In a study published in Cell, researchers investigated the effectiveness of a circular RNA (circRNA) vaccine against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and its Delta and Omicron variants, which elicited significant antibody and T-cell responses in mice and rhesus macaques. The circRNA vaccine elicited a higher immune response than the linear 1mΨ-modified mRNA vaccine.
Since its emergence in 2019, coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2, has led to 493 million infection cases and 6.15 million deaths.
COVID-19 infections haven’t just been caused by the wild-type strain but also other SARS-CoV-2 variants. The Omicron variant is currently the most dominant variant, accounting for about 85% of cases. The neutralizing antibodies against the original strain are less effective towards the Omicron variant, which displays 30 spike mutations, of which 15 are present on the receptor-binding domain (RBD).
As vaccination is the most promising approach to tackle the infection, it is critical to develop vaccines that would be effective against the fast-emerging SARS-CoV-2 variants.
About the study
In the current study, plasmids containing either SARS-CoV-2 RBD antigen, EGFP, nanobody, or hACE2- 899 decoy-coding sequence were constructed to obtain pcircRNA. The circRNAs were produced by the in vitro synthesis (IVT) reaction from linearized circRNA plasmids. The RNA products were digested with DNase I and column purified. Two methods were used to produce circRNAs; group I intron strategy and T4 RNA ligase method. The circRNAs were purified by high-performance liquid chromatography (HPLC).
Linearized mRNAs were produced by using plasmids containing the 5’-UTR, RBD-coding region, 3’-UTR, and -81 nucleotide (nt) polyA elements. These mRNAs were modified with 1-Methylpseudouridine-5 Triphosphate to obtain the 1mΨ-modified mRNA. RNase H and RNase R cleavage assays were performed.
The synthesized circRNAs were transfected in the human embryonic kidney (HEK293T) or mouse NIH3T3 cells for determining the secretion of RBD antigens. Further, the circular RNAs encoding SARS-CoV-2 RBD antigens (circRNARBD) were encapsulated into the lipid nanoparticles (LNPs) to obtain LNP-circRNARBD, which were characterized for loading concentration and size. The effect of storage and temperature on the antigenic activity were also determined. The expression level of SARS-CoV-2 spike RBD protein in vitro was determined by performing ELISA.
For in vivo experiments, BALB/c female mice of six-eight-week-old were vaccinated with LNP-circRNARBD or placebo, i.e., only LNPs with two doses. Subsequently, sera were collected, and IgG endpoint geometric mean titers (GMTs) and neutralizing antibodies were determined by ELISA and neutralization assays, respectively.
The 50 % neutralization titer (NT50) of the immunized mouse sera was determined by SARS-CoV-2 pseudovirus-based neutralization assay in HEK293T-hACE2 and Huh7 cells. Further, the neutralization assay of circRNAnAB or circRNAACE2 decoys was also performed. To determine the authentic SARS-CoV-2 NT50 assay, A549-hACE2 cells were added with the virus/mouse sera, and RT-qPCR was performed to quantify viral load.
In another experiment, mice were immunized with the SARS-CoV-2 Beta variant circRNARBD-Beta, followed by a challenge with the SARS-CoV-2 Beta virus. After three days, the mice were sacrificed, and viral RNA load was determined by RT-qPCR. Moreover, T-cell flow cytometric analysis was also done. Similar experiments were done using circRNARBD-Delta, circRNARBD-Omicron vaccines.
In addition to mice experiments, the researchers also performed vaccination experiments on rhesus macaques. Briefly, a group of 2~4-year-old male rhesus macaques was intramuscularly immunized with different concentrations of LNP-circRNARBD, 100 ug of LNP-circRNACtrl twice, three weeks apart. The plasma was collected at prime and post-boost. T-cell immune responses such as IFN-γ, IL-2, and IL-4 were quantified by ELISpot assay.
The rhesus macaques were further challenged with the native SARS-CoV-2 virus, intranasally and intratracheally. The plasma (cytokine analysis) and major organs (lungs) were collected for analysis.
The plasmids circRNARBD were produced successfully. Immunogenicity was enhanced by fusing peptide sequence of human tissue plasminogen activator (tPA) with N-terminus, while trimerization motif of bacteriophage T4 fibritin protein (Foldon) was fused with C-terminus of RBD. The circRNA was 90 % pure and circular as confirmed by HPLC and reverse transcription-PCR, Sanger sequencing assays, respectively.
The concentration of RBD antigens produced by circRNARBD (produced by the group I intron strategy) showed a ~1,400 ng/ml (600-fold) increase as compared to that produced by its linear precursor RNA. Moreover, with the T4 RNA ligase-based method, about 200-fold higher RBD antigens were produced. The secreted RBD antigens were effective in blocking the SARS-CoV-2 pseudovirus infection.
LNPs encapsulated circRNARBD at an efficiency of 93 % to produce LNP-circRNARBD of size 100 nm. LNPs were found to be an excellent delivery platform.
In the mice immunization experiment, the circRNARBD stimulated higher RBD-specific IgG GMTs with both 10 μg (~1.9×104) and 50 μg (~5.7×105). The sera of immunized mice effectively neutralized the pseudovirus (~4.5×103) and actual SARS-CoV-2 (~7.0×104) virus, with the highest in the native D614G strain. Moreover, the same NT50 results were found with the authentic native (6.0×103) and Beta strain (2.6×104).
The lower viral load in the lung tissues also confirmed the protection of mice immunized with the circRNARBD-Beta vaccine against the Beta infection, indicating its protective activity. The antibodies elicited by circRNARBD-Delta showed effective neutralization of Delta pseudovirus with an NT50 of ~1.4×105 (10 μg dose). The circRNARBD-Delta protected other SARS-CoV-2 variants such as Alpha and Beta, with higher neutralization activity with Delta.
The circRNARBD-Omicron vaccine-induced Omicron spike-specific antibodies with the endpoint GMTs of ~4.7×104 (5 μg dose) and ~2.2×105 (10 μg dose). However, activity was not effective against other variants.
Importantly, the circRNAs produced high levels of antigens and were more durable and stable than the 1mΨ-mRNA and unmodified mRNA. The researchers also found that the LNPs helped increase the protein production and durability of both modified and unmodified linear mRNA. The circRNAs were stable for two weeks at room temperature without losing the antigen production levels, whereas at higher temperatures (37 ° C), all showed reduced activity.
Further, the circRNAs were able to induce immune responses. The total IgG elicited by circRNARBD-Delta was comparable to that produced by 1mΨ-mRNARBD-Delta. However, the ratios of IgG2a/IgG1, IgG2c/IgG1, or (IgG2a + IgG2c)/IgG1 elicited by circRNARBD-Delta were higher than those from 1mΨ-mRNARBD-Delta vaccine. Furthermore, the group also investigated the levels of neutralizing and binding antibodies, wherein the circRNAs showed higher proportions of neutralizing antibodies at 2.5 μg and 10 μg doses as compared to the 1mΨ-mRNA vaccine. This indicated that the circRNA vaccine can evade antibody-dependent enhancement (ADE) produced by the viruses.
Both circRNARBD-Delta, and 1mΨ-mRNARBD-Delta, also elicited the production of IFN-γ, TNF-α, and IL-2 produced by CD8+ T cells, with lower IL-4 responses, suggesting that the circRNA vaccine-induced Th1-biased T cell responses.
Moreover, the circRNARBD-Delta vaccine was found effective against the Omicron variant with NT50 of ~4.7×103, while the NT50 of the circRNARBD-Beta against Omicron reduced to 5×102. However, the neutralizing antibodies produced by circRNARBD-Delta against both the variants were comparable to 1mΨ-mRNARBD-Delta.
Interestingly, when the mice that received two doses of circRNARBD-Delta were immunized with the third booster dose of either circRNARBD-Delta, circRNARBD-Beta, circRNARBD-Omicron, only the circRNARBD-Delta elicited neutralizing antibodies against Delta and Omicron, indicating its potential against all the variants of concern.
In rhesus macaques, the results showed that circRNA produced neutralizing antibodies and Th1-biased T cell responses (FN-γ and IL-2, undetectable IL-4). The IgG endpoint GMTs reached ~2.1×104 (20 μg), ~1.6×104 (100 μg dose), and ~7×103 (500 μg dose) for circRNARBD, whereas, with the PBS and circRNACtrl, the RBD specific antibodies were not induced. Moreover, the circRNARBD also inhibited the native strains and reduced activity towards other variants.
The circRNA protected rhesus macaques from the infection with native SARS-CoV-2 confirmed by the RT-qPCR and histopathology, with no severe clinical adverse effects. This indicated the safety of the circRNA vaccine in non-human primates. In a further next step, the researchers also investigated the therapeutic potential of the circRNA vaccine. The circRNA based nanobodies (circRNAnAB) and hACE2 (circRNAhACE2) decoys prevented SARS-CoV-2 pseudovirus infection. circRNAnAB1-Tri and circRNAnAB3-Tri also prevented Alpha and Beta variants, whereas the circRNAhACE2 failed in prevention.
In conclusion, the researchers suggested the effective role of circular RNA-based vaccines against SARS-CoV-2 and their variants. The circRNA vaccine was effective in eliciting neutralizing antibodies towards respective strains in both mice and rhesus macaque models. Moreover, their therapeutic potential was also explored.
Despite such a large study, the researchers pointed out a few limitations concerning the use of a smaller number of rhesus macaques, with more safety investigations required. A detailed comparison between circRNA and linear mRNA is also warranted.