Abstract
BACKGROUND/OBJECTIVES: Approved mRNA vaccines commonly use sequences modified with pseudouridine to enhance translation efficiency and mRNA stability. However, this modification can result in ribosomal frameshifts, reduced immunogenicity, and higher production costs. This study aimed to explore the potential of unmodified mRNA sequences for varicella-zoster virus (VZV) and evaluate whether codon optimization could overcome the limitations of pseudouridine modification. METHODS: We utilized artificial intelligence (AI) to design several unmodified gE mRNA sequences for VZV, considering factors such as codon preference and secondary structure. The optimized mRNA sequences were assessed for protein expression levels in vitro and were subsequently used to develop a vaccine, named Vac07, encapsulated in a lipid nanoparticle (LNP) delivery system. The immunogenicity of Vac07 was evaluated in mice. RESULTS: Codon-optimized mRNA sequences showed significantly higher protein expression levels in vitro compared to wild-type (WT) sequences. Vaccination with Vac07 demonstrated immunogenicity in mice that was comparable to, or even superior to, the licensed Shingrix vaccine, characterized by a stronger Th1-biased antibody response and a slightly more robust Th1-type cellular response. CONCLUSIONS: Codon-optimized unmodified mRNA sequences may also represent a viable approach for mRNA vaccine development. These optimized sequences have the potential to lower production costs while possibly enhancing the immunogenicity of mRNA vaccines. Vac07, developed using this method, shows promise as a potentially more efficient and cost-effective mRNA vaccine candidate for VZV.