Designing an mRNA-based universal vaccine for influenza A virus using an in silico approach

Abstract

Introduction: The influenza virus undergoes constant mutation, necessitating annual updates to the seasonal flu vaccine. Traditional vaccine methods struggle to keep pace with the rapid mutation rate of this virus. This study aimed to develop a universal mRNA vaccine using immunoinformatic approaches to effectively target various strains of influenza A viruses (IAVs). Specifically, this study identified epitopes capable of eliciting T-cell-mediated immune responses for inclusion in vaccine design.

Methods: Immunoinformatic tools from the Immune Epitope Database (IEDB) were used to direct neutralizing antibodies (nAbs) against hemagglutinin (HA) molecules. T lymphocyte epitopes were predicted using multiple sequence alignment from Clustal Omega. These epitopes were filtered through a pipeline to assess their affinity for major histocompatibility complex (MHC) class 1 molecules and their conservation across various IAV strains.

Results: After data cleaning, 82,200 of the 91,093 HA amino acid sequences of 15 human-infected IAV strains were eligible for subsequent epitope prediction and conservation assessment. The lengths of the HA amino acid sequences across the 15 strains predominantly fell within the range of 560–567 amino acids. MHC class 1 interaction analysis revealed 16 antigenic sequences generally located in the stalk domain of HA. Among those, 11 were found to have a consensus. Five consensus sequences with strong binding scores were converted into universal mRNA sequences that were ready to be used for subsequent vaccine research.

Conclusion: Consesus and antigenic HA sequences were successfully designed for use against diverse IAV strains, particularly those circulating in Vietnam. The use of immune-informatics tools with stalk-directed strategies has been shown to be an effective step in viral vaccine development.