CodonBERT: a BERT-based architecture tailored for codon optimization using the cross-attention mechanism.

MOTIVATION: Due to the varying delivery methods of mRNA vaccines, codon optimization plays a critical role in vaccine design to improve the stability and expression of proteins in specific tissues. Considering the many-to-one relationship between synonymous codons and amino acids, the number of mRNA sequences encoding the same amino acid sequence could be enormous. Finding stable and highly expressed mRNA sequences from the vast sequence space using in silico methods can generally be viewed as a path-search problem or a machine translation problem. However, current deep learning-based methods inspired by machine translation may have some limitations, such as recurrent neural networks, which have a weak ability to capture the long-term dependencies of codon preferences. RESULTS: We develop a BERT-based architecture that uses the cross-attention mechanism for codon optimization. In CodonBERT, the codon sequence is randomly masked with each codon serving as a key and a value. In the meantime, the amino acid sequence is used as the query. CodonBERT was trained on high-expression transcripts from Human Protein Atlas mixed with different proportions of high codon adaptation index codon sequences. The result showed that CodonBERT can effectively capture the long-term dependencies between codons and amino acids, suggesting that it can be used as a customized training framework for specific optimization targets. AVAILABILITY AND IMPLEMENTATION: CodonBERT is freely available on https://github.com/FPPGroup/CodonBERT.

Ferritin-binding and ubiquitination-modified mRNA vaccines induce potent immune responses and protective efficacy against SARS-CoV-2.

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) incessantly engenders mutating strains via immune escape mechanisms, substantially escalating the risk of severe acute respiratory syndrome. In this context, the urgent development of innovative and efficacious mRNA vaccines is imperative. In our study, we synthesized six unique mRNA vaccine formulations: the Receptor Binding Domain (RBD) monomer vaccine, RBD dimer (2RBD) vaccine, RBD-Ferritin (RBD-Fe) vaccine, ubiquitin-modified wild-type Nucleocapsid gene (WT-N) vaccine, rearranged Nucleocapsid gene (Re-N) vaccine, and an epitope-based (COVID-19 epitope) vaccine, all encapsulated within the lipid nanoparticle SM102. Immunization studies conducted on C57BL/6 mice with these vaccines revealed that the RBD monomer, RBD dimer (2RBD), and RBD-Fe vaccines elicited robust titers of specific antibodies, including neutralizing antibodies. In contrast, the wild-type N gene (WT-N), rearrange N gene (Re-N), and COVID-19 epitope vaccines predominantly induced potent cellular immune responses. Protective efficacy assays in golden hamsters demonstrated that vaccinated cohorts showed significant reduction in lung pathology, markedly lower viral loads in the lungs, nasal turbinates, and trachea, and substantially reduced transcriptional and expression levels of pro-inflammatory cytokines. Overall, our vaccine candidates pave the way for novel strategies in vaccine development against various infectious agents and establish a critical foundation for the formulation of advanced vaccines targeting emerging pathogens.

Dynamic Covalent Hydrogel as a Single-Dose Vaccine Adjuvant for Sustained Antigen Release and Significantly Elevated Humoral Immunity.

Vaccine is the most important way for fighting against infection diseases. However, multiple injections and unsatisfied immune responses are the main obstacles for current vaccine application. Herein, a dynamic covalent hydrogel (DCH) is used as a single-dose vaccine adjuvant for eliciting robust and sustained humoral immunity. By adjusting the mass ratio of the DCH gel, 10-30 d constant release of the loaded recombinant protein antigens is successfully realized, and it is proved that sustained release of antigens can significantly improve the vaccine efficacy. When loading SARS-CoV-2 RBD (Wuhan and Omicron BA.1 strains) antigens into this DCH gel, an over 32 000 times and 8000 times improvement is observed in antigen-specific antibody titers compared to conventional Aluminum adjuvanted vaccines. The universality of this DCH gel adjuvant is confirmed in a Nipah G antigen test as well as a H1N1 influenza virus antigen test, with much improved protection of C57BL/6 mice against H1N1 virus infection than conventional Aluminum adjuvanted vaccines. This sustainably released, single-dose DCH gel adjuvant provides a new promising option for designing next-generation infection vaccines.