Computational prediction of multidrug-resistant Acinetobacter baumannii Epitopes: A bioinformatic approach to Acinetobacter vaccine development
Computational prediction of multidrug-resistant Acinetobacter baumannii Epitopes: A bioinformatic approach to Acinetobacter vaccine development
Zahra Shokouhi,1,*
1. Microbial Technology and Products Research Center, University of Tehran, Tehran, Iran
Introduction: Acinetobacter baumannii, a Gram-negative bacterium, has become a prominent nosocomial pathogen, especially in intensive care units. Its capacity to develop resistance to multiple antibiotics, including polymyxins, has resulted in the emergence of multidrug-resistant (MDR) strains. This alarming trend underscores the urgent need for innovative therapeutic strategies, with vaccine development being a particularly promising approach.
The advent of computer-aided vaccine design presents a viable pathway for developing effective vaccines. Research indicates that several proteins are overexpressed in MDR A. baumannii, such as B Barrel Outer Membrane Protein (BAM), Porin, and Lipoproteins. These proteins play a key role in cell membrane formation, maintaining outer membrane integrity, and contributing to the bacterium’s drug resistance mechanisms. Consequently, they represent potential candidates for vaccine development. In this study, epitopes were devised for these proteins using specific bioinformatics tools.
Methods: Protein sequences retrieval
The FASTA-formatted amino acid sequences of outer membrane proteins: BamD (A0A009Q2E5), Putative lipoprotein (B0VSC6), Porin subfamily protein (A0A062IZ25), and Peptidoglycan-associated lipoproteins (A0A009TK41 and A0A454ASL3) were obtained from UniProt (https://www. uniprot.org/) database. And saved for subsequent analysis.
Epitope Prediction
In this step, we aimed to predict linear B cell, Cytotoxic T lymphocyte (CTL), and Helper T lymphocyte (HTL) epitopes. To achieve this, different bioinformatics servers capable of identifying these specific types of epitopes were employed.
Prediction of B-Cell Epitopes
The prediction of linear B cell epitopes was conducted using the IEDB database (http://tools.iedb.org/bcell/). The Bepipred Linear Epitope Prediction 2.0 method was employed for this purpose.
Prediction of MHC class I binding epitopes
we utilized the IEDB web server to predict 9-mer epitopes with potential binding affinity to MHC-I molecules, employing the IEDB recommended method 2020.09 (NetMHCpan EL 4.1). A cut-off value of percentile rank<1 was set to identify high-affinity epitopes, and the reference human HLA allele set was used for prediction.
Prediction of MHC class ІІ binding epitopes
For MHC class II binding epitopes, we used the IEDB web server to identify 15-mer epitopes with potential binding affinity to MHC-II molecules, employing the NetMHCIIpan 4.1 EL prediction method. The human HLA-DR allele set was used as the MHC source species, and high-affinity epitopes were screened by adjusting the percentile rank<1 as the cut-off value.
The antigenicity and allergenicity assessment
The antigenicity of selected epitopes was predicted using VaxiJen v2.0 server (by set of 0.4 threshold). And, their allergenicity was checked by the employment of Allertop server.
Results: The top-scoring HTL and CTL alleles were predicted from MDR A. baumannii proteins using the IEDB database. Specific peptides and their corresponding length, Antigenicity score, and allergenicity nature are presented.
Conclusion: Epitope prediction is the foundational step in the design of multi-epitope vaccines. This study focused on predicting potential B-cell, CTL, and HTL epitopes using various bioinformatics servers and criteria to identify peptides with high binding affinity to human HLAs. The selected epitopes were both immunogenic and non-allergenic.