Department of Physics and Astronomy, McMaster University, Hamilton, Ontario, Canada.
Mol Biol Evol. 2010 Sep;27(9):2129-40. doi: 10.1093/molbev/msq102. Epub 2010 Apr 19.
Most transfer RNAs (tRNAs) can translate more than one synonymous codon, and most codons can be translated by more than one isoacceptor tRNA. The rates of translation of synonymous codons are dependent on the concentrations of the tRNAs and on the rates of pairing of each anticodon-codon combination. Translational selection causes a significant bias in codon frequencies in highly expressed genes in most bacteria. By comparing codon frequencies in high and low-expression genes, we determine which codons are preferred for each amino acid in a large sample of bacterial genomes. We relate this to the number of copies of each tRNA gene in each genome. In two-codon families, preferred codons have Watson-Crick pairs (GC and AU) between the third codon base and the wobble base of the anticodon rather than GU pairs. This suggests that these combinations are more rapidly recognized by the ribosome. In contrast, in four-codon families, preferred codons do not correspond to Watson-Crick rules. In some cases, a wobble-U tRNA can pair with all four codons. In these cases, A and U codons are preferred over G and C. This indicates that the nonstandard UU combination appears to be translated surprisingly well. Differences in modified bases at the wobble position of the anticodon appear to be responsible for the differences in behavior of tRNAs in two- and four-codon families. We discuss the way changes in the bases in the anticodon influence both the speed and the accuracy of translation. The number of tRNA gene copies and the strength of translational selection correlate with the growth rate of the organism, as we would expect if the primary cause of translational selection in bacteria is the requirement to optimize the speed of protein production.
大多数转移 RNA(tRNA)可以翻译一个以上的同义密码子,大多数密码子可以由一个以上的同功接受体 tRNA 翻译。同义密码子的翻译速度取决于 tRNA 的浓度以及每个反密码子-密码子组合配对的速度。翻译选择导致大多数细菌中高表达基因的密码子频率产生显著的偏向。通过比较高表达和低表达基因中的密码子频率,我们确定了在大量细菌基因组中每个氨基酸的首选密码子。我们将这与每个基因组中每种 tRNA 基因的拷贝数联系起来。在两个密码子家族中,首选密码子在第三个密码子碱基和反密码子的摆动碱基之间具有 Watson-Crick 对(GC 和 AU),而不是 GU 对。这表明这些组合被核糖体更快地识别。相比之下,在四个密码子家族中,首选密码子不符合 Watson-Crick 规则。在某些情况下,一个摆动 U tRNA 可以与所有四个密码子配对。在这些情况下,A 和 U 密码子优先于 G 和 C。这表明非标准的 UU 组合似乎被翻译得很好。反密码子摆动位置的修饰碱基的差异似乎是导致两种和四种密码子家族中 tRNA 行为差异的原因。我们讨论了反密码子摆动位置碱基的变化如何影响翻译的速度和准确性。tRNA 基因拷贝数和翻译选择的强度与生物体的生长速度相关,这正如我们所预期的,如果细菌中翻译选择的主要原因是优化蛋白质生产速度的需求。