Designing a multi-epitope subunit vaccine against Orf virus using molecular docking and molecular dynamics

ABSTRACT

Orf virus (ORFV) is an acute contact, epitheliotropic, zoonotic, and double-stranded DNA virus that causes significant economic losses in the livestock industry. The objective of this study is to design an immunoinformatics-based multi-epitope subunit vaccine against ORFV. Various immunodominant cytotoxic T lymphocytes (CTL), helper T lymphocytes (HTL), and B-cell epitopes from the B2L, F1L, and 080 protein of ORFV were selected and linked by short connectors to construct a multi-epitope subunit vaccine. Immunogenicity was enhanced by adding an adjuvant β-defensin to the N-terminal of the vaccine using the EAAAK linker. The vaccine exhibited a significant degree of antigenicity and solubility, without allergenicity or toxicity. The 3D formation of the vaccine was subsequently anticipated, improved, and verified. The optimized model exhibited a lower Z-score of −4.33, indicating higher quality. Molecular docking results demonstrated that the vaccine strongly binds to TLR2 and TLR4. Molecular dynamics results indicated that the docked vaccine-TLR complexes were stable. Immune simulation analyses further confirmed that the vaccine can induce a marked increase in IgG and IgM antibody titers, and elevated levels of IFN-γ and IL-2. Finally, the optimized DNA sequence of the vaccine was cloned into the vector pET28a (+) for high expression in the E.coli expression system. Overall, the designed multi-epitope subunit vaccine is highly stable and can induce robust humoral and cellular immunity, making it a promising vaccine candidate against ORFV

Introduction

Orf, also known as contagious ecthyma, is a globally distributed zoonotic infectious disease that mainly affects sheep and goats, causing significant economic losses in the livestock industry. Orf virus (ORFV), the causative agent of Orf, is a double-stranded DNA virus which belongs to the Parapoxvirus genus within the Poxviridae family [1,2]. The genome of ORFV ranges from 130 kb to 139 kb in size and contains 132 open reading frames (ORFs) [3,4]. The large central coding region (ORFs009-111) contains genes that are essential for viral replication and viral particle formation, while the two terminal inverted repeat regions (ORFs001-008, ORFs112-134) play a crucial role in determining viral virulence, pathogenicity, host range, and immune evasion [5,6].

The B2L (ORF011) and F1L (ORF059) gene are located in the core region of the ORFV genome and exhibit significantly high homology at nucleotide and amino acid levels among various ORFV isolates [7–10]. B2L, a homologue of the vaccinia virus (VACV) F13 protein, is a highly immunogenic major envelope protein of ORFV that is displayed on the surface of extracellular enveloped virus (EEV). The recombinant B2L protein is known to induce a robust antibody response [11,12]. Similarly, F1L, a homologue of the VACV H3L protein, is also an immunodominant major envelope protein of ORFV although it is expressed on the surface of intracellular mature virions (IMV) [13]. Both B2L and F1L have been widely used for vaccine development. Zhao et al. demonstrated that the chimeric DNA vaccine, ORF011/ORF059, induces a significantly stronger immune response than the single ORF011 or ORF059 DNA vaccines [14]. Wassie et al. found that the recombinant B2L and Kisspeptin-54 DNA vaccine induces a significant antibody and cell-mediated immune response against ORFV in rats. Furthermore, the B2L protein can be used as an immunomodulator for kisspeptin-54 to elicit a stronger antibody response [15]. Wang et al. constructed a DNA vaccine expressing the B2L and F1L gene and a subunit vaccine with purified B2L full-length protein and F1L truncated protein as targets. BALB/c mice were administered in different ways including DNA/DNA, protein/protein, and DNA prime-protein boost strategies. Consequently, the heterologous DNA prime-protein boost strategy induced a stronger humoral and cellular immune response [16]. In another study, Wang et al. also found that the DNA prime-protein boost approach elicits more robust humoral and cellular immune responses compared to DNA prime-adenovirus boost and other single-type vaccine immunization strategies [17]. The ORF080 protein of ORFV is a homologue of VACV A4L, which is an immunodominant viral core protein involved in virus assembly and viral morphogenesis [18,19].

Vaccination remains the primary method for preventing and controlling ORFV due to the absence of specific drugs. ORFV vaccines currently available mainly consist of live-attenuated vaccines, inactivated vaccines, and genetically engineered vaccines [20,21]. However, inactivated vaccines tend to induce humoral immunity, with relatively poor ability to induce cellular immunity. Live-attenuated vaccines may have safety concerns, such as virulence regression and genetic recombination between vaccine strains and wild-type strains. Therefore, it is urgently needed to design a safe and efficient vaccine for the prevention and control of ORFV. Multi-epitope vaccines contain multiple dominant cytotoxic T lymphocytes (CTL), helper T lymphocytes (HTL) and B-cell epitopes derived from pathogens, which can overcome the shortcomings of poor immunogenicity and limited immunoprotection induced by a single antigen. Furthermore, multi-epitope vaccines elicit both cellular and humoral immunity to specific pathogens with minimal allergenicity and toxicity compared to conventional vaccines [22,23]. Multi-epitope vaccines have been extensively studied in various viruses, including HIV [24], SARS-Cov2 [25,26], Lumpy skin disease virus [27], and goatpox virus (GTPV) [28]. However, there are no reports of multi-epitope vaccines against ORFV.

The objective of this research was to create a multi-epitope subunit vaccine against ORFV via an immunoinformatics approach. The main immunogenic proteins B2L, F1L, and 080 of ORFV were selected as target antigens, and their immunodominant CTL, HTL, and B-cell epitopes were screened and tandemly linked by flexible junctions to obtain the ORFV multi-epitope vaccine. The physicochemical characteristics, antigenic properties, allergenic potential, toxicity, and the secondary and three-dimensional structure of the developed multi-epitope vaccine were then predicted. Furthermore, the binding affinity with TLRs and protective efficacy of the multi-epitope vaccine against ORFV were evaluated through molecular docking, molecular dynamic simulation, and immune simulation.

Methods

Protein sequence retrieval

The entire amino acid sequences of the B2L(ADF28645.1), F1L(AYN61006.1) and 080(AYN61027.1) proteins of ORFV as well as β-defensin-3 (AAV41025.1) in FASTA format were retrieved from the NCBI database (https://www.ncbi.nlm.nih.gov/).

Screening of CTL epitopes

Cytotoxic T lymphocytes (CTL), also known as CD8⁺ T cells, can directly eradicate the virus and infected cells from the host, thus playing a critical role in cellular immunity. MHC-I-bound epitopes derived from degraded viral protein fragments are recognized by CTLs [29,30]. A key step in the design of vaccines in silico is the prediction of potential CTL epitopes. In the present study, the MHC-I binding server from the Immune Epitope Database (IEDB) (https://tools.iedb.org/mhci/) was used to predict CTL epitopes (9-mer) that are specific for two commonly occurring human MHC-I alleles such as HLA-A*01:01 and HLA-A*02:01 with the prediction method of IEDB NetMHCpan EL4.1 [31]. Epitopes with score > 0.5, percentile rank < 0.5% (higher score and lower percentile rank indicate higher affinity) were retained for further analysis. The VaxiJen v2.0 server (https://www.ddg-pharmfac.net/vaxijen/VaxiJen/VaxiJen.html) was then used to predict the antigenicity of selected epitopes with a threshold value of 0.4 [32].

Screening of HTL epitopes

Antigen-presenting cells, such as macrophages and dendritic cells, present epitopes derived from degraded viral protein fragments as epitope-MHC-II complex to Helper-T cells (HTL) or CD4⁺ T cells. These epitopes elicit an HTL-mediated immune response and activates both the B-cell and cytotoxic T-cell pathways, inducing both humoral and cellular immunity [33]. The IEDB MHC-II binding server was used to predict HTL epitopes (15-mer) specific for two commonly occurring human MHC-II alleles such as HLA-DRB1*01:01 and DRB1*01:02 using the IEDB NetMHCIIpan EL4.1 prediction method [34]. Epitopes with score > 0.5, percentile rank < 0.5% (higher score and lower percentile rank indicate higher affinity) were retained for further analysis. The VaxiJen v2.0 server was then used to predict the antigenicity of selected epitopes with a threshold value of 0.4. IFN-γ plays a significant role in both innate and adaptive immunity by activating macrophages and natural killer cells to fight against invading pathogens. Hence, the IFNepitope server (https://webs.iiitd.edu.in/raghava/ifnepitope/predict.php) was used to screen the HTL epitopes that can induce the release of IFN-γ from CD4 ⁺T cells [35]. Only screened HTL epitopes with a positive IFN-γ score were retained for vaccine construction.

Screening of B-cell epitopes

B lymphocytes differentiate into plasma cells that secrete targeted antibodies to neutralize invading pathogens upon activation by B-cell epitopes [36]. We confidently selected linear B-cell epitopes (16-mer) of selected antigens for vaccine construction using the ABCpred server (https://webs.iiitd.edu.in/raghava/abcpred/index.html) at the default threshold of 0.51 [37,38]. Only screened B-cell epitopes scoring higher than 0.90 were retained for vaccine construction.

Multi-epitope subunit vaccine construction

A multi-epitope subunit vaccine is generated by merging the screened immunodominant CTL, HTL, and B-cell epitopes using the AAY, GPGPG, and KK linkers. To increase immunogenicity, an adjuvant β-defensin-3 was attached to the N-terminal of the vaccine using the EAAAK linker.

Assessment of the physicochemical properties, solubility, antigenicity, allergenicity and toxicity of the multi-epitope subunit vaccine

The ProtParam tool (https://web.expasy.org/protparam/) was employed to predict the physicochemical properties of the constructed multi-epitope vaccine, which include the amino acid count and composition, molecular weight, theoretical isoelectric point (pI), instability index, and the grand average of hydropathicity (GRAVY) [39]. The stability of a protein can be inferred indirectly through its instability index. A protein with a computed instability index below 40 is considered stable, whereas values greater than 40 indicate instability. A GRAVY analysis reveals the amphipathic properties of the proteins, while negative and positive values reveal the hydrophilic and hydrophobic nature of the constructed vaccine, respectively. The Protein-Sol Server (https://protein-sol.manchester.ac.uk/) with a threshold value of 0.45 and the ANTIGENpro server (https://scratch.proteomics.ics.uci.edu/) with a cut-off value of 0.4 were exploited to predict the solubility and antigenicity of the designed multi-epitope vaccine, respectively [40,41]. To predict vaccine toxicity, the ToxinPred server (https://crdd.osdd.net/raghava/toxinpred/) was used [42], while to predict allergenicity, AllerTOP2.0(https://www.ddg-pharmfac.net/AllerTOP/) was used [43].

Modeling, refinement and validation of the multi-epitope subunit vaccine

Using PSIPRED 4.0(http://bioinf.cs.ucl.ac.uk/psipred/), the secondary structure of the constructed multi-epitope vaccine was predicted [44]. Phyre2(http://www.sbg.bio.ic.ac.uk/phyre2/html/page.cgi?id=index)was subsequently employed to determine the 3D arrangement of the multi-epitope vaccine [45,46]. The 3D model underwent further improvement by utilizing GalaxyRefine server (https://galaxy.seoklab.org/cgi-bin/submit.cgi?type=REFINE) [47,48]. The overall quality of the refined model was further verified using ProSA-Web (https://prosa.services.came.sbg.ac.at/prosa.php). A positive Z-score indicates that the generated 3D protein model contains erroneous or erratic sections [49]. The Ramachandran plots showed dominant, anomalous, and rotational isomer regions in the 3D structures before and after optimization via the SWISS-MODEL server (https://swissmodel.expasy.org/assess) [50,51].

Molecular docking

Molecular docking is employed to examine the binding affinity between vaccines and immune receptors. The PDB files for TLR2 (6NIG) and TLR4 (4G8A) were acquired from the RCSB PDB repository. With the aid of the HawkDock server (http://cadd.zju.edu.cn/hawkdock/), we examined how the ORFV-S vaccine interacted with TLR2 and TLR4 [52,53]. Binding ligands, heteroatoms, and chains B, C, and D for TLR2 and TLR4 were removed prior to docking. The Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) analysis was used to predict the binding free energy of the vaccine-TLR complex. Furthermore, the interacted residues between the docked chains were deeply analyzed using PDBsum (http://www.ebi.ac.uk/thornton-srv/databases/pdbsum/) [54].

Molecular dynamic simulation

The iMODS Web server (https//imods.iqfr.csic.es/) was used to perform molecular dynamics (MD) simulation to evaluate the stability of the vaccine-receptor complex [55]. The coordinated movement of the multi-epitope vaccine in internal coordinates was showcased through the normal modal analysis (NMA) approach. Furthermore, the stability of the docked complex was assessed by considering the B-factor, flexibility, covariance, variances, elastic model, and eigenvalues.

Immune simulation

Computerized immune simulations were performed on the C-ImmSim server (https://kraken.iac.rm.cnr.it/C-IMMSIM/) to evaluate the immune response of the multi-epitope vaccine [56,57]. Three injections were administered with a four-week gap between each dose. The total simulation steps were defined as 1050, with each step representing 8 hours. The injection occurred at time steps 1, 84, and 168, respectively.

Codon optimization and in silico cloning

Using the online JCat tool(https://www.jcat.de/Start.jsp), the vaccine construct was reverse translated and codon-optimized in the E.coli K12 host strain [58]. To evaluate the transcription and translation efficiency, we analyzed the percentage of GC content and the codon adaptation index (CAI). A higher level of exogenous gene expression is indicated by a higher CAI value and an ideal GC content ranging between 30% and 70% [59]. Furthermore, the BamHI and XhoI restriction sites were inserted at the N and C termini of the cDNA to allow it to be cloned into the vector pET-28a (+) using SnapGene software.

Results

Screening of immunodominant CTL, HTL and B-cell epitopes

The CTL epitopes for the immunogenic proteins B2L, F1L and 080 of ORFV were predicted by the MHC-I binding server in IEDB. CTL epitopes that met the criteria of MHC-I binding score > 0.5 and percentile rank < 0.5% were screened for further analysis. Furthermore, only selected CTL epitopes with antigenicity score > 0.4 were retained. Finally, a total of 14 immunodominant CTL epitopes of selected proteins were obtained (Table 1).

Table 1.

The screened CTL epitopes for the multi-epitope vaccine construction. (9-mer, score > 0.5, percentile rank < 0.5% and antigenicity score > 0.4).

Protein	Alleles	Peptide sequence	Position	Score	Percentile rank %	Antigenicity score
B2L	HLA-A *01:01	NLDGTHYRY	336-344	0.98	0.01	1.0663
	HLA-A *02:01	SLLSMVPVI	237-245	0.71	0.12	0.6594
	HLA-A *01:01	STIKNLGLY	142-150	0.63	0.12	0.5007
	HLA-A *02:01	TLAKEGVNV	72-80	0.61	0.19	0.4061
	HLA-A *02:01	SMVPVIKHA	240-248	0.58	0.21	0.5074
	HLA-A *02:01	GLYSTNKHL	148-156	0.54	0.24	0.8150
	HLA-A *01:01	VANLDGTHY	334-342	0.50	0.19	0.4185
F1L	HLA-A *01:01	YSDDDFVLV	128-136	0.70	0.1	0.9654
	HLA-A *02:01	RLLWFIAGL	322-330	0.53	0.24	0.7855
	HLA-A *01:01	SAGTVSTKY	120-128	0.50	0.19	1.5142
080	HLA-A *02:01	VLSETSSSL	17-25	0.82	0.07	0.4927
	HLA-A *02:01	GQSPLVPSL	80-88	0.80	0.08	0.4884
	HLA-A *02:01	SLLPRGAKV	87-95	0.74	0.11	0.6864
	HLA-A *02:01	KVQSTPLNV	94-102	0.50	0.26	1.3154

Subsequently, we submitted the selected antigen sequences to the MHC-II binding server in IEDB to predict highly immunogenic HTL epitopes. HTL epitopes that met the MHC-II binding score > 0.5 and the percentile rank < 0.5% were screened for further analysis. Furthermore, the selected HTL epitopes were subjected to the VaxiJen v2.0 server and the IFNepitope server to predict their antigenicity and IFN-γ inducing potential, respectively. Only HTL epitopes with antigenicity score > 0.4 and positive IFN-γ inducer score were retained. Interestingly, there are only three immunodominant HTL epitopes in the B2L protein (Table 2).

Table 2.

The screened HTL epitopes for the multi-epitope vaccine construction. (15-mer, score > 0.5, percentile rank < 0.5%, antigenicity score > 0.4 and positive ifn-γ inducer score).

Protein	Alleles	Peptide sequence	Position	Score	Percentile rank %	Antigenicity score	IFN-γ inducer score
B2L	HLA-DRB1 × 01:01	DDTFAHLTVANLDGT	326-340	0.94	0.28	0.7434	1
	HLA-DRB1 × 01:01	YRYHAFVSVNAEKGD	342-356	0.93	0.31	0.9693	0.6340
	HLA-DRB1 × 01:01	VDDTFAHLTVANLDG	325-339	0.91	0.37	0.6235	1

Then, the linear B-cell epitopes of selected proteins were predicted and screened using the ABCpred server. A total of 8 linear B-cell epitopes with a score >0.90 were retained for the multi-epitope vaccine construction (Table 3). Figure 1 displays a flow chart for the construction of the multi-epitope vaccine.

Table 3.

The screened linear B-cell epitopes for the multi-epitope vaccine construction. (16-mer, ABCpred score ≥0.90).

Protein	Peptide sequence	Position	ABCpred score
B2L	YSMIVEPKVPFTRLCC	168-183	0.95
	LSAVFERDWRSEFCKP	361-376	0.94
	SKDKDADELREAGVNY	88-103	0.90
F1L	KPLIEAMRTNGWYMAQ	150-165	0.92
080	ASPVLEPRVPDKIINA	243-257	0.96
	CSTPVAACSATAVVCP	107-122	0.94
	TACMQRPTTGQSPLVP	71-86	0.91
	PARPAPACPPSTRQCP	188-203	0.91

Figure 1.

Flow chart for the construction of the multi-epitope vaccine.

Construction of multi-epitope subunit vaccine

The multi-epitope vaccine against ORFV was expertly created by combining adjuvant, CTL epitopes, HTL epitopes and linear B-cell epitopes. The screened epitopes were seamlessly merged using the AAY, GPGPG, and KK connectors for CTL, HTL, and B-cell epitopes, respectively. Furthermore, an adjuvant β-defensin was attached to the N-terminal of the vaccine using the EAAAK linker to enhance immunogenicity. Figure 2 clearly displayed the amino acid arrangement of the constructed multi-epitope vaccine.

Figure 2.

Amino acid arrangement of the multi-epitope subunit vaccine. (orange: β-defensin adjuvant; blue: CTL epitopes; red: HTL epitopes; Green: B-cell epitopes. Black residues include the EAAAK connector between β-defensin adjuvant and CTL epitopes, the AAY connector between CTL epitopes, the GPGPG connector between HTL epitopes, and the KK connector between B -cell epitopes).

Physiochemical properties, antigenicity, allergenicity and toxicity of the multi-epitope subunit vaccine

According to the Expasy Protparam results, the constructed ORFV-S vaccine contains 419 amino acid residues with a molecular weight of approximately 45.19 kDa, and a theoretical isoelectric point of 9.48. Its instability index is 38.76 smaller than the threshold value of 40, which indicates that the constructed multi-epitope vaccine is stable. The grand average of hydropathicity (GRAVY) value is −0.215 (<0), indicating that the vaccine is hydrophilic. The vaccine has excellent solubility (0.562), surpassing the threshold value of 0.45. Its high antigenicity (0.91) tends to trigger a robust immune response. Furthermore, the constructed vaccine is non-allergenic and non-toxic based on results from AllerTOP and ToxinPred servers.

Modeling, refinement and validation of the multi-epitope subunit vaccine

The PSIPRED 4.0 analysis confirms that the secondary structure of the vaccine comprises 36.52% α-helix (153/419), 12.41% β-strand (52/419) and 51.07% random coil (214/419) (Figure 3). We utilized the phyre2 online server to forecast the 3D structure of the vaccine (Figure 4a). The 3D model underwent additional refinement through the GalaxyRefine web server, and its quality and potential errors were verified using the ProSA-web server. The GalaxyRefine server presents five optimized 3D models. Within the finest model (Figure 4b), the GDTHA value reached 0.9278, while the RMSD value was 0.482. Additionally, the MolProbity value was recorded as 2.576, and the Clash score amounted to 23.3. The ProSA-web shows that the optimal model has a lower Z-score of −4.33, indicating superior quality (Figure 4c). Figure 4d displays the energy plot of the optimized model. The SWISS-MODEL server evaluation found that the original model had a dominant region of 65.95%, an anomalous region of 18.94% and a rotational isomer region of 15.11% as shown in Figure 4e. After optimization, the dominant, anomalous, and rotational isomer regions accounted for 85.13%, 4.32%, and 10.57%, respectively (Figure 4f).

Figure 3.

Secondary structure of the multi-epitope subunit vaccine.

Figure 4.

Modeling, refinement and validation of the multi-epitope subunit vaccine.

a. 3D structure of the multi-epitope vaccine; b. Optimized 3D structure of the multi-epitope vaccine; c. Z-score of the optimized model was -4.33; d. Energy map of the optimized model; e. The Ramachandran plot showed the original model had 65.95% dominant, 18.94% anomalous, and 15.11% rotational isomer region, respectively; f. The refined model had 85.13% dominant, 4.32% anomalous and 10.57% rotational isomer region, respectively.

Molecular docking between the multi-epitope subunit vaccine with TLR2 and TLR4

The HawkDock server was used to dock the multi-epitope vaccine (ORFV-S) to the human immune receptors TLR2 and TLR4. For each docking, top 100 models were achieved according to the docking score and the binding free energy. The TLR2-ORFV-S complex was found to have an optimal model with a docking score of −5560.92 and a binding free energy of −13.91 kcal/mol, respectively (Figure 5a). The optimal model of the TLR4-ORFV-S complex achieved a docking score of −6473.66 and a binding free energy of −64.87 kcal/mol, respectively (Figure 5b). The best-docked model for the vaccine-TLR complex was visualized and analyzed using PDBsum. There are two salt bridges, six hydrogen bonds, and 158 non-bonded contacts reported at the binding interface of the TLR2-ORFV-S complex (Figures 5c,e). Furthermore, the interfacial residues of TLR2 were 24, while the corresponding ORFV-S had 20 interfacial residues. The interaction analysis of the TLR4-ORFV-S complex showed two salt bridges, six hydrogen bonds, and 205 non-bonded contacts (Figures 5d,f). TLR4 had 25 interfacial residues, while the corresponding ORFV-S had 21 interfacial residues. These results suggest a strong interaction between the ORFV-S vaccine and TLR2 and TLR4.

Figure 5.

Molecular docking between the multi-epitope vaccine and TLR2 and TLR4.

a-b. The binding mode between the multi-epitope vaccine ORFV-S and TLR2 and TLR4. Purple indicates TLR2 or TLR4, while red indicates the ORFV-S vaccine; c, e. The interacting residues between the ORFV-S vaccine and TLR2; d, f. The interacting residues between the ORFV-S vaccine and TLR4.

Molecular dynamics simulation

The stability and movement of the top docked vaccine-TLR2 and vaccine-TLR4 complexes were evaluated through molecular dynamics simulation using the iMODS web server. Figures 6a and 7a demonstrate the movement of the TLR2-ORFV-S and TLR4-ORFV-S complexes, respectively. The deformability plots show that there is very little distortion in the TLR2-ORFV-S and the TLR4-ORFV-S complex, respectively (Figure 6b and 7b). The B-factor plots in Figures 6c and 7c depict the correlation between the mobility of the docked complex NMA and the PDB score (representing the mean RMSD). The eigenvalue is directly related to the energy required to deform a structure. Carbon alpha atoms are more easily deformed at lower eigenvalues. The eigenvalues of the TLR2-ORFV-S and TLR4-ORFV-S complex are 1.324164e-05 (Figure 6d) and 2.824120e-05 (Figure 7d), respectively, indicating the stability of the complex. Each normal mode of the complex is linked to variance plots that depict both individual variance (in purple) and cumulative variance (in green) (Figure 6e and 7e). Covariance plots are used to characterize the motions of correlated (red), non-correlated (white), or anti-correlated (blue) atoms in the dynamic regions of the complex molecules (Figure 6f and 7f). The stiffness of the TLR2-ORFV-S and TLR4-ORFV-S complexes was investigated using the elastic network model of vaccine-TLR complex. Darker grey indicates stiffer regions, while lighter dots indicate flexible regions (Figure 6g and 7g).

Figure 6.

Molecular dynamics simulation of the vaccine-TLR2 complex.

a. Mobility; b. Deformability plot; c. B-factor plot; d. Eigenvalues plot; e. Variance plot. Purple represents individual variance, while green represents cumulative variance; f. Covariance plot. Red represents correlated motion, white represents non-correlated motion, and blue represents anti-correlated motion; G. Elastic network. Darker grey represents stiffer regions.

Figure 7.

Molecular dynamics simulation of the vaccine-TLR4 complex.

a. Mobility; b. Deformability plot; c. B-factor plot; d. Eigenvalues plot; e. Variance plot. Purple represents individual variance while green represents cumulative variance; f. Covariance plot. Red represents correlated motion, white represents non-correlated motion, and blue represents anti-correlated motion; g. Elastic network. Darker grey represents stiffer regions.

Immune simulation of the multi-epitope subunit vaccine

Based on the results of immune simulation, a significant rise in the antibody titers of IgG1+IgG2, IgM and IgM+IgG was observed after the second and third administrations of the ORFV-S vaccine (Figure 8a). Furthermore, there was a remarkable increase in the total number of B cells after the second and third injections of the vaccine (Figure 8b). Active B cells were significantly induced during the vaccination period (Figure 8c). Total helper T cells (TH), and active TH cells were also remarkably induced by vaccination during the simulation time (Figures 8d,e). Additionally, cytotoxic T cells (TC)were observed during the immune simulation, with a peak count of over 1,150 cells per mm³ (Figure 8f). After chimeric antigen injection, the number of active cytotoxic T cells (TC) increased significantly while the number of resting TC cells decreased dramatically, as depicted in Figure 8g. Although IFN-γ production was relatively lower after the third administration, the vaccine still induced significantly elevated levels of IFN-γ (~400,000 ng/mL) after the first and second administrations. Furthermore, the constructed ORFV-S vaccine consistently induced high levels of IL-2 (peaking at 600,000 ng/mL) after each administration, with stronger immune responses observed after the second and third administrations (Figure 8h). The data unequivocally demonstrate that the ORFV-S vaccine induces a robust immune response against ORFV.

Figure 8.

Immune simulation of the multi-epitope subunit vaccine.

a. Changes in IgM and IgG antibody titers after administration of the ORFV-S vaccine; b. B cell count; c. B cell count by state; d. TH cell count; e. TH cell count by state; f. TC cell count; g. TC cell count by state; h. Interleukins and cytokine levels. In the illustration, the letter D is a warning sign.

Codon optimization and in silico cloning

The vaccine construct was reverse-translated and codon-optimized based on the efficiency of codon usage of the E.coli K12 strain using the JCat online tool. The CAI value was determined to be 1.0, and the improved DNA sequence had a GC content of 52.27%, which falls within the optimal range of 30%-70%. These findings indicate that the vaccine holds excellent potential for high expression in the E.coli expression system. Finally, a recombinant plasmid was created by inserting the optimized vaccine DNA sequence into the pET-28a (+) vector with the BamH I and Xho I restriction sites. This process was facilitated by SnapGene software, as shown in Figure 9.

Figure 9.

In silico cloning of the multi-epitope vaccine sequence into the pET-28a(+) vector.

Discussion

Our study successfully developed a multi-epitope vaccine against ORFV by screening immunodominant CTL, HTL, and B-cell epitopes from three significant immunogenic proteins: B2L, F1L, and 080. We confidently identified a total of 7 CTL epitopes from B2L, 3 from F1L, and 4 from 080 protein. Furthermore, we discovered 3 immunodominant HTL epitopes from the B2L protein alone. Our selection process was rigorous, resulting in the confident selection of 8 linear B-cell epitopes from the identified proteins with a score > 0.90. Short connectors AAY, GPGPG, and KK were employed to merge the screened CTL, HTL, and B-cell epitopes, respectively. These linkers play significant roles in generating extended conformations, facilitating protein folding, and segregating functional domains, which enhance the stability of the protein structure [60,61]. Previous research demonstrated that β-defensin can enhance immunogenicity and induce a longer-lasting immune response when administered in combination with antigens [36,62]. Therefore, an adjuvant β-defensin was incorporated into the N-terminal of the vaccine using the EAAAK linker. The vaccine construct comprises 419 amino acids, exhibiting a notable degree of antigenicity and solubility, without allergenicity or toxicity. The 3D structure of the vaccine was predicted and further refined.

Toll-like receptors (TLRs) play a crucial role in the innate immune response, as they detect conserved pathogen-associated molecular patterns (PAMP) originating from different microorganisms [63]. Among all TLRs, only TLR4 and TLR2 possess the capability to recognize the viral proteins [64,65]. A previous study demonstrated that TLR2 was responsible for mediating innate immunity against the vaccinia virus (VACV), a prototypic member of the Orthopoxvirus genus in the Poxviridae family [66]. The TLR4 receptor, which is found on the surface of many types of cells, has been reported to detect structural proteins of different viruses and stimulate the secretion of inflammatory cytokines [67–69]. Therefore, we chose the TLR2 and TLR4 to dock with the designed vaccine construct. Molecular docking results confirmed that the lowest binding free energy of the TLR2-ORFV-S and the TLR4-ORFV-S complex was −13.91 and −64.87 kcal/mol, respectively. It reveals strong binding affinities between the engineered multi-epitope vaccine and TLR2 and TLR4. We confidently chose the E.coli expression system to generate the multi-epitope subunit vaccine due to its high yield, rapid speed, low cost, and ease of purification [70]. The vaccine candidate was first reverse-translated and codon-optimized based on the E.coli K12 strain using the JCat tool. The optimized vaccine sequence was then cloned into the pET-28a (+) vector for high expression in E.coli. Immune simulation results indicated that the constructed multi-epitope vaccine can induce significantly elevated levels of IgG and IgM antibodies, along with IFN-γ and IL-2 cytokines, and the lymphocyte proliferation. It implies that the developed vaccine candidate has the potential to elicit robust humoral and cellular immunity, making it a highly promising candidate against ORFV.

The constructed multi-epitope vaccine against ORFV has several advantages over traditional vaccine types like inactivated, live-attenuated, and genetically engineered vaccines. First, the multi-epitope vaccine targeted multiple epitopes from various immunodominant antigens, offering a more precise and targeted immune response. Second, the designed multi-epitope vaccine contains only specific immunogenic epitopes without allergenicity or toxicity, reducing the likelihood of adverse reactions or side effects often associated with live-attenuated or inactivated vaccines. Third, the designed multi-epitope vaccine is generally safer than live-attenuated vaccines, which carry a risk of causing the disease in immunocompromised individuals, or inactivated vaccines, which may not elicit a strong immune response. Overall, the designed multi-epitope vaccine offers a promising approach to vaccine development, providing enhanced safety, specificity, and flexibility in targeting ORFV.

Nevertheless, the present study does have certain constraints. Due to the absence of goat MHC alleles in the immunoinformatics database, human HLA alleles were used as a substitute to predict T cell epitopes based on previously published studies. The designed multi-epitope subunit vaccine is believed to be immunogenic according to results of various immunoinformatics tools. However, the precision of these approaches is not flawless, and the protection efficacy of the designed vaccine remains uncertain. In subsequent phases, preclinical challenge-protection experiments will be performed in goats to evaluate the safety and protective efficacy of the designed vaccine compared to inactivated vaccines and live-attenuated vaccines against ORFV on the market. Furthermore, vaccine-induced humoral and cellular immune responses will also be evaluated through a series of experiments that included specific antibody detection, IgG subtype classification, virus neutralization test, Th1/2 cytokine detection, and lymphocyte proliferation detection.

Conclusions

A multi-epitope subunit vaccine against ORFV was successfully constructed based on immunoinformatics approaches. The vaccine comprises a β-defensin adjuvant, 14 CTL, 3 HTL, and 8 B-cell epitopes, exhibiting excellent solubility, and antigenicity, without causing sensitization or cytotoxicity. Additionally, it has been found to stably interact with TLR2 and TLR4 with high affinity, and can induce a robust and lasting host immune response, making it a promising vaccine candidate against ORFV. However, the safety and protective efficacy of the designed ORFV multi-epitope subunit vaccine need to be evaluated through in vivo and in vitro experiments in subsequent phases.

Funding Statement

This work was supported by the Guizhou Provincial Science and Technology Project under Grant [2022093].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

Feng Pang: Conceptualization, Writing-original draft, Writing-review & editing, Funding acquisition. Qinqin Long: Software, Visualization, Writing-review & editing; Shaobo Liang: Software，Writing-review & editing. All authors read and approved the submitted version.

Data availability statement

The Data supporting the findings of this study are openly available at repository Science Data Bank at https://doi.org/10.57760/sciencedb.07393