A novel heterologous receptor-binding domain dodecamer universal mRNA vaccine against SARS-CoV-2 variants

Abstract

There are currently approximately 4000 mutations in the SARS-CoV-2 S protein gene and emerging SARS-CoV-2 variants continue to spread rapidly worldwide. Universal vaccines with high efficacy and safety urgently need to be developed to prevent SARS-CoV-2 variants pandemic. Here, we described a novel self-assembling universal mRNA vaccine containing a heterologous receptor-binding domain (HRBD)-based dodecamer (HRBD^dodecamer) against SARS-CoV-2 variants, including Alpha (B.1.1.7), Beta (B.1.351), Gamma (B.1.1.28.1), Delta (B.1.617.2) and Omicron (B.1.1.529). HRBD containing four heterologous RBD (Delta, Beta, Gamma, and Wild-type) can form a stable dodecameric conformation under T4 trimerization tag (Flodon, FD). The HRBD^dodecamer -encoding mRNA was then encapsulated into the newly-constructed LNPs consisting of a novel ionizable lipid (4N4T). The obtained universal mRNA vaccine (4N4T-HRBD^dodecamer) presented higher efficiency in mRNA transfection and expression than the approved ALC-0315 LNPs, initiating potent immune protection against the immune escape of SARS-CoV-2 caused by evolutionary mutation. These findings demonstrated the first evidence that structure-based antigen design and mRNA delivery carrier optimization may facilitate the development of effective universal mRNA vaccines to tackle SARS-CoV-2 variants pandemic.

Graphical abstract

The HRBD^dodecamer containing four heterologous RBD (Delta, Beta, Gamma, and Wild-type) was designed as a safe, effective, and universal mRNA vaccine.

1. Introduction

The SARS-CoV-2 pandemic is characterized by recurring waves of cases driven by the emerging SARS-CoV-2 variants with higher fitness¹. The 1544 PANGO lineages and 2904 nonsynonymous mutations of SARS-CoV-2 have been identified (https://cov-lineages.org/lineage_list.html). Hundreds of variants were formed, including eleven types of variants being monitored (VBM) and five types of variants of concern (VOC)^2,3. SARS-CoV-2 variants, especially Alpha (B.1.1.7), Delta (B.1.617.2), Beta (B.1.351), Gamma (B.1.1.28.1), and Omicron (B.1.1.529), has spread globally causing humongous public health crises⁴. The inactivated vaccines or nucleic acid vaccines based on a single SARS-CoV-2 (e.g., the Wild-type? virus) are highly likely unable to prevent immune evasion caused by constant mutations in emerging viral variants^5,6, while the bivalent vaccine containing Wild-type SARS-CoV-2 and Beta (mRNA-1273.211) presented potent and durable antibody responses against emerging variants⁷. Taken into account, we proposed the hypothesis that a structure-based design based on heterologous antigen integration might be a potential approach to develop a universal vaccine against the immune escape of SARS-CoV-2 variants.

SARS-CoV-2 consists of several structural proteins, spike (S), nucleocapsid (N), envelope (E), and membrane (M) proteins⁸. The receptor-binding domain (RBD) of S is the smallest domain that directly binds to the ACE2 receptor to mediate the entry of virions into target cells, thereby acting as the most promising vaccine target9, 10, 11. RBD monomers have limited size and poor immunogenicity, which are not easily captured by the immune system^12,13. In contrast, multimeric antigens have advantages in interacting with B cell receptors, thereby promoting the production of high-affinity antibodies and exerting stronger humoral and cellular immune responses¹⁴. The dimer or multimer RBD-based vaccines induce higher neutralizing antibody titers than the RBD monomers^13,15,16. This kind of antigen design just focused on a certain RBD, so we tried to construct a potent SARS-CoV-2 vaccine based on a heterologous RBD multimer in this work.

Research has shown that infection with Omicron is not effective in increasing immunity to future reinfection as limited T and B cell immunity is induced^17,18, while the SARS-CoV-2 vaccines designed according to the VOC rather than Omicron also provided protection against Omicron due to the induced memory B cell repertoire19, 20, 21. Collectively, the RBD fusion antigen (HRBD^dodecamer) containing four heterologous RBD (Delta, Beta, Gamma, and Wild-type) was here designed to obtain universal SARS-CoV-2 vaccine, which combined the mutational properties of SARS-CoV-2 and the superiority of RBD multimeric antigens in activating immunity.

The messenger RNA (mRNA)-based vaccines have demonstrated tremendous advantages due to their high efficacy, safety, and low-cost manufacturing processes, and have emerged as a rapid and versatile platform to combat the devastating pandemic coronavirus disease 2019 (COVID-19)^22,23. Additionally, the mRNA vaccine platform has substantial advantages in antigen natural modification for the preparation of vaccines^22,24. Engineering precision delivery systems play a critical role in mRNA encoding antigen expression and initiating durable protective immunity^25,26. In particular, lipid nanoparticles (LNPs) have been shown to enable efficient and safe mRNA delivery, and have been successfully used to treat SARS-CoV-2 infections through the delivery of mRNA (such as SM-102 and ALC-0315 LNPs for mRNA-1273 and BNT162b2)^22,27. Ionizable lipids, one of the key components of LNPs, are critical to enhancing the delivery efficiency and therapeutic efficacy of mRNA vaccines27, 28, 29, 30. Ionized lipids provide a positive charge to ensure maximum encapsulation of negatively charged mRNAs into LNPs through electrostatic interactions and help mRNAs cross cellular barriers²⁵. Importantly, numerous studies have shown that LNPs with multiple charges (such as C12-200, G0-C14, cKK-E12, etc.) significantly improve mRNA encapsulation efficiency, uptake and lysosomal escape²⁶. The rational design of ionizable lipids has great potential to initiate more potently protective immunity of mRNA-LNP vaccines^31,32. We designed a new set of multi-charged ionizable lipids (MIC1-MIC6) and found significant increases in the transfection and expression efficiency of mRNA³⁰.

Here, we designed and developed a heterologous RBD-based dodecamer universal mRNA vaccine (4N4T-HRBD^dodecamer) based on novel LNPs containing optimally ionizable lipid nanomaterials. The 4N4T-HRBD^dodecamer elicited robust and durable protective immunity (CD4⁺ and CD8⁺ T cell immune responses) against five types of VOC SARS-CoV-2 variants in mice, including Alpha, Beta, Gamma, Delta, and Omicron. Importantly, the 4N4T-HRBD^dodecamer induced 5-fold higher RBD-specific IgG titers and neutralizing antibody responses than 4N4T-based monomer RBD or ALC-0315-based S mRNA vaccines. These results highlight that this new approach, combining the structure-based antigen design (heterologous RBD-based dodecamer) and the novel ionizable lipid-based LNPs, has great potential for developing universal mRNA vaccines for broadly targeting SARS-CoV-2 variants.

2. Materials and methods

2.1. Materials

N-Boc-ethylenediamine, triethylamine, dry dichloromethane (DCM), acryloyl chloride, piperazine, trifluoroacetic acid, K₂CO₃, 2-dodecylethylene oxide, and isopropanol solution was obtained from Jinan Daigang Biomaterial Co., Ltd. (Daigang, Jinan, China).

2.2. Methods

2.2.1. Ethics statement

This study was carried out following the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocols were approved by the Institutional Animal Care and Use Committee at Sichuan University. All the animal experimental procedures were performed under anesthesia that was induced and maintained with isoflurane, and all efforts were made to minimize animal suffering.

2.2.2. Animals

BALB/c mice (6–8 weeks) were purchased from Beijing Huafukang Biotechnology Co., Ltd. (Huafukang, Beijing, China). Mice were maintained in the animal facility at Sichuan University for two weeks before the experiment. Male animals were used with random grouping. All animal studies were approved by the Sichuan University Institutional Animal Care and Use Committee and performed following the animal care and institutional guidelines.

2.2.3. Cell lines

Human embryonic kidney 293 cells (HEK293T) and mouse dendritic cells (DC2.4) were obtained from American Type Culture Collection (ATCC, 293T, Manassas, VA, USA) and cultured following the manufacturer's instructions. Mouse bone marrow dendritic cells (BMDCs) were induced from the primary bone marrow cells using GM-CSF (Abcam, ab9742, Cambridge, UK) and cultured in a complete RPMI 1640 medium.

2.2.4. Synthesis of 4N4T lipids

The synthesis steps of 4N4T lipids are shown in Supporting Information Fig. S1A. Briefly, N-Boc-ethylenediamine (10 mmol) and triethylamine (20 mmol) were dissolved in dry dichloromethane (DCM) and acryloyl chloride (12 mmol) was added dropwise on ice. The mixture was stirred in an ice-water bath overnight, and the crude product was purified by silica gel column chromatography to obtain compound 1 (named as t-butyloxy carbonyl-AEAA, Boc-AEAA). A mixture of compound 1 (8 mmol) and piperazine (4 mmol) was stirred in ethanol at 60 °C overnight and evaporated in a vacuum. The crude product was purified by silica gel column chromatography to obtain compound 2 (named as t-butyloxy carbonyl-PAE, Boc-PAE). Four milliliters of trifluoroacetic acid was added to compound 2 (4 mmol) and stirred at room temperature for 2 h. Excess K₂CO₃ and 2-dodecylethylene oxide (12 mmol/L) were added, and the mixture was placed in an isopropanol solution and stirred at 90 °C for 24 h. The product was filtered, evaporated in a vacuum, and purified by a chromatographic column to obtain a novel ionizable lipid named 4N4T. The structures of all new and established compounds were verified by nuclear magnetic resonance (NMR) (Fig. S1). ALC-0315³³ was purchased from Hitgen (Hitgen, ALC-0315, Chengdu, China).

2.2.5. Characterization of 4N4T lipid nanoparticles

GFP mRNA and firefly luciferase (FLuc) mRNA were encapsulated in 4N4T ionizable lipid nanoparticles (4N4T-LNPs) or ALC-0315 LNPs. The morphology of 4N4T lipid nanoparticles was observed by transmission electron microscopy (TEM), and the zeta potential and average size of lipid nanoparticles 4N4T/FLuc mRNA complexes were measured by using a Malvern Laser Particle Size Analyzer (Zetasizer Nano ZS 90, Malvern, UK). The distribution/targeting of 4N4T@FLuc-mRNA and transfection efficiency were measured in vivo and in vitro, respectively. Briefly, 4N4T@FLuc-mRNA was transfected into DC2.4 cells and BMDCs. Transfection efficiency was analyzed by flow cytometry (GFP mRNA) and a microplate reader (FLuc mRNA). Similarly, BALB/c mice were injected intramuscularly with 20 μg of 4N4T@FLuc-mRNA, and then were injected intraperitoneally with 3 mg of luciferin in each mouse (dissolved in 200 μL of PBS) after 8 h. Then, mice were anesthetized with isoflurane (RWD Life Science, R510-22, San Diego, CA, USA), and images were captured with an in vivo imaging system (IVIS) spectrometer (PerkinElmer, 124,262, Waltham, MA, USA) at 15 min after injection of luciferin.

2.2.6. SARS-CoV-2 RBD dodecamer mRNA design and synthesis

The RBD-dodecamer was formed by tandem four heterologous RBD (Delta, Beta, Gamma, and Wild-type) of SARS-CoV-2 (The mutation sites are shown in Supporting Information Fig. S3A and Table S1). The nucleic acid sequence of RBD mRNA was integrated into the open reading frame of plasmid (hCD24) containing the T7 promoter and poly A tail, N-terminal tPA signal peptide (SP), the linker peptides (LP) and the C-terminal Foldon (FD) trimer tag, and was obtained by in vitro synthesis (Jiyu Technology Co., Ltd., Shenzhen, China), and was further capped to obtain in vitro transcribed (IVT) mRNA. The construction of antigen was accurately predicted by AlphaFold 2 (Fig. 1).

Figure 1.

Design of dodecameric SARS-CoV-2 RBD mRNA antigens. The encoding of SARS-CoV-2 RBD antigens from the heterologous S proteins of SARS-CoV-2, including Delta, Beta, Gamma, and Wild-type, were designed and accurately predicted by AlphaFold 2. The per-residue confidence score (pLDDT) of the predicted structure is 78.74, indicating that the predicted results are highly credible.

2.2.7. Synthesis of 4N4T-HRBD^dodecamer

4N4T lipid, 1,2-dioleoyl-sn-glycerol-3-phosphoethanolamine (DOPE) (AVT), cholesterol, and 1,2-dimyristoyl-rac-glycerol-3-methoxy polyethylene glycol-2000 (DMG-PEG2000) were co-dissolved in pure ethanol at a molar ratio of 36/15/46.5/2.5, and mRNA was diluted with 10 mmol/L sodium citrate buffer (pH 6.0). The ratio of SARS-CoV-2 RBD dodecamer mRNA to 4N4T lipid nanomaterials was adjusted to 15:1 (w/w) using a syringe pump (the final concentration of mRNA was 0.1 mg/mL), and the mixture was mixed in a 1:3 ratio in a microfluidic chip device to generate the 4N4T SARS-CoV-2 RBD dodecamer universal mRNA vaccine (4N4T-HRBD^dodecamer).

2.2.8. Quantification of SARS-CoV-2 RBD dodecamer antigen expression

SARS-CoV-2 RBD dodecamer mRNA was transfected into HEK293T cells by Lipofectamine™ 2000 according to the manufacturer's instructions (Thermo Fisher Scientific, 11668027, Waltham, MA, USA). Supernatants and cell lysates were obtained, and the expression of SARS-CoV-2 RBD dodecamer antigen in vivo was detected by SARS-CoV-2 RBD protein detection ELISA kit (Vazyme, 7E510D1, Nanjing, China) and Western blotting (WB). Briefly, cells were lysed with Cell Lysis Reagent (Sigma–Aldrich, C3228-50 ML, St. Louis, MO, USA) containing protease inhibitor (Thermo Fisher Scientific, 78,430, Waltham, MA, USA), denatured at 100 °C for 10 min, separated by electrophoresis on 10% SDS-PAGE gels, and then transferred to nitrocellulose transfer membranes (GE Amersham Biosciences, 10-6000-01, Pittsburgh, PA). Membranes were incubated with RBD primary Abs overnight (Proteintech Europe, 67758-1-Ig, Deansgate, UK), and then incubated with corresponding secondary Abs conjugated to HRP (Santa Cruz Biotechnology, sc-2005, Dallas, TX, USA). Finally, the relative expression levels of protein were detected using ECL reagents (Santa Cruz Biotechnology, sc-2048, Dallas, TX, USA) and quantified by Quantity One software [Bio-Rad Laboratories, Quantity One®1-D, Hercules, CA, USA)].

2.2.9. Animals vaccinated

Male BALB/c mice (aged 6–8 weeks, n = 6) were selected for the immunological evaluation of the 4N4T-HRBD^dodecamer. Mice were immunized i.m. once with three different doses (1, 5  or 10 μg per mouse) as the prime immunization, and phosphate-buffered saline (PBS) served as the placebo control. The mice were immunized with two (boost) doses at 2-week intervals. Serum samples were collected on Days 14, 28, and 35 after immunization. Serum antibody titers were measured on Days 14 and 28. Pseudovirus neutralization experiments were performed on Day 35. Spleen samples and organs were acquired on Day 40, and ELISpot and intracellular cytokine staining (ICS) assays were performed.

2.2.10. Binding antibody responses using ELISA

ELISA plates were coated with 0.1 μg of SARS-CoV-2 and its variants RBD recombinant proteins, including Wild-type, Alpha, Delta, Beta, Gamma, and Omicron (BA.1) (Vazyme, Nanjing, China) variants in coating buffer (Sangon Biotech, Shanghai, China) overnight at 4 °C. The coated plates were washed with wash buffer and blocked with 2% BSA (BioFroxx, Guangzhou, China) at 25 °C for 4 h. Serially diluted serum samples were added to ELISA plates and incubated overnight at 4 °C. The plates were washed and incubated with anti-mouse IgG-HRP antibodies (Thermo Fisher Scientific, 88-50400-86, Waltham, MA, USA) that were diluted in wash buffer containing 0.2% BSA (1:50,000). Plates were incubated for 2 h at 25 °C. TMB (Solarbio, PR1200, Beijing, China) was added and the reaction was stopped with 2 mol/L sulfuric acids, and the absorbance was measured at 450 nm using a microplate reader (Tecan Group Ltd., Männedorf, Switzerland). End-point titers were used to define serum antibody titers against SARS-CoV-2 and its variants.

2.2.11. Pseudovirus neutralization assays

SARS-CoV-2 and its variants pseudovirus (Wild-type, Delta, Beta, Gamma, and Omicron) were purchased from Genomeditech (Shanghai, China). 293T-ACE2 cells were cultured in complete DMEM at 37 °C and 5% CO₂ in 96-well plates. Serially diluted serum samples were added to plates. Add a total of 0.1 μL of pseudovirus (diluted in 50 mL of complete medium) to each well. The mixture of serum and pseudovirus was incubated for 1 h at 37 °C. Then, 293T-ACE2 (1.5 × 10⁴) cells were added to each well and incubated at 37 °C for 48 h. Luciferase substrate (100 μL) (Vazyme, Nanjing, China) was added to each well and shaken for 2 min. Luminescence was measured using a microplate reader (Tecan Group Ltd., Männedorf, Switzerland). Neutralizing titers (NT) were defined as the dilutions of serum required to inhibit 50% of luciferase activity compared to viral controls.

2.2.12. Intracellular cytokine staining (ICS)

Functional responses of SARS-CoV-2 RBD dodecamer specific CD8⁺ and CD4⁺ T cells in vaccinated animals were measured using peptide pools and an intracellular cytokine staining (ICS) assay. Splenocytes were isolated in complete RPMI 1640 medium (HyClone Laboratories Inc., SH30197.02, Grand Island, NY, USA). Cells were incubated overnight with a peptide pool of Delta S proteins (DG peptides) and 2 μmol/L monensin (YEASEN, HB210319, Shanghai, China) was added after 2 h. Cells were harvested and stained with anti-CD4 FITC (Biolegend, 100,405, San Diego, CA, USA) and anti-CD8a APC (Biolegend, 162,305, San Diego, CA, USA), and then fixed and permeabilized in fixation buffer (Biolegend, 420,801, San Diego, CA, USA) and intracellular staining buffer (Biolegend, 421,002, San Diego, CA, USA), followed by anti-interferon γ (IFN-γ) PE/Cy7, anti-interleukin-2 (IL-2) PE, and anti-interleukin-4 (IL-4) PerCP/Cy5.5 staining, and were analyzed and quantified by flow cytometry (NovoCyteTM, Eisen Bioscience, USA).

2.2.13. Enzyme-linked immunospot (ELISpot) assay

IFN-γ ELISpot assay of splenocyte was performed using the mouse IFN-γ ELISpot^PLUS kit (3321-4APW-10, Mabtech, Stockholm, Sweden) according to the manufacturer's instructions. The 96-well plates precoated with IFN-γ antibodies were washed with PBS and incubated with RPMI 1640 medium for 30 min at room temperature. Splenocyte cells (5 × 10⁵) were incubated with the DG peptides at 37 °C for 48 h, and then the plates were incubated with biotinylated anti-IFN-γ antibody and streptavidin-ALP at room temperature for 2 h after washing with PBS. IFN-γ ELISpot was measured by an ELISpot reader (Autoimmun Diagnostika GmbH, Strassberg, Germany), and T cell activation was assessed by measuring the expression of IFN-γ.

2.2.14. Safety evaluation of the universal mRNA vaccine

Male BALB/c mice were i.m. administered with 30 μg of 4N4T SARS-CoV-2 RBD dodecamer universal mRNA vaccine per mouse, and serum was extracted and used to examine the blood biochemistry indexes 12 h after immunization. Likewise, tissues were fixed in 10% formalin (Sigma–Aldrich, PR1200, St. Louis, MO, USA) for 48 h at 4 °C and then embedded in paraffin using a routine histologic procedure. H&E staining was carried out according to the standard staining procedure.

2.2.15. Quantification and statistical analysis

All experiments were performed at least three times unless stated otherwise. Data were presented as mean ± standard error means (SEM). All data were analyzed with GraphPad Prism 8.0 software using One-Way ANOVA plus Tukey's post-hoc test; Statistically significant differences are indicated as ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001.

3. Results

3.1. Dodecamer SARS-CoV-2 RBD mRNA immunogen conformation

The RBD of the SARS-CoV-2 spike directly mediates host cell entry by combining with the cellular receptor angiotensin-converting enzyme 2 (ACE2)^34,35. SARS-CoV-2 RBD contains multiple distinct antigenic sites and antibodies binding to the RBD can potently block the attachment of the virus to ACE2³⁶. The RBD is an attractive vaccine target but has limited immunogenicity¹⁵. There are significant differences in the specific antigens induced by heterologous SARS-CoV-2 RBD, namely, with typical antigen–antibody specificity. To acquire the multimer conformation to enhance the immunogenicity of the SARS-CoV-2 RBD and antibody specificity, the gene sequences of SARS-CoV-2 RBD mRNA antigens from the heterologous S proteins of SARS-CoV-2, including Delta, Beta, Gamma, and Wild-type, were integrated as a fusion gene. To model the structure of SARS-CoV-2 RBD mRNA antigens, we used AlphaFold 2 to accurately visualize its conformation in vitro, showing a fused symmetrical rosette-like tetrameric structure and presenting a stable dodecameric under FD tag (pLDDT 78.74) (Fig. 1). These results demonstrate that the structure-based design of self-assembling mRNA immunogens formed stable and natural multimeric structures in vitro.

3.2. Characterization of SARS-CoV-2 RBD dodecameric mRNA immunogen

The efficient expression of mRNA-encoded antigens reveals a critical role in initiating durable protective immunity^15,23. The RBD dodecameric immunogens were integrated into the open reading frame of the plasmid (hCD24) containing T7 promoter and poly A tail, SP, LP, and the C-terminal Foldon (FD) trimer tag (Fig. 2A). We confirmed the expression of the heterologous fusion RBD protein in HEK293T cells by using Western blotting under reduced conditions (Fig. 2B). Likewise, to confirm SARS-CoV-2 RBD dodecameric formation, we also examined its expression by using Western blotting under nonreduced conditions (Fig. 2C). These results showed that the heterologous RBD protein can be expressed in the intracellular and supernatant of 293T cells. The heterologous RBD protein (HRBD) was further confirmed by gel electrophoresis to be ∼120 kDa, and the fusion RBD protein containing the C-terminal FD trimer tag (HRBD-F) was confirmed to be ∼300 kDa (Fig. 2D). These results demonstrated that four heterologous RBD fusion proteins formed dodecameric structures under T4 trimerization.

Figure 2.

Dodecameric SARS-CoV-2 RBD mRNA immunogen expression in vitro. (A) Flow chart of the experimental design. The various RBDs (R391-D541) of the SARS-CoV-2 spike (S) protein containing the N-terminal tPA signal peptide (SP), the linker peptides (LP), and C-terminal Foldon (FD) trimer tag were synthesized in hCD24 DNA plasmid by codon optimization. The modified mRNA was synthesized by in vitro transcribed (IVT). (B) 4N4T-based-LNPs containing 3-μg mRNA is transfected into HEK293T, the expression of the dodecameric SARS-CoV-2 RBD protein in the whole-cell lysate (WCL) and supernatant (C) was quantified by Western blotting under reduced conditions and ELISA. (D) The expression of the dodecameric SARS-CoV-2 RBD protein was analyzed by Western blotting under nonreduced conditions. HRBD, heterologous-RBD mRNA antigen; HRBD-F, heterologous-RBD-Foldon mRNA antigen. Data are presented as mean ± SD (n = 3). ∗∗∗∗P < 0.0001.

3.3. High mRNA transfection and expression efficiency of novel 4N4T ionizable lipid nanoparticles

LNPs-based mRNA delivery systems typically consist of four components: ionizable lipids, phospholipids, cholesterol, and PEGylated lipids^37,38. Ionizable lipids are positively charged by protonation at low pH, which facilitates the endosomal escape of nanoparticles³⁹. Likewise, ionizable lipids are characterized by pH-sensitivity, which results in less interaction with the anionic membrane of blood cells or negatively charged plasma proteins, thereby enhancing the biocompatibility of lipid nanoparticles, which plays an important role in improving transfection efficiency and mRNA expression39, 40, 41. We pioneered the design of a positively charged 4N4T lipid composed of four tertiary amine nitrogen atoms (4N) and four hydrophobic tails (4T) after screening a large number of ionizable lipids. The synthetic roadmap is shown in Fig. S1A. The synthesized product was identified by NMR (Fig. S1B and S1C). Next, the 4N4T-LNPs-based mRNA were prepared by squeezing the mixture of alcoholic lipid solution and aqueous mRNA solution in the microfluidic chip, entrapping mRNA occurs when amphiphilic lipids form nanostructures with a hydrophilic outer layer and a hydrophobic inner layer in water. We evaluate the stability of 4N4T-LNPs-based GFP mRNA (4N4T@GFP-mRNA) (Fig. 3A and Supporting Information Fig. S2A). The results showed that the particle size, particle size polydispersity index and encapsulation efficiency of 4N4T remained stable within 180 days at 5 and ‒20 °C. Additionally, the mRNA encapsulation efficiency was up to 85%, and the mRNA integrality (%) is more than 80% (Fig. 3B and Fig. S2A-2C). Importantly, the results showed that the mRNA transfection and expression efficiency of 4N4T-LNPs into DC2.4 cells and BMDCs were significantly higher than that of ALC-0315 LNPs (Fig. 3C–E). Transmission electron microscopy (TEM) images of 4N4T-LNPs showed that uniform and stable nanoparticles were formed (Fig. 3F). Furthermore, the expression efficiency and endosome escape ability of 4N4T-LNPs was stronger than ALC-0315 LNPs in vivo (Fig. 3G and Fig. S2D). These data demonstrated that the novel 4N4T ionizable lipid nanoparticles have high mRNA transfection and expression efficiency and favorable stability for hugely potential increasing mRNA-based drug efficacy.

Figure 3.

Novel 4N4T-mRNA LNPs construction and characterization. (A) Average size and ζ potential of 4N4T-LNPs were measured by DLS. (B) mRNA encapsulation efficiency was measured by using agarose gel electrophoresis stained with ethidium bromide and quantified by the RiboGreen assay, and mRNA integrity of 4N4T@HRBD-mRNA and 4N4T@HRBD-F-mRNA were evaluated by Qsep1 Automatic Bioanalyzer. (C) The transfection and expression efficiency of 4N4T@FLuc-mRNA were analyzed using SpectraMax MiniMax 300 in HEK293T. (D) The 4N4T@GFP mRNA was transfected into DC2.4 cells and BMDCs, the expression images of GFP⁺ cells were captured by SpectraMax MiniMax 300, scale bar = 150 μm. (E) The expression level of GFP⁺ cells were analyzed and quantified by using flow cytometry. (F) TEM images of 4N4T-LNPs, scale bar = 100 nm. (G) The distribution and expression of 4N4T@FLuc-mRNA were measured by an IVIS imaging system in vivo. 4N4T, 4N4T@GFP-mRNA; ALC-0315, ALC-0315@GFP mRNA. Data are presented as mean ± SD (n = 3). ∗∗∗∗P < 0.0001.

3.4. 4N4T-HRBD^dodecamer is a universal mRNA vaccine against SARS-CoV-2 variants

To maximize the efficacy of the mRNA vaccine, we encapsulated fusion RBD mRNA into the novel ionizable lipid nanoparticles 4N4T-LNPs to form the self-assembling SARS-CoV-2 RBD dodecamer universal mRNA vaccine (4N4T-HRBD^dodecamer). Primarily, the SARS-CoV-2 spike mRNA (Delta) were generated with 4N4T-LNPs or ALC-0315 LNPs. The results showed that the 4N4T-S protein induced 2-fold higher specific S protein-binding antibody responses than ALC-0315-S protein (Fig. S3B), indicating that delivery vector optimization can significantly enhance the efficacy of mRNA vaccines. To further test whether the structure-based antigen design have remarkable immunogenicity and protective potential, a series of vaccine candidates expressing heterologous RBD structure patterns were generated with 4N4T-LNPs, and were assessed in immunized BALB/c mice. To our delight surprise, the 4N4T-HRBD^dodecamer induced 5-fold higher specific Delta RBD-binding antibody responses than those induced by the SARS-CoV-2 S protein and RBD monomer or tetramer (Fig. S3C). These data illustrated that structure-based antigen design and mRNA delivery carrier optimization significantly improved mRNA vaccine immunogenicity and efficacy. To further explore the protective potential of 4N4T-HRBD^dodecamer for SARS-CoV-2 and its variants, RBD-specific IgG titers were detected in 4N4T-HRBD^dodecamer-vaccinated mice. The results showed that the 4N4T-HRBD^dodecamer induced higher antibody titers than the 4N4T-based heterologous-RBD (no C-terminal Foldon) against the five VOCs (Supporting Information Fig. 4B–G). Our analysis supports that the 4N4T-HRBD^dodecamer, as a universal mRNA vaccine, induced robust specific RBD-binding antibodies against the immune escape of SARS-CoV-2 caused by evolutionary mutation. The high binding antibody titers supported subsequent assessment of the neutralizing antibody titers in 4N4T-HRBD^dodecamer-immunized BALB/c mice. Furthermore, the neutralizing antibody titer induced by 4N4T-HRBD^dodecamer was evaluated by pseudovirus neutralization experiments in SARS-CoV-2 and its variants containing the primary mutation to increase transmissibility⁴². Apparently, the 4N4T-HRBD^dodecamer elicits potent neutralizing antibodies against SARS-CoV-2 and its variants in 10-μg dose, including the newly emerging Omicron variant. Likewise, it is higher than 4N4T-based heterologous-RBD (no C-terminal Foldon) (Fig. 4H–K). Collectively, these findings indicate that 4N4T-HRBD^dodecamer-induced antibodies possess neutralizing activity and can be used as a universal mRNA vaccine against diverse SARS-CoV-2 variants.

Figure 4.

Immune response in 4N4T-HRBD^dodecamer-vaccinated mice. (A) Flow chart of the experimental design. Male BALB/c mice were i.m. administered different doses of 4N4T-HRBD^dodecamer (PBS administered as control). Samples (serum, lung tissue, spleen tissue and associated primary cells) were collected at various time points after immunization. (B–G) RBD-specific IgG titers were detected by ELISA, anti-WT RBD IgG titer (B), anti-Delta RBD IgG titer (C), anti-Beta RBD IgG titer (D), and anti-Gamma RBD IgG titer (E), anti-Alpha RBD IgG titer (F), anti-Omicron RBD IgG titer (G). (H–K) The neutralization titer induced by the 4N4T-HRBD^dodecamer was evaluated by pseudovirus neutralization experiments, anti-WT RBD neutralizing antibody titer (H), anti-Beta RBD neutralizing antibody titer (I), anti-Delta RBD neutralizing antibody titer (J), and anti-Omicron RBD neutralizing antibody titer (K). HRBD, heterologous-RBD; HRBD-F, heterologous-RBD-Foldon; 14-HRBD, heterologous-RBD-induced IgG titers on Day 14; 14-HRBD-F, heterologous-RBD-Foldon-induced IgG titers on Day 14; 28-HRBD, heterologous-RBD-induced IgG titers on Day 28; 28-HRBD-F, heterologous-RBD-Foldon-induced IgG titers on Day 28. Data are presented as mean ± SD (n = 3). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗∗P < 0.0001.

3.5. 4N4T-HRBD^dodecamer activates Th1-biased T-cell responses

Primary immunization with the mRNA vaccine generates potent anti-SARS-CoV-2 neutralizing antibodies and CD4⁺ and CD8⁺ T cell responses, ultimately generating memory plasma cells against future infections⁴³. T cells responses (helper T lymphocytes CD4⁺ and cytotoxic T lymphocytes CD8⁺) confer durable immune memory to prevent re-infection with SARS-CoV-2, such as CD8⁺ T cells persisting for 6–11 years in SARS-CoV-1 survivors⁴⁴. CD4⁺ initiates type 1 T helper (Th1) and type 2 T helper (Th2) immune responses against bacterial and viral infections⁴⁵. The Th1 immune response activates cellular immune responses and inhibits viral replication by secreting the pro-inflammatory cytokines IFN-γ, TNF-β, IL-2, and TNF-α^46,47. Th2 cells activate humoral immune responses and balance Th1 immune responses by secreting the anti-inflammatory cytokines IL-2, IL-4, and IL-6⁴⁸. Understanding the activation of Th1 and Th2 immune responses is crucial for maintaining adaptive immune responses⁴⁹. To further study whether a SARS-CoV-2-specific T cells immune response were elicited by two doses of immunization with the 4N4T-HRBD^dodecamer, we detected SARS-CoV-2 RBD-specific CD4⁺ and CD8⁺ T cells in splenocytes by flow cytometry. The results showed that the specific CD4⁺ and CD8⁺ effector T cells were significantly increased in splenocytes from 4N4T-HRBD^dodecamer-vaccinated mice compared to the 4N4T-heterologous RBD upon stimulation with peptide pools covering the SARS-CoV-2 RBD (Fig. 5A). Furthermore, the SARS-CoV-2 RBD-specific cytokines IL-4, IL-2 and IFN-γ produced by CD4⁺ and CD8⁺ T cells were measured in 10-μg 4N4T-HRBD^dodecamer-vaccinated mice by ICS. ELISpot assay was also performed to visually analyze the expression of RBD-specific IFN-γ. Our results showed that the secretion of the RBD-specific cytokines IL-2 and IFN-γ in splenocytes from 10-μg 4N4T-HRBD^dodecamer-immunized mice was significantly higher than that in splenocytes from 4N4T-heterologous RBD-immunized mice. There was no significant difference in IL-4 secretion between 4N4T-RBD^dodecamer-immunized animals and 4N4T-HRBD, indicating a favorable Th1 profile and no potentially deleterious Th2 immune response (Fig. 5B and C and Fig. S4). Our results indicate that the 4N4T-HRBD^dodecamer mRNA vaccine successfully induces a Th1-biased specific cellular immune response.

Figure 5.

SARS-CoV-2-specific T cell immune response in 4N4T-HRBD^dodecamer vaccinated mice. (A) SARS-CoV-2 RBD-specific CD4⁺ and CD8⁺ T cells in splenocytes were detected upon stimulation with peptide pools covering the SARS-CoV-2 RBD by flow cytometry. (B) The SARS-CoV-2 RBD-specific cytokines IL-4, IL-2 and IFN-γ produced by CD4⁺ and CD8⁺ T cells were measured upon stimulation with peptide pools covering the SARS-CoV-2 RBD by ICS. (C) ELISpot assay for IFN-γ in splenocytes. HRBD: 10 μg heterologous-RBD, HRBD-F: 10 μg heterologous-RBD-Foldon, 5HRBD: μ5 μg heterologous-RBD. Data are presented as mean ± SD (n = 3). ∗P < 0.05, ∗∗P < 0.01.

3.6. 4N4T-HRBD^dodecamer is a safe, effective, and universal mRNA vaccine

To further evaluate in vivo safety, we employed blood biochemistry indexes and H&E staining in 10-μg 4N4T-RBD^dodecamer-immunized BALB/c mice. There was no significant difference in toxicity as measured by inflammation, tissue injury and so on in 10-μg 4N4T-RBD^dodecamer-immunized BALB/c mice in comparison with 4N4T-HRBD or sham control (Fig. 6A and B). Consistent with in vitro and in vivo observations, these results support that the 4N4T-HRBD^dodecamer is a safe, effective, and universal mRNA vaccine that elicits robust protective immunity against SARS-CoV-2 and its variants. We may conclude that customizable structure-based antigen design and vector-optimized delivery have potential research value in the field of mRNA vaccines and have promising clinical application prospects (Fig. 7).

Figure 6.

Safety evaluation of 4N4T-HRBD^dodecamer. (A) Blood biochemistry indexes, including alanine aminotransferase (ALT), alkaline phosphatase (ALP), total protein (TP), albumin (ALB), creatine kinase MB isoenzyme (CKMB), urea (UREA), creatinine (CRE), aspartate aminotransferase (AST), in immunized mouse serum were measured to assess the safety of 4N4T-HRBD^dodecamer. (B) H&E staining of major organ pathologies. The scale bar is 50 nm and the photograph was magnified 40 times. HRBD: 10 μg heterologous-RBD, HRBD-F: 10 μg heterologous-RBD-Foldon.

Figure 7.

Schematic illustration of the 4N4T-HRBD^dodecamer elicits robust and durable protective immunity against SARS-CoV-2 and its variants. In vitro transcribed (IVT) 4N4T-HRBD^dodecamer is encapsulated into 4N4T-LNPs and is endocytosed by antigen-presenting cells. mRNA is released into the cytoplasm after escaping from endosomes and then translated into RBD tetramer proteins by ribosomes. Subsequently, endogenous antigens are degraded into polypeptides by the proteasome and are presented by MHC I, and then activate cytotoxic T cells (CD8⁺ T cells). Additionally, HRBD proteins were self-assembled into dodecamers under FD (HRBD^dodecamer) and secreted outside the cell, where they can initiate potent humoral immunity response by directly interacting with B cells. Additionally, it also was taken up by cells, degraded inside endosomes, and presented on the cell surface to helper T cells (CD4⁺ T cells) by MHC class II proteins, stimulating B cells to produce RBD-specific neutralizing antibodies against SARS-CoV-2 and its variants.

4. Discussion

The initial protective immunity elicited by mRNA vaccines may not provide comprehensive protection with the continued emergence of new variant strains of SARS-CoV-2 thanks to immune escape⁵⁰. An increasing number of mutations in the RBD and NTD of spike proteins reduced the sensitivity of SARS-CoV-2 to antibody neutralization⁵¹. The D614G variant increases viral infectivity and transmissibility and reduces viral susceptibility to neutralization⁵². Variants carrying E484K (Alpha and Beta) slightly reduced the protection provided by the BNT162b2 mRNA vaccine⁵³, and the N501Y mutation (Alpha, Beta, and Gamma) reduced virus susceptibility to neutralizing monoclonal antibodies and vaccine-induced polyclonal antibodies⁵⁴. L452R (Epsilon, Iota, and Kappa) promotes viral replication and infectivity, and resistance to monoclonal antibodies⁵⁵. Neutralizing antibodies against SARS-CoV-2 mostly arise from IgG⁺ B cells and target the RBD, and most of them (antibodies) are typically non-polyreactive^16,56. Omicron variants have robust immune escape for vaccine-associated neutralization antibodies despite attenuated replication and pathogenicity⁵⁷. These findings showed that universal vaccines are urgently needed to counter emerging SARS-CoV-2 variants. Here, we designed and developed an ionizable lipid nanoparticles-based self-assembling SARS-CoV-2 RBD dodecamer universal mRNA vaccine and tested its immunogenicity and efficacy in vivo and in vitro. Our findings indicate that the 4N4T-HRBD^dodecamer elicits robust diverse specific RBD binding and neutralizing antibodies for SARS-CoV-2 and its variants, including Wild-type, Delta (L452R, T478K), Beta (K417N, E484K, N501Y), Gamma (K417T, E484K, N501Y), Alpha (N501Y), and Omicron (BA.1) (G339D, S371L, S373P, S375F, K417N, N440K, G446S, S477N, T478K, E484A, Q493K, G496S, Q498R, N501Y, Y505H, T547K) (Fig. S3A), and induced a potent Th1-biased specific cellular immune response (Fig. 5).

Multimerization RBD protein or prefusion-stabilized spike (S) glycoprotein have been shown to induce potent neutralizing antibody responses^6,58. SARS-CoV-2 mRNA vaccination induces functionally diverse antibodies to the nucleocapsid protein N-terminal domain (NTD), RBD, and spike 2⁵⁹. The SARS-CoV-2 RBD is the target of 90% of the neutralizing activity present in COVID-19 convalescent sera and has been a focus of therapeutic and vaccine design efforts⁵⁶. Importantly, the SARS-CoV-2 RBD elicits a potent neutralizing response without antibody-dependent enhancement (ADE) that has been reported in SARS-CoV and MERS-CoV^10,11, suggesting that the RBD-based vaccine is safe and effective⁶⁰. However, monomeric RBD antigens have limited ability on engaging interactions with B cell receptors and thereby initiate potent protective immunity^56,61,62. Various strategies have been developed to improve the immunogenicity of RBD protein^63,64. An RBD-Fc-based COVID-19 vaccine candidate induces highly potent SARS-CoV-2 neutralizing antibody response⁶⁵. Multimerization of the SARS-CoV-2 RBD using SpyCatcher induces potent neutralizing antibody responses⁶⁶. Structure-guided immunogen designs RBD dimers as universally increased neutralizing antibody titers compared to conventional RBD monomers¹⁵. The RBD trimer mRNA vaccine induces higher neutralizing antibody titer than monomers at doses as low as 1 μg^14,67. Multivalent displays of RBD to protein ferritin nanocages elicit robust and durable protective immunity against SARS-CoV-2^13,63. Additionally, the structure-based design of self-assembling protein nanoparticle immunogens stimulates robust B cell responses and neutralizing antibodies targeting multiple epitopes in mice and a nonhuman primate, suggesting that they may not be easily susceptible to escape mutations, which are of higher quality than those obtained from immunization with the prefusion-stabilized S–2P trimer (mRNA-1273 and BNT162b2) and monomeric RBD⁵⁶. However, the disadvantage of protein vaccines is the instability and complex manufacturing process compared to mRNA vaccines68, 69, 70. Our data further showed that the structure-guided design of SARS-CoV-2 RBD mRNA antigens for the heterologous S proteins (sequences) as a tandem repeat single-chain yielded a stable symmetrical rosette-like tetrameric structure and presented a stable dodecameric conformation under the T4 trimerization tag (Fig. 1), enhancing neutralizing antibody titers (Fig. 4 and Fig. S3). The framework of immunogen design can be universally applied to other mRNA vaccines.

mRNA vaccines have focused on antigen structural design and delivery systems optimization to improve its immunogenicity, instability, and translation efficiency, initiating a potent immune response against viral infections⁷¹. Safe, effective, and stable delivery systems are critical in mediating mRNA that protects the nucleic acid from degradation, cellular uptake, and intracellular mRNA release^72,73. Ionizable lipid nanoparticles have the most promising clinical transformation prospects among numerous mRNA delivery systems^26,74,75, and have been investigated and successfully entered the clinic for the delivery of two authorized mRNA vaccines, ionizable SM-102 LNPs-based-mRNA-1273⁷⁶ and ionizable ALC-0315 LNPs-based-BNT162b21⁷⁷. The transfection efficiency and intracellular translation efficiency of mRNA are keys for the successful clinical translation of mRNA vaccines⁷⁸. Ionizable lipids are pH-sensitive and can be protonated to be positively charged or electrically neutral at physiological pH, which improves their biocompatibility and in vivo delivery efficiency79, 80, 81. A variety of ionizable lipids have been explored for mRNA delivery, including DOGS⁸², MVL5⁸³, GL67⁸⁴, DLin-DMA³⁸, MC3^75,85, SM-102^78,86, and ALC-0315³². MC3-based lipid nanoparticles have been shown to be safe and efficient, and have been widely used for mRNA therapeutics87, 88, 89, 90. The biodegradable lipids SM-102 and ALC-0315 have better in vivo delivery efficacy and pharmacokinetics than MC3^86,91. LNPs of MIC1, MIC2, and MIC5 showed higher expression levels than SM-102-LNPs³⁰, indicating that mRNA delivery carrier optimization may facilitate the development of efficient and safe mRNA vaccines. Our data showed that the novel 4N4T lipid nanoparticles were uniform and stable, and had a higher efficiency of mRNA transfection and expression than ALC-0315 (Fig. 3). Importantly, no significant toxicity was observed in 10-μg 4N4T-RBD^dodecamer-immunized BALB/c mice (Fig. 6). Consistent with these observations, we conclude that the novel 4N4T lipid nanoparticle optimizes mRNA transfection and translation efficiency thus increasing mRNA efficacy. Additionally, it has previously been reported that lipid nanoparticles have a self-adjuvant effect that is critical for improving the efficacy of mRNA and protein subunit vaccines by inducing robust T follicular helper cell and humoral responses^38,92. Our results showed that 4N4T-HRBD^dodecamer induces potent specific RBD-binding and neutralizing antibody responses, and provides protection against diverse SARS-CoV-2 and its variants. Mechanistically, 4N4T-HRBD^dodecamer induced Th1-biased CD4 T cell response, which reveals a critical role in controlling SARS-CoV-2 infection in humans (Fig. 5 and Fig. S4).

5. Conclusions

In summary, we pioneered a novel approach to prepare universal mRNA vaccine for halting SARS-CoV-2 and its variant infection with customizable structure-based antigen design and vector-optimized delivery. Our data confirmed that the ionizable lipid nanoparticles-based self-assembling SARS-CoV-2 RBD dodecamer mRNA vaccine is a safe, effective, and universal mRNA vaccine for SARS-CoV-2 and its variants, representing a potential vaccine design strategy to control the SARS-CoV-2 pandemic (Fig. 7). Our work provides the initial framework to continue the development of a universal mRNA vaccine to treat intractable infectious disease.

Author contributions

Shugang Qin, Hai Huang, Wen Xiao, Kepan Chen and Xiangrong Song designed the research. Shugang Qin, Hai Huang, Wen Xiao and Kepan Chen carried out the experiments and performed data analysis. Xi He, Xiaoshan Tang, Zhiying Huang, Yupei Zhang, Xing Duan, Na Fan, Qian Zheng and Min Wu participated part of the experiments. Guangwen Lu, Yuquan Wei and Xiawei Wei provided experimental drugs and quality control. Shugang Qin and Xiangrong Song wrote the manuscript. Yuquan Wei, Xiawei Wei and Xiangrong Song revised the manuscript. All of the authors have read and approved the final manuscript.

Conflicts of interest

The authors have no conflicts of interest to declare.

Appendix A. Supplementary data

The following is the Supplementary data to this article: