Immunoinformatics Prediction of an Epitope Based Peptide Vaccine for Neisseria Gonorrhea Dihydrolipoamide Acetyltransferase Protein

Sexually transmitted infections (STIs) such as Gonorrhea is associated with serious morbidity and mortality rates in the world considering the multiple virulence factors possessed. The disease is manifested as salpingitis, pelvic inflammatory disease (PID), and bacteremia and is controlled by macrophages, dendritic cells, neutrophils, T cells, epithelial cells and cytokines. Dihydroliponamide acetyltransferase, a component of the mitochondrial pyruvate complex can be used as immunogenic target. Recent changes in the strain allowed the bacteria to acquire resistance against antibiotics.Vaccination remains an alternative to prevention against the disease. This study predicts an effective epitope-based vaccine against dihydrolipoamide acetyltransferase of Neisseria Gonorrhea using immunoinformatics approaches. Sequences retrieved from NCBI were passed on several prediction tests to analyze for possible B-cell, T-cell MHC class I epitopes and class II. Two epitopes showed high binding affinity for B-cells, while thirteen epitopes showed high binding affinity for MHCI and forty-five for MHCII. A population coverage of 100% for combined MHC I and II dictates the huge number of individuals who will benefit from formulating the vaccine. We recommend invivo and invitro studies to prove our prediction results.


INTRODUCTION
Sexually transmitted infections(STIs) such as Gonorrhea is considered one of the major threat to the public health. It is also associated with serious morbidity and mortality [1][2][3]. Almost one million persons suffered from a curable STI daily [4]. Neisseria Gonorrhea which is the causative agent of Gonorrhea, is a gram-negative diplococcus. It poses a growing threat to the public health concern, as evident by the estimated 78 million infections per year [5].N. Gonorrhea infections usually start at the urogenital, rectal, and nasopharyngeal mucosa. The infection can ascend, causing salpingitis, pelvic inflammatory disease (PID), bacteremia, and it can cross the endothelial barrier even disseminated gonococcal infection (DGI) [6]. The pathogenesis of gonorrhea is regulated by macrophages, dendritic cells, neutrophils, T cells, epithelial cells and cytokines [7].N. gonorrhoeae has multiple virulence factors like lipooligosaccharide (LOS), type IV pili, opacity associated (Opa) proteins, and an outer membrane porin PorB, that among other roles will help the bacteria to survive in the presence of serum. [5]Nucleic acid amplification tests (NAAT) are the preferred method used for diagnosis of N. Gonorrhea infection. [8] Furthermore, Neisseria Gonorrhea is able to accumulate different mutations leading to the emergence of clinical isolates with clinically significant levels of resistance to currently or previously used antibiotics such as sulfadrugs, penicillins and tetracyclines which has led the WHO to consider adding N. Gonorrhea infections to the category of untreatable infections [9][10][11][12]. With the absence of effective vaccines, and the lack of prolonged immunity after infection due to multiple evading mechanisms, society relies on antibiotics to reduce the spread of the gonococcus in the community [6,13,14].
The aim of this study is to predict an effective epitope-based vaccine against dihydrolipoamide acetyltransferase enzyme of N. gonorrhea using immunoinformatics approaches. No previous reports were found for N. gonorrhea epitope-based vaccine so this is considered the first study to our knowledge to use insilico approach to design an epitope-based vaccine.
Literature showed that there are twenty one potentially immunogenic proteins in the N. Gonorrhea genome that could be used as immunogenic targets for vaccine design. Dihydrolipoamide acetyltransferase (DLAT) is one of those targets [15]. It is a component of the mitochondrial pyruvate complex (PDC), and is considered as a novel metabolic longevity factors for many species [16].
Vaccination is generally considered to be the most cost effective method of preventing infectious diseases [17]. Understanding the relation between epitope and antibody interaction is the key to design potent and safe vaccines, and effective new model of diagnostics [18,19]. These epitopes are capable of inducing B cell and T cell-mediated immunity [17]. Immunoinformatics research focuses on the design and study of algorithms for mapping possible B-and T-cell epitopes. This type of peptide-based vaccine is easier to construct and have a more chemical stability than traditional vaccines. Using such techniques(reverse vaccinology) to analyze the sequence areas with potential binding sites can lead to the development of new vaccines [20].

Protein Sequence Retrieval
Neisseria Gonorrhea owns a total of 11 strains were retrieved from National Center for Biotechnology Information (NCBI) database on July 2019 in FASTA format. The retrieved protein strains had length of 529 with name dihydrolipoamide acetyltransferase.

Determination of conserved regions
The retrieved sequences of Neisseria gonorrhea were showed by multiple sequence alignment (MSA) using ClustalW tool of BioEdit Sequence Alignment Editor Software version 7.2.5 to define the conserved regions.

Sequenced-Based Method
The reference sequence of Neisseria gonorrhea was (YP_207709.1). The reference sequence is submitted to different prediction tools at the Immune Epitope Database (IEDB) analysis resource (http://www.iedb.org/) to predict various B and T cell epitopes. The epitope for B and T cell would be filtered form the conserved epitope

B Cell Epitope Prediction
B cell epitopes is the portion of the vaccine that interacts with B lymphocytes. Candidate epitopes were analysed bya number of B cell prediction methods from IEDB (http://tools.iedb.org/bcell/), to identify the surface accessibility, antigenicity and hydrophilicity with the aid of random forest algorithm, a form of unsupervised learning. The Bepipred Linear Epitope Prediction 2 was used to predict linear B-cell epitope with default threshold value of 0.500 (http://tools.iedb.org/bcell/result/). The Emini Surface Accessibility Prediction tool was used to discover the surface accessibility with default threshold value of 1.000 (http://tools.iedb.org/bcell/result/). The Kolaskar and Tongaonker Antigenicity method was used to find the antigenic places of candidate epitope with default threshold value 1.037 (http://tools.iedb.org/bcell/result/). The Parker Hydrophilicity Prediction tool was used to find the hydrophilic, accessible, or movable regions with default threshold value of 1.621.

T-Cell Epitope Prediction MHC Class I Binding
The portion of the vaccine is T cell that cooperates with T lymphocytes. Analysis of peptide binding to the MHC (Major Histocompatibility complex) class I molecule was assessed by the IEDB MHC I prediction tool (http://tools.iedb.org/mhci/) to predict cytotoxic T cell epitopes. The presentation of peptide complex to T lymphocyte undergoes a number of steps. Artificial Neural Network (ANN) 4.0 prediction method was used to predict the binding affinity. Before the prediction, all human allele lengths were selected and set to 9 amino acids. The half-maximal inhibitory concentration (IC50) value required for all conserved epitopes to bind at score less than 100 were selected. [21][22][23][24][25][26][27]

T-Cell Epitope Prediction MHC Class II Binding
T-cell epitopes interacting with MHC Class II were assessed by the IEDB MHC II prediction tool (http://tools.iedb.org/mhcii/) for helper T cells. Human allele reference set was used to regulate the interaction potentials of T cell epitopes and MHC Class II allele (HLA DR, DP and DQ). NN-align method was used to predict the binding affinity. IC50 values at score less than 500 were selected. [28][29][30][31]

Population Coverage
In IEDB, the population coverage link was selected to analyse the epitopes. This tool gives an illustration about the fraction of individuals predicted to respond to a given set of epitopes with known MHC restrictions (http://tools.iedb.org/population/iedbinput). The appropriate checkbox for calculation was checked based on MHC I, MHC II separately and combination of both.

Homology Modelling
The 3D structure was modelled using raptorX (http://raptorx.uchicago.edu) i.e. a protein structure prediction server developed by Xu group, outstanding at predicting 3D structures for protein sequences without close homologs in the Protein Data Bank (PDB). USCF chimera (version 1.13.1rc) was the program used for visualizing and editingthe molecular structure of the promising epitopes (http://www.cgl.uscf.edu/chimera).

Multiple Sequence Alignment
Eleven N. Gonorrhea species dihydrolipoamide acetyltransferase (DLAT) protein were selected for Multiple Sequence Alignment to show areas of conservation. (Figure 1)

B-cell epitope prediction
The reference sequence of N. Gonorrhea DLAT was subjected to Bepipred linear epitope 2, EMINI surface accessibility, Kolaskar&Tongaonkar antigenicity and Parker hydrophilicity prediction methods to test for various immunogenicity parameters (Table 1). Two epitopes have successfully passed the three tests. (Table 1)

T-Cell epitope predictions MHC class-I binding peptides
The reference sequence was analyzed using (IEDB) MHC-1 binding prediction tool to predict T cell epitopes interacting with different types of MHC Class I alleles. Seventy peptides were predicted to interact with different MHC-I alleles. The most promising epitopes with their corresponding MHC-1 alleles and IC50 scores as shown in (Table 2)

MHC class-II binding peptides
The reference sequence was analyzed using (IEDB) MHC-II binding prediction tool there were 438 predicted epitopes found to interact with MHC-II alleles. The most promising epitopes with their corresponding alleles and IC50 scores are shown in (Table 3).  Figure 2). For MHC class II, the epitopes that showed highest population coverage were GMRFTVFLANLLKDF (99.93%), AGMRFTVFLANLLKD (99.86%), AAGMRFTVFLANLLK (99.84%), RFTVFLANLLKDFRR (99.81%), and MRFTVFLANLLKDFR (99.81%)(Table 5, Figure 3). As for combined MHC II and I peptides, the global population coverage percentage was approximately 100% (figure 4).    PDB ID 4n72A was predicted using raptor X software, and visualized using Chimera (version 1.13.1rc) for the most promising peptides that showed low IC50 values, high Global population coverage percentages and high numbers of HLA-allele hits. (Figures 5, 6and 7)

Discussion
This study proposes candidate peptides to design a peptide-based vaccine against Neisseria gonorrheae targeting its immunogenic dihydrolipoamide acetyltransferase protein. Two epitopes (HVTVHEEAD, and VTVHEEAD) are suggested to be promising B-cell epitopes for passing antigenicity, hydrophlicity and surface accessibility tests. Eight epitopes were predicted to elicit T-cell response. (FAADTPNGL, FAPRLMCPL, and KLSPLAFII) interacts with MHC-I allele while, (GMRFTVFLANLLKDF, AGMRFTVFLANLLKD, AAGMRFTVFLANLLK, RFTVFLANLLKDFRR, and MRFTVFLANLLKDFR) interacted with MHC-II alleles.
Two B-cell peptides passed Bepipred linear epitope 2, Emini surface accessibility (threshold of 1), Kolaskar & Tongaonkar antigenicity (threshold of 1.03) and Parker hydrophilicity (threshold of 1.43) prediction tests. These peptides also share the same amino acid positions starting from 328, 329, respectively and ending at position 336. However, HVTVHEEAD having the ideal peptide length of nine amino acids makes it a stronger peptide candidate.
Seventy T-cell epitopes with half-maximum inhibitory concentration (IC50) <100 were predicted to interact with MHC I alleles. The coverage of worldwide population for all of the MHC I epitopes was 97.4%.Threeepitopes were the most promising according to their affinity to bind the highest number of MHC1 alleles, global population coverage, and low IC50 scores. FAADTPNGL is the strongest candidate due to its binding capacity to six HLA-alleles enriched with a low IC50 score of 3.32 and coverage of 35.02% of the global population. KLSPLAFII and FAPRLMCPL are also strong candidates with high global population coverage percentages (44.14% and 22.29%, respectively), and low IC50 scores of (25.74 and 8.48, respectively) Four hundred thirty-eight T-cell Epitopes were predicted to interact with MHC II alleles with IC50 < 500.Fortyfive of them had the highest efficacy and affinity to bind to the highest number of MHC II alleles, excluding certain alleles for MHC II. The best epitope was GMRFTVFLANLLKDF (99.93%), with 27 HLA hits and an IC50 of 7.5.The five strongest peptide candidates are sharing similar positions starting from 508 and ending at position 526, indicating high conservancy and antigenicity of that area. The total population coverage was massive with a percentage of 99.99% for all conserved MHC II epitopes.
As for combined MHC II and I peptides, the total global population coverage percentage was a 100%.These epitopes have the ability to induce a T-cell immune response by strongly interacting with MHC I & MHC II alleles. Thus, effectively generating a cellular and humoral immune response to the invading pathogen.
DLAT protein had been reported by previous studies to be able to induce an immune response. A study by Christen et al., 1994, suggests that the E2 subunit of pyruvate dehydrogenase complex is an autoantigen in halothane hepatitis and reports reactivity of the protein with anti-CF 3 CO antibodies. [32] Another study by Fussey et al., 1990, reported that dihydrolipoamide acetyltransferase was also recognized as one of the autoantigens that causes primary biliary cirrhosis (PBC), which is a chronic cholestatic liver disease characterized by the presence of antimitochondrial autoantibodies in the serum [33].A study by Corona et  This study was limited by being strictly computational. In addition, MHC-II alleles (HLA-DRB5*01:01, HLA-DRB3*01:01, HLA-DRB4*01:01, and HLA-DPA1*01) had not given results when analyzed by population coverage tool. Future in-vivo and in-vitro studies are highly recommended for the promising peptides.

Conclusion
There is a growing need of a vaccination for N. Gonnorhoeae with the increasing drug resistant gonococcal infections. Using insilico prediction methods to design vaccines is highly appreciated due to the marked