Efficient design and production of glycosylated PEDV spike S1 protein Authors: Eva-Maria Uhlemann, Rinku Patel, Brittany Thivierge, Glenn Hamonic, Wolfgang Köster, Trina Racine, Qiang Liu Published: 2026 ABSTRACT The spike protein is a structural protein of coronaviruses and can be divided into an amino terminal S1 domain and a carboxyl terminal S2 domain. The S1 domain binds to receptors on the cell surface, so it contains major epitopes for neutralizing antibodies and is an important immunogen for vaccine development. Vaccine research and protein structure—function studies require pure protein preparations, ideally with no tag appended. In this study, we designed constructs encoding porcine epidemic diarrhea virus (PEDV) S1 protein with tissue plasminogen activator signal peptide and a Protein Select™ tag at the amino terminus. We showed the expression of PEDV S1 protein in the supernatants of mammalian cell cultures after transfection. We further showed the purification of tag-free PEDV S1 protein upon tag cleavage by the Protein Select™ resin from Cytiva. The purified PEDV S1 protein was then characterized by deglycosylation, dynamic light scattering, size-exclusion chromatography with multi-angle light scattering, one- and two-dimensional native-PAGE, and circular dichroism. The immune reactivity of the purified PEDV S1 protein to a hyper-immune serum was demonstrated in an enzyme-linked immunosorbent assay (ELISA). Taken together, we have expressed, purified, and comprehensively characterized PEDV spike S1 protein without a tag.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-56192-3. Introduction Porcine epidemic diarrhea virus (PEDV) is the etiological agent of PED in pigs1. PED was reported in the 1970s in England and the virus was isolated a few years later2,3. Since then, PEDV has been circulating across Europe and Asia with sporadic outbreaks4. Highly pathogenic PEDV strains were identified in 2010 when severe PED outbreaks were reported in China and the virus quickly spread into other countries, resulting in significant economic losses4. Porcine epidemic diarrhea is characterized by vomiting, diarrhea, dehydration, and anorexia, causing up to 100% mortality in suckling piglets and growth rate damage in finishing pigs4,5. PEDV is an enteric coronavirus in the genus Alphacoronavirus of the Coronaviridae family1. The PEDV genome is approximately 28 kb in length, encoding four structural proteins and numerous non-structural proteins1. The spike protein, one of the structural proteins, sits at the outer surface of the virion. Structurally, the spike protein can be divided into an amino terminal S1 domain and a carboxyl terminal S2 domain6. The S1 domain is responsible for binding receptors on the cell surface. As such, S1 contains major epitopes for neutralizing antibodies7,8. The S2 subunit is more conserved and mediates membrane fusion to enable entry of the virus9. Thus, the spike protein, especially the S1 protein, is a preferred target for subunit vaccine development1,8,10–13. In addition, PEDV S1 protein has been demonstrated to be a diagnostic antigen for detecting anti-PEDV antibodies14. Most vaccine research and protein structure—function studies require highly purified protein preparations. Achieving sufficient purity for recombinant tag-free proteins typically involves combining several general protein fractionation methods, such as differential precipitation, ion-exchange chromatography, and size-exclusion chromatography, into a multistep procedure tailored to the physicochemical properties of the target protein. A common strategy to simplify purification is the use of affinity tags15,16. Adding a tag enables detection with tag-specific antibodies and rapid purification using affinity chromatography. Among these, the His-tag comprising a short sequence of histidine residues is widely used due to its small size and nearly neutral charge at physiological pH, which generally results in limited impact on protein structure, function, or immunogenicity. However, several cases show that His-tags can interfere with conformation and functions of the target proteins in some cases17–20. Moreover, the His-tag may affect the immune responses to a protein antigen21. In addition, metal affinity chromatography alone often does not achieve the purity required for manufacturing, necessitating further purification steps. Protein affinity tags, such as maltose-binding protein (MBP) or glutathione-S-transferase (GST), may provide higher purity but require additional tag-removal and separation steps22. Combining a protein-based affinity tag with a self-cleaving intein allows producing highly purified target proteins in a single chromatographic step23, but the large size of the protein fusion constructs can reduce expression yields. An ideal affinity tag would be small and self-cleaving, allowing for efficient expression and purification of tag-free proteins in quantities needed for intended studies. In this study, we report the expression and purification of a tag-free PEDV S1 protein from mammalian cells using a split self-cleaving intein based chromatography system that meets these requirements24. The purified protein was extensively characterized as a first step to using it as an antigen in vaccine development. Materials and methods Construct design and plasmid DNA preparation The PEDV S1 protein expression construct was designed with a 5′ sequence encoding tissue plasminogen activator (TPA) signal peptide for secretion and a Protein Select™ tag (Cytiva) fused to the coding sequence for the S1 protein consisting of amino acids 21 to 734 of the spike protein of PEDV Colorado-2013 (GenBank Accession number KF272920, https://www.ncbi.nlm.nih.gov/nuccore/KF272920). We inserted a Ser-Arg-Gln tripeptide at the amino-terminus of the PEDV S1 protein during the cloning process. This sequence is optimal for Protein Select tag cleavage as per manufacturer information. To improve solubility and stabilize the protein structure, a Gly-Ser-Gly-Ser sequence was added at the C-terminus. In a separate construct, we inserted a hexahistidine-tag between the TPA signal peptide and the Protein Select™ tag to allow assessing Protein Select tag cleavage. The gene sequence was codon optimized for mammalian expression, synthesized, and cloned into the pcDNA3.4 vector (ThermoFisher Scientific) using a proprietary protocol (GenScript Biotech). Transfection grade plasmid DNA was prepared via Endotoxin Free Maxi prep kit (QIAGEN). Cell culture and transient transfection Expi293 and ExpiCHO cells cultured in their respective expression media according to manufacturer’s instructions (ThermoFisher Scientific). Cells were maintained in vented, non-baffled shake flasks on a 19 mm-throw orbital shaker at 125 rpm in a humidified incubator st to 37 °C, ≥ 90% relative humidity, and 8% CO₂. Cultures grown under these conditions were used for transfection and protein expression. One day prior to transfection, ExpiCHO cells were seeded to a density of 3–4 × 106 viable cells/ mL and Expi293 cells were seeded to a density of 2.5–3 × 106 viable cells/ mL. On the day of transfection, 100 mL of ExpiCHO cell culture was diluted to 6 × 106 cells /mL and transfected with 0.8 µg of plasmid DNA using ExpifectamineCHO reagent diluted in OptiPro medium. In parallel, 100 mL of Expi293 cells were diluted to 3 × 106 cells/mL and transfected with 1 µg of plasmid DNA with Expifectamine293 reagent diluted in OptiMEM medium (ThermoFisher Scientific). ExpiCHO cultures were maintained following the ExpiCHO™ Max Titer Expression protocol (ThermoFisher Scientific), including a temperature shift to 32 °C at 18–22 h post-transfection. Protein expression was monitored by daily harvesting of culture supernatants containing secreted proteins (days 1–6 for Expi293 cultures and days 1–7 for ExpiCHO cultures). Supernatants were collected by centrifugation at 4000× g for 30 min at 8 °C. For protein purification, Expi293 cultures were harvested on day 3 post-transfection. ExpiCHO cultures were harvested on day 3 for non-His-tagged proteins and on day 5 for His-tagged proteins. Cell cultures were centrifuged at 4000× g for 30 min at 8 °C, and the clarified supernatants were filtered through a 0.2 µm polyethersulfone membrane. Filtered supernatants were stored at − 20 °C until further use. Protein purification on protein select resin The PEDV S1 protein was purified from the cell culture supernatant at room temperature (RT) on a 1 mL HiTrap Protein Select resin column according to the manufacturer’s instructions (Cytiva) using an ÄKTA Pure™ 25 system (Cytiva). In brief, the column was equilibrated with 10 mL phosphate buffered saline (PBS), pH 7.4 (Gibco) at a flow rate of 4 mL/min. Then, 40 mL of cell culture supernatant was diluted with 10 mL PBS and immediately applied to the column at a flow rate of 4 mL/min. The column was washed at the same flow rate with 10 mL PBS and then incubated for 22 h at RT to allow tag cleavage. The protein was eluted with 5 mL PBS, pH 7.4, at 2 mL/min and collected in 1.5 mL fractions. SDS-PAGE, Coomassie blue staining, silver staining, and western blotting Protein samples were separated on a 4–20% gradient Mini-PROTEAN TGX gel (BioRad) with BioRad Tris–glycine-SDS running buffer at 200 V for 50 min. Gels were stained using Coomassie Blue staining (ThermoFisher Scientific) or silver staining (Pierce) kits. For Western blotting after SDS-PAGE, the proteins were transferred onto a 0.2 µm nitrocellulose membrane using Trans Turbo Transfer pack (BioRad) or wet tank blotting with Towbin transfer buffer. The membrane was blocked with 3% ECL Prime blocking agent (Cytiva) in PBS buffer, pH 7.4, washed twice with PBS, pH 7.4, supplemented with 0.05% Tween 20, and then twice with PBS, pH 7.4. The membrane was probed with an in-house anti-PEDV S1 pig serum at 1:5,000 dilution for 1 h at RT, washed as described above, and incubated with ReserveAP™ labeled anti-Swine IgG (H + L) antibody (SeraCare-KPL) at 1:2,500. The protein bands were visualized with the alkaline phosphatase substrate kit (BioRad) according to manufacturer’s instructions. Alternatively, the membrane was incubated with an anti-His-tag monoclonal antibody (MilliporeSigma) at 1: 5,000 dilution for 1 h at RT, washed as described above and incubated with IRDye 680CW anti-mouse antibody (Li-Cor Biosciences) at 1:10,000. The reaction was visualized with a Li-Cor NIR scanner. One- and two-dimensional native-PAGE analyses One-dimensional non-denaturing PAGE analysis of the purified PEDV S1 was performed using 4–12% gradient Bis–Tris NativePAGE system according to the manufacturer’s instructions (Life Technologies). For Western blotting, the native gel was incubated 2 × 15 min in a NuPAGE Transfer Buffer (Life Technologies) prior to transfer. For 2D analysis, 5 µg of purified protein were subjected to BN-PAGE as described above. Then a single full lane was cut out of the gel and successively incubated in reducing and alkylating solutions as per manufacturer’s instructions (Life Technologies). The gel strip was then applied to an IPG well of a 4–12% Bis–Tris Zoom™ gel and electrophoresed with the MOPS/SDS running buffer (ThermoFisher Scientific). Dynamic light scattering (DLS) and size exclusion chromatography with multi-angle light scattering (SEC-MALS) The purified PEDV S1 protein was analyzed on a Wyatt Technology multi-angle light scattering instrument with refractive index (RI) detector in-line with a Cytiva Superdex 200 Increase 10/300 GL column, which was calibrated using bovine serum albumin, connected to a BioRad NGS FPLC system. Size-exclusion chromatography was performed on a sample containing 0.9 mg protein in 500 µL PBS, pH 7.4 supplemented with 300 mM NaCl at a flow rate of 0.5 mL/min. Fractions were analyzed by SDS-PAGE and Western blotting. Acquired data was analyzed using Wyatt software package ASTRA 7.3.2. Glycoprotein analysis was carried out using the Wyatt/Waters conjugate analysis algorithm. Circular dichroism spectroscopy and thermal stability assay The secondary structure content of PEDV S1 protein was analyzed by circular dichroism (CD) spectroscopy using Applied Photophysiscs CHIRASCAN Plus CD spectrometer. Measurements were obtained from 194 to 280 nm with a bandwidth setting of 2 nm in a 0.5 mm quartz glass cell. The CD spectrum was a mean of three measurements. The resulting CD data was analyzed using the Beta Structure Selection (BeStSel) online tool (https://bestsel.elte.hu/ssfrompdb.php). For the thermal stability assay, CD spectra were successively recorded while the sample was gradually heated at a rate of 1 °C/min, from 5 to 95 °C, and then cooled back from 95 to 20 °C. The changes in the secondary structure were assessed at 215 nm and analyzed with GraphPad Prism 10 to determine the melting temperature of the protein. The experimentally determined secondary structure content was compared to the predicted from the high resolution PEDV S1 CryoEM structure (PDB ID: 6VV5), using PyMOL. Protein deglycosylation To remove the carbohydrate moiety, PEDV S1 protein was treated with PNGaseF under both non-denaturing and denaturing conditions according to manufacturer’s instructions (New England Biolabs). For non-denaturing reactions, 160 µg of purified protein were incubated with 25,000 U of PNGaseF for 21 h at 30 °C in parallel with a no-enzyme control. For denaturing reactions, the deglycosylation was performed with 2 µg of purified protein at 37 °C for 1 h. Samples were analyzed by SDS-PAGE with Coomassie staining and by Western blotting using an anti-PEDV S1 antibody as described previously. Mass spectrometry analysis A sample containing 5 µg of purified protein was subjected to SDS-PAGE on a 10% TGX gel (BioRad). The gel was stained with Bio-Safe Coomassie stain G-250 (BioRad). Alternatively, the protein was subjected to 2D BN-PAGE followed by silver staining. Protein bands or 2D spots were excised, and analyzed by ESI-LC–MS following in-gel digestion with trypsin at the Alberta Proteomics and Mass Spectrometry Facility, University of Alberta. ELISA Immulon-2 HB U Bottom Microtiter plates were coated with 0.5 µg/mL of one-step purified tag-free S1 protein in 100 µL coating buffer and incubated overnight at 4˚C. Next day, the plates were washed 5 times with ultrapure water and then blocked with 5% skim milk in Tris-buffered saline, pH 7.5, supplemented with Tween 20 (TBST) for an hour. After blocking the plates were washed again and incubated for 2 h at RT with hyperimmune mouse sera or saline-vaccinated control sera fourfold serially diluted, starting at 1:100 in TBST with 1% skim milk. The hyperimmune serum was generated by vaccinating mice with a subunit antigen containing PEDV S1 protein as we previously described25. All animal procedures were approved by the Animal Research Ethics Board, University of Saskatchewan (Protocol # 2024-0009)25. All the methods were performed in accordance with the relevant guidelines and regulations. After washing with ultrapure water, 100 μL of biotinylated goat anti-mouse IgG antibody (ThermoFisher Scientific) diluted 1:5000 with TBST/1% skim milk were added and the plate was incubated for 1 h. The plate was washed 5 times with ultrapure water and 100 μL of streptavidin-AP (Jackson ImmunoResearch) were added to each well and incubated for 1 h at RT. After the incubation, the wells were washed 5 times with ultrapure water followed by incubation with p-nitrophenyl phosphate substrate (Sigma) for 15 min at RT. The reaction was stopped with 0.3 M EDTA, and absorbance was measured at 405 nm with a reference wavelength of 490 nm. The cut-off value was defined as the mean absorbance of the negative controls plus 2 standard deviations. Immunoreactivity was determined as the highest serum dilution that produced an absorbance value above the cut-off. Results and discussion Construct design The aim of this study was to express a secreted tag-free PEDV S1 protein. To achieve this, we fused the coding sequences for TPA signal peptide and Protein Select™ tag (Cytiva) to the sequence encoding the N-terminus of PEDV S1 protein in the pcDNA3.4 expression vector, generating plasmid PS-PEDV S1 (Fig. 1A; GenBank accession number: PZ052629, https://www.ncbi.nlm.nih.gov/nuccore/PZ052629). Because an antibody against the Protein Select tag is not available, we generated a His6-PS-PEDV S1 construct with a His-tag inserted between the TPA signal peptide and the Protein Select tag (Fig. 1B; GenBank accession number: PZ052630, https://www.ncbi.nlm.nih.gov/nuccore/PZ052630). This will allow monitoring protein expression level and Protein Select tag removal during protein purification using an anti-His tag antibody.Figure 1PEDV spike S1 expression plasmids constructed in this study. (A) Coding sequences for tissue plasminogen activator (TPA) signal peptide and the Protein Select tag (Cytiva) followed by Ser-Arg-Gln tripeptide were fused to PEDV S1 gene and cloned into the expression vector pcDNA3.4. The plasmid was designated PS-PEDV S1. (B) A hexahistidine-tag was inserted between the TPA signal peptide and the Protein Select tag, generating plasmid His6-PS-PEDV S1. PEDV spike S1 expression plasmids constructed in this study. (A) Coding sequences for tissue plasminogen activator (TPA) signal peptide and the Protein Select tag (Cytiva) followed by Ser-Arg-Gln tripeptide were fused to PEDV S1 gene and cloned into the expression vector pcDNA3.4. The plasmid was designated PS-PEDV S1. (B) A hexahistidine-tag was inserted between the TPA signal peptide and the Protein Select tag, generating plasmid His6-PS-PEDV S1. Expression of PEDV S1 protein To express PEDV S1 protein in mammalian cells, Expi293 and ExpiCHO cells were transfected with plasmids PS-PEDV S1 or His6-PS-PEDV S1. Western blotting analysis of the cell culture supernatants with an anti-PEDV S1 antibody showed an increasingly intense protein band around 130 kDa over time, reaching maximum on Day 3 for Expi293 cells for both constructs, on Day 5 for the non-his-tagged construct in ExpiCHO cells, and on Day 3 for the his-tagged construct in ExpiCHO cells after plasmid transfection (Fig. 2A). When an anti-His-tag antibody was used in Western blotting, protein bands with the same apparent molecular weight were detected after His6-PS-PEDV S1 transfection (Fig. 2B). Minor bands between 70 and 80 kDa likely represented protein degradation products. It is noteworthy that the apparent molecular weight of ~ 100–130 kDa is much higher than the theoretical value of 73.85 and 74.69 kDa without and with the His-tag, respectively. This difference is probably due to protein glycosylation as we showed subsequently. These results indicate that both the PS-PEDV S1 and His6-PS-PEDV S1 proteins are expressed in the transfected Expi293 and ExpiCHO cells and secreted into the cell culture supernatants.Figure 2PEDV spike S1 protein expression in mammalian cells. Expi293 and ExpiCHO cells were transfected with constructs PS-PEDV S1 or His6-PS-PEDV S1. Cell culture supernatants collected at different time points after transfection were subjected to Western blotting with an anti-PEDV S1 antibody (A) or anti-His-tag antibody (B). The expressed PEDV S1 protein was marked with an arrow on the blots. Harvest day is shown above each lane. Original blots/gels for A and B are presented in Supplementary Fig. 1. PEDV spike S1 protein expression in mammalian cells. Expi293 and ExpiCHO cells were transfected with constructs PS-PEDV S1 or His6-PS-PEDV S1. Cell culture supernatants collected at different time points after transfection were subjected to Western blotting with an anti-PEDV S1 antibody (A) or anti-His-tag antibody (B). The expressed PEDV S1 protein was marked with an arrow on the blots. Harvest day is shown above each lane. Original blots/gels for A and B are presented in Supplementary Fig. 1. Purification of PEDV S1 protein on protein select resin We next attempted to purify the PS-PEDV S1 and His6-PS-PEDV S1 proteins from transfected Expi293 cell culture supernatants on the Protein Select resin (Cytiva). Flow-through and elution fractions were collected during the chromatography monitored by UV light absorption A280 (Fig. 3A,D). A single protein peak was recorded during elution (fractions E1 and E2) for both PS-PEDV S1 (Fig. 3A) and His6-PS-PEDV S1 (Fig. 3D) proteins. The two elution fractions contained a major protein band with an apparent molecular weight between 100 and 130 kDa, as shown by SDS-PAGE followed by Coomassie blue staining (Fig. 3B,E). The identity of the major band as PEDV S1 protein was confirmed by Western blotting with anti-PEDV S1 serum (Fig. 3C) and by mass spectrometry following tryptic digest (Supplementary Fig. 4).Figure 3PEDV S1 protein purification with Protein Select resin. The supernatants of Expi293 cells transfected with plasmid PS-PEDV S1 (A, B and C) or His6-PS-PEDV S1 (D, E, F and G) were used to purify the PEDV S1 protein on Protein Select resin. Purification was monitored by UV light absorption (A280) (A and D). Collected fractions were analyzed by SDS-PAGE followed by Coomassie blue staining (B and E) or Western blotting with a polyclonal anti-PEDV S1 antibody (C and G) or with an anti-His tag antibody (F). Lanes: M, molecular weight marker with molecular weight (kDa) shown on the left; L, column load; FT, column flow through; W, column wash; E1 and E2, elution fractions; R, Protein Select resin after elution. Original blots/gels for C, G and F are presented in Supplementary Figs. 2 and 3, respectively. PEDV S1 protein purification with Protein Select resin. The supernatants of Expi293 cells transfected with plasmid PS-PEDV S1 (A, B and C) or His6-PS-PEDV S1 (D, E, F and G) were used to purify the PEDV S1 protein on Protein Select resin. Purification was monitored by UV light absorption (A280) (A and D). Collected fractions were analyzed by SDS-PAGE followed by Coomassie blue staining (B and E) or Western blotting with a polyclonal anti-PEDV S1 antibody (C and G) or with an anti-His tag antibody (F). Lanes: M, molecular weight marker with molecular weight (kDa) shown on the left; L, column load; FT, column flow through; W, column wash; E1 and E2, elution fractions; R, Protein Select resin after elution. Original blots/gels for C, G and F are presented in Supplementary Figs. 2 and 3, respectively. Due to the lack of Protein Select tag-specific antibody, we used the His-tagged version of the protein to assess the extent of Protein Select tag cleavage during resin binding and elution. On the Western blot, the crude cell culture supernatant showed a strong reaction with the anti-His antibody at the apparent molecular weight corresponding to His6-PS-PEDV S1 (Fig. 3F). In contrast, there were no reactive bands in the elution fractions containing the purified PEDV S1 (Fig. 3F), while the eluted protein identity was confirmed by reaction with anti-PEDV S1 pig serum (Fig. 3G). Since the Protein Select tag is positioned between the His-tag and PEDV S1 protein, this indicated efficient removal of the Protein Select tag. We also purified PEDV S1 proteins from transfected ExpiCHO cell culture supernatants (data not shown). When the yields of purified proteins from Expi293 and ExpiCHO cell supernatants were compared, we found that 64 mg/L was achieved from ExpiCHO cells, whereas the yield in Expi293 cells was approximately 30% lower (Table 1). Interestingly, the addition of a His-tag at the N-terminus decreased the protein yields in both cell lines (Table 1). The reason for the yield difference is not clear. However, it may be due to the reduced secretion efficiency caused by introducing the bulky positively charged histidine residues immediately after the signal peptide, as the first two amino acid residues immediately following the signal peptide are known to affect protein secretion efficiency26.Table 1Purified PEDV S1 protein yields from Expi293 and ExpiCHO cell culture supernatants.Expression ConstructExpi293 Yield (mg/L)aExpiCHO Yield (mg/L)aExpiCHO / Expi293His₆-PS-PEDV S128.951.81.79PS-PEDV S164.082.21.28aProtein concentration was measured by absorbance at 280 nm (A280). Protein was purified from 45 mL cultures. Table 1Purified PEDV S1 protein yields from Expi293 and ExpiCHO cell culture supernatants.Expression ConstructExpi293 Yield (mg/L)aExpiCHO Yield (mg/L)aExpiCHO / Expi293His₆-PS-PEDV S128.951.81.79PS-PEDV S164.082.21.28aProtein concentration was measured by absorbance at 280 nm (A280). Protein was purified from 45 mL cultures. Purified PEDV S1 protein yields from Expi293 and ExpiCHO cell culture supernatants. aProtein concentration was measured by absorbance at 280 nm (A280). Protein was purified from 45 mL cultures. Deglycosylation of PEDV S1 protein The purified PEDV S1 protein migrated as diffuse bands with an apparent molecular weight around 130 kDa, whereas the calculated molecular weight is only 78.3 kDa. Coronaviral spike proteins are well known to be modified by glycosylation9. To determine whether the observed difference between the calculated and apparent molecular weights could be attributed to glycosylation, we performed a protein deglycosylation assay. Indeed, following PNGaseF enzyme treatment, the PEDV S1 protein migrated as a band at ~ 84 kDa, which aligns with the theoretical molecular weight (Fig. 4). Interestingly, the deglycosylated PEDV S1 appeared to react weaker with the PEDV S1 antibody compared to the native glycosylated protein on Western blotting (Fig. 4B). Although the reason for the observed difference in immunoreactivity is not known, it may indicate that the glycan moiety contributes to the immunogenicity of the PEDV S1 protein.Figure 4PEDV S1 protein deglycosylation. Purified PEDV S1 protein was treated with PNGaseF under the denaturing or non-denaturing conditions. Subsequently, the samples were analyzed by Coomassie blue staining (A) and Western blotting with an anti-PEDV S1 antibody (B). Asterisk * denotes the PNGaseF protein band (A). Original blot/gel for B is presented in Supplementary Fig. 5. PEDV S1 protein deglycosylation. Purified PEDV S1 protein was treated with PNGaseF under the denaturing or non-denaturing conditions. Subsequently, the samples were analyzed by Coomassie blue staining (A) and Western blotting with an anti-PEDV S1 antibody (B). Asterisk * denotes the PNGaseF protein band (A). Original blot/gel for B is presented in Supplementary Fig. 5. Biophysical characterization of PEDV S1 protein We characterized aggregation status, purity and stability of the purified PEDV S1 protein using dynamic light scattering (DLS), size-exclusion chromatography with multi-angle light scattering (SEC-MALS) and circular dichroism (CD) spectroscopy. DLS showed that the PEDV S1 protein purified via Protein Select resin was mostly monomeric, with 99% of the protein contained in a single peak corresponding to a polydisperse molecular mass between 115 and 150 kDa (Fig. 5), likely due to heterogeneous protein glycosylation. A minor population of high–molecular weight aggregates was observed in the sample formulated in PBS with 155 mM NaCl (arrow in Fig. 5A); these aggregates were no longer detected when the sample was formulated in PBS containing 300 mM NaCl (Fig. 5B).Figure 5Molecular mass determination of purified PEDV S1 protein by dynamic light scattering (DLS). (A) Mass distribution histogram of PEDV S1 protein measured in physiological PBS containing 155 mM NaCl. (B) Mass distribution histogram of PEDV S1 protein measured in PBS containing 300 mM NaCl. Molecular mass determination of purified PEDV S1 protein by dynamic light scattering (DLS). (A) Mass distribution histogram of PEDV S1 protein measured in physiological PBS containing 155 mM NaCl. (B) Mass distribution histogram of PEDV S1 protein measured in PBS containing 300 mM NaCl. The SEC profile showed a single and nearly symmetric main peak at the approximate retention volume corresponding to the monomeric protein with a minor front shoulder, consistent with the DLS results (Fig. 6A).Figure 6Assessment of the oligomeric state of purified PEDV S1 protein by size-exclusion chromatography (SEC). (A). SEC of the purified PEDV S1 protein. (B and C). Protein-containing fractions were analyzed by Coomassie blue staining (B) or Western blotting with an anti-PEDV S1 antibody (C). Lanes: M, molecular weight marker with molecular weights shown on the left (kDa); L, column load; 22–30, elution fractions. Original blot/gel for C is presented in Supplementary Fig. 6. Assessment of the oligomeric state of purified PEDV S1 protein by size-exclusion chromatography (SEC). (A). SEC of the purified PEDV S1 protein. (B and C). Protein-containing fractions were analyzed by Coomassie blue staining (B) or Western blotting with an anti-PEDV S1 antibody (C). Lanes: M, molecular weight marker with molecular weights shown on the left (kDa); L, column load; 22–30, elution fractions. Original blot/gel for C is presented in Supplementary Fig. 6. The MALS analysis of this peak showed an average molecular mass of 108.7 kDa with minor component distribution in the range of 93–115 kDa (Fig. 7B). This was consistent with the molecular weight distribution observed in the corresponding SEC-fractions by SDS-PAGE (Fig. 6B,C) and with the DLS data (Fig. 5A,B). On the Western blot, all the fractions reacted to the anti-PEDV S1 antibody (Fig. 6C), and the band pattern aligned well with the Coomassie blue stained gel (Fig. 6B). Therefore, the major band in all the peak fractions represented PEDV S1.Figure 7Molecular mass distribution of purified PEDV S1 protein analyzed by multi-angle light scattering (MALS). (A). The chromatogram shows the light-scattering signal (red), refractive index signal (blue), and UV absorbance at 280 nm (green). (B). Molecular weight distribution determined by MALS with the horizontal orange trace representing the total conjugate mass (glycosylated protein), conjugate modifier analysis showing the protein mass (horizontal black trace) and the glycan modifier mass (horizontal magenta trace). Molecular mass distribution of purified PEDV S1 protein analyzed by multi-angle light scattering (MALS). (A). The chromatogram shows the light-scattering signal (red), refractive index signal (blue), and UV absorbance at 280 nm (green). (B). Molecular weight distribution determined by MALS with the horizontal orange trace representing the total conjugate mass (glycosylated protein), conjugate modifier analysis showing the protein mass (horizontal black trace) and the glycan modifier mass (horizontal magenta trace). To further determine if the observed molecular weight variation was due to the glycosylation of the protein, we performed conjugate modifier analysis of the SEC-MALS data (Fig. 7B and Table 2). The results indicated that the protein component with molecular weight of 71 kDa was attached to heterogeneous glycan with a molecular weight of 37 kDa.Table 2Molecular weight analysis of PEDV S1 protein.SampleTheoreticala, kDaSEC-MALS, average MW, kDaSEC-MALS, Top of Peak, kDaSDS-PAGE, Apparent MW, kDaPEDV S1 (glycosylated)–108.7108 ~ 90–120PEDV S1 (without glycan modifier)78.3567.6571 ~ 84Glycan modifier–41.0537–aTheoretical values were calculated using ProtParam. Table 2Molecular weight analysis of PEDV S1 protein.SampleTheoreticala, kDaSEC-MALS, average MW, kDaSEC-MALS, Top of Peak, kDaSDS-PAGE, Apparent MW, kDaPEDV S1 (glycosylated)–108.7108 ~ 90–120PEDV S1 (without glycan modifier)78.3567.6571 ~ 84Glycan modifier–41.0537–aTheoretical values were calculated using ProtParam. Molecular weight analysis of PEDV S1 protein. aTheoretical values were calculated using ProtParam. One- and two-dimensional native-PAGE analyses We showed that Protein Select resin could be used to purify PEDV S1 protein in one-step procedure (Fig. 3). Size-exclusion chromatography, which we used for SEC-MALS analysis, could also serve as an additional purification step (Fig. 8A). To compare the purity of the PEDV S1 protein after one-step or two-step purification procedure using a more stringent test than 1D PAGE, we performed 2D electrophoresis.Figure 8One-dimensional (1D) native-PAGE analysis of the purity of PEDV S1 protein preparations. (A). Schematic overview of PEDV S1 purification by a single affinity chromatography step (1-step) or by affinity chromatography followed by size-exclusion chromatography (2-step). (B, C and D). Purified PEDV S1 protein preparations analyzed by 1D native-PAGE and visualized by Coomassie blue staining (B), silver staining (C), or Western blotting using an anti-PEDV S1 antibody (D). Lane 1 contains the one-step purified sample, and lane 2 contains the two-step purified sample. Molecular weight marker positions are shown on the left. Original blot/gel for D is presented in Supplementary Fig. 7. One-dimensional (1D) native-PAGE analysis of the purity of PEDV S1 protein preparations. (A). Schematic overview of PEDV S1 purification by a single affinity chromatography step (1-step) or by affinity chromatography followed by size-exclusion chromatography (2-step). (B, C and D). Purified PEDV S1 protein preparations analyzed by 1D native-PAGE and visualized by Coomassie blue staining (B), silver staining (C), or Western blotting using an anti-PEDV S1 antibody (D). Lane 1 contains the one-step purified sample, and lane 2 contains the two-step purified sample. Molecular weight marker positions are shown on the left. Original blot/gel for D is presented in Supplementary Fig. 7. The 1D native-PAGE showed that PEDV S1 protein purified via either one-step or two-step procedure contained a major band around 230 kDa by Coomassie blue staining, silver staining, and Western blotting (Fig. 8B–D), and a minor component at approximately 400 kDa, detectable by silver staining only in 1-step purification (Fig. 8C, lane1). In the second dimension, under the denaturing conditions, the minor component migrated at the same apparent molecular weight as the major one (Fig. 9B) and reacted with the anti-PEDV S1 antibody (Fig. 9C), suggesting that it represented PEDV S1 dimer. The PEDV S1 protein purified by the two-step procedure showed a single and nearly symmetric spot on the 2D gel, even after the highly sensitive silver staining (Fig. 9E). Mass-spectrometry showed that the protein contained no detectable peptides originating from the Protein Select tag. These results indicated that the purified protein was homogeneous. Thus, the two-step procedure produces essentially pure protein, but requires an additional chromatography step, while the one-step procedure is faster, and the protein purity may be sufficient for many applications. For manufacturing purposes, the one-step procedure therefore may be more cost-effective.Figure 9Two-dimensional (2D) native-PAGE analysis of PEDV S1 protein purified by 1-step or 2-step procedures. Purified PEDV S1 preparations were subjected to 2D native-PAGE followed by Coomassie blue staining (A), silver staining (B), or Western blotting with an anti-PEDV S1 antibody (C). Molecular weight marker positions (kDa) are shown on the left. The corresponding first dimension gel stained with Coomassie is shown above panel A. Original blots/gels for C and F are presented in Supplementary Fig. 8. Two-dimensional (2D) native-PAGE analysis of PEDV S1 protein purified by 1-step or 2-step procedures. Purified PEDV S1 preparations were subjected to 2D native-PAGE followed by Coomassie blue staining (A), silver staining (B), or Western blotting with an anti-PEDV S1 antibody (C). Molecular weight marker positions (kDa) are shown on the left. The corresponding first dimension gel stained with Coomassie is shown above panel A. Original blots/gels for C and F are presented in Supplementary Fig. 8. Secondary structure analysis We experimentally determined the secondary structure content and stability of PEDV S1 via CD spectroscopy. PEDV S1 secondary structure determined by CD consisted of 38.2% β-sheet, 4.7% α-helix, and various other conformations (Fig. 10A,B). This secondary structure is consistent with the high-resolution structure of PEDV spike protein (PDB ID: 6VV5)6, which shows several domains with high β-sheet content, connected by extended loop regions in the S1 domain (Fig. 10C). The CD spectrum also contained a characteristic disulfide bond peak at 225 nm, consistent with multiple disulfide bridges present in the native protein, indicating an intact tertiary structure (Fig. 10D). We also conducted a thermal denaturation experiment to assess conformational stability of the PEDV S1 protein. The PEDV S1 melting point determined from the CD at 215 nm as a function of temperature by non-linear regression was found to be 58.4 °C with 95% confidence interval of 57–60.1 °C (Fig. 10E). The protein denatured at 95 °C showed partial refolding upon cooling the sample to 20 °C with some irreversible denaturation and aggregation (Fig. 10D). These results demonstrated that the tag-free PEDV S1 protein purified via the Protein Select resin was properly folded and stable under the standard purification and storage conditions.Figure 10Secondary structure and thermal stability of purified PEDV S1 protein analyzed by circular dichroism (CD) spectroscopy. (A). Thermal denaturation profile of the purified PEDV S1 protein. (B). Estimated secondary structure composition of PEDV S1 protein. (C). A high-resolution structure of PEDV S1 (PDB ID: 6VV5). (D). CD spectra of the purified PEDV S1 protein before and after thermal denaturation at 95 °C recorded at 20 °C, showing the changes in the secondary structure and the loss of disulfide bonds upon melting. (E). Determination of the melting temperature of purified PEDV S1 protein. Secondary structure and thermal stability of purified PEDV S1 protein analyzed by circular dichroism (CD) spectroscopy. (A). Thermal denaturation profile of the purified PEDV S1 protein. (B). Estimated secondary structure composition of PEDV S1 protein. (C). A high-resolution structure of PEDV S1 (PDB ID: 6VV5). (D). CD spectra of the purified PEDV S1 protein before and after thermal denaturation at 95 °C recorded at 20 °C, showing the changes in the secondary structure and the loss of disulfide bonds upon melting. (E). Determination of the melting temperature of purified PEDV S1 protein. Immunoreactivity of recombinant PEDV S1 protein Using mouse serum samples obtained after immunization with a different subunit vaccine containing PEDV S1 protein or saline performed in our laboratory25, we determined the immune reactivity of the tag-free PEDV S1 in ELISA assay. The purified tag-free PEDV S1 protein reacted with the hyperserum down to 105 dilution, while no reactivity was detected with control serum after saline injections (Fig. 11). These results are consistent with what we reported previously25.Figure 11Immune reactivity of PEDV S1 protein. Using purified tag-free PEDV S1 protein as the coating antigen, PEDV S1-specific IgG titres in hyperimmune and saline-vaccinated mouse sera were determined in ELISA. Immune reactivity of PEDV S1 protein. Using purified tag-free PEDV S1 protein as the coating antigen, PEDV S1-specific IgG titres in hyperimmune and saline-vaccinated mouse sera were determined in ELISA. In summary, we demonstrated that mammalian cells transfected with the engineered PEDV-S1 construct efficiently expressed and secreted the highly glycosylated protein of interest with a cleavable affinity tag, producing robust yields for downstream purification. Using the Protein Select tag technology, we achieved a fast and successful one-step purification of the PEDV S1 protein, yielding a 110 KDa glycosylated product and producing a tag-less final protein of approximately 98–99% purity as shown by SDS-PAGE analysis. We further performed comprehensive biophysical characterization of the purified S1 protein, demonstrating its structural integrity and suitability for downstream applications. The PEDV S1 protein produced in this study provides a valuable reagent for future functional analyses and vaccine development, including applications in vaccine manufacturing. The integrated workflow from rapid expression in mammalian cells to the generation of a highly pure, tag-less protein through a simple and cost-effective one-step purification offers a practical and efficient approach for antigen production. This streamlined process is well suited for vaccine manufacturing pipelines that require speed, scalability, and reproducibility. Moreover, the purification and characterization strategy described here should be readily adaptable for producing spike protein antigens of other coronaviruses. Supplementary Information Supplementary Information. Supplementary Information.