Relative synonymous codon usage (RSCU): Its role in the virulence of CSFV
Text by Lauro Velazquez-Salinas
Relative synonymous codon usage (RSCU) refers to the phenomenon during which some synonymous codons are used more often than others. This preference for a particular codon can vary between species. Since viruses are obligate intracellular parasites, they often exploit and co-evolve with host molecular mechanisms to prosper. We determined that the E2 glycoprotein of classical swine fever virus (CSFV) had a similar codon usage bias of Sus Scrofa, its natural host. We wanted to know the effect of just switching native codons in E2 for the less frequently used codons. In fact, changing just the codons in E2 resulted in a full attenuation in pigs. Interestingly, when we explored the potential use of this phenotype as a vaccine, we found that it was able to protect animals against the disease. By not changing a single amino acid, this potential vaccine leaves all natural antigenic epitopes intact.
The motivation for this project came from Lauro Velazquez-Salinas who spent many hours analyzing viral sequences. After reading about the potential for attenuating viruses by de-optimizing RSCU, wondered if this could be applied to CSFV. The benefits of producing vaccines by only changing the nucleotide composition allow the antigenic mosaic in the virus to remain intact. Additionally by changing the nucleotide composition, allows for a genetic marker that can be used to differentiate between vaccinated and infected animals.
The first problem that we faced was to establish a methodology to carry out this process. Since different variables are playing a role shaping RSCU, we decided to test different approaches. The next important decision was to determine the place in the genome to conduct the de-optimization. Since E2 has been shown to have important antigenic epitopes, we decided to explore this target, but honestly we don’t know what would have been the outcome in other parts of the genome. In our first attempt, we were unable to recover virus, and it appeared that we made so many mutations that the virus was unable to grow. Our next two attempts had fewer mutations but we didn’t observe any differences in virulence when we tested in swine. Finally we were able to produce a virus that had enough but not too many deoptimized mutations, where the virus was fully attenuated, with no clinical symptoms, and was able to fully protect swine against CSFV.
Codon pair bias (CPB) scores of swine genes and CSFV E2 from BICv and CSFm1-4 mutants. Each mark represents the calculated CPB score of a gene plotted against its amino acid length. CPB scores of the mutants used in this study (CSFm1-4) are indicated.
Introducing the author
Lauro Velazquez-Salinas, behind different codon usage sequences.