key: cord-0971142-4eayvrvt
authors: Fernandes, Bárbara; Castro, Rute; Bhoelan, Farien; Bemelman, Denzel; Gomes, Ricardo A.; Costa, Júlia; Gomes-Alves, Patrícia; Stegmann, Toon; Amacker, Mario; Alves, Paula M.; Fleury, Sylvain; Roldão, António
title: Insect Cells for High-Yield Production of SARS-CoV-2 Spike Protein: Building a Virosome-Based COVID-19 Vaccine Candidate
date: 2022-04-13
journal: Pharmaceutics
DOI: 10.3390/pharmaceutics14040854
sha: 65e095cab452c4b67d54ee907a063b723b60324d
doc_id: 971142
cord_uid: 4eayvrvt

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) homotrimeric spike (S) protein is responsible for mediating host cell entry by binding to the angiotensin-converting enzyme 2 (ACE2) receptor, thus being a key viral antigen to target in a coronavirus disease 19 (COVID-19) vaccine. Despite the availability of COVID-19 vaccines, low vaccine coverage as well as unvaccinated and immune compromised subjects are contributing to the emergence of SARS-CoV-2 variants of concern. Therefore, continued development of novel and/or updated vaccines is essential for protecting against such new variants. In this study, we developed a scalable bioprocess using the insect cells-baculovirus expression vector system (IC-BEVS) to produce high-quality S protein, stabilized in its pre-fusion conformation, for inclusion in a virosome-based COVID-19 vaccine candidate. By exploring different bioprocess engineering strategies (i.e., signal peptides, baculovirus transfer vectors, cell lines, infection strategies and formulation buffers), we were able to obtain ~4 mg/L of purified S protein, which, to the best of our knowledge, is the highest value achieved to date using insect cells. In addition, the insect cell-derived S protein exhibited glycan processing similar to mammalian cells and mid-term stability upon storage (up to 90 days at −80 and 4 °C or after 5 freeze-thaw cycles). Noteworthy, antigenicity of S protein, either as single antigen or displayed on the surface of virosomes, was confirmed by ELISA, with binding of ACE2 receptor, pan-SARS antibody CR3022 and neutralizing antibodies to the various epitope clusters on the S protein. Binding capacity was also maintained on virosomes-S stored at 4 °C for 1 month. This work demonstrates the potential of using IC-BEVS to produce the highly glycosylated and complex S protein, without compromising its integrity and antigenicity, to be included in a virosome-based COVID-19 vaccine candidate.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a highly pathogenic virus responsible for the coronavirus disease 2019 . In March 2020, COVID-19 was declared a pandemic by the World Health Organization (WHO) and, as of March 2022, Insect Sf -9 (Invitrogen, Waltham, MA, USA), ExpiSf9™ (Thermo Fisher Scientific, Waltham, MA, USA) and SuperSf9-2 (Oxford Expression Technologies, Oxford, UK) cells were routinely sub-cultured at 0.4-1 × 10 6 cell/mL every 3-4 days when cell density reached 2-5 × 10 6 cell/mL in 500 mL shake flasks (10% working volume, w/v) in a Innova 44R incubator (orbital motion diameter of 2.54 cm-Eppendorf) at 27 • C and 100 rpm. Sf-900 TM II SFM (Thermo Fisher Scientific, Waltham, MA, USA) and ExpiSf TM CD (Thermo Fisher Scientific, Waltham, MA, USA) media were used to culture Sf -9 and SuperSf9-2 cells, and ExpiSf9™ cells, respectively. SARS-CoV-2 spike (S) protein sequence (GenBank MN 908947) is described in Supplementary Information: The S protein sequence was modified by eliminating the furin cleavage site between S1 and S2, introducing the mutations K968P and V987P, truncation of the protein after Q1208 and introduction of the HIV gp160 derived fusion clamp (Watterson et al., 2021) after Q1028, followed by the 6 histidine tag (6his-tag), for an S protein with an expected MW of 140 kD. Three signal peptides were tested: (1) honeybee melittin (BVM) insect-derived; (2) gp67 of Autographa californica nuclear polyhedrosis virus (AcMNPV); and (3) native S signal peptide. Three baculovirus transfer vectors were tested: (1) pOET1, the S protein gene is under the control of AcMNPV polyhedrin (polh) promoter; (2) pOET3, the S protein gene is under the control of AcMNPV basic (p6.9) promoter; and (3) pOET5, one copy of the S protein gene is under the control of AcMNPV polh promoter and another copy is under the control of p10 promoter. The five different expression plasmids evaluated in this study, synthetized by GenScript, are described in Table 1 . 

Recombinant baculoviruses (rBac) were generated using the flashback ULTRA TM system (Oxford Expression Technologies, Oxford, UK) according to the manufacturer's instruction. Amplification of baculovirus stocks was performed as described elsewhere [23] . Briefly, Sf -9 cells were infected at cell concentration of 1 × 10 6 cells/mL and at a multiplicity of infection (MOI) of 0.01-0.1 pfu/cell. When cell viability reached 80-85%, cultures were harvested, centrifuged at 200× g for 10 min at 4 • C. The pellet was discarded, and the supernatant was centrifuged at 2000× g for 20 min at 4 • C. The resulting supernatant was stored at 4 • C until further use.

S protein production using Sf -9, ExpiSf9™ and SuperSf9-2 was performed in shake flasks (SF; 500 mL, 10% w/v) and in 20 L stirred tank bioreactors (STB). Cells were seeded at 0.3-0.6 × 10 6 cells/mL and infected with rBac at different cell concentrations at the time of infection (CCI, 1 × 10 6 and 2 × 10 6 ), and MOI (0.1 and 1 pfu/cell).

Bioreactor cultures were performed in a computer-controlled BIOSTAT ® Cplus 20 L vessel (Sartorius, Göttingen, Germany) equipped with two Rushton impeller and a ringsparger for gas supply. The pH was monitored (not controlled) along culture time. The partial pressure of oxygen (pO 2 ) was set to 30% of air saturation and was maintained by varying the agitation rate (70 to 250 rpm), the percentage of O 2 in the gas mixture (0 to 100%) and the gas flow rate was set to 0.01 vessel volumes per minute (vvm). The temperature was kept at 27 • C and the working volume was 20 L. Cell concentration and viability were measured daily, and culture samples processed and stored as described elsewhere [24] . The bioreactor culture was harvested 2-3 days post-infection when cell viabilities reached 70%.

Purification of secreted S protein was carried out on ÄKTA systems (Cytiva, Marlborough, MA, USA) as described elsewhere (Castro et al., 2021) . Cell culture bulk was harvested by filtering through 0.45 and 0.22 µm Sartopore 2 (Sartorius, Göttingen, Germany). Tangential flow filtration (TFF) was used to concentrate and dialyze the clarified supernatants to 50 mM sodium phosphate supplemented with 500 mM NaCl and 20 mM imidazole, at pH 7.4 (binding buffer). S protein was purified by immobilized metal ion affinity chromatography on a Histrap HP or Nickel Sepharose HP column (5 mL volume; Cytiva, Marlborough, MA, USA) equilibrated with binding buffer. Two washing steps with 35 and 50 mM imidazole were performed, and S protein was eluted with a linear gradient up to 500 mM imidazole. The eluate was concentrated using a Vivaflow ® 200 of 50 kDa crossflow device (Sartorius, Göttingen, Germany), incubated 30 min at 4 • C with 5 mM EDTA and purified by size exclusion chromatography (SEC) using Superdex 200 (GE Healthcare, Chicago, IL, USA) previously equilibrated with 10 mM HEPES, 150 mM NaCl, at pH 7.2. The eluate was concentrated using Vivaflow ® 200 of 50 kDa crossflow device and filtered through a polyethersulfone membrane with 0.2 µm. Purified S protein was formulated in 10 mM HEPES, 150 mM NaCl at pH 7.2 with 10% glycerol, and stored at −80 • C until further analysis.

Virosomes were prepared as described earlier [25] . Briefly, inactivated influenza virus A/Brisbane/59/2007 (Seqirus, Parkville, Australia) was solubilized with the detergent octaethyleneglycol-mono(n-dodecyl)ether (OEG) and the viral nucleocapsid was removed by centrifugation. To the supernatant, the lipid 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) (Merck) and the click chemistry lipid dicyclobenzooctyl-phosphatidylethanolamine (DBCO-PE) (Avanti Polar Lipids, Alabaster, AL, USA) were dissolved in OEG, and added OEG was then removed by batch chromatography on polystyrene beads (BioBeads SM2) as described; resulting homogenous virosomes were sterilized by 0.22 µm filtration.

2-azidoethyl thiophosphodichlorate (ATPD) was synthesized and purified as described (Jia, S. et al., 2020) by Acme Bioscience (Palo Alto, CA, USA). S protein was dialyzed against 50 mM HEPES pH 8.5 for 4 h and then mixed with ATPD at a 200:1 ratio of ATPD to protein for 1 h at RT. The product was dialyzed overnight against 2000 volumes of buffer (145 mM NaCl, 5 mM HEPES, 1 mM EDTA, pH 7.4). The resulting S-azide conjugate product was incubated with virosomes for at least 24 h at 25 • C resulting in covalent coupling of S to virosomes through azide-DBCO-PE click chemistry.

Cell concentration and viability were analyzed by trypan blue dye exclusion method [26] using a Cedex HiRes Analyzer (Roche, Basel, Switzerland).

Baculovirus titers were determined using the MTT assay [27, 28] .

The concentration of purified S protein was determined by spectrophotometry at 280 nm on the mySPEC (VWR).

Culture samples were centrifuged at 200× g for 10 min and supernatants collected and stored at −80 • C until further analysis. Western blot analysis was performed as reported elsewhere [29] . For S identification, the human monoclonal antibody SARS-CoV-2 spike protein (MA5-35948, Thermo Scientific, Breda, The Netherlands) was used at dilution of 1:3000. For 6his-tag recognition on S proteins identification, a mouse monoclonal antibody anti-6His tag (MA1-21315, Thermo Scientific, Breda, The Netherlands) was used at a dilution of 1:1000. As secondary antibody, an anti-mouse IgG (A3438, Sigma, Amsterdam, The Netherlands) and an anti-human IgG antibody (A9544, Sigma, Amsterdam, The Netherlands) conjugated with alkaline phosphatase were used at a dilution of 1:5000. 

Purified S protein was analyzed using a HPLC system equipped with Photodiode Array Detector (Waters, Milford, MA, USA). S protein samples were injected in a XBridge BEH 450 Å SEC 3.5 µm HPLC column (Waters) equilibrated in 10 mM HEPES with 150 mM NaCl at pH 7.2. The system flow rate was maintained at 0.86 mL/min and eluted proteins were detected at 280 nm. Twenty micrograms of protein was injected in each HPLC run.

Purified S protein was analyzed by LC-MS as described in [30] . For glycans identification, the N-glycans database described in CFG Functional Glycomics Gateway (http: //www.functionalglycomics.org/fg/, accessed on 2 October 2021) with Spodoptera taxonomic restriction was used. MS data were analyzed using the BioPharmaView software (BPV, Version 3.0, SCIEX) and the protein sequence of spike. Glycans were identified using MS1 data (considering a peptide deconvolution tolerance of 10 ppm, XIC m/z width of 0.025 Da and m/z tolerance of 5 ppm) and fragmentation MSMS data (considering a MSMS tolerance of 0.03 Da). All MSMS data were manually examined for the presence of MSMS specific glycan marker ions. The data was also manually examined for consistence in retention time information and spectrum quality.

For the epitope mapping ELISA on S protein, ELISA plates (Nunc high-binding, Thermo-Fisher) were coated overnight with purified S protein at 0.5 µg/mL in phosphate buffered saline (PBS). The plates were then washed with PBS containing 0.05% Tween-20 (PBST, Sigma, Amsterdam, The Netherlands) and blocked with 5% Protifar (Nutrica, Utrecht, The Netherlands) in PBST for 2 h at RT. After washing in PBST, human monoclonal antibodies and ACE-2-Fc chimeric protein were added for one hour at RT, and revealed with goat anti-human antibodies coupled to horse radish peroxidase (HRP).

Coupling of the S protein to the virosomes was assessed by an ELISA in which the mouse monoclonal antibody 395-F2-04/03 (CePower) toward the hemagglutinin on virosomes was used to coat ELISA plates for capturing virosomes. The plates were then washed with PBST and blocked with 5% Protifar (Nutrica, Utrecht, The Netherlands) in PBST, then incubated with virosomes for 1 h at RT and further processed as described above. During the ELISA, the virosomes remained intact.

Antibodies tested were: ACE2-NN-IgGFc (Absolute Antibody, Redcar, UK), CR3022 (Abcam, Cambridge, UK), all SARS-CoV-2 antibodies (courtesy of prof. Sanders, Academic Medical Center, Amsterdam, The Netherlands) [31] .

Data were expressed as mean ± standard deviation. Differences were tested by One-Way ANOVA with post hoc Tukey's multiple comparison analysis method and Dunnett's multiple comparison test (adjusted p-value < 0.05 was considered statistically significant) or by t-test unpaired assuming Gaussian distribution (adjusted p-value < 0.05 was considered statistically significant).

Aiming to improve S protein production yields in insect cells, different signal peptides, baculovirus transfer vectors, cell lines, infection strategies and formulation buffers were evaluated.

To identify the best infection strategy, Sf -9 cells were infected at cell concentrations at infection (CCI) of 1 and 2 × 10 6 cell/mL with recombinant baculovirus rBac 1 ( Table 1) using multiplicities of infection (MOI) of 0.1 and 1 pfu/cell, and their growth and S protein expression kinetics assessed in small-scale shake flasks (SF) (Figure 1 ). Traditional profiles of insect cell growth and viability upon infection were observed, with S protein being identified by Western blot only in experiments at CCI of 2 × 10 6 cell/mL (Supplementary Figure S1A ). Maximum S protein titers and specific production rates were achieved for CCI = 2 × 10 6 cell/mL and MOI = 1 pfu/cell ( Figure 1A ) and, therefore, this infection strategy was used in subsequent studies.

Three different signal peptides were explored: the insect BVM (rBac 1), the rBac gp67 (rBac 2), and the native SARS-CoV-2 S protein signal peptide (rBAC 3) (Table 1) . Insect Sf -9 cells were infected at CCI = 2 × 10 6 cell/mL with each rBac at MOI = 1 pfu/cell, and their growth and S protein expression kinetics assessed in small-scale SF (Figure 1 ). Traditional profiles of insect cell growth and viability upon infection were observed (Supplementary Figure S1B) , with S protein being only identified by Western blot in samples following infection with rBac 1 ( Figure 1B) ; thus, baculovirus constructs using the BVM signal sequence were used for subsequent experiments.

Three baculovirus transfer vectors, i.e., pOET1, pOET3 and pOET5 (Table 1) , and three insect cell lines, i.e., Sf -9, SuperSf9-2 and ExpiSf9™, were evaluated for S protein production in small-scale SF using CCI = 2 × 10 6 cell/mL and MOI = 1 pfu/cell (Figure 1 ). While baculovirus transfer vectors seem to have negligible impact on cell growth kinetics, the same does not withstand for the cell lines tested (Supplementary Figure S1C ). Despite these differences, the S protein could be identified by Western blot in all experiments performed. Noteworthy, maximum S protein titer and specific production rate was achieved using SuperSf9-2 cells and rBac 5, where the S protein gene is duplicated and under the AcMNPV very late promoter polyhedrin (polh) and the p10 ( Figure 1C ).

Three formulation buffers were evaluated for S protein storage (Table 1) : Buffer (A) 10 mM HEPES + 150 mM NaCl at pH 7.2; Buffer (B) 10 mM HEPES + 150 mM NaCl at pH 7.2, 10% glycerol; and Buffer (C) 10 mM HEPES + 150 mM NaCl at pH 7.2, 10% sucrose. S protein thermal denaturation was analyzed by DSF, with Buffers B and C showing consistently higher melting temperatures than Buffer A irrespective of storage time (up to 30 days) and temperature (−80 and 4 • C) ( Figure 2) . Noteworthy, S protein stored in Buffer A showed extensive degradation after two freeze-thaw cycles, being impossible to estimate its melting temperature, contrasting with S protein stored in Buffers B and C that only showed a 1 • C reduction in its melting temperature. For being slightly better, Buffer B was selected as formulation buffer for subsequent experiments. days) and temperature (−80 and 4 °C) ( Figure 2) . Noteworthy, S protein stored in Buffer A showed extensive degradation after two freeze-thaw cycles, being impossible to estimate its melting temperature, contrasting with S protein stored in Buffers B and C that only showed a 1 °C reduction in its melting temperature. For being slightly better, Buffer B was selected as formulation buffer for subsequent experiments. 

The feasibility of producing S protein in SuperSf9-2 cells was demonstrated in controlled, scalable 20 L stirred-tank bioreactors (STB); small-scale SF were used as control.

Cell growth and viability kinetics were comparable in both culture systems ( Figure 3A ). In addition, S protein could be identified by Western blot in both STB and SF cultures, with apparent similar band intensities ( Figure 3B ).

S protein produced in the 20 L STB was purified and a final production yield of 4.1 mg/L could be achieved, with removal of >95% of infectious particles, total deoxyribonucleic acid (DNA) and baculovirus genome copies. Purified S protein showed purity above 95% and a molecular weight of 383 kDa suggesting that S protein is in trimer conformation.

Purified S protein was characterized by LC-MS for determination of site-specific glycosylation and glycan composition for all N-linked glycan sites previously described in the literature [15, 17, 30] . The glycosylation sites of S protein and their main glycan processing, subdivided into high mannose and complex/paucimannose-type glycosylation, are presented in Figure 3C . 21 N-glycosylation sites were found occupied and the detailed glycan compositions are described in Supplementary Table S1. A mixture of high mannoseand complex/paucimannose-type glycans was found at glycosylation sites N 68_81, N172, N241, N1081; the remaining 15 sites were dominated by processed, complex-type glycans. 7 kDa) . (E) S protein thermal stability using differential scanning fluorimetry; data are expressed as mean ± standard deviation (relative to three replicates measurements, n = 3). (F) Binding of non-overlapping human neutralizing antibodies recognizing epitopes in the receptor binding domain of S protein (i.e., ACE2-NN-IgGFc, CR3022, all SARS-CoV-2 antibodies) or an ACE2-Fc chimeric protein binding to S protein bound to ELISA plates, and developed with goat anti-human HRP.

Mid-term storage stability of purified S protein was assessed by HPLC-SEC and DSF. HPLC-SEC analysis revealed a single peak in all conditions tested, thus, suggesting that S protein trimer conformation is maintained up to 90 days when stored at −80 • C and 4 • C or after 5 freeze-thaw cycles ( Figure 3D ). Stability of S protein was further confirmed by DSF data, in which a marginal variation (~1.5 • C) in S protein melting temperatures could be observed between all conditions explored ( Figure 3E ).

The quality of purified S protein was confirmed by epitope mapping. ELISA plates were coated with the protein and epitopes from non-overlapping antigenic clusters on the protein were detected with human monoclonal antibodies known to neutralize SARS-CoV2 with high affinity [31] . The pan-SARS antibody CR3022 [32] and an ACE2-Fc chimeric protein were used to test S protein binding to its receptor ( Figure 3F ). ELISA data indicate that insect-derived S protein is capable of binding to ACE2 receptor and, importantly, it is recognized by CR3022 monoclonal antibody and by all the other tested anti-S neutralizing antibodies directed toward various epitope clusters.

Purified S protein was covalently coupled to virosomes through DBCO-azide click chemistry, and the presence of S protein on the virosomes through the exposure of key epitopes on the protein and the binding of an ACE2-Fc were confirmed by ELISA (Figure 4) . Results indicate that the S protein as displayed on the surface of the virosomes is capable of binding to the ACE2 receptor and is also recognized by CR3022 and by all the tested neutralizing antibodies toward various epitope clusters. This binding capacity is also preserved on virosomes-S stored at 4 • C for 1 month, as similar IC50 values were obtained for t = 0 and t = 1 month, thus demonstrating an enhanced stability of virosomes-S. 

The production of the full-length recombinant S protein of SARS-CoV-2 S has been attempted in mammalian (e.g., HEK293, CHO) and insect (e.g., Sf-9 and High Five) cells; however, despite several bioprocess strategies being explored to date e.g., low culture temperature, new cell lines, most induce modest to no improvements in production yields [33, 34] .

In this study, we have explored different signal peptides, baculovirus transfer vectors, insect cell lines and infection strategies for S protein production. Specific production rates were maximized using BVM signal peptide, pOET5 baculovirus transfer vector, Su-perSf9-2 cell line and CCI = 2 × 10 6 and MOI = 1 pfu/cell. While the cell line is known to enhance expression of highly unstable proteins [20, 35] , its improved phenotype has never been demonstrated for S protein production; all other variables herein studied have not yet been explored or reported yet to date. Noteworthy is that the production yield achieved was ~4 mg/L, which, to the best of our knowledge, is the highest value obtained for full-length S protein production using IC-BEVS.

In this study, aiming to obtain a stable and folded trimer form of S protein in its prefusion conformation, we have combined the elimination of the S1 and S2 furin cleavage site with the addition of double proline mutation to prevent unfolding [36] and the HIV gp160 derived fusion clamp [37] . To further increase S protein stability, particularly upon storage, a screening study of formulation buffers was performed. Buffers B and C, which include glycerol and sucrose as cryoprotectants, respectively, have shown to outperform 

The production of the full-length recombinant S protein of SARS-CoV-2 S has been attempted in mammalian (e.g., HEK293, CHO) and insect (e.g., Sf -9 and High Five) cells; however, despite several bioprocess strategies being explored to date e.g., low culture temperature, new cell lines, most induce modest to no improvements in production yields [33, 34] .

In this study, we have explored different signal peptides, baculovirus transfer vectors, insect cell lines and infection strategies for S protein production. Specific production rates were maximized using BVM signal peptide, pOET5 baculovirus transfer vector, SuperSf9-2 cell line and CCI = 2 × 10 6 and MOI = 1 pfu/cell. While the cell line is known to enhance expression of highly unstable proteins [20, 35] , its improved phenotype has never been demonstrated for S protein production; all other variables herein studied have not yet been explored or reported yet to date. Noteworthy is that the production yield achieved was 4 mg/L, which, to the best of our knowledge, is the highest value obtained for full-length S protein production using IC-BEVS.

In this study, aiming to obtain a stable and folded trimer form of S protein in its prefusion conformation, we have combined the elimination of the S1 and S2 furin cleavage site with the addition of double proline mutation to prevent unfolding [36] and the HIV gp160 derived fusion clamp [37] . To further increase S protein stability, particularly upon storage, a screening study of formulation buffers was performed. Buffers B and C, which include glycerol and sucrose as cryoprotectants, respectively, have shown to outperform buffer A (without cryoprotectant) in all conditions tested (i.e., −80 and 4 • C, and after 2 freeze-thaw cycles). Glycerol and sucrose act as stabilizers by inhibiting protein aggregation, shifting the native protein ensemble to more compact states and reducing local backbone fluctuations, resulting in protein stabilization in extreme thermal or chemical environments [38, 39] . The list of approved drug products using glycerol as cryoprotectant is extensive as it facilitates drug uptake by the cells while having low toxicity associated [40] and as such, Buffer B (containing glycerol) was selected as the formulation buffer for S protein in our study. S protein thermal stability was assessed by DSF, with melting temperatures varying between 44-46 • C in agreement with literature data [30, 41, 42] . This method is commonly used to assess protein stability and readily applied for formulation buffer screening [43] . S protein aggregation was assessed using HPLC-SEC, which is one of the more robust and reproducible methods for tracing protein aggregates [44] . The chromatograms showed only one main form of S protein with approximately 400 kDa, consistent with its tertiary structure. Noteworthy, S trimeric form could be maintained for up to 90 days at −80 and 4 • C or after five freeze-thaw cycles, contrasting to other reports in which S protein aggregation was observed after 1 day at 4 • C [30, 42] .

The results obtained in this study reveal that most N-glycosylation sites in S protein were occupied and that N-glycans included complex/paucimannose glycans and high mannose glycans. In addition, five sites had more than 40% oligomannose type, similar to previous studies using Sf -9 cells [15] and mammalian HEK293 cells [17] . The S protein produced in insect cells, either as single antigen or displayed on the surface of the virosomes is bound by ACE2 receptor, pan-SARS antibody CR3022 [32] and neutralizing antibodies to the various epitope clusters [31] in ELISA, as proxy for its antigenicity integrity and biological activity [45] . Additionally, binding capacity was also maintained on virosomes-S stored at 4 • C for 1 month. These results suggest that S protein covalently coupled via its His tag to a click chemistry lipid present in the virosomal membrane results in an oriented display of the protein and properly exposes its receptor binding domain (RBD) involved in ACE2 binding on target cells, thus, theoretically favoring the induction of relevant neutralizing antibodies toward RBD for blocking cell infection.

This study demonstrates the potential of IC-BEVS for the expression of high-quality SARS-CoV-2 S protein to be included into a virosome-based COVID-19 vaccine candidate. The bioprocessing engineering strategies herein adopted allowed the production of~4 mg/L of full-length S protein, which, to the best of our knowledge, is the highest value achieved to date using insect cells. In addition, the insect Sf -9 cells derived S protein exhibited glycan processing similar to mammalian cells and mid-term stability upon storage. Furthermore, the S protein displayed on the surface of the virosomes was capable of binding to the ACE2 receptor and was recognized by a broad array of neutralizing antibodies, even after storage of the virosomes-S at 4 • C for 1 month. To validate these particles as a COVID-19 vaccine candidate, immunogenicity and safety-toxicology studies in adequate animal models should be performed.

Coronavirus Graphs: Worldwide Cases and Deaths-Worldometer

A Novel Coronavirus from Patients with Pneumonia in China

COVID-19: Coronavirus Vaccine Development Updates. Front

COVID-19: Novavax vaccine efficacy is 86% against UK variant and 60% against South African variant

World Health Organization. Interim Recommendations for Use of the Novavax NVX-CoV2373 Vaccine against COVID-19: Interim Guidance

Epaxal: A virosomal vaccine to prevent hepatitis A infection

Eleven years of Inflexal ® V-A virosomal adjuvanted influenza vaccine

Influenza virosomes as a combined vaccine carrier and adjuvant system for prophylactic and therapeutic immunizations

Influenza virosomes as vaccine adjuvant and carrier system

Vaccination with virosomally formulated recombinant CyRPA elicits protective antibodies against Plasmodium falciparum parasites in preclinical in vitro and in vivo models

A SARS-CoV-2 Wuhan spike virosome vaccine induces superior neutralization breadth compared to one using the Beta spike

A serological assay to detect SARS-CoV-2 seroconversion in humans

Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

Structural analysis of full-length SARS-CoV-2 spike protein from an advanced vaccine candidate

Glycosylation is a key in SARS-CoV-2 infection

Site-specific glycan analysis of the SARS-CoV-2 spike

Rapid, high-yield production of full-length SARS-CoV-2 spike ectodomain by transient gene expression in CHO cells

Technology transfer and scale-up of the Flublok ® recombinant hemagglutinin (HA) influenza vaccine manufacturing process

Thirty years of baculovirus-insect cell protein expression: From dark horse to mainstream technology

Protein N-glycosylation in the baculovirus-insect cell system

Two-Component Nanoparticle Vaccine Displaying Glycosylated Spike S1 Domain Induces Neutralizing Antibody Response against SARS-CoV-2 Variants

Triple layered rotavirus VLP production: Kinetics of vector replication, mRNA stability and recombinant protein production

Enabling PAT in insect cell bioprocesses: In situ monitoring of recombinant adeno-associated virus production by fluorescence spectroscopy

New GMP manufacturing processes to obtain thermostable HIV-1 gp41 virosomes under solid forms for various mucosal vaccination routes

Evaluation of the trypan blue technique for determination of cell viability

Titration of Non-Occluded Baculovirus Using a Cell Viability Assay

Error assessment in recombinant baculovirus titration: Evaluation of different methods

Improving Influenza HA-Vlps Production in Insect High Five Cells via Adaptive Laboratory Evolution

Production of high-quality SARS-CoV-2 antigens: Impact of bioprocess and storage on glycosylation, biophysical attributes, and ELISA serologic tests performance

Potent neutralizing antibodies from COVID-19 patients define multiple targets of vulnerability

A highly conserved cryptic epitope in the receptor binding domains of SARS-CoV-2 and SARS-CoV

Optimizing high-yield production of SARS-CoV-2 soluble spike trimers for serology assays

A thermostable, closed SARS-CoV-2 spike protein trimer

Production of recombinant proteins in insect cells

Virus vaccines: Proteins prefer prolines

Preclinical development of a molecular clamp-stabilised subunit vaccine for severe acute respiratory syndrome coronavirus 2

Effects of sugars on the thermal stability of a protein

Mechanisms of Protein Stabilization and Prevention of Protein Aggregation by Glycerol

Principles of cryopreservation

Characterization of the SARS-CoV-2 S Protein: Biophysical, Biochemical, Structural, and Antigenic Analysis

Characterization of the SARS-CoV-2 Spike in an Early Prefusion Conformation

Formulation screening by differential scanning fluorimetry: How often does it work?

Strategies for the Assessment of Protein Aggregates in Pharmaceutical Biotech Product Development

Immunogenicity assessment of biotherapeutic products: An overview of assays and their utility

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data Availability Statement: The sensitive nature of some of the reagents used in this study (e.g., cell lines, plasmids, baculoviruses, virosomes and antibodies) means that they are only readily available internally to the author's institutions staff for R&D purposes. For external researchers, approval of reagents request may be obtained via email addressed to the corresponding author.