc tests. * p<0.05, ** p<0.01, *** p<0.001. challenge (Figures 4C–F). These results indicate that cross-reactive to humane endpoints (Figure 4H). No significant weight loss was observed neutralizing antibodies were generated in those individuals. in vaccinated GSHs, indicating that these animals were protected from Unexpectedly, the neutralizing activity against the SARS-CoV-2 disease development (Figures 4G, H). Beta variant was higher in challenged control animals than in S-21- To evaluate viral replication in tissues, we determined the levels of and S-29-vaccinated GSH by day 7 (Figure 4D). According to the gRNA and sgRNA by RT-qPCR. We did not identify any differences binding ELISA data, neutralization titers also increased in among study groups in the levels of gRNA and sgRNA in nasal immunized GSH by day 7 after viral challenge (p<0.05; Conover´ turbinate, lung, in oropharyngeal samples were detected (Figure 5A s post-hoc test), indicating that infection boosts vaccine-induced and Supplementary Figure 2C). However, vaccinated animals humoral neutralizing responses (Figures 4C–F). exhibited a decreasing trend in their nasal turbinate levels of both To determine whether vaccination protected GSHs from SARS-CoV- gRNA and sgRNA over time after challenge (p=0.061; Peto & Peto 2-induced disease, we monitored animal weight over time after viral Left-censored k-sample test) (Figure 5A and Supplementary challenge (Figure 4G). Challenged control GSHs showed progressive Figure 2C). In addition, the analysis of nasal turbinate samples on weight reduction until day 6, which was indicative of disease day 7 post-challenge showed that vaccinated GSH displayed lower progression. One animal from this group suffered a weight reduction gRNA and sgRNA levels tendency compared with unvaccinated- greater than 20% by day 6 post-inoculation and was euthanized according challenged controls (gRNA: p=0.061; sgRNA: p=0.056; Peto & Peto Frontiers in Immunology 10 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 A B C FIGURE 5 Histopathology and viral loads in tissues from vaccinated GSH after challenge with SARS-CoV-2 Beta variant. SARS-CoV-2 viral loads were analyzed in oropharyngeal swabs, and samples from nasal turbinate and lung of vaccinated and challenged GSH. Virus distribution and tissue damage was analyzed by immunohistochemistry and histopathology, respectively. (A) Levels of SARS-CoV-2 gRNA, expressed as cycles threshold (CTs), in oropharyngeal swabs, nasal turbinate, and lung during infection. Dotted line indicates limit of positivity (40 CTs). Differences among groups were analyzed using Peto & Peto left-censored k-sample test with FDR correction. (B) Detection of SARS-CoV-2 Nucleocapsid protein in lung and nasal turbinate by immunohistochemistry. Staining score: (0) no antigen, (1) low, (2) moderate, and (3) high viral antigen. (C) Histopathological analysis of lung and nasal turbinate by hematoxylin and eosin staining. Lesion score: (0) no lesion, (1) mild lesion, (2) moderate lesion, and (3) severe lesion. Differences among groups were analyzed using Asymptotic Generalized Pearson Chi-Squared test and FDR. Left-censored k-sample test) (Figure 5A and Supplementary resistance to the immunity generated by the first generation of Figure 2C). To confirm RT-qPCR data, the presence of NP was COVID-19 vaccines, which were based on the ancestral WH1 analyzed in nasal turbinate and lung by IHC. NP was not detected in sequence (37–39). Although additional immunizations proved to nasal turbinate samples from immunized animals on day 7 increase the protection level against new emerging SARS-CoV-2 (Figure 5B). These results confirm the decreasing trend observed variants (40, 41), this protection remains transient (42). when gRNA and sgRNA were analyzed over time. Similarly, SARS- Particularly, the levels of NAbs elicited against the newest CoV-2 replication associated lesions were hardly detected in nasal variants (i.e. Omicron and subvariants) wane overtime (43, 44), turbinate samples on day 7 (Figure 5C). However, despite NP was not pointing out the importance of developing novel vaccines that detected in lung of vaccinated GSHs on day 7, low levels of tissue increase coverage and duration of immunity. Thus, the adaptation lesions were still present (Figure 5C). No significant differences in of vaccines to the new variants has shown encouraging results (45– tissue damage were observed in lung samples among study groups, 47), and several studies performed in animal models indicate that probably due to the low number of animals per group. intranasal immunization may also improve protection (48, 49). Overall, these results confirm that all three S-2P, S-21 and S-29 Besides these two complementary approaches, S immunogenicity immunogens showed an equivalent immunogenicity and can be enhanced by protein stabilization strategies. Studies prophylactic activity in GSHs, protecting these animals from the performed with the S glycoprotein of MERS and with functional development of severe SARS-CoV-2-induced disease. analogous of other viruses have shown that the introduction of mutations that stabilize these proteins in a prefusion state may increase its production and immunogenicity (13, 50). Accordingly, 4 Discussion the introduction of two prolines (K986P and V987P) into the S2 subunit of SARS-CoV-2 S was promptly confirmed to enhance its The implementation of SARS-CoV-2 vaccines became an stability and immunogenicity (50), and was successfully inflexion point on the course of the COVID-19 pandemic. implemented in several widely used commercial vaccines (e.g. However, new SARS-CoV-2 variants have shown partial BNT162b2, mRNA-1273 and Ad26.COV2.S). However, it is still Frontiers in Immunology 11 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 possible to improve the current S-2P strategy since the target vaccine delivery, these advances could significantly contribute to the recombinant protein is produced with low yield and shows some generation of novel COVID-19 mucosal vaccines that prevent viral degree of instability (18). Initial attempts to stabilize the S in its infection, irrespectively of the circulating variants. Our results, closed conformation yielded low production, suggesting that the including our new computational pipeline, may also contribute to open conformation of the RBD or its intrinsic motility might play a the development of novel vaccines for other pathogenic viruses. role in protein expression (20). Recently, Juraszek and colleagues showed that the incorporation of the D614N, A892P, A942P, and V987P mutations were able to stabilize the S glycoprotein in its Data availability statement closed conformation and increase its yield by six-fold compared to the original S-2P protein (19). Interestingly, Hsieh et al. improved The original contributions presented in the study are included the S stability and production by ten-fold, after introducing four in the article/Supplementary Material. Further inquiries can be additional proline mutations into the S-2P backbone (18). directed to the corresponding authors. Here, we designed and produced a set of S-2P mutated variants whose yield increased between two and five-fold using our new computational pipeline. Unlike proline substitutions, we selected Ethics statement mutations that generated hydrogen bonds, ionic interactions filling hydrophobic pockets, or other well-defined interaction between The animal study was approved by Committee on the Ethics of different chains of the S trimer. We selected two representative S Animal Experimentation of the IGTP and counted with the mutants based on their production levels and RBD exposure. Thus, authorization of Generalitat de Catalunya (Code: 10965 and S-21 showed the highest RBD exposure and a moderate increased 11094). The study was conducted in accordance with the local production, while S29 showed the highest production but a legislation and institutional requirements. moderate increased RBD exposure compared to S-2P. S-21, S-29 and S-2P were then compared in terms of immunogenicity and capacity to protect against the SARS-CoV-2 Beta variant, one of the Author contributions most virulent SARS-CoV-2 variants tested in the K18-hACE2 mouse model (34). To substantiate our results, we used two CÁ -N: Data curation, Formal analysis, Investigation, different animal models: K18-hACE2 transgenic mice and GSHs. Writing – original draft. JV-A: Conceptualization, Data curation, The K18-hACE2 is a transgenic mouse model that develops severe Funding acquisition, Investigation, Methodology, Supervision, disease after SARS-CoV-2 challenge (51, 52). Most mice succumb Writing – review & editing. PA-R: Investigation, Software, Writing – after viral challenge, mainly due to the infection of the central review & editing. EA-E: Investigation, Methodology, Writing – review nervous system (32). On the other hand, GSHs develop a moderate & editing. MB: Investigation, Writing – review & editing. MR: disease, and animals spontaneously recover two weeks after Investigation, Writing – review & editing. NP-L: Investigation, challenge (31, 36). Our results showed that despite S-2P, S-21, Writing – review & editing. JR: Investigation, Writing – review & and S-29 showed equivalent immunogenicity and protected both editing. VU: Formal analysis, Writing – review & editing. EP: animal models against disease progression, they differed in the Investigation, Writing – review & editing. SM: Investigation, Writing degree of protection. The S-29 protein induced an immune – review & editing. EB: Data curation, Investigation, Writing – review response that protected 100% of K18-hACE2 transgenic mice & editing. ER-M: Investigation, Writing – review & editing. MP: after challenge with the SARS-CoV-2 Beta variant. On the Investigation, Writing – review & editing. NR: Investigation, Writing contrary, one mouse in both S-2P and S-21 groups developed – review & editing. FT-F: Investigation, Writing – review & editing. JC: severe disease and had to be euthanized on days 8 and 14 after Investigation, Writing – review & editing. GC: Investigation, Writing – challenge, respectively. Therefore, our results suggested that the S review & editing. AP-G: Investigation, Writing – review & editing. CR: stabilization may impact on the capacity of this protein to induce a Investigation, Writing – review & editing. CA-G: Investigation, protective immune response, particularly against heterologous Writing – review & editing. RO: Investigation, Writing – review & SARS-CoV-2 variants. According to the in vivo protection data, editing. AB: Investigation, Writing – review & editing. BT: S-29- and S-21-immunized animals showed a faster viral clearance Investigation, Writing – review & editing. RL: Investigation, in nasal turbinate than S-2P immunized mice. Software, Writing – review & editing. JM-B: Investigation, Writing In summary, we described a novel set of mutations that – review & editing. DP-Z: Investigation, Writing – review & stabilized the S glycoprotein, increasing its production in vitro, editing. NI-U: Conceptualization, Writing – review & editing. and improving its protective capacity against SARS-CoV-2-induced AV: Conceptualization, Writing – review & editing. JB: disease in vivo. Despite all these immunogens were based on the Conceptualization, Data curation, Funding acquisition, original WH1 S sequence, S-29 protein showed 100% protection Writing – review & editing. BC: Conceptualization, Funding against the SARS-CoV-2 Beta variant. The inclusion of these acquisition, Writing – review & editing. VG: Conceptualization, mutations on the next generation of variant-adapted S-based Software, Writing – review & editing. JS: Conceptualization, Data COVID-19 vaccines could enhance the degree of protection to curation, Investigation, Writing – review & editing. JC: new emerging variants. In addition to an improvement in mucosal Conceptualization, Data curation, Funding acquisition, Frontiers in Immunology 12 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 Investigation, Methodology, Writing – original draft, Writing – Abad, Ivan Cordon, Anna Pou, Oscar Garcı́a, Joanna Wiacek, review & editing. Maria Angeles Osuna, Luı́s Ribas and Claudia Pereira Sunyé ) for their technical assistance with in vivo animal studies. Funding Conflict of interest The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work Author PA-R is currently employed by the company Sanofi, was supported by Grifols pharmaceutical, the CERCA Program (2021 which has no association with any content related to this work. SGR 00452; Generalitat de Catalunya), Direcció General de Recerca i Unrelated to the submitted work, JB and JC are founders and Innovació en Salut (Generalitat de Catalunya) (projects SLD0015 and shareholders of AlbaJuna Therapeutics, S.L. BC is founder and SLD0016), the Carlos III Health Institute (PI17/01518 and PI18/ shareholder of AlbaJuna Therapeutics, S.L. and AELIX 01332), and the crowdfunding projects “YomeCorono”, BonPreu/ Therapeutics, S.L, and VG is founder and shareholder of Esclat, and Correos. JB is supported by the Health Department of the Nostrum Biodiscovery. Unrelated to the submitted work, NI-U is Catalan Government (Generalitat de Catalunya). CÁ -N, AP-G, and supported by institutional funding from Pharma Mar, HIPRA, PA-R were supported by predoctoral grants from Generalitat de Amassence and Palobiofarma. Catalunya and Fons Social Europeu (2020 FI_B_0742; 2022 The remaining authors declare that the research was conducted FI_B_00698 and 2020FI_B2_00138, respectively). EP was supported in the absence of any commercial or financial relationships that by a doctoral grant from National Agency for Research and could be construed as a potential conflict of interest. Development of Chile (ANID: 72180406). NI-U is supported by the The author(s) declared that they were an editorial board Spanish Ministry of Science and Innovation (grant PID2020- member of Frontiers, at the time of submission. This had no 117145RB-I00), EU HORIZON-HLTH-2021-CORONA-01 (grant impact on the peer review process and the final decision. 101046118). This study was also supported by CIBER - Consorcio Centro de Investigació n Biomédica en Red (CB 2021), Carlos III Health Institute, Ministerio de Ciencia e Innovació n and Unió n Publisher’s note Europea – NextGenerationEU. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of All claims expressed in this article are solely those of the authors this article, or the decision to submit it for publication. and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its Acknowledgments manufacturer, is not guaranteed or endorsed by the publisher. We would like to thank Foundation Dormeur that support the acquisition of the QuantStudio-5 real time PCR system, an Eclipse Supplementary material Ts2R-FL Inverted Research Microscope, and an ÄKTA go protein purification system. We thank to the CMCiB´s staff (Sara The Supplementary Material for this article can be found online Capdevila, Jordi Grifols, Rosa Maria Ampudia, Jorge Diaz, Yaiza at: https://www.frontiersin.org/articles/10.3389/fimmu.2023. Rosales and Sergi Sunyé ) and the BSL3 IRTA-CReSA staff (Xavier 1291972/full#supplementary-material References 1. Greenwood B. The contribution of vaccination to global health: past, present and 8. Walls AC, Park Y-J, Tortorici MA, Wall A, McGuire AT, Veesler D. Structure, future. Philos Trans R Soc Lond B Biol Sci (2014) 369:20130433. doi: 10.1098/ function, and antigenicity of the SARS-CoV-2 spike glycoprotein. Cell (2020) 181:281– rstb.2013.0433 92.e6. doi: 10.1016/j.cell.2020.02.058 2. Strassburg MA. The global eradication of smallpox. Am J Infect Control (1982) 9. Hoffmann M, Kleine-Weber H, Schroeder S, Krüger N, Herrler T, Erichsen S, 10:53–9. doi: 10.1016/0196-6553(82)90003-7 et al. SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a 3. Krammer F. SARS-CoV-2 vaccines in development. Nature (2020) 586:516–27. clinically proven protease inhibitor. Cell (2020) 181:271–80.e8. doi: 10.1016/ doi: 10.1038/s41586-020-2798-3 j.cell.2020.02.052 4. Poland GA, Ovsyannikova IG, Kennedy RB. SARS-CoV-2 immunity: review and 10. Premkumar L, Segovia-Chumbez B, Jadi R, Martinez DR, Raut R, Markmann AJ, applications to phase 3 vaccine candidates. Lancet (2020) 6736: 1595-1606. et al. The receptor-binding domain of the viral spike protein is an immunodominant doi: 10.1016/s0140-6736(20)32137-1 and highly specific target of antibodies in SARS-CoV-2 patients. Sci Immunol (2020) 5:1–10. doi: 10.1126/SCIIMMUNOL.ABC8413 5. Watson OJ, Barnsley G, Toor J, Hogan AB, Winskill P, Ghani AC. Global impact of the first year of COVID-19 vaccination: a mathematical modelling study. Lancet 11. Wrapp D, Wang N, Corbett KS, Goldsmith JA, Hsieh C-L, Abiona O, et al. Cryo- Infect Dis (2022) 22:1293–302. doi: 10.1016/S1473-3099(22)00320-6 EM structure of the 2019-nCoV spike in the prefusion conformation. Science (2020) 367:1260–3. doi: 10.1126/science.abb2507 6. Chen X, Huang H, Ju J, Sun R, Zhang J. Impact of vaccination on the COVID-19 pandemic in U.S. states. Sci Rep (2022) 12:1554. doi: 10.1038/ 12. Cai Y, Zhang J, Xiao T, Peng H, Sterling SM, Walsh RM, et al. Distinct S41598-022-05498-Z conformational states of SARS-CoV-2 spike protein. Science (2020) 369:1586–92. doi: 10.1126/science.abd4251 7. Alexandridi M, Mazej J, Palermo E, Hiscott J. The Coronavirus pandemic – 2022: Viruses, variants & vaccines. Cytokine Growth Factor Rev (2022) 63:1–9. doi: 10.1016/ 13. Graham BS, Gilman MSA, McLellan JS. Structure-based vaccine antigen design. j.cytogfr.2022.02.002 Annu Rev Med (2019) 70:91–104. doi: 10.1146/annurev-med-121217-094234 Frontiers in Immunology 13 frontiersin.org Ávila-Nieto et al. 10.3389/fimmu.2023.1291972 14. Pallesen J, Wang N, Corbett KS, Wrapp D, Kirchdoerfer RN, Turner HL, et al. 38. Hoffmann M, Krüger N, Schulz S, Cossmann A, Rocha C, Kempf A, et al. The Immunogenicity and structures of a rationally designed prefusion MERS-CoV spike Omicron variant is highly resistant against antibody-mediated neutralization: antigen. Proc Natl Acad Sci U S A (2017) 114:E7348–57. doi: 10.1073/pnas.1707304114 Implications for control of the COVID-19 pandemic. Cell (2022) 185:447–56.e11. 15. Walsh EE, Frenck RW, Falsey AR, Kitchin N, Absalon J, Gurtman A, et al. Safety doi: 10.1016/j.cell.2021.12.032 and immunogenicity of two RNA-based Covid-19 vaccine candidates. New Engl J Med 39. Pouwels KB, Pritchard E, Matthews PC, Stoesser N, Eyre DW, Vihta KD, et al. Effect (2020) 383:2439–50. doi: 10.1056/nejmoa2027906 of Delta variant on viral burden and vaccine effectiveness against new SARS-CoV-2 16. Polack FP, Thomas SJ, Kitchin N, Absalon J, Gurtman A, Lockhart S, et al. Safety infections in the UK. Nat Med (2021) 27:2127–35. doi: 10.1038/s41591-021-01548-7 and efficacy of the BNT162b2 mRNA Covid-19 vaccine. New Engl J Med (2020) 40. Choi A, Koch M, Wu K, Chu L, Ma LZ, Hill A, et al. Safety and immunogenicity 383:2603–15. doi: 10.1056/nejmoa2034577 of SARS-CoV-2 variant mRNA vaccine boosters in healthy adults: an interim analysis. 17. Sadoff J, le Gars M, Shukarev G, Heerwegh D, Truyers C, de Groot AM, et al. Nat Med (2021) 27:2025–31. doi: 10.1038/s41591-021-01527-y Interim results of a phase 1–2a trial of Ad26.COV2.S Covid-19 vaccine. New Engl J Med 41. Atmar RL, Lyke KE, Deming ME, Jackson LA, Branche AR, el Sahly HM, et al. (2021) 384(19):1824–35. doi: 10.1056/nejmoa2034201 Homologous and heterologous Covid-19 booster vaccinations. New Engl J Med (2022) 18. Hsieh C-L, Goldsmith JA, Schaub JM, DiVenere AM, Kuo H-C, Javanmardi K, 386:1046–57. doi: 10.1056/nejmoa2116414 et al. Structure-based design of prefusion-stabilized SARS-CoV-2 spikes. Sci (1979) 42. Patalon T, Saciuk Y, Peretz A, Perez G, Lurie Y, Maor Y, et al. Waning (2020) 369:1501–5. doi: 10.1126/science.abd0826 effectiveness of the third dose of the BNT162b2 mRNA COVID-19 vaccine. Nat 19. Juraszek J, Rutten L, Blokland S, Bouchier P, Voorzaat R, Ritschel T, et al. Commun (2022) 13:3203. doi: 10.1038/s41467-022-30884-6 Stabilizing the closed SARS-CoV-2 spike trimer. Nat Commun (2021) 12:1–8. 43. Pajon R, Doria-Rose NA, Shen X, Schmidt SD, O’Dell S, McDanal C, et al. SARS- doi: 10.1038/s41467-020-20321-x CoV-2 Omicron Variant Neutralization after mRNA-1273 Booster Vaccination. N Engl 20. Xiong X, Qu K, Ciazynska KA, Hosmillo M, Carter AP, Ebrahimi S, et al. A J Med (2022) 386:1088–91. doi: 10.1056/NEJMc2119912 thermostable, closed SARS-CoV-2 spike protein trimer. Nat Struct Mol Biol (2020) 44. Lassaunière R, Polacek C, Frische A, Boding L, Sækmose SG, Rasmussen M, et al. 27:934–41. doi: 10.1038/s41594-020-0478-5 Neutralizing antibodies against the SARS-CoV-2 omicron variant (BA.1) 1 to 18 weeks 21. Riley TP, Chou HT, Hu R, Bzymek KP, Correia AR, Partin AC, et al. Enhancing after the second and third doses of the BNT162b2 mRNA vaccine. JAMA Netw Open the prefusion conformational stability of SARS-CoV-2 spike protein through structure- (2022) 5:e2212073. doi: 10.1001/jamanetworkopen.2022.12073 guided design. Front Immunol (2021) 12:660198. doi: 10.3389/fimmu.2021.660198 45. Chalkias S, Harper C, Vrbicky K, Walsh SR, Essink B, Brosz A, et al. A bivalent 22. Tan TJC, Mou Z, Lei R, Ouyang WO, Yuan M, Song G, et al. High-throughput omicron-containing booster vaccine against Covid-19. N Engl J Med (2022) 387:1279– identification of prefusion-stabilizing mutations in SARS-CoV-2 spike. Nat Commun 91. doi: 10.1056/NEJMoa2208343 (2023) 14:2003. doi: 10.1038/s41467-023-37786-1 46. Xu K, Gao P, Liu S, Lu S, Lei W, Zheng T, et al. Protective prototype-Beta and 23. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. Delta-Omicron chimeric RBD-dimer vaccines against SARS-CoV-2. Cell (2022) SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic 185:2265–78.e14. doi: 10.1016/j.cell.2022.04.029 Acids Res (2018) 46:W296–303. doi: 10.1093/nar/gky427 47. Chalkias S, Eder F, Essink B, Khetan S, Nestorova B, Feng J, et al. Safety, 24. Schymkowitz J, Borg J, Stricher F, Nys R, Rousseau F, Serrano L. The FoldX web immunogenicity and antibody persistence of a bivalent Beta-containing booster vaccine server: An online force field. Nucleic Acids Res (2005) 33: 382–8. doi: 10.1093/nar/ against COVID-19: a phase 2/3 trial. Nat Med (2022) 28:2388–97. doi: 10.1038/s41591- gki387 022-02031-7 25. Perez-Zsolt D, Muñoz-Basagoiti J, Rodon J, Elosua-Bayes M, Raïch-Regué D, 48. Bricker TL, Darling TL, Hassan AO, Harastani HH, Soung A, Jiang X, et al. A Risco C, et al. SARS-CoV-2 interaction with Siglec-1 mediates trans-infection by single intranasal or intramuscular immunization with chimpanzee adenovirus-vectored dendritic cells. Cell Mol Immunol (2021) 18:2676–8. doi: 10.1038/s41423-021- SARS-CoV-2 vaccine protects against pneumonia in hamsters. Cell Rep (2021) 00794-6 36:109400. doi: 10.1016/j.celrep.2021.109400 26. Rodon J, Muñoz-Basagoiti J, Perez-Zsolt D, Noguera-Julian M, Paredes R, Mateu 49. Hassan AO, Kafai NM, Dmitriev IP, Fox JM, Smith BK, Harvey IB, et al. A L, et al. Identification of plitidepsin as potent inhibitor of SARS-CoV-2-induced single-dose intranasal ChAd vaccine protects upper and lower respiratory tracts against cytopathic effect after a drug repurposing screen. Front Pharmacol (2021) 12:646676. SARS-CoV-2. Cell (2020) 183:169–84.e13. doi: 10.1016/j.cell.2020.08.026 doi: 10.3389/fphar.2021.646676 50. Corbett KS, Edwards DK, Leist SR, Abiona OM, Boyoglu-Barnum S, Gillespie 27. Kamala T. Hock immunization: A humane alternative to mouse footpad RA, et al. SARS-CoV-2 mRNA vaccine design enabled by prototype pathogen injections. J Immunol Methods (2007) 328:204–14. doi: 10.1016/j.jim.2007.08.004 preparedness. Nature (2020) 586:567–71. doi: 10.1038/s41586-020-2622-0 28. Pradenas E, Trinité B, Urrea V, Marfil S, Á vila-Nieto C, Rodrı́guez de la 51. McCray PB, Pewe L, Wohlford-Lenane C, Hickey M, Manzel L, Shi L, et al. Concepció n ML, et al. Stable neutralizing antibody levels 6 months after mild and Lethal infection of K18-hACE2 mice infected with severe acute respiratory syndrome severe COVID-19 episodes. Med (2021) 2:313–20.e4. doi: 10.1016/j.medj.2021.01.005 coronavirus. J Virol (2007) 81:813–21. doi: 10.1128/jvi.02012-06 29. Connor RI, Chen BK, Choe S, Landau NR. Vpr is required for efficient 52. Dong W, Mead H, Tian L, Park J-G, Garcia JI, Jaramillo S, et al. The K18-human replication of human immunodeficiency virus type-1 in mononuclear phagocytes. ACE2 transgenic mouse model recapitulates non-severe and severe COVID-19 in Virology (1995) 206:935–44. doi: 10.1006/viro.1995.1016 response to an infectious dose of the SARS-CoV-2 virus. J Virol (2022) 96:e0096421. doi: 10.1128/JVI.00964-21 30. Wölfel R, Corman VM, Guggemos W, Seilmaier M, Zange S, Müller MA, et al. Virological assessment of hospitalized patients with COVID-2019. Nature (2020) CITATION 581:465–9. doi: 10.1038/s41586-020-2196-x Ávila-Nieto C, Vergara-Alert J, Amengual-Rigo P, Ainsua-Enrich E, Brustolin M, Rodríguez de la Concepción ML, Pedreño-Lopez N, Rodon J, 31. Brustolin M, Rodon J, Rodrı́guez de la Concepció n ML, Á vila-Nieto C, Cantero Urrea V, Pradenas E, Marfil S, Ballana E, Riveira-Muñoz E, Pérez M, Roca N, G, Pé rez M, et al. Protection against reinfection with D614- or G614-SARS-CoV-2 Tarrés-Freixas F, Carabelli J, Cantero G, Pons-Grífols A, Rovirosa C, isolates in golden Syrian hamster. Emerg Microbes Infect (2021) 12:1–36. doi: 10.1080/ 22221751.2021.1913974 Aguilar-Gurrieri C, Ortiz R, Barajas A, Trinité B, Lepore R, Muñoz-Basagoiti J, Perez-Zsolt D, Izquierdo-Useros N, Valencia A, 32. Vidal E, Ló pez-Figueroa C, Rodon J, Pé rez M, Brustolin M, Cantero G, et al. Blanco J, Clotet B, Guallar V, Segalés J and Carrillo J (2023) Novel Chronological brain lesions after SARS-CoV-2 infection in hACE2-transgenic mice. Vet Pathol (2022) 59:613–26. doi: 10.1177/03009858211066841 Spike-stabilized trimers with improved production protect K18-hACE2 mice and golden Syrian hamsters from the highly pathogenic 33. Mrabet NT, van den Broeck A, van den Brande I, Stanssens P, Laroche Y, SARS-CoV-2 Beta variant. Lambeir AM, et al. Arginine residues as stabilizing elements in proteins. Biochemistry (1992) 31:2239–53. doi: 10.1021/bi00123a005 Front. Immunol. 14:1291972. doi: 10.3389/fimmu.2023.1291972 34. Tarré s-Freixas F, Trinité B, Pons-Grı́fols A, Romero-Durana M, Riveira-Muñoz E, Á vila-Nieto C, et al. Heterogeneous infectivity and pathogenesis of SARS-CoV-2 COPYRIGHT variants beta, delta and omicron in transgenic K18-hACE2 and wildtype mice. Front © 2023 Ávila-Nieto, Vergara-Alert, Amengual-Rigo, Ainsua-Enrich, Brustolin, Microbiol (2022) 13:840757. doi: 10.3389/fmicb.2022.840757 Rodríguez de la Concepción, Pedreño-Lopez, Rodon, Urrea, Pradenas, Marfil, 35. Pradenas E, Trinité B, Urrea V, Marfil S, Tarré s-Freixas F, Ortiz R, et al. Clinical Ballana, Riveira-Muñoz, Pérez, Roca, Tarrés-Freixas, Carabelli, Cantero, Pons- course impacts early kinetics,magnitude, and amplitude of SARS-CoV-2 neutralizing Grífols, Rovirosa, Aguilar-Gurrieri, Ortiz, Barajas, Trinité, Lepore, Muñoz- antibodies beyond 1 year after infection. Cell Rep Med (2022) 3:100523. doi: 10.1016/ Basagoiti, Perez-Zsolt, Izquierdo-Useros, Valencia, Blanco, Clotet, Guallar, j.xcrm.2022.100523 Segalés and Carrillo. This is an open-access article distributed under the terms 36. Sia SF, Yan LM, Chin AWH, Fung K, Choy KT, Wong AYL, et al. Pathogenesis of the Creative Commons Attribution License (CC BY). The use, distribution and transmission of SARS-CoV-2 in golden hamsters. Nature (2020) 583:834–8. or reproduction in other forums is permitted, provided the original doi: 10.1038/s41586-020-2342-5 author(s) and the copyright owner(s) are credited and that the original 37. Cele S, Jackson L, Khoury DS, Khan K, Moyo-Gwete T, Tegally H, et al. Omicron publication in this journal is cited, in accordance with accepted academic extensively but incompletely escapes Pfizer BNT162b2 neutralization. Nature (2022) practice. No use, distribution or reproduction is permitted which does not 602:654–6. doi: 10.1038/s41586-021-04387-1 comply with these terms. Frontiers in Immunology 14 frontiersin.org