Production and Immunogenicity Assessment of LTp50: An Escherichia coli-Made Chimeric Antigen Targeting S1- and S2-Epitopes from the SARS-CoV-2/BA.5 Spike Protein

Abstract

Subunit vaccines stand as a leading approach to expanding the current portfolio of vaccines to fight against COVID-19, seeking not only to lower costs but to achieve long-term immunity against variants of concern and have the main attributes that could overcome the limitations of the current vaccines. Herein a chimeric protein targeting S1 and S2 epitopes, called LTp50, was designed as a convenient approach to induce humoral responses against SARS-CoV-2. LTp50 was produced in recombinant Escherichia coli using a conventional pET vector, recovering the expected antigen in the insoluble fraction. LTp50 was purified by chromatography (purity > 90%). The solubilization and refolding stages helped to obtain a stable protein amenable for vaccine formulation. LTp50 was adsorbed onto alum, resulting in a stable formulation whose immunogenic properties were assessed in BALB/c mice. Significant humoral responses against the S protein (BA.5 variant) were detected in mice subjected to three subcutaneous doses (10 µg) of the LTp50/alum formulation. This study opens the path for the vaccine formulation optimization using additional adjuvants to advance in the development of a highly effective anti-COVID-19 vaccine directed against the antigenic regions of the S protein, which are less prone to mutations.

Keywords: COVID-19; built-in adjuvant; chimeric antigen; humoral response; linear epitopes.

Figure 1

(A) Schematic representation of the elements included in the LTp50 chimeric antigen. (B) Predicted structure for LTp50. The LTp50 structure predicted shows the classic structure of the heat-labile enterotoxin B subunit of E. coli; this structure consists of two sets of three anti-parallel beta sheets and alpha helices (red). The SARS-CoV-2 omicron epitopes (purple) are linked to each other with glycine linkers (yellow) linker. The DSFKEELDKYFKNHTS epitope (S_1146–1161) forms a four-turn alpha helix, while the DPSKRSFIEDLLFNKV epitope (S_812–826) is predicted to be a partially ordered alpha helix, the HADQLTPTWRVYSTGSNV epitope (S_625–642) is mostly unstructured with a small portion that could be forming a beta-sheet. The epitopes TESNKKFLPFQQFGRDIA (S_553–570), SYQTQTKSHRRA (S_673–684) and YVNNSYECDIPIGAGICA (S_655–672) are predicted to be completely unstructured.

Figure 2

Analysis of protein extracts recovered after flask expression of LTp50. (A) SDS-PAGE. Lanes: 1, uninduced soluble fraction; 2, soluble fraction from induced culture; 3, insoluble fraction from uninduced culture; and 4, insoluble fraction from induced culture. (B) Western blot. Lanes: 1, insoluble fraction from induced culture; +, positive control (100 ng of recombinant LTB). The theoretical molecular weight for LTp50 sequence is 25.4 kDa.

Figure 3

Downstream processing of LTp50. (A) Analysis by SDS-PAGE of the recovery phases of the LTp50 protein. Lanes: 1, extract from uninduced culture; 2, soluble fraction from induced culture; 3, supernatant recovered from the penultimate solubilization stage; 4, supernatant recovered from the last solubilization stage and used for chromatography; 5, mixture of the positive fractions recovered after purification of the protein by IMAC; 6, LTp50 after the downstream process; and MW: molecular weight marker. The theoretical molecular weight for LTp50 sequence is 25.4 kDa. (B) Chromatogram obtained after injecting S4 to an IMAC column, the LTp50 protein was recovered after increasing the imidazole concentration to 5 mM.

Figure 4

Cultivation in batch of E. coli Rosetta in Luria–Bertani (LB) broth. Conditions registered in a bioreactor continuously stirred at 400 rpm, initially with 1.4 L at 37 °C, 95 ± 1.5% of dissolved oxygen, pH of 7.1. Starting with OD = 0.148, the induction occurred after 5.25 h (OD = 0.879) at 28 °C, continuing for 10 h. The consumption of oxygen corresponded to the lag, exponential, and stationary phases of growth.

Figure 5

Analysis by SDS-PAGE of total protein extracts obtained at different expression times of the LTp50 protein in a 1.5 L culture using a batch, stirred tank bioreactor. (A) soluble fraction, (B) insoluble fraction, MW: molecular weight marker, +: LTp50 protein recovered in the flask scale. The theoretical molecular weight for the LTp50 sequence is 25.4 kDa.

Figure 6

Densitometry to determine purity and concentration of LTp50. Lanes: 1–5, BSA standard curve (1, 2, 3, 4, 5 µg/lane); 6, mixture of the positive fractions recovered after purification of the protein by IMAC. The theoretical molecular weight for the LTp50 sequence is 25.4 kDa.

Figure 7

Assessment of the LTp50 adsorption onto alum. The LTp50 antigen was incubated with different LTp50:alum mass ratios (1:2, 1:8, and 1:16) at 4 °C after 1 and 8 days, and a DLS analysis was performed to measure the hydrodynamic diameter (A) and ζ potential (B) of the antigen-adjuvant complexes. Antigen adsorption was verified by measuring the protein remnants in the supernatant (for all LTp50:alum ratios) after an electrophoretic analysis (C), showing almost complete protein adsorption. The first lane shows the soluble LTp50 (which was not contacted with alum), and the last lane is the molecular weight marker.

Figure 8

Anti-Spike IgG responses induced in mice by LTp50. Anti-S protein antibody levels were measured in serial dilutions from mice sera immunized with 10 µg doses of LTp50 plus alum or the vehicle plus alum. Mice were subjected to s.c. immunizations on days 1, 14, and 28, and sera were analyzed on day 45. Total IgG levels (A) or the IgG1/IgG2a subclasses (B) were determined using appropriate secondary antibodies. Reactivity of the test sera was assessed against the individual epitopes from the LTp50 antigen (C). The asterisks (* p < 0.05) indicate statistically significant differences versus the groups treated with alum alone.

Figure 9

Neutralization assay using SARS-CoV-2 BA.4/5 S-pseudotyped lentivirus. Serial two-fold dilutions of sera were mixed with 1 × 10⁶ relative light units (RLU) of BA.4/5-pseudotyped lentivirus before the addition of HEK293T-hACE2 cells. A virus pre-mixed with media alone (no serum) was used as a positive control to define 0% neutralization while cell-only wells were used as negative controls to define 100% neutralization. After 48 h, Bright-Glo substrate was added to each well and chemiluminescent signals were read using a Synergy HTX microplate reader. PVNT50 values were calculated in the Graphpad Prism 8.0 software using the reciprocal of the dilution that resulted in 50% neutralization. The dotted line indicates the negative cut-off value of 40.