You are viewing the site in preview mode

Skip to main content
Download PDF

A novel immunoinformatic approach for design and evaluation of heptavalent multiepitope foot-and-mouth disease virus vaccine

BMC Veterinary Research volume 21, Article number: 152 (2025) Cite this article

Abstract

Background

Foot-and-mouth disease virus (FMDV) vaccine development can be a laborious task due to the existence of various serotypes and lineages and its quasi-species nature. Immunoinformatics provide effective and promising avenue for the development of multiepitope vaccines against such complex pathogens. In this study, we developed an immunoinformatic pipeline to design a heptavalent multi-epitope vaccine targeting circulating FMDV isolates in Egypt.

Result

B and T-cell epitopes were predicted and selected epitopes were proved to be non-allergenic, non-toxic, with high antigenicity, and able to induce interferon-gamma response. The epitopes were used to construct a vaccine by adding suitable linkers and adjuvant. Prediction, refinement, and validation of the final construct proved its stability and solubility, having a theoretical isoelectric point (PI) of 9.4 and a molecular weight of 75.49 kDa. The final construct was evaluated for its interaction with bovine toll-like receptor (TLR) 2 and 4 using molecular docking analysis and molecular dynamic simulation showed high binding affinity, especially toward TLR4. MM/GBSA energy calculation supported these findings, confirming favorable energetics of the interaction. Finally, the DNA sequence of the vaccine was cloned in pET-30a (+) for efficient expression in Escherichia coli.

Conclusion

The inclusion of computational and immunoinformatic approaches will ensure cost-effectiveness and rapid design of FMDV vaccine, decrease wet lab experimentation, and aid the selection of novel FMDV vaccines. While the vaccine demonstrates promising in-silico results, experimental assessment of vaccine efficiency is required.

Peer Review reports

Background

Foot-and-mouth disease virus (FMDV) is one of the highly endemic viruses that can infect various cloven-hoofed species. FMDV causes Foot-and-mouth disease (FMD), a non-fatal disease in adult animals, but has a high mortality rate in younglings and causes significant animal abortions [1, 2]. Despite the low level of viral mortality in adult animals, there is a high proportion of infectivity between animals in proximity, due to the easy spread of the virus between them [3]. FMD was reported to cause 8–22 billion USD in economic losses each year in endemic countries [4].

FMDV consists of positive-sense single-stranded RNA with around ~ 7000 nucleotides (nt) open reading frame, that encodes for all the structural and non-structural proteins. Other regions for the single-stranded RNA are designated for the 5` and 3` untranslated regions that measure around 1300 and 90 nt in length, respectively [5, 6]. The viral open reading frame is translated upon entering the mammalian cell into a polyprotein that is cleaved into 15 mature proteins divided into structural (VP1, VP2, VP3, and VP4) -that assemble to form the viral capsid- and non-structural proteins (Lab, Lb, 2 A, 2B, 2 C, 3 A, 3B1, 3B2, 3B3, 3 C, and 3D) [7].

FMDV has seven different serotypes (A, O, Asia1, C, SAT1, SAT2, and SAT3) where SAT serotypes are localized mainly in Africa. This classification of FMDV was developed based on serological analysis [8] and supported by VP1 sequence analysis [9, 10]. RNA replication errors [11], genome recombination [12, 13], and selection pressure imposed by the host immune system [14] are responsible for the high sequence variability of the FMDV genome, sequentially caused the emergence of different serotype variants. Most sequence variability is present in the structural protein-coding sequence, with VP1 showing the lowest identity scores between different serotypes ranging from 50 to 70%. Whereas the most variabilities were found in the exposed parts of proteins (VP1-3) of the viral capsid, VP4 protein showed less variability within structural proteins [14].

The production of an effective FMDV vaccine that induces sterile and solid immunity has not been achieved despite the ongoing research effort during the last decades, resulting in intensive viral spread, especially in enzootic countries [15]. The lack of an effective vaccine can be reasoned for viral quasi-species and the emergence of new FMDV variants every year. The high variabilities among FMDV serotypes restrain the ability to reach a universal vaccine that provides reliable protection from the seven known viral serotypes [16]. Authorities resort to use vaccines consisting of different formulations of inactivated virus serotypes [17]. Current approaches for developing FMDV vaccine aim to produce multi-epitope vaccines (MEV) that provide a solid immune response with a low risk of infectivity.

The wide availability of different databases for most biological information, such as molecular biology and genetics, enabled the development of various computational tools for in-silico experimentation to be conducted before application in a laboratory environment [18, 19]. Improvement of computational power and introduction of Immunoinformatics traverse the paradigm of vaccine design to sequence-based techniques instead of traditional methodologies [20]. In addition to recombinant DNA technology, immunoinformatics helped in the preparation of new peptides with better antigenic and immunological characteristics increasing the efficiency of the designed vaccine [21, 22].

Immunoinformatics, with the aid of machine learning and artificial neural network models, have a great role in the selection of those peptides for different immunocytes with high prediction specificity and sensitivity [23,24,25]. In-silico prediction of effective epitopes can have a huge impact in decreasing wet lab experiments and in turn, increasing vaccine design efficiency [26]. Different statistical models were developed and validated to evaluate the designed peptides based on the genomic and amino-acid sequence variability [27,28,29].

In this study, we harnessed the power of immunoinformatics to develop and evaluate a new FMDV multiepitope-based vaccine for circulating serotypes in Egypt. Our workflow pipeline (Fig. 1) starts with the identification of efficient structural protein epitopes for B and T cells. Selected epitopes are connected through specific protein linkers to each other and to the adjuvant for the generation of MEV. The proposed pipeline provides a good pathway for the development of the FMDV vaccine industry and increases the knowledge of FMDV control.

Fig. 1
figure 1

Illustration of the study workflow

Full size image

Methods

Protein sequence retrieval

Complete FMDV proteomes of serotypes A, O and SAT2 were retrieved from the National Center of Biotechnology Information (NCBI) protein database in FASTA format (https://www.ncbi.nlm.nih.gov, accessed on 15 February 2022) (Table 1). All amino acid sequences were trimmed to include only the structural proteins (VP1, VP2, and VP3). The trimmed amino acid sequences were later used for epitope prediction.

Table 1 Protein sequences retrieved from the NCBI protein database
Full size table

Epitope prediction

Cytotoxic T lymphocytes (CTL)

Epitopes of the CD8 + T cell were predicted using major histocompatibility I (MHC-I) binding prediction tool, immune epitope database (IEDB) webserver, (http://tools.iedb.org/mhci/). The web tool depends on the (NetMHCpan EL 4.1) updated version, which has been trained in extended MHC-I binding affinity, and multi and single-allelic eluted ligands data from different public domains [30]. The tool’s machine learning algorithm was also updated from NNAlign to NNAlign_MA to include the multi-allelic MHC data [30, 31]. Available Bovine MHC-I alleles in the IEDB database (Table s1) were aligned against the query FMDV protein sequences, and data less than the percentile rank threshold (< 0.05) were selected.

Helper T lymphocytes (HTL)

NetMHCIIpan 2.1 server (https://services.healthtech.dtu.dk/service.php?NetMHCIIpan-2.1) was used to predict CD4 + T cell epitopes. The server uses artificial neural networks (ANN) that have been trained in more than 200,000 randomly selected peptides [32]. The server was selected because it includes Bovine leukocyte alleles (BoLA) of MHC-II (i.e., BoLA-DRB3) (Table s2). One of three parameters was used for the selection of epitopes that may have a strong binding affinity with selected MHC-II alleles. These parameters are binding affinity (IC50), percentile rank, and prediction score with thresholds of < 50 nM, < 0.5, and > 0.9, respectively.

B-cells

Linear B-cell epitopes were predicted using the ABCpred web tool (http://crdd.osdd.net/raghava/abcpred/). ABCpred uses ANN to predict B-cell epitopes depending on four metrics generated during the prediction process, precision, sensitivity, positive predictive value, and specificity. The prediction accuracy of the webtool model is 65.93% with the use of the cross-validation fold [33]. The FASTA sequences of all selected FMDV serotypes in the study were used to predict the B-cell epitopes, with a threshold of > 0.5 and peptide length of 16 mer.

Allergenicity, toxicity, antigenicity, and conservancy analysis of epitopes

Selected epitopes were evaluated for their antigenicity using the Vaxijen v2.0 server (http://www.ddg-pharmfac.net/vaxijen/) by default score of > 0.4 for the antigenic epitope [34, 35]. Peptides that showed no antigenicity or value near the threshold were discarded.

Allergenicity and toxicity of the selected epitopes were identified using the AllerTop server (http://www.ddg-pharmfac.net/AllerTop/) [36] and ToxinPred server (http://crdd.osdd.net/raghava/toxinpred/) [37], respectively. Epitopes that were identified as allergenic or toxic were discarded.

High-level antigenic epitope conservancy is required to ensure that the vaccine can provide wider protection against different strains. Hence, the predicted epitopes were compared against 150 amino acid sequences for each serotype retrieved from the NCBI protein database using the BLASTp tool against the FMDV serotype P1 protein. The epitope conservancy analysis tool from the IEDB web server (http://tools.iedb.org/conservancy/) was used for the analysis of the epitope conservancy pattern with a threshold of 0.8 for sequence identity [38].

Design of MEV

The MEV was designed by fusing B-cell and T-cell epitopes. These epitopes were highly conserved within the same serotype and showed high antigenicity, non-allergenicity, and non-toxicity. “KK”, “AAY” and “GPGPG” linkers were used to connect epitopes from B-cell, cytotoxic T-cell, and helper T-cells, respectively. Mycobacterium tuberculosis heparin-binding hemagglutinin (HBHA) was used as an adjuvant in the vaccine construct as it was proved to be an efficient TLR-4 agonist, enhancing immune activation. For proper separation and flexibility between the adjuvant and the multiepitope region, an “EAAAK” linker was used.

MEV construct, TLR-4, and TLR-2 structure modeling, refinement, and validation

The tertiary structure of MEV was predicted using the Robetta server (Robetta server, https://robetta.bakerlab.org/). The Robetta server is a protein prediction server, its core is the Rosetta macromolecular modeling suite [39]. The RoseTTAFold option, a deep learning-based method, was chosen for the prediction of the MEV tertiary structure. The web server generated multiple models for the tertiary structure of the MEV. The bovine TLR-4 protein sequence (accession number NP_776623) and TLR-2 protein sequence (accession number ALL55248.1) were aligned against the SWISS-Model database [40] to build their tertiary structures using homology modeling. The web server ranked the model according to their highest accuracy scores.

The generated tertiary structures models were then refined and enhanced using GalaxyRefine (https://galaxy.seoklab.org/cgi-bin/submit.cgi?type=REFINE) web server [41, 42]. The residual geometry of the refined vaccine construction was analyzed and improved followed by the prediction of construction stereochemical quality using the PROCHECK web tool from the SAVES v6.0 server (https://saves.mbi.ucla.edu/) [43]. Data generated from the Ramachandran plot and ERRAT scores were used for the selection of the best tertiary model. The final structure of the selected model was tested for the presence of errors using the ProsA server (https://prosa.services.came.sbg.ac.at/prosa.php) [44]. The final optimized protein structure was evaluated for its physicochemical properties using ProtParam, Expasy webserver (https://web.expasy.org/protparam/).

Discontinuous B-cell epitope prediction

Discontinuous B-cell epitopes of MEV 3D structure were predicted using the Ellipro tool in the IEDB webserver (http://tools.iedb.org/ellipro/). The Ellipro tool can predict both linear and discontinuous epitopes. It uses three main algorithms to predict the discontinuous epitopes. These algorithms work on identification of the epitopes by clustering protein residues based on certain values called Protrusion Index (PI), their spatial proximity, and giving them scores determined by the PI of their residues. The web tool was used with its default parameters with a minimum score of 0.5 and a maximum distance of 6 angstroms.

Molecular docking of TLR-4 and TLR-2 with MEV

Protein-protein interaction between the MEV and both bovine TLR-4 and TLR-2 were predicted using Clustpro server (https://cluspro.org/login.php?redir=/models.php?job=1177853) [45]. The structures with interaction within the extracellular region of TLR-4 or TLR-2 were selected for further analysis of the docked complex. Binding affinities (\(\:\varDelta\:G)\) and Equilibrium dissociation constant (Kd) of selected models was determined using the PRODIGY webserver (https://wenmr.science.uu.nl/prodigy/) [46]. The interacting atoms were identified and visualized using the PDBsum server (https://www.ebi.ac.uk/thorntonsrv/databases/pdbsum/Generate.html) [47]. The final docked complexes were visualized using Discovery Studio 2023 (https://www.3ds.com/products/biovia/discovery-studio).

Molecular dynamic simulation and energy minimization

The topology parameters of the proteins were generated using GROMACS 2024.2 with AMBER99SB force field [48]. Each protein complex was solvated in a cubic box with SPC water model [49] and neutralized by adding Cl− counter ions. The neutralized systems underwent energy minimization using the steepest descent algorithm in 50,000 steps or Fmax < 400 kJ/mol. Subsequently, the minimized systems were equilibrated in two steps: first, NVT ensemble where the temperature was controlled at 300 K for 200 ps using the V-rescale thermostat, followed by NPT ensemble where the pressure was controlled by Parrinello-Rahman barostat algorithm for another 200 ps. Finally, the MD simulations were performed for 100 ns under NPT ensemble with time step of 2 fs. Long-range electrostatic interactions were calculated using Particle Mesh Ewald (PME) algorithm [50]. Hydrogen bond lengths were constrained using Linear Constraint Solver (LINCS) algorithm [51]. The trajectories from the production run were treated for removing the periodic boundary conditions (PBC) and then used in the analysis.

Following the MD simulation, the root mean square deviations (RMSD) from initial equilibrated positions of C-α atoms of each chain of TLRs was investigated to test the stability of respective proteins. The fluctuations in the side chain atoms of residue of the tested TLR protein and were analysed as root mean square fluctuation (RMSF) of C-α atoms of each protein in systems. The compactness of system and consequent stability was analysed in terms of radius of gyration (Rg) for each protein of the TLRs. The hydrogen bonds formed between vaccine chain and the TLR chains were analysed using appropriate index files of respective chains. the trajectories were analysed using the GROMACS tools rms for RMSD, rmsf for RMSF, gyrate for RG, sasa for SASA and hbond to calculate the number of hydrogen bonds.

The docked protein-vaccine complexes were imported into the HawkDock server V2 (http://cadd.zju.edu.cn/hawkdock/) for VD-MM/GBSA analysis [52]. The binding free energy of each complex was calculated, and key residues contributing to the interaction were identified through per-residue energy decomposition.

MEV codon optimization and in-silico cloning

Reverse translation and codon optimization were generated using the Reverse translate tool, sequence manipulation suite (https://www.bioinformatics.org/sms2/rev_trans.html), and codon optimization tool, VectorBuilder (https://en.vectorbuilder.com/tool/codon-optimization.html), respectively. The generated cDNA sequence was optimized for expression in the E. coli K-12 strain. The DNA sequence was inserted into the pET-30a (+) vector using the SnapGene tool.

Results

Strains selection and epitope prediction

Protein sequences representing Egyptian circulating FMDV strains for O, A, and SAT2 serotypes were retrieved from the NCBI protein database. Whole genomes or full P1 sequences for all strains were obtained and in case, the full-length sequences were not found for a certain strain, the full sequence of the nearest strain was retrieved. For instance, the ETH/9/2019 genomic sequence was used as the nearest replacement for the circulated Egyptian East Africa 3 (EA-3) strain. Those sequences were later prepared and subjected to epitope prediction.

Predicted epitopes that showed high binding affinity to B- and T-cell alleles were considered for the vaccine design. epitope conservancy analysis that showed high conservancy within its serotype (> 80%) with high antigenicity were selected. (Fig. 2).

Fig. 2
figure 2

Conservancy and antigenicity analysis for the predicted surface antigen epitopes for (1) BCL, (2) CTL and (3) HTL

Full size image

CTL predicted epitopes

T-cell epitopes were included in preventive vaccines to naturally mimic the immune response triggered by infection. T-cell epitopes help in the induction of durable cellular-mediated immunity. Epitopes with the highest binding scores and the lowest percentile rank (< 0.05) against 103 BoLA were selected as strong binders (Table 2). More than 62 epitopes were higher than 0.85 binding score, and lower than 0.05 in percentile rank. After antigenic and allergenic filtration, fifteen different alleles showed high binding affinity toward 16 different epitopes in all tested serotypes. Epitope starts at position 402 of P1 has been predicted in the three serotypes and has a binding affinity toward four different MHC-I alleles (BoLA-1:02301, BoLA-6:01301, BoLA-D18.4, and BoLA-HD6). Although the position of the epitope was conserved in all the serotypes, its amino acid sequence was different, particularly in the SAT2 serotype. On the other hand, epitope “DVAEACPTL” was predicted to have a high affinity toward MHC-I from SAT2 VP3 protein. Although it was only predicted within the VP3 protein sequence, the conservancy analysis of this epitope showed to be conserved in the three serotypes.

Table 2 List of CTL (MHC-I) predicted peptides including conservancy and antigenicity score
Full size table

HTL predicted epitopes

HTL epitopes are essential for the generation of both humoral and cell-mediated immunity. These epitopes induce CD4 + responses that help in the formation of CD8 + T-cell memory and activation of antibody production by B-cells [53, 54]. We selected epitopes with the highest prediction score, the lowest percentile rank, and a binding affinity of less than 50 nM. The final selected epitopes were found to be mostly in VP1 protein (11 epitopes) and only three epitopes were found in VP2 protein (Table 3). Serotype A and SAT-2 showed epitopes within the same location of VP2 with different amino acid sequences, these epitopes had an affinity toward the same MHC-II alleles. Interestingly, the epitope sequences from the A serotype showed conservancy in O serotype isolates.

Table 3 List of HTL (MHC-II) predicted peptides including conservancy and antigenicity score
Full size table

B-cell predicted epitopes

B-cell epitopes were predicted using the ABCpred server where the algorithm sorts the predicted peptides according to their score in different ranks. Epitopes were filtered by selecting highly antigenic and non-toxic epitopes. Moreover, B-cell epitopes were tested for the presence of signal peptides. After filtration, only four epitopes showed promising B-cell epitopes from VP3 and VP2 proteins (Table 4). One peptide from SAT2 VP3 protein showed conservancy with the three serotypes.

Table 4 BCL-predicted epitopes for FMDV serotypes including conservancy and antigenicity scores
Full size table

MEV structure modeling, refinement, and validation

Vaccine construction involved 17 MHC-I epitopes, 11 MHC-II epitopes, and four B-cell epitopes. Those epitopes have been separated by AAY, GPGPG, and KK linkers, respectively. After the inclusion of HBHA adjuvant separated from B-cell epitopes with EAAAK linker and 6X his-tag, the final multiepitope protein length was 704 amino acids. The molecular weight of MEV was 75.49 KDa, theoretical PI 9.4, and the extension coefficient at 280 nm in water was 95,370 M− 1 cm− 1. The protein was predicted to be stable with an instability index of 27.5. The aliphatic index was 82.12 and the grand average of hydropathicity (GRAVY) was − 0.292, indicating that the protein is hydrophilic and soluble in water.

The tertiary structure of the protein was generated using the Robetta server. The generated models were further refined using the GalaxyRefine webserver. The validation scores of both original and refined models were used for the best model selection. The final selected model had an ERRAT score of 92.7, and its Ramachandran plot showed 92.9% residues within most favored regions, 6.1% residues in allowed regions, 0.2% residues in generously allowed regions, and 0.8% residues in disallowed regions. The final selected model Vaxigen score was 0.527, proving that the MEV is antigenic. The MEV was non-allergenic, non-toxic, and had no signal peptide nor transmembrane helices, so no specific protein localization nor difficulties could be faced during protein production (Fig. 3).

Fig. 3
figure 3

Modeled and refined structure of MEV. (A) amino acid sequence of MEV. (B) the tertiary structure of MEV. (C) Ramachandran plot analysis of predicted MEV structure

Full size image

Discontinuous B-cell epitopes prediction

A total of 359 residues were predicted to be involved in discontinuous B-cell epitopes. The length of predicted epitopes ranged from 9 to 147 residues with 37 residues having the highest binding score of 0.972 (Table 5) (Fig. 4). HBHA domain showed the highest epitope score and lowest epitope score located within MHCI epitopes.

Table 5 Discontinuous B-cell epitope prediction results showing involved residues within each epitope with its score
Full size table
Fig. 4
figure 4

Discontinuous B-cell epitope prediction. Results showing epitope 1 to epitope 6 (A–F). The protein backbone is represented in blue and the yellow-highlighted residues represent the predicted discontinuous B-cell epitopes

Full size image

TLR-4 and TLR-2 structural prediction

Bovine TLR-4 and TLR-2 sequences were used to predict their tertiary structure using homology modeling by SWISS-Model from the Expasy webserver. SWISS-Model generated different tertiary structures based on sequence alignment with PDB and AlphaFold database. TLR-4 tertiary structure showed 84% similarity with the TLR-4 Neophocaena asiaeorientalis model from the AlphaFold database. TLR-2 highest model showed 95% similarity with Capra ibex TLR-2. Selected models were refined using GalaxyRefine and RMSD values were 0.338 and 0.3 for TLR-4 and TLR-2, respectively. Ramachandran plot and quality score were generated using ERRAT and PROCHECK from the SAVES V6.0 website. The overall quality factor for TLR-4 and TLR-2 were 88.3 and 91.6, respectively. TLR-4 Ramachandran plot showed that the refined model is reliable with 89.4% atoms in the favored regions, 10.1% in the allowed regions, and 0.4% in disallowed regions. TLR-2 had 90.3% of the molecular atoms in the favored regions, 9.5% in the allowed regions, and 0.1% in the disallowed regions.

Molecular docking analysis

The docking analysis was performed using the ClusPro web server for interaction between TLR-4 or TLR-2 and MEV. ClusPro generated 30 models for each interaction, each model was visualized and assessed based on type, location, and number of interacting atoms between the TLRs and MEV. Models that attain a high number of bonds between MEV and extracellular domains of TLRs were selected for further evaluation. Molecular docking analysis of the selected MEV and TLR-4 predicted the formation of 62 bonds with 10 salt bridges and 26 hydrogen bonds (Fig. 5). TLR-2 molecular docking analysis revealed the formation of 60 bonds that contained 6 salt bridges and 23 hydrogen bonds (Fig. 6). The selected docked models were further evaluated based on their binding affinity (\(\:\varDelta\:G)\) and dissociation factor (Kd). The final selected models for MEV-TLR-4 \(\:\varDelta\:G\) value − 22 and Kd 2.8e-16, MEV-TLR-2 model \(\:\varDelta\:G\) was − 17.8 and Kd 3e-13. These values indicated that selected models are energetically viable.

Fig. 5
figure 5

Molecular docking analysis of MEV and Bovine TLR-4. (A) The docked complex of bovine TLR-4 and MEV. (B) interacting amino acids between TLR-4 (Chain A) and MEV (Chain B). (C) analysis of formed bonds between docked complexes. (D) Key for interactions between docked molecules

Full size image
Fig. 6
figure 6

Molecular docking analysis of MEV and Bovine TLR-2. (A) The docked complex of bovine TLR-2 and MEV. (B) interacting amino acids between TLR-2 (Chain A) and MEV (Chain B). (C) analysis of formed bonds between docked complexes. (D) Key for interactions between docked molecules

Full size image

Molecular dynamic simulation

RMSD analysis of TLR-2 and TLR-4 protein-vaccine complexes revealed distinct patterns of structural stability throughout the simulation. Initially, TLR-4 exhibited higher RMSD than TLR-2 up to 5.5 ns (0.9110 ± 0.3025 nm vs. 0.7330 ± 0.2265 nm, respectively). However, beyond 5.5 ns, the trend reversed, with TLR2 showing higher RMSD. Both systems converged at approximately 40 ns, with TLR-4 displaying a lower RMSD compared to TLR-2 during the final phase of the simulation (40–100 ns), indicating enhanced structural stability (Fig. 7A).

Fig. 7
figure 7

Molecular dynamic simulations graph representation of the performance of MEV construct with TLR-4/TLR-2 complexes. (A) RMSD plot of MEV-receptor complexes showing their stability within the 40 ns of the simulation. (B) RMSF plot shows good flexibility of construct with high fluctuation within the c-terminus of both receptors. (C) The RG plot indicates the compactness of the receptors as we progress in the simulation. (D) SASA plot shows higher solvent exposure of TLR-4 protein complex than TLR-2. (E) Hydrogen bond analysis plot showing higher number of hydrogen bond formation between MEV and TLR-4 compared to MEV and TLR-2

Full size image

RMSF analysis showed TLR-2 showed a slightly higher flexibility (0.99 ± 0.4) than TLR-4 (0.93±. 0.54). At the N-terminus, TLR-2 showed significantly greater flexibility (1.53 ± 0.14), while of TLR-4 exhibited pronounced fluctuations in the C-terminus (3.06 ± 0.23) (Fig. 7B). TLR-2 exhibited a lower RG (3.97 ± 0.56) than TLR-4 (4.47 ± 0.58), indicating a more compact structure for TLR-2. During the initial phase (up to 10 ns), both proteins showed higher RG values as they equilibrated, with TLR-2 and TLR-4 having RGs of 5.478 ± 0.28 and 5.408 ± 0.215, respectively. After 40 ns, the RG values stabilized, reflecting equilibrium conformations; TLR-2 maintained a more compact structure (3.678 ± 0.121), while TLR-4 stabilized at a higher RG (4.046 ± 0.089), with a more expanded conformation (Fig. 7C). SASA analysis showed TLR-4 had a consistently higher exposed surface area (434.43 ± 13.97) compared to TLR-2 (415.86 ± 8.03), aligned with the RG trends (Fig. 7D).

Hydrogen bonds analysis reveals distinct interaction profiles that highlight differences in binding strength and stability. On average, the interaction between TLR-4 and the vaccine formed a significantly higher number of H bonds (29.64 ± 9.00) compared to TLR-2 and the vaccine (15.01 ± 3.50). The larger but variable H bond number in the TLR4-vaccine interaction suggests a more dynamic and robust binding interface (Fig. 7E).

The MMGBSA analysis provided insights into the binding energetics of the MEV vaccine construct with TLR-2 and TLR-4. The TLR4-MEV complex exhibited a stronger total binding energy of -226.51 kcal/mol, mostly due the favorable van der Waals (-304.67 kcal/mol) and electrostatic (-425.19 kcal/mol) interactions, partially encountered by an unfavorable polar solvation energy (543.15 kcal/mol) and a stabilizing nonpolar solvation energy (-39.8 kcal/mol). In contrast, the TLR2-MEV complex demonstrated a weaker total binding energy of -177.09 kcal/mol, with a lower van der Waals (-238.44 kcal/mol) and electrostatic (-234.54 kcal/mol) interactions contributing significantly, and a lesser desolvation penalty (326.77 kcal/mol).

In-silico molecular cloning

The protein sequence of MEV was reverse-translated to its respective DNA sequence using the Reverse translate tool, sequence manipulation suit. The codon optimization index was 0.91 and the GC percentage was 59.94%, indicating a good abundance of commonly used codons. (Fig. 8) shows the optimized DNA sequence ligated with the pET-30a (+) vector using EcoRI and HindIII restriction digestion.

Fig. 8
figure 8

Illustration of vaccine DNA sequence cloning in pET30a(+) vector. (A) shows the insertion of DNA sequence in pET30 vector using EcoRI and HindIII restriction enzyme. (B) shows the final recombinant plasmid containing MEV sequence in green arrow

Full size image

Discussion

Vaccination remains one of the most essential measures for the prevention and control of FMD. Traditional viral-based vaccines have great immunogenicity, but there are several drawbacks, including high manufacturing costs linked to the need for a dedicated manufacturing facility, along with the possibility of insufficient inactivation or viral leakage [55, 56]. Recombinant DNA technology provides a solution for those problems with a chance of producing a vaccine that attains long-lasting immunity. Recent publications indicated the success of using recombinant epitopes from different FMDV structural proteins to provide better viral immunity and fewer hazardous production process [57,58,59].

Immunoinformatics and computational biology have raised the bar in the vaccine production industry through enabling a safer and more innovative way to design viral vaccines [60, 61]. The huge amount of available biological data can be used to identify promising epitopes and help suggest vaccine candidates to be validated in wet lab [62]. Therefore, the current study aimed to develop MEV with high immunogenicity against FMDV. During this work, we identified the potential dominant B-cell and T-cell epitopes from different circulating Egyptian serotypes to help design an efficient FMDV vaccine. The suggested vaccine is predicted to provide immunity against different FMDV lineages and serotypes based on the inclusion of conserved and immunogenic epitopes. Additionally, the designed vaccine is computationally predicted to activate both humoral and cell-mediated immunity, though experimental validation is required to confirm the efficacy of these findings.

During this study, we showed the significance of employing immunoinformatics for FMDV vaccine development. Previous studies such as those by Ahmed et al. 2022 [63] and Bhutta et al. 2024 [64] utilized immunoinformatics for vaccine design focused on specific serotypes or polyprotein regions. For instance, these studies have focused only on developing FMDV vaccine against serotype O. In contrast, our study includes epitopes from multiple serotypes increasing the vaccine’s capability to provide cross-serotype immunity. Like Bhutta, we have specifically predicted epitopes for FMDV against BoLA. However, Bhutta has studied the interaction of multiepitope vaccine with specific BoLA using molecular docking. our work focused on studying the interaction of developed vaccines with different cellular receptors such as TLR-4 and TLR-2. We also focused on the use of variable structural proteins of virus capsid when other studies such as Ahmed et al. used the full polyprotein for prediction of vaccine epitopes. Using structural proteins ensures the inclusion of FMDV antigenic sites.

Several antigenic sites have been identified in highly variable structural proteins and have been divided into five main antigenic sites. Sites 1, 3, and 5 are βG-βH loop and carboxy terminus, residues 566 and 567 located in the βΒ-βC loop, and residue 672 of VP1, respectively. Antigenic site 2 is composed of residues at positions 155–158, 160,162, and 216 of VP2, whereas residues at positions 164 and 219 are responsible for interaction with monoclonal antibodies. Residues 359 and 361 of VP3 were found to be crucial for antigenic site 4 [17]. Even, VP1 is considered a highly variable protein, it contains conserved regions throughout different serotypes. These regions are crucial for cellular integration and interaction, such as G-H loop containing the “Arg-Gly-Asp, or RGD” motif [12, 65], or capsid processing such as the YCPRP sequence [66].

During this work, we predicted the linear B cell epitopes in FMDV structural proteins (VP1-3). Peptides from the VP3 protein were selected for all tested serotypes. Selected VP3 epitopes have not been experimentally discovered or identified in previous studies. VP3-131 and 132 [67] residues were predicted to have a significant antigenic property. Those residues were found within predicted B-cell epitopes with high rank and binding affinity toward B-cell receptors. However, those epitopes were tested non-antigenic by the Vaxigen webserver. A single peptide was selected from SAT2 VP2 protein that have commonly known antigenic residue in position 133 [68].

Helper T cell epitopes were identified in multiple FMDV proteins, with abundance in conserved sequences of VP4 and non-structural proteins [68]. A specific epitope in the VP4 protein was identified to have binding affinity to four different BoLA alleles [69, 70]. Being a very conserved protein among structural proteins, VP4 was beyond the scope of this study, as we were looking to enhance the FMDV vaccine with different epitopes with higher antigenicity and conservancy within FMDV isolates. A recent study was able to generate memory HTL from FMDV variable structure proteins [71]. Epitopes presented by the study were predicted in the present work, yet their binding affinity and percentile rank were lower than our selection threshold. Two epitopes, VP1(560–575) and VP1(561–576) were predicted with the required antigenicity and allergenicity from serotype A. Those epitope ranges were following experimentally identified epitopes [73]. HTL epitopes were also identified in non-structural proteins [72, 73], however, the inclusion of these epitopes into the predicted vaccine would decrease the efficiency of differentiating infected from vaccinated animals (DIVA).

For cytotoxic T-cell epitopes, fewer reports were able to identify FMDV peptides for bovine and swine MHC-1. In a study focused on identifying swine MHC-1 epitopes, authors identified high binding affinity between swine leukocyte antigen-2 and epitopes from VP1 protein [74]. Herein, we identified a similar position with high conservancy and antigenicity in the VP1 region (602–611) that has a binding affinity toward BoLA-2:00801. BoLA-2:00801 was also identified as the second most frequent MHC-1 allele in Holsten cattle [75]. Other epitopes were predicted in VP2 and VP3 proteins that have a binding affinity toward MHC-I. Such epitopes as VP3 (402–410), VP2 (256–264), and VP2 (275–283) epitopes were reported to induce T-cell proliferation [67]. The presence of different epitopes that can activate cytotoxic response indicates the need for further investigation of different structural protein (VP1-VP3) peptides in wet lab.

This study has faced two major limitations in predicting and selecting effective epitopes for FMDV. First, there are limited prediction tools that align the protein of interest against commonly identified BoLA alleles [76]. Second, since FMDV presents a high genomic variation between its serotypes and even within each serotype, this variation affects epitope selection. This variation decreased the chance of selecting common epitopes for FMDV serotypes. It is important to notice using different FMDV proteins, strains, serotypes, or updated versions of the same or another web tool may yield different epitope predictions.

HBHA was used as an adjuvant, and has been tested experimentally with MEV against FMDV. HBHA has been identified as a TLR-4 agonist with no toxicity, strong immunogenicity, and the ability to induce dendritic cell maturation [77]. HBHA would increase the chance of MEV induction of cell-mediated immunity.

Adjuvant and predicted epitopes are connected through different types of linkers. EAAAK linker forms a rigid α-helix that provides a fixed distance between different domains, hence providing good separation between them [78, 79]. EAAAK linker was used to link adjuvant and B-cell epitopes. KK linker is one of the target sequences for lysosomal protease cathepsin B that works on the antigen presentation process [80]. Besides, the KK linker increases the immunogenicity of the MEV [81]. Likewise, AAY linker can effectively separate cytotoxic T-cell epitopes, as it targets different mammalian proteasomes [82]. GPGPG linkers could induce helper T lymphocytes, a critical characteristic for MEV [81, 83]. The presence of protein linkers is important to prevent and reduce the chance of junctional immunogenicity, ensures flexible structures and low rigidity [84–86]. The final construct contained a 6x His tag to facilitate the purification of the final protein.

Constructed vaccine physicochemical properties were evaluated using the ProtParam server [87]. The vaccine’s basic nature predicted by its theoretical pI and its low GRAVY score indicates the solubility of the vaccine within the physiological environment. Previous studies such as Foroutan et al. [88] and Azami et al. [89] have used the same tools for in-silico assessment of their designed vaccine that have proved to induce a cellular and humoral response in mice through laboratory validation. The physicochemical properties predicted for the FMDV vaccine in our study were comparable with these previous studies. However, our predicted vaccines have a higher aliphatic index ensuring more thermal stability of the protein. The high pI of the FMDV ensures the presence of positively charged residues capable of interacting with negatively charged cellular receptors [90]. Toll-like receptors are recognition molecules that can be found on different cell surfaces of many cell types. Toll-like receptors contain a leucine-rich repeat domain as an extracellular domain and an intracellular toll/IL-1 receptor-like (TIR) domain. FMDV was reported to interact with different Toll-like receptors [25, 91]. For instance, TLR-3,7 and 8 were reported to interact with FMDV endosomal RNA [92, 93], on the other hand, viral VP1 and VP3 proteins interact with TLR-2 and TLR-4, respectively. The interaction between FMDV and TLR-2 induces IL-6 production, while the interaction of VP3 and TLR-4 activates the immune response toward FMDV [94, 95]. In this study, we analyzed the interaction between the designed MEV and TLR-4 and TLR-2 using molecular docking analysis. The molecular docking with TLR-4 showed many interactions between the extracellular domain of TLR-4 and MHCII and part of MHCI epitopes. In the same manner, TLR-2 showed interaction with MHCI and MHCII epitopes. The analysis of dissociation and binding affinity energies exhibited a highly stable structure between the MEV and TLRs.

Molecular dynamic simulation revealed a distinct pattern of structural stability and flexibility of binding between the MEV construct and TLR-2/TLR-4. TLR-4 RMSD analysis showed greater stability of its complex with MEV compared to TLR-2, specifically during final phase of the simulation. RG and SASA proved that TLR-4 adopts more open and exposed conformation during interaction with the vaccine, while TLR-2 maintains a more compact structure. Furthermore, hydrogen bond and MMGBSA analysis demonstrated that TLR-4-vaccine complex forms stronger and more interactions with high binding affinity (-226.51 kcal/mol) compared to TLR-2 (-177.09 kcal/mol). These results suggest that the MEV construct would have stronger and more dynamic interaction with TLR-4. This finding supports its potential role in activating TLR-4-mediated immunity.

The use of immunoinformatics and in-silico design is an important step in evaluating and predicting a new candidate vaccine, especially for highly heterogenic viruses such as FMDV. Including in-silico evaluation of vaccine candidates is essential to avoid time and financial-consuming in wet lab experimentations. Herein, we proposed a possible pipeline for designing the FMDV multiepitope vaccine. The multiepitope vaccine contained B-cell and T-cell epitopes that can induce immune system reaction in both its format, cell-mediated and humoral immunity. This approach has been applied in the design of vaccines against different pathogens such as SARS-CoV-2 [54], lumpy skin virus [25], avian leukosis virus [96], human papillomavirus [97] and Acinetobacter baumannii [98].

Conclusion

The application of vaccine strategies for eradicating FMDV infections showed its effect in many countries worldwide. The process requires the application of an effective vaccine to eliminate circulating strains and prevent the introduction of new ones. In this study, we have used an immunoinformatic approach for the design and evaluation of a polyvalent multiepitope vaccine for FMDV. The introduced pipeline can be generalized to different FMDV serotypes, strains, or isolates based on circulating strains. The final construct presented in this study could protect against seven isolates of three FMDV serotypes. Multiple computational tools, including discontinuous B-cell epitope prediction, molecular docking, and molecular dynamic simulation, predicted the stability and the strong interaction of the vaccine with bovine TLR-4 and TLR-2. The physicochemical properties of the vaccine underscore its potential as a promising candidate for further experimental validation. Future work should focus on testing the proposed vaccine experimentally to ensure its protective efficacy and the capability to induce humoral and cellular immune responses.

Data availability

Additional file can be found in the Git-Hub repository through this link: https://github.com/mustafarzaher/A-novel-immunoinformatic-approach-for-design-and-evaluation-of-hepta-valent-multiepitope-FMDV.git.

Abbreviations

FMDV:

Foot-and-mouth disease virus

FMD:

Foot-and-mouth disease

MEV:

Multi-epitope vaccine

ML:

Machine learning

NCBI:

National center of biotechnology information

CTL:

Cytotoxic T Lymphocytes

HTL:

Helper T Lymphocytes

MHC:

Major histocompatibility

IEDB:

Immune epitope database

ANN:

Artificial neural network

BoLA:

Bovine leukocyte alleles

EA-3:

Egyptian East Africa 3

TLR:

Toll-like receptor

HBHA:

Mycobacterium tuberculosis heparin-binding hemagglutinin

References

  1. Arzt J, Baxt B, Grubman MJ, Jackson T, Juleff N, Rhyan J, et al. The pathogenesis of Foot-and-Mouth Disease II: viral pathways in swine, small ruminants, and Wildlife; myotropism, chronic syndromes, and molecular virus–host interactions. Transbound Emerg Dis. 2011;58:305–26.

    Article  CAS  PubMed  Google Scholar 

  2. Arzt J, Juleff N, Zhang Z, Rodriguez LL. The pathogenesis of Foot-and-Mouth Disease I: viral pathways in cattle. Transbound Emerg Dis. 2011;58:291–304.

    Article  CAS  PubMed  Google Scholar 

  3. Belsham GJ. Towards improvements in foot-and-mouth disease vaccine performance. Acta Vet Scand. 2020;62:20.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Knight-Jones TJD, Rushton J. The economic impacts of foot and mouth disease - what are they, how big are they and where do they occur? Prev Vet Med. 2013;112:161–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Belsham GJ, Jamal SM, Tjørnehøj K, Bøtner A. Rescue of Foot-and-Mouth Disease viruses that are pathogenic for cattle from preserved viral RNA samples. PLoS ONE. 2011;6:e14621.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Belsham GJ, Bostock CJ. Studies on the infectivity of foot-and-mouth disease virus RNA using microinjection. J Gen Virol. 1988;69(Pt 2):265–74.

    Article  CAS  PubMed  Google Scholar 

  7. Danielle D, Alison B, Stuart C, Stephen B, HM J, HI R, et al. Foot-and-Mouth Disease Virus forms a highly stable, EDTA-Resistant complex with its principal receptor, integrin αvβ6: implications for infectiousness. J Virol. 2008;82:1537–46.

    Article  Google Scholar 

  8. Bachrach HL. Foot-and-Mouth Disease. Annu Rev Microbiol. 1968;22:201–44.

    Article  CAS  PubMed  Google Scholar 

  9. Hagag NM, Hassan AM, Zaher MR, Elnomrosy SM, Shemies OA, Hussein HA, et al. Molecular detection and phylogenetic analysis of newly emerging foot-and-mouth disease virus type A, Lineage EURO-SA in Egypt in 2022. Virus Res. 2023;323:198960.

    Article  CAS  PubMed  Google Scholar 

  10. Yoon SH, Park W, King DP, Kim H. Phylogenomics and molecular evolution of foot-and-mouth disease virus. Mol Cells. 2011;31:413–21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Korboukh VK, Lee CA, Acevedo A, Vignuzzi M, Xiao Y, Arnold JJ, et al. RNA Virus Population Diversity, an optimum for maximal fitness and virulence. J Biol Chem. 2014;289:29531–44.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Bachanek-Bankowska K, Di Nardo A, Wadsworth J, Mioulet V, Pezzoni G, Grazioli S, et al. Reconstructing the evolutionary history of pandemic foot-and-mouth disease viruses: the impact of recombination within the emerging O/ME-SA/Ind-2001 lineage. Sci Rep. 2018;8:14693.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Jamal SM, Nazem Shirazi MH, Ozyoruk F, Parlak U, Normann P, Belsham GJ. Evidence for multiple recombination events within foot-and-mouth disease viruses circulating in West Eurasia. Transbound Emerg Dis. 2020;67:979–93.

    Article  CAS  PubMed  Google Scholar 

  14. Carrillo C, Tulman ER, Delhon G, Lu Z, Carreno a. Comparative genomics of Foot-and-Mouth Disease Virus. J Virol. 2005;79:6487–504.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Kenubih A. Foot and mouth disease vaccine development and challenges in inducing long-lasting immunity: trends and current perspectives. Vet Med Res Rep. 2021;12:205–15.

    Google Scholar 

  16. Paton DJ, Valarcher JF, Bergmann I, Matlho OG, Zakharov VM, Palma EL, et al. Selection of foot and mouth disease vaccine strains–a review. Rev Sci Tech. 2005;24:981–93.

    Article  CAS  PubMed  Google Scholar 

  17. Mahapatra M, Parida S. Foot and mouth disease vaccine strain selection: current approaches and future perspectives. Expert Rev Vaccines. 2018;17:577–91.

    Article  CAS  PubMed  Google Scholar 

  18. Dubey KK, Luke GA, Knox C, Kumar P, Pletschke BI, Singh PK, et al. Vaccine and antibody production in plants: developments and computational tools. Brief Funct Genomics. 2018;17:295–307.

    Article  CAS  PubMed  Google Scholar 

  19. Orozco A, Morera J, Jiménez S, Boza R. A review of bioinformatics training applied to research in molecular medicine, agriculture and biodiversity in Costa Rica and Central America. Brief Bioinform. 2013;14:661–70.

    Article  PubMed  Google Scholar 

  20. Seib KL, Dougan G, Rappuoli R. The key role of genomics in modern vaccine and drug design for emerging infectious diseases. PLoS Genet. 2009;5:e1000612–1000612.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Paladino A, Marchetti F, Rinaldi S, Colombo G. Protein design: from computer models to artificial intelligence. WIREs Comput Mol Sci. 2017;7:e1318.

    Article  Google Scholar 

  22. Sinha S, Kuo C-Y, Ho JK, White PJ, Jazayeri JA, Pouton CW. A suicidal strain of Listeria monocytogenes is effective as a DNA vaccine delivery system for oral administration. Vaccine. 2017;35:5115–22.

    Article  CAS  PubMed  Google Scholar 

  23. Boehm KM, Bhinder B, Raja VJ, Dephoure N, Elemento O. Predicting peptide presentation by major histocompatibility complex class I: an improved machine learning approach to the immunopeptidome. BMC Bioinformatics. 2019;20:7.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Fathollahi M, Fathollahi A, Motamedi H, Moradi J, Alvandi A, Abiri R. Silico vaccine design and epitope mapping of New Delhi metallo-beta-lactamase (NDM): an immunoinformatics approach. BMC Bioinformatics. 2021;22:458.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Kar PP, Araveti PB, Kuriakose A, Srivastava A. Design of a multi-epitope protein as a subunit vaccine against lumpy skin disease using an immunoinformatics approach. Sci Rep. 2022;12:19411.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. María RR, Arturo CJ, Alicia JA, Paulina MG, Gerardo AO. The impact of bioinformatics on vaccine design and development. In: Arturo CJ (ed). Vaccines. InTech, Rijeka; 2017. Ch. 7, pp 123–46. http://dx.doi.org/10.5772/intechopen.69273.

    Google Scholar 

  27. Qiu J, Qiu T, Dong Q, Xu D, Wang X, Zhang Q, et al. Predicting the antigenic relationship of Foot-and-Mouth Disease Virus for Vaccine Selection through a computational model. IEEE/ACM Trans Comput Biol Bioinform. 2021;18:677–85.

    Article  CAS  PubMed  Google Scholar 

  28. Rahman T, Mahapatra M, Laing E, Jin Y. Evolutionary non-linear modelling for selecting vaccines against antigenically variable viruses. Bioinformatics. 2015;31:834–40.

    Article  CAS  PubMed  Google Scholar 

  29. Reeve R, Blignaut B, Esterhuysen JJ, Opperman P, Matthews L, Fry EE et al. Sequence-based prediction for vaccine strain selection and identification of antigenic variability in foot-and-mouth disease virus. PLoS Comput Biol. 2010;6.

  30. Reynisson B, Alvarez B, Paul S, Peters B, Nielsen M. NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Res. 2020;48:W449–54.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Alvarez B, Reynisson B, Barra C, Buus S, Ternette N, Connelley T, et al. NNAlign_MA; MHC peptidome deconvolution for accurate MHC binding motif characterization and improved T-cell epitope predictions. Mol Cell Proteom. 2019;18:2459–77.

    Article  CAS  Google Scholar 

  32. Jensen KK, Andreatta M, Marcatili P, Buus S, Greenbaum JA, Yan Z, et al. Improved methods for predicting peptide binding affinity to MHC class II molecules. Immunology. 2018;154:394–406.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Saha S, Raghava GPS. Prediction of continuous B-cell epitopes in an antigen using recurrent neural network. Proteins Struct Funct Bioinform. 2006;65:40–8.

    Article  CAS  Google Scholar 

  34. Doytchinova IA, Flower DR. Identifying candidate subunit vaccines using an alignment-independent method based on principal amino acid properties. Vaccine. 2007;25:856–66.

    Article  CAS  PubMed  Google Scholar 

  35. Flower DR, Doytchinova I, Zaharieva N, Dimitrov I. Immunogenicity prediction by VaxiJen: a ten year overview. J Proteom Bioinform. 2017;10.

  36. Dimitrov I, Bangov I, Flower DR, Doytchinova I. AllerTOP v.2—a server for in silico prediction of allergens. J Mol Model. 2014;20:2278.

    Article  PubMed  Google Scholar 

  37. Gupta S, Kapoor P, Chaudhary K, Gautam A, Kumar R, Consortium OSDD, et al. In Silico Approach for Predicting toxicity of peptides and proteins. PLoS ONE. 2013;8:e73957.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Bui H-H, Sidney J, Li W, Fusseder N, Sette A. Development of an epitope conservancy analysis tool to facilitate the design of epitope-based diagnostics and vaccines. BMC Bioinformatics. 2007;8:361.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Baek M, DiMaio F, Anishchenko I, Dauparas J, Ovchinnikov S, Lee GR et al. Accurate prediction of protein structures and interactions using a three-track neural network. Science (1979). 2021;373:871–6.

  40. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018;46:W296–303.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Heo L, Park H, Seok C. GalaxyRefine: protein structure refinement driven by side-chain repacking. Nucleic Acids Res. 2013;41:W384–8.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Lee GR, Heo L, Seok C. Effective protein model structure refinement by loop modeling and overall relaxation. Proteins Struct Funct Bioinform. 2016;84:293–301.

    Article  Google Scholar 

  43. Lovell SC, Davis IW, Arendall WB, de Bakker PIW, Word JM, Prisant MG, et al. Structure validation by Cα geometry: ϕ,ψ and Cβ deviation. Proteins Struct Funct Bioinform. 2003;50:437–50.

  44. Wiederstein M, Sippl MJ. ProSA-web: interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Res. 2007;35 suppl2:W407–10.

    Article  Google Scholar 

  45. Desta IT, Porter KA, Xia B, Kozakov D, Vajda S. Performance and its limits in rigid body protein-protein docking. Structure. 2020;28:1071–e10813.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Xue LC, Rodrigues JP, Kastritis PL, Bonvin AM, Vangone A. PRODIGY: a web server for predicting the binding affinity of protein–protein complexes. Bioinformatics. 2016;32:3676–8.

    Article  CAS  PubMed  Google Scholar 

  47. Laskowski RA, Jabłońska J, Pravda L, Vařeková RS, Thornton JM. PDBsum: structural summaries of PDB entries. Protein Sci. 2018;27:129–34.

    Article  CAS  PubMed  Google Scholar 

  48. Abraham MJ, Murtola T, Schulz R, Páll S, Smith JC, Hess B, et al. GROMACS: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX. 2015;1–2:19–25.

    Article  Google Scholar 

  49. Němec T. Nucleation parameters of SPC/E and TIP4P/2005 water vapor measured in NPT molecular dynamics simulations. J Mol Model. 2022;28.

  50. Durrant JD, McCammon JA. Molecular dynamics simulations and drug discovery. BMC Biol. 2011;9:1–9.

    Article  Google Scholar 

  51. Bailey AG, Lowe CP. Milch shake: an efficient method for constraint dynamics applied to alkanes. J Comput Chem. 2009;30:2485–93.

    Article  CAS  PubMed  Google Scholar 

  52. Weng G, Wang E, Wang Z, Liu H, Zhu F, Li D, et al. HawkDock: a web server to predict and analyze the protein–protein complex based on computational docking and MM/GBSA. Nucleic Acids Res. 2019;47:W322–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Wilson CC, Palmer B, Southwood S, Sidney J, Higashimoto Y, Appella E, et al. Identification and antigenicity of broadly cross-reactive and conserved human immunodeficiency virus type 1-Derived helper T-Lymphocyte epitopes. J Virol. 2001;75:4195–207.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Kar T, Narsaria U, Basak S, Deb D, Castiglione F, Mueller DM, et al. A candidate multi-epitope vaccine against SARS-CoV-2. Sci Rep. 2020;10:10895.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Zhang L, Zhang J, Chen H, Zhou J, Ma L, Ding Y, et al. Research in advance for FMD novel vaccines. Virol J. 2011;8:268.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Park J-H. Requirements for improved vaccines against foot-and-mouth disease epidemics. Clin Exp Vaccine Res. 2013;2:8–18.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Chathuranga WAG, Hewawaduge C, Nethmini NAN, Kim T-H, Kim JH, Ahn Y-H, et al. Efficacy of a novel multiepitope vaccine candidate against foot-and-mouth disease virus serotype O and A. Vaccines (Basel). 2022;10.

  58. Jo H, Kim BY, Park SH, Kim HM, Shin SH, Hwang SY, et al. The HSP70-fused foot-and-mouth disease epitope elicits cellular and humoral immunity and drives broad-spectrum protective efficacy. NPJ Vaccines. 2021;6:42.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Hardham JM, Krug P, Pacheco JM, Thompson J, Dominowski P, Moulin V, et al. Novel foot-and-mouth disease vaccine platform: formulations for safe and DIVA-compatible FMD vaccines with improved potency. Front Veterinary Sci. 2020;7.

  60. Cherryholmes GA, Stanton SE, Disis ML. Current methods of epitope identification for cancer vaccine design. Vaccine. 2015;33:7408–14.

    Article  CAS  PubMed  Google Scholar 

  61. Kazi A, Chuah C, Majeed ABA, Leow CH, Lim BH, Leow CY. Current progress of immunoinformatics approach harnessed for cellular- and antibody-dependent vaccine design. Pathog Glob Health. 2018;112:123–31.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Liu F, Wang X, Zheng M, Xiong F, Liu X, Zhou L, et al. Immunization with DNA prime-subunit protein boost strategy based on influenza H9N2 virus conserved matrix protein M1 and its epitope screening. Sci Rep. 2020;10:4144.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Ahmed RMS, Almofti YA, Abd-elrahman KA. Analysis of foot and mouth disease virus polyprotein for multi peptides vaccine design: an in silico strategy. J Pure Appl Microbiol. 2022;16:2083–98.

    Article  Google Scholar 

  64. Bhutta MS, Awais M, Sadaqat S, Zanchi FB, Shahid N, Qayyum Rao A. A novel immunoinformatics approach for developing a poly-epitope vaccine targeting foot and mouth disease virus, exploiting structural VP proteins. J Biomol Struct Dyn. 2024;1–17.

  65. Abubakar M, Syed Z, Manzoor S, Arshed MJ. Deciphering molecular dynamics of Foot and Mouth Disease Virus (FMDV): a looming threat to Pakistan’s dairy industry. Dairy. 2022;3:123–36.

    Article  Google Scholar 

  66. Kristensen T, Belsham GJ. Identification of a short, highly conserved, motif required for picornavirus capsid precursor processing at distal sites. PLoS Pathog. 2019;15:e1007509.

    Article  PubMed  PubMed Central  Google Scholar 

  67. Pérez Filgueira M, Wigdorovitz A, Romera A, Zamorano P, Borca MV, Sadir AM. Detection and characterization of functional T-Cell epitopes on the structural proteins VP2, VP3, and VP4 of Foot and Mouth Disease Virus O1 Campos. Virology. 2000;271:234–9.

    Article  PubMed  Google Scholar 

  68. Forner M, Cañas-Arranz R, Defaus S, de León P, Rodríguez-Pulido M, Ganges L, et al. Peptide-based vaccines: Foot-and-Mouth Disease Virus, a paradigm in Animal Health. Vaccines (Basel). 2021;9:477.

    Article  CAS  PubMed  Google Scholar 

  69. Gerner W, Denyer MS, Takamatsu H-H, Wileman TE, Wiesmüller K-H, Pfaff E, et al. Identification of novel foot-and-mouth disease virus specific T-cell epitopes in c/c and d/d haplotype miniature swine. Virus Res. 2006;121:223–8.

    Article  CAS  PubMed  Google Scholar 

  70. Rangel G, Martín V, Bárcena J, Blanco E, Alejo A. An adenovirus vector expressing FMDV RNA polymerase combined with a chimeric VLP harboring a neutralizing epitope as a prime boost strategy to induce FMDV-specific humoral and cellular responses. Pharmaceuticals (Basel). 2021;14.

  71. Mitoma S, Carr BV, Harvey Y, Moffat K, Sekiguchi S, Charleston B, et al. The detection of long-lasting memory foot-and-mouth disease (FMD) virus serotype O-specific CD4 + T cells from FMD-vaccinated cattle by bovine major histocompatibility complex class II tetramer. Immunology. 2021;164:266–78.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  72. Blanco E, Garcia-Briones M, Sanz-Parra A, Gomes P, De Oliveira E, Valero ML, et al. Identification of T-cell epitopes in nonstructural proteins of foot-and-mouth disease virus. J Virol. 2001;75:3164–74.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  73. Garcı́a-Briones MM, Russell GC, Oliver RA, Tami C, Taboga O, Carrillo E, et al. Association of bovine DRB3 alleles with immune response to FMDV peptides and protection against viral challenge. Vaccine. 2000;19:1167–71.

    Article  PubMed  Google Scholar 

  74. Feng L, Gao Y-Y, Sun M, Li Z-B, Zhang Q, Yang J et al. The parallel presentation of two functional CTL epitopes derived from the O and Asia 1 serotypes of foot-and-mouth disease virus and swine SLA-2*HB01: implications for universal vaccine development. Cells. 2022;11.

  75. Pandya M, Rasmussen M, Hansen A, Nielsen M, Buus S, Golde W, et al. A modern approach for epitope prediction: identification of foot-and-mouth disease virus peptides binding bovine leukocyte antigen (BoLA) class I molecules. Immunogenetics. 2015;67:691–703.

    Article  CAS  PubMed  Google Scholar 

  76. Pyasi S, Sharma V, Dipti K, Jonniya NA, Nayak D. Immunoinformatics approach to design multi-epitope-subunit vaccine against bovine ephemeral fever disease. Vaccines (Basel). 2021;9:1–20.

    Google Scholar 

  77. Lei Y, Shao J, Ma F, Lei C, Chang H, Zhang Y. Enhanced efficacy of a multi-epitope vaccine for type A and O footand-mouth disease virus by fusing multiple epitopes with Mycobacterium tuberculosis heparin-binding hemagglutinin (HBHA), a novel TLR4 agonist. Mol Immunol. 2020;121:118–26.

    Article  CAS  PubMed  Google Scholar 

  78. Arai R, Ueda H, Kitayama A, Kamiya N, Nagamune T. Design of the linkers which effectively separate domains of a bifunctional fusion protein. Protein Eng Des Selection. 2001;14:529–32.

    Article  CAS  Google Scholar 

  79. Chen X, Zaro JL, Shen W-C. Fusion protein linkers: property, design and functionality. Adv Drug Deliv Rev. 2013;65:1357–69.

    Article  CAS  PubMed  Google Scholar 

  80. Yano A, Onozuka A, Asahi-Ozaki Y, Imai S, Hanada N, Miwa Y, et al. An ingenious design for peptide vaccines. Vaccine. 2005;23:2322–6.

    Article  CAS  PubMed  Google Scholar 

  81. Li X, Guo L, Kong M, Su X, Yang D, Zou M, et al. Design and evaluation of a Multi-epitope peptide of human metapneumovirus. Intervirology. 2016;58:403–12.

    Article  Google Scholar 

  82. Yang Y, Sun W, Guo J, Zhao G, Sun S, Yu H, et al. In silico design of a DNA-based HIV-1 multi-epitope vaccine for Chinese populations. Hum Vaccin Immunother. 2015;11:795–805.

    Article  PubMed  PubMed Central  Google Scholar 

  83. Livingston B, Crimi C, Newman M, Higashimoto Y, Appella E, Sidney J, et al. A rational strategy to Design Multiepitope Immunogens based on multiple th lymphocyte Epitopes1. J Immunol. 2002;168:5499–506.

    Article  CAS  PubMed  Google Scholar 

  84. Meza B, Ascencio F, Sierra-Beltrán AP, Torres J, Angulo C. A novel design of a multi-antigenic, multistage and multi-epitope vaccine against Helicobacter pylori: an in silico approach. Infect Genet Evol. 2017;49:309–17.

    Article  CAS  PubMed  Google Scholar 

  85. Shey RA, Ghogomu SM, Esoh KK, Nebangwa ND, Shintouo CM, Nongley NF, et al. In-silico design of a multi-epitope vaccine candidate against onchocerciasis and related filarial diseases. Sci Rep. 2019;9:4409.

    Article  PubMed  PubMed Central  Google Scholar 

  86. Bhattacharya M, Sharma AR, Patra P, Ghosh P, Sharma G, Patra BC, et al. A SARS-CoV-2 vaccine candidate: In-silico cloning and validation. Inf Med Unlocked. 2020;20:100394.

    Article  Google Scholar 

  87. Lewis J. The protein protocols handbook; 1997.

  88. Foroutan M, Ghaffarifar F, Sharifi Z, Dalimi A. Vaccination with a novel multi-epitope ROP8 DNA vaccine against acute Toxoplasma gondii infection induces strong B and T cell responses in mice. Comp Immunol Microbiol Infect Dis. 2020;69:101413.

    Article  PubMed  Google Scholar 

  89. Azami Z, Farahmand M, Kavousi M. A new multi-epitope DNA vaccine against Helicobacter Pylori infection in a BALB/c mouse model. Heliyon. 2024;10:e39433.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  90. Tokmakov AA, Kurotani A, Sato K-I. Protein pI and intracellular localization. Front Mol Biosci. 2021;8.

  91. Wang X, Luoreng Z, Xu S, Gao X, Li J, Ren H, et al. The structure and sequence analysis of TLR4 gene in cattle. Agric Sci China. 2009;8:632–7.

    Article  CAS  Google Scholar 

  92. Borrego B, Rodríguez-Pulido M, Revilla C, Álvarez B, Sobrino F, Domínguez J, et al. Synthetic RNAs mimicking structural domains in the Foot-and-Mouth Disease Virus Genome Elicit a Broad Innate Immune Response in Porcine cells triggered by RIG-I and TLR activation. Viruses. 2015;7:3954–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  93. Lannes N, Python S, Summerfield A. Interplay of foot-and-mouth disease virus, antibodies and plasmacytoid dendritic cells: virus opsonization under non-neutralizing conditions results in enhanced interferon-alpha responses. Vet Res. 2012;43:64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  94. Zhang J, Li D, Yang W, Wang Y, Li L, Zheng H. Foot-and-mouth disease virus VP3 protein acts as a critical proinflammatory factor by promoting toll-like receptor 4-mediated signaling. J Virol. 2021;95.

  95. Lin Y-T, Chen Y-P, Fang C-H, Huang P-Y, Liang S-M. Capsid proteins of foot-and-mouth disease virus interact with TLR2 and CD14 to induce cytokine production. Immunol Lett. 2020;223:10–6.

    Article  CAS  PubMed  Google Scholar 

  96. Elshafei SO, Mahmoud NA, Almofti YA. Immunoinformatics, molecular docking and dynamics simulation approaches unveil a multi epitope-based potent peptide vaccine candidate against avian leukosis virus. Sci Rep. 2024;14:2870.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  97. Guo N, Niu Z, Yan Z, Liu W, Shi L, Li C, et al. Immunoinformatics design and in vivo immunogenicity evaluation of a conserved CTL multi-epitope vaccine targeting HPV16 E5, E6, and E7 proteins. Vaccines (Basel). 2024;12:392.

    Article  CAS  PubMed  Google Scholar 

  98. Jeffreys S, Tompkins MP, Aki J, Papp SB, Chambers JP, Guentzel MN, et al. Development and evaluation of an Immunoinformatics-based Multi-peptide Vaccine against Acinetobacter baumannii infection. Vaccines (Basel). 2024;12:358.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

Authors would like to thank the three reviewers for their insightful comments which significantly improved the manuscript. We would like to thank Dr. Dalia M. El-Husseini for her valuable insights and comments.

Funding

Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). This research was supported by the Science, Technology, and Innovation Funding (STDF) Authority (No. 43225), STDF, Ministry of Scientific Research, Cairo, Egypt.

Author information

Authors and Affiliations

  1. Genome Research Unit, Animal Health Research Institute, Agriculture Research Center (ARC), Giza, 12618, Egypt

    Mostafa R. Zaher & Naglaa M. Hagag

  2. Department of Biotechnology, Faculty of Science, Cairo University, Giza, 12613, Egypt

    Mostafa R. Zaher & Azza M. El-Amir

  3. Reference Laboratory for Veterinary Quality Control on Poultry Production, Animal Health Research Institute, Agriculture Research Center (ARC), Giza, 12618, Egypt

    Mohamed H. El-Husseiny & Naglaa M. Hagag

  4. Department of Microbiology and Immunology, Faculty of Pharmacy, Ahram Canadian University, Giza, Egypt

    Mohamed E. El Zowalaty

  5. Department of Chemistry, Faculty of Science, Cairo University, Giza, 12613, Egypt

    Reham H. Tammam

Authors
  1. Mostafa R. Zaher
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Mohamed H. El-Husseiny
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Naglaa M. Hagag
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Azza M. El-Amir
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Mohamed E. El Zowalaty
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Reham H. Tammam
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

MRZ, AMA and NMH participated in the study conception and design. MHE and RHT study supervision. MRZ conducted data selection and retrieval. MRZ and MHE contributed Epitope prediction, data curation and tertiary structure analysis. MRZ conducted homology modeling and molecular docking. NMH, MHE and MEZ conducted result validation and project supervision, MHE and MEZ conducted funding acquisition, MRZ wrote the initial draft manuscript. MHE, AMA, RHT reviewed and edited the manuscript. MRZ, NMH, MEZ critically revised the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Mohamed E. El Zowalaty or Reham H. Tammam.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zaher, M.R., El-Husseiny, M.H., Hagag, N.M. et al. A novel immunoinformatic approach for design and evaluation of heptavalent multiepitope foot-and-mouth disease virus vaccine. BMC Vet Res 21, 152 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12917-025-04509-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12917-025-04509-1

Keywords

BMC Veterinary Research

ISSN: 1746-6148

Contact us