In Silico Design Of A Promiscuous Chimeric Multi-epitope Vaccine Against Mycobacterium Tuberculosis Part 1

Abstract

Tuberculosis (TB) is a global health threat, killing approximately 1.5 million people each year. The eradication of Mycobacterium tuberculosis, the main causative agent of TB, is increasingly challenging due to the emergence of extensive drug-resistant strains. Vaccination is considered an effective way to protect the host from pathogens, but the only clinically approved TB vaccine, Bacillus Calmette-Guérin (BCG), has limited protection in adults. Multi-epitope vaccines have been found to enhance immunity to diseases by selectively combining epitopes from several candidate proteins. 

Tuberculosis is an infectious disease caused by the bacterium Mycobacterium tuberculosis that mainly affects the lungs but may also affect other organs. Immunity plays an important role in the development and treatment of TB.

In the case of insufficient immunity, Mycobacterium tuberculosis can easily invade the human body and develop into tuberculosis. For example, a compromised immune system from certain diseases or medical treatments increases the risk of TB infection. In addition, factors such as malnutrition, poor quality of life, and excessive stress may also weaken immunity, leading to the development of TB.

Therefore, maintaining good health and strengthening immunity is very important to prevent tuberculosis. Immunity can be boosted through a healthy diet, regular lifestyle, proper exercise, and enough sleep, which can effectively reduce the risk of contracting various diseases, including tuberculosis.

Immunity is also important for treatment and recovery when TB has already occurred. The stronger the immune system, the more effectively the body can defend itself against an attack by the TB bacillus and produce enough antibodies to fight the disease. Therefore, maintaining adequate nutrition, proper exercise, and a positive attitude can improve immunity and promote recovery.

In conclusion, there is a strong relationship between immunity and TB. Maintaining good health and enhancing immunity is one of the important ways to prevent and treat tuberculosis. Let us face life positively, maintain a healthy lifestyle, and stay away from diseases. From this point of view, we need to improve our immunity. Cistanche can significantly improve immunity, because meat ash contains a variety of biologically active components, such as polysaccharides, two mushrooms, Huang Li, etc. These components can stimulate the immune system Various types of cells in the system, increase their immune activity.

Click health benefits of cistanche

This study aimed to design a multi-epitope vaccine against TB using an immuno-informatics approach. Through functional enrichment, we identified eight proteins secreted by M. tuberculosis that are either required for pathogenesis, secreted into extracellular space, or both. We then analyzed the epitopes of these proteins and selected 16 helper T lymphocyte epitopes with interferon-γ inducing activity, 15 cytotoxic T lymphocyte epitopes, and 10 linear Bcell epitopes, and conjugated them with adjuvant and Pan HLA DR-binding epitope (PADRE) using appropriate linkers. 

Moreover, we predicted the tertiary structure of this vaccine, its potential interaction with Toll-Like Receptor-4 (TLR4), and the immune response it might elicit. The results showed that this vaccine had a strong affinity for TLR4, which could significantly stimulate CD4+ and CD8+ cells to secrete immune factors and B lymphocytes to secrete immunoglobulins, to obtain good humoral and cellular immunity. Overall, this multi-epitope protein was predicted to be stable, safe, highly antigenic, and highly immunogenic, which has the potential to serve as a global vaccine against TB.

1. Introduction

Tuberculosis (TB), a highly contagious disease caused by Mycobacterium tuberculosis, is ranked by World Health Organization (WHO) as the top cause of death from a single infectious agent [1–3]. In 2021, the estimated number of TB deaths and new cases reached 1.6 million and 10.6 million, respectively [4]. Currently, the clinical treatment of TB is relatively scarce, and the combination of multiple antimicrobial drugs is mainly used.
This chemotherapy cycle is very long, usually taking nine to twelve months, or even longer [5], which increases the risk of drug-resistant mutations in M. tuberculosis [6,7]. In recent years, chemotherapy has become less effective because of the emergence and increasing proportion of multi-drug and extensively drug-resistant M. tuberculosis [6]. Preventing TB from developing may be more effective than treating it. Vaccination is well known to be an effective way to protect the host from pathogenic bacteria [8]. 

Currently, Bacillus Calmette-Guérin (BCG), developed over 100 years ago, is the only clinically approved TB vaccine [9]. Unfortunately, BCG only protects newborns and infants and is largely ineffective against adolescents and adults [2,10], although WHO reports that 89% of TB cases in 2021 were adults [4]. Therefore, there is an urgent need to develop a novel and effective anti-TB vaccine, especially for adolescents and adults.

TB vaccine development is complicated by multiple features of mycobacteria, such as latent infection, persistence, and immune evasion [11–13]. An ideal TB vaccine should be designed to target the proteins/pathways responsible for these properties in M. tuberculosis and be able to efficiently induce CD4+ and CD8+ T cell-mediated immune responses [14]. 

Moreover, an effective vaccine should also target the host's major histocompatibility complexes (MHC), which are highly polymorphic [15]. These characteristics put forward very high requirements for the versatility of the vaccine, which obviously cannot be achieved by a single natural protein. Multi-epitope vaccine, a recombinant protein consisting of a series of overlapping epitopes (peptides) [16], is a novel type of vaccine candidate that may address the above issues. 

In recent years, multi-epitope vaccines have attracted much attention due to their advantages of higher immunity and lower allergenicity than conventional vaccines [17,18]. Currently, multi-epitope vaccines have been designed against many pathogenic microorganisms, including Shigella spp. [19], foot-and-mouth disease virus [20], Helicobacter pylori [21,22], hepatitis B virus [23], Toxoplasma gondii [24], Leishmania infantum [25], Nipah virus [26], Onchocerca volvulus [27], Pseudomonas aeruginosa [28], and leukosis virus [29]. 

In particular, the emergence of the COVID-19 pandemic has strengthened the application of this technology [16,30–32]. As for TB, several multi-epitope vaccines have been designed to target inherently active TB [33–39] and latent TB [40,41]. Among them, three vaccine candidates were designed in the form of DNA [34,36,40], and two of them incorporated epitopes into the protein backbones to generate recombinant vaccines [34,36]. 

It should be noted that the candidate proteins for some of the above multi-epitope vaccines are randomly selected, and the population coverage of these vaccines requires further studies. 

Moreover, two multi-epitope TB vaccine candidates with broad population coverage were designed, one epitope was selected from immunogenic exosomes vesicle proteins with pathogenic properties [39], and the other does not focus on candidate proteins, but directly selects highly conserved and experimentally validated epitopes from the Immune Epitope Database (IEDB) [38]. However, these candidate proteins lack functional enrichment, and the ability of vaccine candidates to induce interferon-γ (IFN-γ) secretion remains to be improved.

A previous study has deduced that rational optimization of epitopes can be achieved by a combination of MHC binding capacity and the epitope’s ability to react with T cell receptors [42]. Furthermore, they predicted that vaccines with cytotoxic T lymphocyte (CTL) A1, A2, A3, A24, and B7 binding epitopes would have coverage of nearly 100% in the major ethnic groups (Blacks, Asians, Hispanics, and Caucasians). 

However, until now there has been no similar approach to design a TB vaccine. In this study, we designed a highly promiscuous multi-epitope TB vaccine using various antigenic features of eight function-enriched proteins. The chimeric vaccine candidate possesses 15 CTL epitopes, 16 helper T lymphocyte (HTL) epitopes with IFN-γ-inducing properties, and 10 linear B-cells epitopes. Immuno-informatics analysis demonstrated that this vaccine candidate was ‘all-encompassing’, making it a potential cornerstone to achieving the ‘The End TB strategy’.

2. Materials and method

2.1. Protein selection and sequence retrieval

To construct a multi-epitope vaccine against TB, we first selected proteins of the M. tuberculosis complex, which are deposited in the IEDB database [43] and have been validated as MHC class I and II binding epitopes. Amino acid sequences (primary structure) of proteins from the M. tuberculosis H37Rv strain were obtained from the UniProt database [44]. Alignment-independent predictions of prospective antigens based on physicochemical properties were obtained from the VaxiJen 2.0 server [45], which underwent automatic and cross-covariance (ACC) transformation of protein sequences into a uniform vector of major amino acid properties, with the antigenicity threshold set at 0.4 for each bacterial protein [45,46]. 

Functional annotation of proteins was assessed using Database for Annotation, Visualization, and Integrated Discovery (DAVID) 6.8 [47]. Secreted proteins were further enriched using two categories: extracellular space and pathogenesis through the DAVID and BioCyc [48] databases, respectively. The proteome of Homo sapiens GRCh38.p13 was downloaded in FASTA format from the National Centre for Biotechnology Information (NCBI) database [49]. BLASTp was used to predict homology (E-value =1e-5) between secreted proteins and H. sapiens proteins.

2.2. T-cell epitope prediction

Prediction and selection of epitopes are crucial steps in the construction of multi-epitope vaccines. MHC I molecules bind short peptides (9–11 amino acids) because the peptide-binding cleft of MHC I molecules consisting of a single α chain is closed [50]. The freely accessible NetMHCpan-4.1 [51] was used for CTL epitopes prediction, which uses NNAlign_MA to generate percentage ranks (% rank) based on a combination of MHC I binding affinities and eluted ligands. 

The "% Ranking" of a query sequence was determined by comparing its prediction score to the distribution of prediction scores for the relevant MHC calculated using a set of randomly chosen native peptides. Epitopes with a % ranking < 0.5% were considered strong binders, while epitopes with a % ranking < 2% were considered weak binders [51]. 

Although up to 12 supertype MHC class I epitopes can be predicted on the server, we only used A1, A2, A3, A24, and B7 because these five supertypes cover 100% of the major human races [42]. We selected strong binders and predicted their antigenicity using VaxiJen2.0 [45], then, we predicted class I immunogenicity using the International Epitope Database (IEDB) [52], which uses 3-fold cross-validation. 

Finally, we arranged epitopes that were both antigenic and immunogenic according to % ranking and selected 15 low-scoring epitopes, three for each supertype and at least one for each candidate protein, except for a candidate protein that could not have a strong CTL binding epitope that is antigenic and immunogenic. Finally, IC50 values for each CTL epitope were predicted from NetMHC-4.0 [53].
Class II MHC molecules bind to antigenic peptides, and the resulting complex can be recognized by HTL. Typically, antigenic peptides range in length from 12 to 20 amino acid residues, but peptides between 13 and 16 residues in length are frequently observed [54]. 

The 15-mers were the most abundant MHC II epitopes for M. tuberculosis and have been deposited in IEDB. As a result, we used the NetMHCIIpan-4.0 [51,55] to predict the binding of 15-mer peptides to Human Leukocyte Antigen-DR (HLA-DR), HLADQ, HLA-DP, and H-2–1 allele. The prediction was also based on NNAlign_MA with a % ranking of < 2% and < 10% considered as strong and weak binders, respectively [51]. 

Also, we predicted 15-mer IFNγ inducing epitopes for candidate proteins using the IFNepitope server [56], which uses a support vector machine hybrid approach that allows virtual screening of IFN-γ-inducing peptide/epitope in a peptide library consisting of IFN-γ-inducible and non-inducible MHC II binders that activate T-helper cells. We then predicted the antigenicity of the IFN-γ inducing epitopes [45], and finally, we selected the 16 most promiscuous epitopes that were strong MHC-II binding, IFN-γ inducing, and antigenic.

It is important to note that signal peptides were removed from candidate proteins before the epitope prediction. In this study, signal peptides were screened using SignalP 5.0 [57] and TargetP2.0 [58].

2.3. Linear B-cell epitope prediction

Linear B-cells epitopes (16-mers) were predicted using ABCpred [59,60] with a default threshold of 0.51. Moreover, to increase the reliability of the prediction results, we also used BepiPred 2.0 [61] to predict linear B-cells epitopes. Epitopes obtained from these two software were further subjected to antigenicity prediction using VaxiJen2.0 [45]. Finally, we selected ten linear B-cell epitopes based on high ABCpred scores and antigenicity, with at least one epitope selected for each candidate protein.

2.4. Construction of the multi-epitope vaccine candidate with chimeric properties

The designed multi-epitope vaccine contains one HBHA (heparin-binding hemagglutinin) adjuvant, one Pan HLA DR-binding epitope (PADRE), 15 CTL, 16 HTL, 10 linear B-cells epitopes, and one His× 6 tag (Fig. 3). Linkers were used to join epitopes, prevent the production of junction epitopes, and enhance the procession and regeneration of individual epitopes in chimeric vaccines [62]. 

For the construction of this vaccine candidate, the HBHA adjuvant (UniProt ID: P9WIP9) was located at the N-terminus and linked to the downstream PADRE via an EAAAK linker. Then, the HTL epitopes joined by the GPGPG linkers were linked to PADRE. Moreover, CTL epitopes joined by the AAY linker were connected to HTL epitopes via the HEYGAEALERAG linker, which also joined the CTL epitope unit to linear B-cell epitopes linked using KK linkers. Finally, a His× 6 tag was attached to the C-terminus of the chimeric protein.

2.5. Antigenicity, allergenicity, and physicochemical properties

The antigenicity of the multiple-epitope vaccine and the eight component proteins were predicted by the VaxiJen 2.0 server [45], while the allergenicity of these proteins was predicted by the AllerTOP 2.0 server [63]. AllerTOP 2.0 uses amino acid E-descriptors, ACC transformation of protein sequences, and k-nearest neighbors (kNN) for allergen classification. 

The method achieved 85.3% accuracy with 5-fold cross-validation. For the prediction of physicochemical properties such as half-life, isoelectric point, instability index, aliphatic index, and grand average of hydropathicity (GRAVY) of this multiple-epitope vaccine, the ExPASy ProtParam server [64] was used.
Further, the solubility of multi-epitope vaccine peptide was assessed using the proteinSol (PROSO II) server [65] based on a classifier exploiting the subtle differences between the well-known insoluble proteins from TargetDB and the soluble proteins from both TargetDB and PDB [66]. When evaluated using 10-fold cross-validation, it achieved 71.0% accuracy (area under ROC curve = 0.785).

2.6. Immune simulation

To characterize the immune response profile and immunogenicity of the vaccine, in silico immune simulations were performed using the C-ImmSim server [67]. C-ImmSim predicts immune interactions using position-specific scoring matrices derived from machine learning techniques for peptide prediction.
It concurrently simulates three compartments representing three separate anatomical regions found in mammals: (i) the bone marrow, where hematopoietic stem cells were simulated to produce new lymphocytes and myeloid cells; (ii) the thymus, where naive T cells were selected to avoid autoimmunity; and (iii) the lymphatic organ such as lymph nodes. 

To effectively prime and boost the vaccine, we followed the approach of [68] where two injections were administered four weeks apart. All simulation parameters were set to default values, with time steps set to 10 and 94 (each time step is eight hours).

2.7. Disordered region prediction

Intrinsically disordered regions (IDRs) are present in many proteins. The disordered region was predicted using DISOPRED3 [69], which uses DISOPRED2 and two other machine-learning-based modules trained on large IDRs to identify disordered residues. They were then anno

2.8. Secondary and tertiary structure prediction

The secondary structure of the designed vaccine was predicted by the PSIPRED 4.0 server [70], which first uses PSI-BLAST to identify sequences closely related to the query protein. The tertiary structure of this vaccine was predicted using the Iterative Threading Assembly Refinement (I-TASSER) server [71]. 

There are four key steps in ITASSER modeling; a) threading template identification; b) iterative structure assembly simulation; c) model selection and refinement; and d) structure-based functional annotation [72,73]. 

I-TASSER generated five models, which were screened using ProSA-web [74], and the model with the lowest Z-score was selected for further refinement. ProSA-web compares the model scores obtained from experimentally verified structures deposited in PDB. A local quality score plot helps identify problematic areas in the model, and the same scores were represented using a color code on the presentation of the 3D structure. This is useful for early structural determination and refinement.

2.9. Tertiary structure refinement

The “coarse” 3D model of the vaccine candidate obtained by ITASSER was refined in two steps using two servers; first with ModRefiner [75] followed by GalaxyRefine [76]. ModRefiner uses Cα traces to affect the construction and refinement of proteins obtained by two-step atomic-level energy minimization. 

First, the Cα traces were used to construct the main chain, followed by the refinement of side chain rotamers and backbone atoms using physics- and knowledge-based composite force fields. GalaxyRefine utilizes multiple templates to generate reliable core structures, while unreliable loops or terminals were generated by optimization-based modeling.

2.10. Tertiary structure validation

The refined structure of the vaccine candidate was validated by Ramachandran plots generated from the PROCHECK [77] and MolProbity [78] databases. Ramachandran plots evaluate the backbone conformation of proteins by dividing amino acid residues into two regions: allowed and disallowed. PROCHECK utilizes stereochemistry to assess the net quality of protein structures by comparing them to the refined structures at the same resolution and then presenting regions requiring further analysis. 

Molprobity validates local and global macromolecule (proteins and nucleic acids) models by a mix of X-ray, NMR, computational, and cryoEM criteria [79]. The power and sensitivity to optimize hydrogen placement and all-atom contact analysis are widely used in an updated version of the covalent geometry and torsion angle criteria [80].

2.11. Discontinuous B-cell epitopes

Discontinuous B-cell epitopes in the native protein structure were predicted using ElliPro [81]. ElliPro implements three algorithms to approximate protein shape as an ellipsoid, calculates the residue protrusion index (PI), and clusters neighboring residues based on their PI values. ElliPro provides each output epitope with a score described as the averaged PI value for the epitope residue. An ellipsoid with a PI value of 0.9 consists of 90% of the contained protein residues, while the remaining 10% of residues lie outside the ellipsoid. For each epitope residue, the PI value is calculated from the center of mass of the residue lying outside the largest possible ellipsoid.

2.12. Molecular docking of chimeric proteins

Molecular docking of the designed vaccine (ligand) with Toll-Like Receptor-4 (TLR4) (PDB ID: 3FXI) immune receptor was performed using Patchdock [82]. The top 10 models were then refined using FireDock [83]. PatchDock replaces the Connolly dot surface representation of the molecules with concave, convex, and flat patches. 

The models were then scored based on geometric fit and atomic desolvation. [82]. FireDock optimizes side chain conformations and orientation of the rigid body and generates an output of a 3D refined complex based on the binding energy [83]. We selected the first model of Firedock based on global energy as the docking complex. Finally, the binding energy and dissociation content within the docking complex was predicted using the PRODIGY server [84].

2.13. Molecular dynamics simulation

Molecular dynamic simulations were performed on proteins using the fast and freely accessible web server, internal coordinates normal mode analysis server (iMODS) [85], and consistent and optimal docking results were obtained from the PatchDock-FireDock server. In internal coordinates, Normal Mode Analysis (NMA) generates collective motions critical for macromolecular function. iMODS presents mechanisms for exploring these modes as vibration analysis, motion animations, and morphing trajectories that were carried out almost interactively at different resolutions [85].

2.14. Reverse translation, codon optimization, and in silico cloning of the vaccine

To effectively express the vaccine candidate in Escherichia coli cells, cDNA was generated in silico through codon optimization and reverse translation using the Java Codon Adaptation Tool (JCAT) [86]. 

Optimization involved (i) avoiding rho-independent transcriptional terminators, ii) avoiding prokaryotic ribosome binding sites, (iii) avoiding cleavage site of restriction enzymes NcoI and XhoI, which serves as N-terminal and C-terminal restriction sites for the insertion of cDNA template of vaccine, and (iv) only partial optimization to apply site-directed mutagenesis. Codon Adaptation Index (CAI) and GC content predicted the quality of the cDNA with an opal stop codon (TGA) inserted after the His× 6 tag. Then, the optimized DNA fragment of the chimeric vaccine candidate was integrated into the reverse strand of pET-28a(+) using the SnapGene tool [87].

For more information:1950477648nn@gmail.com