Codon usage revisited: Lack of correlation between codon usage and the number of tRNA genes in enterobacteria

Abstract

It is widely believed that if a high number of genes are found for any tRNA in a rapidly replicating bacteria, then the cytoplasmic levels of that tRNA will be high and an open reading frame containing a higher frequency of the complementary codon will be translated faster. This idea is based on correlations between the number of tRNA genes, tRNA concentration and the frequency of codon usage observed in a limited number of strains as well as from the fact that artificially changing the number of tRNA genes alters translation efficiency and consequently the amount of properly folded protein synthesized. tRNA gene number may greatly vary in a genome due to duplications, deletions and lateral transfer which in turn would alter the levels and functionality of many proteins. Such changes are potentially deleterious for fitness and as a result it is expected that changes in tRNA gene numbers should be accompanied by a modification of the frequency of codon usage. In contrast to this model, when comparing the number of tRNA genes and the frequency of codon usage of several Salmonella enterica and Escherichia coli strains we found that changes in the number of tRNA genes are not correlated to changes in codon usage. Furthermore, these changes are not correlated with a change in the efficiency of codon translation. These results suggest that once a genome gains or loses tRNA genes, it responds by modulating the concentrations of tRNAs rather than modifying its frequency of codon usage.

An improved estimation of tRNA expression to better elucidate the coevolution between tRNA abundance and codon usage in bacteria

“…Additionally, tRNA-mediated codon usage bias has been broadly observed in a variety of organisms, including the gram-negative bacterium Salmonella enterica serovar typhimurium 10 , the gram-positive Bacillus subtilis 11 , eukaryotes such as yeast 10,12 , a variety of fungal and invertebrate mitochondrial genomes 13,14 , and viruses including HIV 15 , and bacteriophages 16,17 . Nonetheless, the nature of the relationship between tRNA abundance and codon usage across bacterial species has been the subject of debate among researchers, with some suggesting that tRNA availability is the main driving force of codon usage bias 18,19 and others contending that the two are weakly correlated 20,21 .…”

Design of typical genes for heterologous gene expression

“…This method of course does not work for synonymous codons that are recognized by tRNAs with equal copy numbers. In addition, decoding is highly redundant and wobble decoding is often as efficient as decoding of cognate codons 37 – 39 . In many codon boxes that codon, for which there are the most tRNA gene copies, is more used in highly expressed genes than any of the other synonymous codons of the box.…”

Oxidative stress strongly restricts the effect of codon choice on the efficiency of protein synthesis in Escherichia coli

“…Reporter plasmids enriched in codons for Gly, Glu, and Ala have been previously constructed ( Rojas et al, 2018 ; Leiva et al, 2020 ). Reporter plasmids enriched in codons coding for the remaining 17 amino acids as well as two control plasmids with four randomly selected non-identical codons were cloned using a similar procedure.…”