Approaches to Improve the Immunogenicity of Plasmid DNA-Based Vaccines against COVID-19

Abstract

Plasmid DNA-based vaccines are attracting considerable interest because of their potential as a platform technology that can be used for a variety of purposes from prevention to therapy. The COVID-19 pandemic has stimulated the development of this platform. The DNA vaccine against COVID-19, developed by Zydus Cadila, was the world’s first DNA vaccine approved for human vaccination. However, the problem of low immunogenicity of DNA vaccines has not yet been completely solved. This article will describe the authors’ experience in creating plasmid DNA-based vaccines against COVID-19, including the design of target antigens, artificial polyepitope T-cell immunogens, delivery of the resulting plasmid constructs using polycationic biodegradable polymers, and producing artificial self-assembled particles incorporating the recombinant protein and DNA vaccine.

1. Introduction

Plasmid DNA vaccines have attracted considerable interest as a platform technology that can be used for a variety of purposes from prevention to therapy. DNA vaccines are advantageous because, like recombinant viral vectors and mRNA vaccines, they induce T-cell and humoral immunity with a good safety profile. The immune response induced by the DNA vaccine is antigen-specific and is not produced to the construction itself, which allows its repeated administration, including prime-boost regimens. DNA vaccines are not pathogenic agents; hence, there is no risk of infection when administered, and numerous clinical studies have shown their high safety [1, 2, 3, 4]. An important advantage is also the stability of DNA in a wide temperature range and, hence, the lack of requirements for cold chain maintaining during storage and transportation of DNA vaccines [5, 6]. Production of plasmid DNA is much faster and easier than the production of live, inactivated, or subunit vaccines; it is much safer, especially compared to vaccines based on live or inactivated pathogens. When new challenges arise, DNA vaccines can be easily adapted to new targets, and they are stable at room temperature, which positively distinguishes them from mRNA vaccines, requiring storage at low temperatures [7].

The COVID-19 pandemic stimulated the development of this platform. The ZyCoV-D DNA vaccine against COVID-19, developed by Cadila Healthcare, was the world’s first DNA vaccine approved for human vaccination [8], showing 66.6% efficacy against symptomatic COVID-19 and 100% efficacy against the development of severe disease in phase III clinical trials [9].

Although DNA vaccines have demonstrated the ability to induce humoral and cellular immunity in small laboratory animal models, intramuscular administration of naked DNA to humans involves increasing the DNA dose to several mg to achieve high efficacy. Low immunogenicity, when administered with naked form, is the main disadvantage of DNA vaccines. This is due to the fact that DNA delivery into cells is difficult because of the large size of the plasmids, their hydrophilic nature, and the overall negative charge of the nucleic acid. In addition, DNA entering the intercellular space can be degraded by nucleases. All this reduces the immunogenicity of plasmids. A large number of studies have focused on developing strategies to improve the immunogenicity of DNA vaccines.

One strategy is the optimization of the coding sequence itself: for example, codon-optimization has become a routine procedure for creating target genes in vaccine plasmids, for which a large number of tools have been created [10, 11]. Another strategy involves incorporating different sequences into the target genes: various regulatory sequences (to ensure high expression of target genes, increase stability of mRNA molecules, and enhance translation [12, 13, 14, 15]), leader sequences (for efficient secretion of a protein product from the cell [10, 16]), signal sequences, such as ubiquitin and LAMP (to enable protein processing in the cell and increase the efficiency of antigen presentation by MHC I and MHC II molecules [17, 18]).

The low immunogenicity of DNA vaccines is largely associated with low transfection activity; therefore, various methods are being actively developed to increase the efficiency of the penetration of injected DNA into cells. This has led to the creation of a wide range of physical and chemical methods of DNA-vaccine delivery [19]. Through the analysis of the publicly available data on clinical studies of DNA vaccines against COVID-19, we can see that most researchers inject plasmids using special devices: electroporators or jet injectors [8, 20, 21, 22, 23, 24]. An interesting trend is the use of a live bacterial vector, such as lactic acid bacteria or Salmonella, which provides not only a strong systemic immune response but also a mucosal one [25, 26]. Being noninvasive, this method is preferred for most patients because it is not associated with pain [27, 28].

Another trend is the development of delivery vehicles based on cationic carriers. The most common approach to packaging DNA vaccines is using of cationic liposomes, which increase the efficiency of cell transfection during intramuscular injection but have serious drawbacks such as toxicity due to the ability to destabilize the cell membrane [29]. Various cationic polymers of natural (chitosan, gamma-polyglutamic acid, and hyaluronic acid) and synthetic origin (polylactideglycolide, polyethylenimine, and others) are alternatives to lipids [30].

This article describes the authors’ original approaches to constructing plasmid DNA vaccines against COVID-19, including the development of target antigens and artificial polyepitope T-cell immunogens, delivery of the resulting plasmid constructs using polycationic biodegradable polymers, and production of artificial self-assembled particles, including the recombinant protein and DNA vaccine.

2. Methodology

The sequence encoding the receptor-binding domain (RBD) of SARS-CoV-2 spike protein published in GenBank (320 V-542 N, MN908947) was used to design the RBD immunogen. The sequence encoding leader peptide MMRTLILAVLLVYFCATVHC was added to the RBD sequence using standard gene engineering methods. Selection of preferred codons, that is, avoiding rare codons with low utilization, and optimizing the secondary RNA structure for efficient expression in mammalian cells have been performed using the GeneOptimizer software [31]. The resulting nucleotide sequence was synthesized and cloned into the pVAX1 vector (DNA-Syntez LLC, Moscow, Russia) under the early promoter of human cytomegalovirus (CMV). After confirming the structure of the construct by Sanger sequencing, it was designated as pVAXrbd.

A strategy based on combining conserved fragments from the main viral proteins S, N, M, and E containing clusters of overlapping epitopes into a single molecule was used to design the T-cell immunogen. T-cell epitope prediction was performed using NetMHCpan-4.1 software and the Immune Epitope Database 2.22, IEDB 2.22 (Tables 1 and 2). To study the DNA vaccine in an animal model, additional mouse MHC I and II restriction epitopes were included in the immunogen sequence. After combining the fragments into one sequence, the universal T-helper epitope PADRE was added to its N-terminus and the marker epitope EPFRDYVDRFYKTLR recognized by 29F2 monoclonal antibodies was added to the C-terminus. The sequence encoding ubiquitin was added to the artificial immunogen sequence, and its whole sequence was optimized taking into account the codons that ensure efficient expression in mammalian cells using the Jcat program [32]. The resulting gene was synthesized and cloned as part of the pVAX1 vector plasmid. The corresponding plasmid was named pBSI-COV-Ub.

MHC II Allele	Peptide	Adjusted Rank
H2-Ied	DDQIGYYRRATRRIR	0.11
H2-Ied	NYNYLYRLFRKSNLK	0.17
H2-Ied	VKPSFYVYSRVKNLN	0.33
H2-Iad	EMIAQYTSALLAGTI	2.80
H2-Ied	ASAFFGMSRIGMEVT	4.35
H2-Ied	QYIKWPWYIWLGFIA	7.05
H2-Ied	CFVLAAVYRINWITG	7.30
H2-IEd	GTWLTYTGAIKLDDK	10.15

Table 1.

Th-epitopes from SARS-CoV-2 proteins restricted by various BALB/c mouse MHC II molecules (the lower the value in the rightmost column, the stronger the binding between the MHC II molecule and the peptide) were predicted using the IEDB 2.22 tool.

The peptides with an adjusted rank value of 10.5 or less are listed in the table.

MHC I Allele	Peptide	Score
H-2-Kd	AYSNNSIAI	0.837637
H-2-Kd	QYIKWPWYI	0.738706
H-2-Kd	YYRRATRRI	0.414253
H-2-Ld	LPPLLTDEM	0.384921
H-2-Dd	SAPHGVVFL	0.370438
H-2-Ld	WPWYIWLGF	0.355595
H-2-Kd	NWITGGIAI	0.347568
H-2-Ld	TPSGTWLTY	0.32614
H-2-Ld	FPQSAPHGV	0.282244
H-2-Dd	FAPSASAFF	0.242852
H-2-Kd	GFIAGLIAI	0.237561
H-2-Kd	QFAPSASAF	0.218262
H-2-Kd	KHIDAYKTF	0.172696
H-2-Ld	FPRGQGVPI	0.160547
H-2-Ld	NSIAIPTNF	0.148747
H-2-Kd	TWLTYTGAI	0.144436
H-2-Ld	SPDDQIGYY	0.126441
H-2-Ld, H-2-Kd	IAIPTNFTI	0.119486
H-2-Ld	QSAPHGVVF	0.159378
H-2-Kd	NFKDQVILL	0.113114
H-2-Ld	YNYKLPDDF	0.101053

Table 2.

CTL-epitopes from SARS-CoV-2 proteins that are restricted by different BALB/c mouse MHC I molecules (the higher the value in the rightmost column, the stronger the binding between the MHC I molecule and the peptide) predicted with the NetMHCpan-4.1 tool.

Peptides with a score value above 0.1 are listed in the table.

To evaluate target gene expression, HEK 293-T cells were transfected with plasmids encoding RBD and BSI-COV-Ub proteins. Gene expression was determined by assessing mRNA level and protein content using RT-PCR and Western blot analysis, respectively.

RBD protein was conjugated to polyglucin-spermidine, which is a polycationic carrier used to deliver DNA vaccines [33, 34, 35, 36]. PGS and PGS-RBD conjugates were used to generate two types of particles (Figures 1 and 2). For the formation of CCV-RBD particles, DNA was mixed with the PGS-RBD conjugate at a ratio of 50 μg of protein for every 100 μg of DNA, incubated for 5 min at room temperature, then PGS was added in 10-fold excess and incubated for another 1 h at 4°C. To form pVAXrbd-PGS particles, DNA was mixed with PGS conjugate in the mass ratio 1:10 and incubated for 1 h at 4°C. The efficiency of complex formation was assessed by changes in electrophoretic mobility of DNA in 1% agarose gel, as well as by gel-filtration on a Sepharose CL-6B column [34]. Samples for gel filtration were applied to the column in equimolar amounts of nucleotide material.

Figure 1.

Schematic representation of the pVAXrbd-PGS particle assembly and interaction between polycationic polymer PGS and DNA.

Figure 2.

Schematic representation of the PGS-RBD conjugation and assembly of the combined CCV-RBD vaccine particle.

To evaluate the immunogenicity of the experimental vaccine constructs, BALB/c mice were twice or thrice immunized intramuscularly or intramuscularly in combination with electroporation. All experiments involving animals were carried out in compliance with the principles of humanity according to the protocols approved by the Bioethical Committee of the SRC VB Vector. Electroporation was performed using a CUY21 EDIT II electroporator and LF650P5 5-mm forceps electrode (BEX CO, Ltd.). The following electroporation protocol was used: rectangular direct and reverse polarity direct current with 3 pulses at 12 V for 30 and 950 ms interval with a current limit of 45 mA [37]. Ten to fourteen days after the second/third immunization, the mice were taken out from the experiment by cervical dislocation, sera were collected from the blood samples, and the spleens were isolated to study the immune response. Preparation of the material for immunological studies (ELISA, VN, ELISpot, ICS) was performed as described previously [34].

ELISA was carried out using recombinant RBD protein sorbed onto 96-well plates in 1 mkg per ml concentration. Viral neutralizing activity of the obtained sera was assessed in an in vitro neutralization assay using live nCoV/Victoria/1/2020 strain of the SARS-CoV-2.

The T-cell immune response was evaluated using IFN-γ-ELISpot technology corresponding manufacturers’ protocol. Splenocytes were stimulated with a pool of peptides from the RBD protein or artificial T-cell immunogen. The ICS procedure was carried out as described previously [34].

The results were statistically processed using nonparametric Mann-Whitney analysis in GraphPad Prism 6.0 program (GraphPad Software) and presented as group mean values, mean with standard deviation (for graphic representation of ELISA, VN, and ELISpot), mean with total variation (for graphic representation of ICS), and medians with 95% confidence interval (for graphic presentation of protectivity). The differences were considered statistically significant at p < 0.05.

3. Results and discussion

3.1 Design of the immunogen

3.1.1 Design of an immunogen capable of inducing a humoral and cellular response

For the design of the first immunogen, the sequence encoding the receptor-binding domain (RBD) of SARS-CoV-2 spike protein published in GenBank (320 V-542 N, MN908947) was used. For SARS-CoV-2, the RBD protein is a good antigen candidate because this relatively small protein is the primary target of neutralizing antibodies [38, 39, 40, 41, 42, 43, 44].

In general, DNA vaccines can induce strong cell-mediated immunity and relatively weak humoral immunity. This is due to the fact that antigen synthesis occurs inside the cell, where it is processed and presented by MHC molecules. To enhance the humoral response, it was decided to increase antigen secretion from the cell. The original signal sequence MMRTLILAVLLVYFCATVHC, a hybrid of leader sequences of two actively secreted proteins: luciferase and fibroin (the leader peptide was developed by Shcherbakov [33]), was introduced to provide efficient transport of RBD protein out of the cell. Obtained construction was confirmed and designated as pVAXrbd.

3.1.2 Design of an artificial polyepitope T-cell immunogen

An effective vaccine against coronavirus must induce a potent T-cell response in addition to neutralizing antibodies. Virus-specific T-cell immunity (against S, N, and M proteins) was detected in people who had contracted the related SARS-CoV more than 10 years after the disease [45]. Using only the RBD as part of the DNA vaccine cannot ensure a long-term effective cellular response. Therefore, the strategy of using also the sequences that are parts of the minor virus proteins as T-cell immunogens seems to be more promising [46]. In addition, these proteins are least susceptible to variability due to less selection pressure.

The peptides listed in Tables 1 and 2 were analyzed for conservativity for different variants of SARS-CoV-2, particularly Wuhan, Gamma, Delta, and Omicron. For each peptide, more than 87% homology to the proteins of the indicated SARS-CoV-2 strains was established.

It is worth noting that the fragments selected for inclusion in the final construct also contained epitopes restricted by a broad spectrum of HLA I (Figure 3) and HLA II, which confirmed immunodominance in the study of SARS-CoV-2-specific cellular immunity in humans and were conserved for different variants of the virus [47, 48, 49]. After combining the fragments into one sequence, the universal T-helper epitope PADRE was added to its N-terminus, and the marker epitope EPFRDYVDRFYKTLR recognized by 29F2 monoclonal antibodies was added to the C-terminus (Figure 4).

Figure 3.

Schematic arrangement of CTL epitopes predicted using the NetMHCpan-4.1 tool from SARS-CoV-2 proteins restricted by different HLA I molecules. Epitopes with a score value greater than 0.3 were taken for the schematic.

Figure 4.

General schematic structure of the polyepitope immunogen BSI-COV-Ub.

In addition, it was decided to attach ubiquitin to the N-terminus of the immunogen. The attachment of ubiquitin to the protein immunogen promotes its targeting to the proteasome, which leads to efficient processing, the release of peptides-epitopes that are presented by MHC I on the surface of the antigen-presenting cell and promote cytotoxic T lymphocyte activation [18, 50, 51], as well as activate T-helper cells as a result of cross-presentation.

The nucleotide sequence corresponding to the designed amino acid sequence was named BSI-COV-Ub (BSI is for Bazhan Sergey Ivanovich, deceased creator of the artificial immunogen) and the corresponding plasmid was named pBSI-COV-Ub.

3.2 The assessment of DNA vaccines’ biological activity

Once the immunogens had been constructed and the plasmids encoding RBD and BSI-COV-Ub had been obtained, their ability to ensure synthesis of the target immunogens in eukaryotic cells needed to be assessed. It was shown that the sizes of the amplified fragments detected using RT-PCR correspond to the theoretically calculated lengths of the target genes: 750 bps for RBD and 1500 bps for BSI-COV.

Western blot assay demonstrated that RBD was detected in both cell lysate and culture medium, indicating a high degree of protein production and secretion provided by the inclusion of a special leader sequence [33]. The polyepitope BSI-COV-Ub protein was detected only in the cell lysate as a set of discrete proteins, indicating the effective processing of T-cell immunogen in the cell, which is ensured, inter alia, by the attachment of ubiquitin to it [52].

3.2.1 Immunogenicity of the DNA vaccines

To evaluate the immunogenicity of the DNA vaccine constructs, BALB/c mice were immunized with pVAXrbd and pBSI-COV-Ub. Fourteen days after a double intramuscular injection of the vaccines (100 μg/dose), blood and spleens were collected from the animals for analysis. According to ELISA results, the average titers of specific antibodies in animals immunized with pVAXrbd were approximately 1:2300, which is pretty well for a naked DNA vaccine. The immunogen encoded by pBSI-COV-Ub did not contain B-cell epitopes, so it induced no virus-specific antibodies as expected.

IFN-γ-ELISpot data showed that the group received 100 μg pVAXrbd had an average number of IFN-γ-secreting T lymphocytes (SFUs) per 10⁶ splenocytes of 238, while the group received 100 μg pBSI-COV-Ub had 122 SFUs. Intracellular cytokine staining (ICS) showed a significant level of IL-2-producing CD4+ and CD8+ T lymphocytes, indicating the formation of a strong virus-specific T-cell response.

Thus, the pVAXrbd DNA vaccine construct, which encodes the RBD gene with a secretory domain, provides induction of the humoral immune response due to the efficient secretion of RBD protein from the cell. Both pVAXrbd and pBSI-COV-Ub DNA vaccines induce a virus-specific T-cell response. It should be noted, in a mouse model, it is difficult to fully assess the potential of the T-cell response of pBSI-COV-Ub because this plasmid encodes epitopes that are restricted by both mouse and human MHC.

The protective efficacy of the designed DNA vaccines was evaluated in the model of mice challenged with the SARS-CoV-2 gamma variant (strain hCoV-19/Russia/SA-17620-080521/2021). Infection was performed 2 weeks after the second immunization with an intranasal dose of 50 ID₅₀. As shown in Figure 5, minimal viral load was detected in mice of the group immunized with pBSI-COV-Ub. The decrease in viral RNA amount compared to the control was 3.45 lg (32.48 Ct vs. 21.00 Ct (p < 0.01)). The lungs of mice immunized with pVAXrbd showed a close value of 3.18 lg (31.53 Ct vs. 21.00 Ct (p < 0.05)) decrease in the amount of viral RNA compared to control.

Figure 5.

Scheme of the experiment to assess the protective efficacy of DNA vaccines. Viral load in lung tissues of BALB/c mice on day 4 after challenge with SARS-CoV-2 gamma variant. Individual Ct values (threshold virus detection cycle) in PCR-RT are represented by dots; medians in the groups by histogram vertices, with numerical Ct values indicated at the base of the histograms; 95% confidence interval limits are indicated by vertical lines. We express our deep gratitude to Pyankov O.V. and his team for the challenge performing.

Thus, both immunogens in DNA format have the ability to induce protective immunity in mice against gamma variant of the SARS-CoV-2, which confirms the importance of further work with these vaccines.

3.3 DNA-vaccine delivery

3.3.1 Delivery of the DNA vaccine using polyglucin-spermidine conjugate

The use of cationic biopolymers for plasmid delivery is one of the techniques that is used to increase the efficiency of DNA vaccines [30]. The most widely used strategy is the association of DNA and the biopolymer through electrostatic interaction between the negatively charged DNA phosphate groups and the positive charges of the carrier material.

The original cationic polymer used in this work was a polyglucin-spermidine conjugate (PGS) [35]. It is important that the components of the resulting polymer are biodegradable and safe for humans, animals, and the environment; polyglucin is a glucose polymer with a molecular weight of 40,000. It is a licensed plasma substitute and is nontoxic. Spermidine is a polyamine that is found in the cells of almost all living organisms and is involved in many biological processes. The low cost, safety, and ability to lyophilize with long-term storage at 4°C provide additional technological advantages in the production and transportation of vaccine preparations containing PGS [53].

The polyglucin-spermidine conjugate can be used both for the delivery of proteins, including the recombinant RBD protein [54] and nucleic acids. Previously, polyglucin-spermidine has proven to be an effective delivery vehicle for experimental DNA vaccines against Ebola and HIV-1 [35, 36, 55]. As a component of anti-HIV vaccine, PGS passed the first phase of clinical trials, demonstrating a high safety level [56].

The optimum ratio of the components for the formation of DNA-polymer complexes was selected based on the analysis of the complexation degree obtained earlier [35]. The ratio of DNA and PGS weights to form complexes was 1:10 (Figure 1). The particles’ assembling was assessed by a shift in electrophoretic mobility in agarose gel: the encapsulated plasmids lost their mobility in the electric field (Figure 6A). Electron microscopy was used to determine the size of the particles, whose values were in the range of 50–1000 nm. It is shown that such size of the particles is optimal for vaccine creation since they accumulate in B-cell follicles and induce a strong immune response [57].

Figure 6.

Assessment of the CCV-RBD particle formation. (A) DNA encapsulation in the PGS and PGS-RBD shells by electrophoresis in a 1% agarose gel: 1, CCV-RBD; 2, pVAXrbd-PGS; and 3, naked plasmid pVAXrbd. (B) Electron micrograph of CCV-RBD particles. We express our gratitude to Zaitsev B.N. For electron microscopy results.

After three-time immunizations on days 0, 14, and 28, blood was collected from the animals for analysis 6 weeks after the start of the experiment. The sera were examined for the presence of RBD-specific antibodies by ELISA and in the SARS-CoV-2 neutralize reaction. The sera of unimmunized mice were used as a control. According to ELISA results, the mean titers of specific antibodies in sera of animals immunized with pVAXrbd-PGS were approximately 40-fold higher compared with animals immunized with pVAXrbd (p < 0.01) and 10,000-fold higher than titers in controls (p < 0.01) (Figure 7A).

Figure 7.

Immune response induced by the pVAXrbd-PGS and pVAXrbd constructs in mice. (A) RBD-specific IgG antibody titers were assessed in ELISA. Reciprocal titer values are shown in the plot. (B) Virus neutralization activity of sera was assessed using the SARS-CoV-2 (nCoV/Victoria/1/2020 strain) neutralization reaction. Reciprocal titer values are provided in the plot. (C) T-cell response was assessed by the ELISpot method. The number of spot-forming units is shown in the plot. We express our gratitude to Pyankov O.V. and his team for the virus neutralization performing.

The neutralizing ability of the obtained sera was evaluated in an in vitro neutralization assay using live SARS-CoV-2. Sera from mice immunized with pVAXrbd-PGS neutralized the nCoV/Victoria/1/2020 SARS-CoV-2 strain at a dilution of 1:200, whereas sera from the group immunized with pVAXrbd neutralized it at a dilution of 1: 12 (p < 0.01). Sera from control animals showed no neutralizing activity (Figure 7B).

The T-cell immune response was assessed by the number of splenocytes producing IFN-γ using ELISpot technology. Splenocytes were stimulated with a pool of peptides from RBD protein. It was shown that splenocytes isolated from the spleens of mice immunized with pVAXrbd in the PGS envelope or none responded more strongly to the stimulation with viral peptides with IFN-γ release than the control group (p < 0.01) but without statistically significant differences between each other (Figure 7C).

Thus, the use of polyglucin-spermidine as a delivery vehicle enhances the immunogenicity of the DNA vaccine, including increased levels of neutralizing antibodies and T-cell response [33].

3.3.2 Combined DNA/protein particles

DNA vaccines are considered to be most effective in inducing a T-cell response and less effective in inducing a humoral response; therefore, it was decided to further enhance the humoral component by incorporating the protein immunogen-RBD protein into the design [56].

For this purpose, we decided to use polyglucin-spermidine-protein RBD conjugate (PGS-RBD) for DNA vaccine packaging. The synthesis technique was proposed by Lebedev et al. [54, 55, 58, 59]. We previously used this approach to increase the immunogenicity of an experimental vaccine based only on the recombinant RBD protein [54]. We attached the protein to a polyglucine-spermidine conjugate and mixed it with dsRNA (as adjuvant). As a result, complexes presenting RBD on the surface and dsRNA inside were obtained due to self-assembly of conjugate and RNA. Such complexes induced virus neutralizing antibody level against SARS-COV-2 much more effectively than RBD with AL(OH)3 or RBD without adjuvant [54].

In this work, we used the PGS-RBD conjugate to deliver the pVAXrbd DNA vaccine. CCV-RBD particles (CCV-RBD stands for CombiCoronaVac-RBD) were obtained by mixing PGS-RBD with plasmid pVAXrbd as described earlier [34]. The assembling of the complexes was confirmed by a loss of the encapsulated plasmids’ mobility in the electric field (Figure 6A). Preservation of plasmid DNA structure in CCV-RBD and pVAXrbd-PGS complexes was conserved by UV spectroscopy; the spectra of these preparations had a characteristic DNA peak at 260 nm. Using electron microscopy, the particle sizes appeared to be in the range of 50–1000 nm (Figure 6B) are comparable to the particles formed by pVAXrbd and PGS without protein [34].

To assess changes in the binding kinetics of RBD after including in CCV-RBD particles with the known mAb iB14, a study was carried out using biolayer interferometry performed on the Octet K2 device (Pall Fortebio, USA). The dissociation constants of RBD and CCV-RBD particles with iB14 showed close values, suggesting protein exposure on the particle surface (Figure 8) and its spatial multimerization, which can potentially significantly enhance the RBD-specific humoral immune response.

Figure 8.

(A1) biolayer interferometry assay of the RBD to SARS-CoV-2 nAb binding. (A2) biolayer interferometry assay of the CCV-RBD to SARS-CoV-2 nAb binding. (B1, B2) schematic representation of the interaction between RBD protein or CCV-RBD particle and antibody. We express our gratitude to Taranin A.V. and Baranov K.O. For the biolayer interferometry performing.

Female BALB/c mice weighing 16–18 g were used to evaluate the immunogenicity of the construct. All experiments with animals were performed in compliance with the principles of humanity according to the protocols approved by the Bioethics Committee of Vector State Research Center. Mice were divided into groups of eight animals each and immunized as follows: CCV-RBD group - a combined vaccine containing 100 μg of DNA and 50 μg of protein; pVAXrbd group - 100 μg of “naked” DNA vaccine; RBD-PGS group - 50 μg of RBD protein conjugated with polyglucin-spermidine. The intact group contained unimmunized animals. Mice were immunized intramuscularly twice at three-week intervals into the hind thigh. Ten days after the second immunization, blood and spleens were drawn from the animals to examine the virus-specific immune response.

According to the results of the RBD-specific ELISA at the endpoint of the experiment, it can be concluded that the use of the protein as a humoral component significantly increases the average titers of specific antibodies in the animals. Thus, in the group immunized with “naked” pVAXrbd DNA, the titer of specific IgG was 1:2617, in the group that received RBD protein was 1:22280, whereas in the group that received the combined preparation, this figure reached the value 1: 369900 (Figure 9A). Thus, the combination of DNA and protein resulted in a synergistic enhancement of the specific humoral immune response compared with the administration of the individual components: 17-fold relative to protein alone, 140-fold relative to “naked” DNA (p < 0.01).

Figure 9.

Design of the experiment and analysis of immunogenicity and protectivity. (A) RBD-specific IgG titers were determined by ELISA. (B) Viral neutralization of murine sera was determined using SARS-CoV-2 nCoV/Victoria/1/2020 strain (100 TCID₅₀). (C) T-cell response: Number of splenocytes producing IFN-γ in response to specific stimulation, per 10⁶ cells determined using ELISpot. (D) Viral load in lung tissues of BALB/c mice on day 4 after infection with gamma variant SARS-CoV-2.

Sera from all groups except intact mice and mice immunized with “naked” pVAXrbd demonstrated the ability to neutralize the nCoV/Victoria/1/2020 strain on cell culture in vitro. However, in the CCV-RBD group, the average neutralizing titer was 1:376, which was significantly higher than the titers shown in the RBD group (1:21) (p < 0.05) (Figure 9B).

The synergistic effect with respect to the humoral response may Indicate the effective capture of CCV virus-like particles by APCs, as well as the stimulation of the T-helper response due to the DNA component. Multiple representations of RBD protein on the particle surface may also play a role in enhancing the humoral immune response to CCV-RBD.

Thus, it has been shown that administration of the DNA/protein vaccine results in a stronger humoral response compared to administration of DNA alone or protein alone.

The number of splenocytes producing IFN-γ in response to specific stimulation by the peptide pool of RBD protein was assessed using the ELISpot method. Splenocytes were isolated from the spleens of the animals participating in the experiment. It was shown that a high rate of cellular immunity was recorded in the group immunized with the DNA vaccine in the absence of protein (Figure 9C). The addition of the protein component to the vaccine resulted in the formation of a lower cellular response, which may be due to the increased uptake of the combined particles by APC and their conduction through the lysosomal cleavage pathway with a low percentage of escaping DNA and its subsequent penetration into the cell nucleus.

The pVAXrbd DNA vaccine and the combined CCV-RBD vaccine were chosen for protection studies using the SARS-CoV-2 strain hCoV-19/Russia/SA-17620-080521/2021, belonging to the P.1 line (gamma variant). Protein conjugate of RBD and polyglucin-spermidine was used as a protein control. Infection was performed 2 weeks after the second immunization, and the intranasal dose was 50 ID₅₀. As shown in Figure 9D, minimal viral load was detected in mice immunized with the combined CCV-RBD vaccine. The decrease in viral RNA amount compared to the control was 3.43 lg (32.41 Ct vs. 21.00 Ct (p < 0.01)). Decrease in the amount of viral RNA in lungs of mice immunized with the pVAXrbd DNA vaccine and RBD protein was 3.18 and 2.79 lg, respectively, compared with control (31.53 and 30.30 Ct vs. 21.00 Ct (p < 0.05)).

Thus, all experimental groups showed a tendency to reduce the viral load, but the vaccination with the combined DNA/protein vaccine provided good protection to the lung tissues.

4. Conclusion

DNA vaccines are a promising vaccine platform that has been rapidly developing in recent years. Five vaccines are currently approved in the U.S. veterinary medicine [60, 61, 62, 63, 64]. DNA vaccines against a number of viral infections, cancers, and autoimmune diseases are in clinical trials nowadays [65].

The rapid worldwide spread of SARS-CoV-2 has led to the unprecedented development of various vaccine creation platforms, such as nucleic acid-based and vector vaccines, and to the approval of the first human DNA vaccine, ZyCoV-D. However, problems including the low immunogenicity of DNA vaccines still need to be solved.

In our work, we proposed a number of strategies to improve the immunogenicity of DNA vaccines against COVID-19: one encoding the receptor-binding domain of protein S of the SARS-CoV-2 and the other encoding an artificial polyepitope T-cell immunogen.

The attachment of the secretory domain to the RBD (in pVAXrbd) ensured a high level of secretion of the protein product from the cell and, consequently, a level of humoral immunity. Addition of ubiquitin to the artificial polyepitope construct (in pBSI-COV-Ub) provided efficient processing of the product inside the cell and its presentation to MHC I.

The polyglucine-spermidine conjugate was used previously to enhance the immunogenicity of the RBD protein [54], and here, we demonstrate the efficacy of using a PGS conjugate to deliver the pVAX-rbd DNA vaccine.

The use of the cationic polymer polyglucin-spermidine for pVAXrbd DNA-vaccine delivery resulted in an increase in its immunogenicity, especially with respect to the humoral response. The PGS envelope protects the DNA from degradation by nucleases, and the particles formed by the interaction of DNA and PGS are more efficiently taken up by antigen-presenting cells. Nanoparticles composed of such polymers show good biocompatibility, water solubility, biodegradability, and absence of toxicity, and they are also easy and inexpensive to produce and stable over a wide temperature range.

Particles obtained using a polyglucin-spermidine-RBD conjugate containing pVAXrbd inside and RBD protein on the surface demonstrated high immunogenicity. Attachment of RBD protein to the polycationic envelope resulted in synergistic enhancement of the humoral immune response compared to its individual components, as well as in the formation of protective immunity in mice, allowing a significant reduction in the viral load in the lungs of animals after virus challenge.

We hope that the approaches described above will be useful to researchers engaged in the design of DNA vaccines.

Acknowledgments

The study was supported by the Ministry of Science and Higher Education of the Russian Federation (agreement # 075-15-2019-1665).

We would like to sincerely thank the entire team of the Bioengineering Department and the team of the Microorganism Collection Department of the SRC VB “Vector” for their well-coordinated work and the most competent performance of the task set.