Developing an Effective Peptide-Based Vaccine for COVID-19: Preliminary Studies in Mice Models

Coronavirus disease 2019 (COVID-19) has caused massive health and economic disasters worldwide. Although several vaccines have effectively slowed the spread of the virus, their long-term protection and effectiveness against viral variants are still uncertain. To address these potential shortcomings, this study proposes a peptide-based vaccine to prevent COVID-19. A total of 15 B cell epitopes of the wild-type severe acute respiratory coronavirus 2 (SARS-CoV-2) spike (S) protein were selected, and their HLA affinities predicted in silico. Peptides were divided into two groups and tested in C57BL/6 mice with either QS21 or Al(OH)3 as the adjuvant. Our results demonstrated that the peptide-based vaccine stimulated high and durable antibody responses in mice, with the T and B cell responses differing based on the type of adjuvant employed. Using epitope mapping, we showed that our peptide-based vaccine produced antibody patterns similar to those in COVID-19 convalescent individuals. Moreover, plasma from vaccinated mice and recovered COVID-19 humans had the same neutralizing activity when tested with a pseudo particle assay. Our data indicate that this adjuvant peptide-based vaccine can generate sustainable and effective B and T cell responses. Thus, we believe that our peptide-based vaccine can be a safe and effective vaccine against COVID-19, particularly because of the flexibility of including new peptides to prevent emerging SARS-CoV-2 variants and avoiding unwanted autoimmune responses.


Introduction
Coronavirus disease 2019  is caused by the severe acute respiratory coronavirus 2 (SARS-CoV-2), which is a single-strand positive RNA virus that enables the host cell to directly synthesize proteins upon viral entry and uncoating [1]. This enables viral replication and the rate of viral spread to occur much more rapidly than other types of viruses [2]. Given this alarming rate of transmission, many countries were unable to implement an effective and timely approach to curb the spread of the virus, and the worldwide economic impacts of COVID- 19 have been historically devastating [3].
Other viral outbreaks, such as SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV), have led to the development of several treatment strategies [4][5][6]. Drugs While DNA vaccines can elicit a good T cell response, the B cell response is relatively low compared to protein-based vaccines [53][54][55][56]. Alternatively, mRNA-based vaccines are a novel approach for infectious diseases and are currently the dominant approach in fighting COVID-19 [57]. However, it remains unclear as to how effective they are in stimulating T cell and B cell responses in comparison to a more traditional protein-based vaccine or to natural infection.
Here, we developed a peptide-based vaccine for COVID-19 prevention. There are four primary advantages of peptide-based vaccines: (1) large-scale peptide synthesis is relatively inexpensive and the technology is well-established, (2) peptides can fold into three-dimensional epitopes capable of inducing antibody responses to linear and conformational structures, (3) unique epitopes can be selected to avoid autoimmune responses that could be generated by the whole protein, and (4) new epitopes can be easily added to the peptide mixture as new viral mutations are identified [58][59][60].

Antigen and Cell Selection
The SARS-CoV-2 spike (S) protein of the wild-type "Wuhan" strain (Accession: YP_009724390.1) was submitted to DNAStar (Madison, WI, USA) for in silico antigen (peptide) identification. Nine epitopes were selected from the S protein's subunit 1 (S1) domain, and six sequences were selected from the subunit 2 (S2) domain. The selected peptide sequences were submitted to GenBank for homology analysis, which confirmed that the peptides were unique. Peptide antigenicity was confirmed by DNAStar.
Human Leukocyte Antigen-I (HLA-I), Mouse Major Histocompatibility Complex-I (MHC-I), and HLA-II affinity analyses were conducted with the T Cell Epitope Prediction Tools, using the Immune Epitope Database (IEBD) Analysis Resource at http://tools.iedb. org/main/tcell/, accessed on 21 July 2021. The specificity of the peptides to SARS-CoV-2 were confirmed using the protein BLAST module at https://blast.ncbi.nlm.nih.gov/Blast. cgi, accessed on 21 July 2021. All peptides were provided by Biomer Technology Inc. (Pleasanton, CA, USA).
For SARS-CoV-2 pseudovirus production, 4 × 10 6 293T cells were seeded in a 100 mm cell culture dish. After 48 h, 10 µg NL4-3 mCherry Luciferase plasmid and 10 µg SARS-CoV-2 spike plasmid were mixed in 1.5 mL Opti-MEM medium and incubated at RT for 5 min. Then, 65 µL of Lipo3000 was mixed in 1.5 mL Opti-MEM medium at RT for 5 min, and the plasmid mixture was then mixed with the Lipo3000 mixture and incubated at RT for 20 min. The mixture was added dropwise into a dish and incubated in the cell culture incubator. After 48 h, cells were centrifuged at 500× g for 10 min, cell supernatant was collected, and 9 mL of supernatant was transferred into a separate tube with a 3 mL Lenti-X concentrator. The supernatant was stored at 4 • C for 4 h and centrifuged at 1500× g for 45 min. The supernatant was aspirated, and the precipitate was resuspended with 90 µL cold DPBS and then frozen at −80 • C.

Mice
C57BL/6 mice were used for this study. The mice were 6 months old, gender balanced in each group, and housed at the animal facility at the University of South Florida (USF) Health Center. All animal experiments were approved (IS00008370) by the Institutional Animal Care and Use Committee (IACUC) at USF and performed according to the National Institute of Health (NIH) guidelines.

Vaccine Preparation
The peptides were split into two groups by S protein subdomain; each group contained peptides with high affinity to MHC (HLA) class I and class II to ensure that the peptide mixture would cover most MHC typing. HLA-I was predicted against all available allele types, and the top 1% was considered high affinity. HLA-II was also tested against all allele types in the database, and the selected standard was set at 10% rank as recommended.
A total of 15 peptides were identified against the S protein. Each peptide was dissolved with DMSO at 50 mg/mL, grouped into two groups, and mixed at 50 µg each. The two most commonly utilized adjuvants-QS21 and Al(OH) 3 (0.4% Alhydrogel from InvivoGen (Cat no# vac-alu-50, San Diego, CA, USA)-were selected for the vaccine formula. QS21 was purchased from Desert King International (San Diego, CA, USA), and Al(OH) 3 was purchased from Invitrogen (Waltham, MA, USA). Common adjuvants include Al(OH) 3 , which can promote earlier antibody production, and QS21, which promotes the release of Th1 cytokines participating in the elimination of intracellular pathogens [61][62][63]. The concentration of adjuvant is 25 µg per dose.

Vaccination and Sample Collection
Two total studies were conducted in this report, with peptide mixture groups 1 and 2 using different groups of mice. Group 1 mice (C57BL/6) were 6 months old and injected with A1-G1 (8 peptides); group 2 mice were 8 months old and received H1-E2 (7 peptides) at equal mixture amounts (50 µg/peptide) with either QS21 or Al(OH) 3 as the adjuvant via bi-weekly subcutaneous injections for 6 injections.
Blood samples were obtained with EDTA vacutainer tubes 7 days after each injection, starting from the 2nd injection. Additionally, 50 µL of whole blood was taken for flow cytometry analysis and the rest of the blood was centrifuged at 300× g for 5 min. Plasma was then transferred into a new tube and stored at −80 • C for future application. The immunization and blood drawing schedule was conducted as shown in Scheme 1.
An indirect ELISA was conducted to measure antibody response. First, 96-well plates were coated with 50 µL of peptides from 10 µg/mL S1 and S2 and 1 µg/mL recombinant protein. The plates were washed 4 times with PBST, 180 µL/well of 0.05% BSA buffer was added, and the plates were incubated at 37 • C for 1 h. The plates were then washed with 1× wash buffer, and diluted plasma was added to each well at 1:200, 1:800, and 1:3200-fold dilutions and incubated for 1 h at room temperature (RT). Anti-human or anti-mouse HRP-conjugated antibodies were then added at 1:10,000 dilutions, incubated for 1 h at RT, and washed with 1× wash buffer. Then, 100 µL of TMB substrate was added, followed by 0.4 M sulfuric acid to stop the reaction. Plates were read at OD450, with OD650 as the reference.
For epitope mapping of mouse and human plasma, plasma was added to the antigencoated 96-well ELISA plate for indirect ELISA as described above at 1:200, 1:800, and 1:3200-fold dilutions and incubated for 1 h at RT.

T Cell Detection
T cell population detection post vaccination was assessed using flow cytometry assay and 50 µL whole EDTA blood. Red blood cells were lysed with ACK buffer and washed twice with 1% (v/v) FBS in 1× PBS. The fluorophore-conjugated antibody mixture was added to the sample and incubated on ice for 20 min. FITC anti-mouse CD3ε Antibody (Cat#10036), APC/Cyanine7 anti-mouse CD8a Antibody (Cat#100714), APC anti-mouse CD4 Antibody (Cat#100412), and PE anti-mouse CD19 Antibody (Cat#115508) were bought from Biolegend (San Diego, CA, USA). Cells were then resuspended with 1% FBS in 1× PBS after washing 3 times. All samples were run on the flow cytometer.

Neutralization Assay
ACE2-293T-GFP cells were seeded into a poly-lysine-coated 96-well plate at a density of 1.2 × 10 4 cells/well. After 24 h, 6 µL of SARS-CoV-2 pseudovirus and diluted human plasma or mouse samples were added at the designated dilutions, and the medium was changed after 6 h of incubation. The plates were incubated in a CO 2 incubator for 48 h, the medium was removed, and the cells were washed with Hank's balanced salt solution (HBSS). HBSS was aspirated and 20 µL of Promega lysis buffer was added at RT for 5 min. Luminescence was measured after adding 50 µL of Luciferase substrate into 20 µL of lysis buffer.
First, 100 µL of assay buffer was added to each well of the assay plate and mixed on a plate shaker for 10 min at RT. The assay buffer was then discarded, and the remaining residue was removed from the wells by inverting the plate and tapping it numerous times on an absorbent towel. Then, 50 µL of the assay buffer was added to the background wells, and 50 µL of the standard, control, and diluted samples were added to the appropriate wells. Afterwards, the MILLIPLEX MAP Anti-Mouse Multi-Immunoglobulin beads were vortexed at medium speed for 15 s and sonicated for 15 s using a sonication bath. Then, 25 µL of the bead solution was added into each well, and the plate was covered and incubated on a plate shaker for 15 min at RT. The plate was then washed twice with 200 µL/well of washing buffer. After washing, 25 µL/well of prepared Anti-Mouse κ Light Chain PE was added, covered, and incubated on a plate shaker at RT for 15 min. Finally, the fluid was aspirated, and beads were resuspended in 150 µL/well of sheath fluid. The plate was then read on the Bio-Rad Bio-Plex MAGPIX Multiplex Reader.

Cytokine Detection with Multiplex Assay
First, 200 µL of washing buffer was added to each well, and the plate was sealed and incubated on the plate shaker for 10 min at RT. The wash buffer was then aspirated, and 25 µL of each standard or control was introduced into the appropriate wells, with the assay buffer utilized for the 0 pg/mL standard (background). Then, 25 µL of the assay buffer was added to the sample wells, and 25 µL of the serum matrix solution was added to the background, standard, and control wells. Afterwards, 25 µL of the diluted samples were introduced into the appropriate wells. Then, the mixing bottle was vortexed, and 25 µL of the mixed or premixed beads were added to each well and shaken for 10 min at RT. The plate was then sealed with a plate sealer, wrapped with aluminum foil, and incubated on a plate shaker overnight at 2-8 • C. The following day, well contents were gently removed, and the plate was washed twice. Then, 25 µL of the detection antibodies were added into each well and incubated on a plate shaker for 1 h at RT. The well contents were removed, the plates were washed twice, and 150 µL of sheath fluid was added to all wells. The beads were then resuspended on a plate shaker for 5 min, and the plate was read on the Bio-Rad Bio-Plex MAGPIX Multiplex Reader.

Selection of SARS-CoV-2 Peptides
Sixteen peptides were predicted to have high antigenicity with 100% homology to the SARS-CoV-2 virus were split across two peptide mixtures (Group 1, Group 2) that covered all Class I and II allele types of C57B/6 mice. The peptide sequences and T cell epitope results are listed in Table 1 and Supplementary Table S1, respectively.

Antibody Response Post Vaccination Using Group 1 Peptides with Different Adjuvants
First, 50 ug/peptide of peptides A-G of the S1 protein were mixed with either the QS21 or Al(OH) 3 adjuvant and injected into mice subcutaneously at two-week intervals. Blood was collected 10 days after each injection and used for antibody detection. It is evident that, after two injections, both groups could detect an antibody response, but the Al(OH) 3 group exhibited a higher antibody response than the QS21 group. With continuous boosting, antibody response to each peptide differed between the two adjuvants ( Figure 1).

Antibody Response to Group 2 Peptides with Different Adjuvants
After the Group 1 peptide test, another group of peptides were injected into mice, with either QS21 or Al(OH) 3 adjuvants. There were a total seven peptides in this group-two in the S1 domain and five in the S2 domain. The ELISA results showed that antibodies were produced against five peptides in both QS21 and Al(OH) 3 groups but were relatively lower in the QS21 group than the Al(OH) 3 adjuvant group; Al(OH) 3 -2D produce higher antibodies than the QS21 adjuvant group (p < 0.001). No antibodies were induced against peptide 2A and 1J after two injections ( Figure 2). Figure 2. Antibody response to peptide mixture after 2 vaccines. In total, 7 peptides were used in this injection, with either QS21 or Al(OH) 3 as adjuvants. Peptides S1I and S1J are selected from S1 domain. Peptide S2A, S2B, S2D, S2E and S2F are selected from S2 domain. (N = 3, p < 0.001).

The Similarity of Epitope Mapping between Recovered Human Sera and Vaccinated Mouse Sera
Four mice plasma (two from each peptide mixture group) and four human plasma samples (two recovered patient blood and two from our tissue bank collected in 2008 as negative controls) were selected for epitope mapping. A total of 16 peptides and recombinant S1 and S2 proteins were predicted in this assay. JK2 and JK3 (recovered patients) display a pattern similar to our peptide vaccine, indicating that the vaccine has included all major epitopes (binding to C, D, H1, I1) ( Figure 3). Coating peptides A-J are peptides from the S1 protein, A2-F2 are peptides from the S2 protein. S1 is the recombinant S1 of the S protein and S2 is the recombinant S2 of the S protein. Ctr is a control peptide of human alpha-synuclein protein, which served as a negative control. Plasma was diluted at a 1:800 dilution with the sample, and ELISA results were shown in the graphs.

The Duration of Antibody Response Post-Vaccination
When examining the duration of the antibody response post-vaccination, Al(OH) 3 appears to generate a longer antibody response than the QS21 adjuvant. Nonetheless, QS21 induced a higher antibody response to some antigens, such as the ones seen in peptide S1-B (Figure 4). . Antibody duration between two adjuvants. The graphs above are the results of antibody testing via ELISA on blood from two time points (1 and 2): 1 is 10 days after the 6th vaccination and 2 is three months after 1. Al(OH) 3 can generate a longer antibody response than QS21 (B,C). However, QS21 can induce a higher antibody response to some antigens, such as that seen in S1-B (A).

Cell Population Analysis to Different Adjuvant Vaccines
Mice blood samples were collected after the third vaccination of peptide mixture group 1 with different adjuvants to compare the B and T cell population difference. Flow cytometry assay was used to analyze the T and B cell population for each vaccine adjuvant. CD3+ and CD8+ T cells ( Figure 5A,C) were significantly higher in Al(OH) 3 groups than QS21 groups (p < 0.05; N = 3 per group). CD4+ T cells ( Figure 5B) were also significantly higher in the Al(OH) 3 adjuvant groups compared to both the QS21 and control groups (p < 0.05; n = 3 per group). Interestingly, the B cell population in the Al(OH) 3 adjuvant group was significantly decreased when compared to the QS21 adjuvant group (N = 3 per group, α = 0.05). These results imply that the Al(OH) 3 adjuvant is ideal for obtaining the desired cellular response ( Figure 5).

Neutralizing Activity to Pseudovirus System
The pseudovirus system in vitro was compared with recovered patient plasma to determine the neutralization potential of post-vaccinated mice plasma. The results showed that mice plasma had better neutralization function than human plasma in this system ( Figure 6).

The Immunoglobulin Isotyping of Plasma from Vaccinated Mice
There are no differences in IgG levels among all three groups. IgM levels are higher in the vaccine group than control mice, indicating that the adjuvant-based peptide vaccine could not change the immunotype of the mice (Figure 7).

Discussion
Vaccines are undeniably the most effective solution to stopping the spread of COVID-19, and several vaccines have already been distributed to contribute to this effort. Pfizer and Moderna have developed nanolipid-based mRNA particle vaccines, while Astrazeneca and Johnson & Johnson have produced adenoviral vector-based DNA vaccines [26,64]. While these have become the most widely used vaccines to prevent COVID-19, other types of vaccines have also been developed, such as an inactivated whole viral particle vaccine (Kexing) and recombinant subunit vaccine (Novavax) [65][66][67]. Although the formulation of each vaccines varies, they all aim to block the binding of the S protein to ACE2 and inhibit viral entry into the host cell [1]. Virologists have proved that the COVID-19 virus surface has spike (S) proteins that can specifically bind to ACE2 through a receptor binding domain (RBD) [39]. The RBD on the S protein of COVID-19 serves as the ligand for ACE2 on the cell surface. It is important to note that most human organs produce the ACE2 protein [40], which may have contributed to the many deaths caused by COVID-19 that were also attributed to multiple organ failure [68].
The COVID-19 vaccines that are currently approved have been reported to have high rates of effectiveness, ranging from 66.3% to 95%, and both B and T cell responses have been observed [69][70][71][72][73][74][75][76][77][78]. Moreover, no large-scale and severe adverse effects have been reported. However, long term monitoring is still necessary before an affirmative assessment can be made regarding vaccine safety. The human immune system works diligently to protect the body from viruses [79]. During viral infection, immune cells produce large amounts of cytokines, which can result in cytokine storm syndrome [80]. Normally, a virus has a specific tissue receptor, meaning the virus can only infect certain tissues, while causing no harm to other tissues [81]. For instance, HIV targets the CD4 receptor, only infecting and killing CD4 T cells [82,83]. However, SARS-CoV-2 is markedly different from other viruses because it utilizes ACE2 receptors, and ACE2 is widely expressed by many types of cells across different organs [40]. ACE2 bound by the SARS-CoV-2 S protein can be considered a viral particle and will trigger an immune response [84]. This scenario complicates vaccine development because the abundance of ACE2 increases the likelihood of potential adverse responses caused by the vaccine and must therefore be carefully considered.
Most of the current vaccines use the S protein as the antigen, and target for blocking viral entry by sealing RBD on S1 of COVID-19, which induces our immune system to attack cells that bind to the S protein [21]. To induce an immune response, antigen-presenting cells (APCs) must uptake the S protein, process it, and then present it to CD4 or CD8 T cells [85]. Although this process is highly effective with other vaccines, it can be problematic for S gene-or S protein-based vaccines, since the S protein will bind to the ACE2 of the host cell.
To combat the potential issues of current COVID-19 vaccines, our study uses peptides as opposed to whole S protein to develop a vaccine for COVID-19. We selected 15 peptides for the whole S protein utilizing B and T cell epitope prediction methods. We have also used two of the most common adjuvants in our vaccine formula-QS21 and Al(OH) 3 -to enhance the immune response and tested this vaccine on a mouse model. The immunization results indicate that antibody production against each peptide varies with time and adjuvant used. In the QS21 adjuvant group, peptide B was the first antibody produced, followed by antibodies against peptides C, F, and D. In the Al(OH) 3 adjuvant group, antibodies against B, C, and F were detected at the same time after the second immunization, which was quicker than in the QS21 adjuvant group (Figure 1). Another major difference between the two adjuvants is that the Al(OH) 3 group produced antibodies against peptide A much earlier than the QS21 group. Since it is imperative for the body to generate a sufficient T cell response to combat infectious diseases [86], the T and B cell population of each adjuvant was further analyzed. The results indicate that Al(OH) 3 is the ideal adjuvant because it can induce a more robust T cell response than QS21 ( Figure 5), although both adjuvants were able to induce an antibody response when tested using the ELISA assay. Since a high antibody titer does not always signify satisfactory protection, we also tested the protection of plasma with the pseudoviral system [87]. This showed that the plasma from the vaccinated mice can effectively neutralize viral entry ( Figure 6). Another issue related to the anti-viral activity of plasma is determining how many epitopes are adequate to completely neutralize the viral entry that can be used as target sites for vaccine development [88]. To collect this information, epitope mapping with plasma from vaccinated mice and from infected patients was conducted. Epitope mapping revealed a similar antibody pattern from vaccinated mice to human samples, which confirmed that our epitope covered most B cell epitopes of the S protein ( Figure 2). Thus, our results demonstrated that development of a S protein peptide-based vaccine is likely feasible to prevent COVID-19.

Conclusions
The authors show that a peptide-based COVID-19 vaccine using S protein peptides with the Al(OH) 3 adjuvant generates a high antibody and T cell response. We also demonstrate that this peptide-based vaccine includes most major B cell epitopes of the S protein by comparing the serological antibody profiles of mice injected with our peptide-based vaccine with humans who were previously infected with COVID-19. Further investigation into peptide-based COVID-19 vaccines is warranted to evaluate important factors such as long-term efficacy.

Data Availability Statement:
The data presented in this article are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.