The unequal usage of synonymous codons, often termed as codon usage bias (CUB), is generally thought to be an intricate combined outcome of mutation pressure, natural selection, and genetic drift. To improve translational efficiency and accuracy, highly expressed genes are suggested to use optimal codons more frequently and thus show highly biased codon usage. This effect is called translational selection.  

Recently, ZHANG Zhang’s Lab from Beijing Institute of Genomics, Chinese Academy of Sciences (BIG) in collaboration with Dr. ZHANG Zhihua from the same institute, Dr. CUI Peng from King Abdullah University of Science and Technology, and Dr. ZHU Jiang from Massachusetts General Hospital and Harvard Medical School, revealed translational selection is more pronounced in human housekeeping genes. 

Translational selection has been widely documented in prokaryotes and unicellular eukaryotes, as typified by Escherichia coli and Saccharomyces cerevisiae, but less is known in human. Albeit not changing protein sequence, silent mutation will influence gene expression during translation process and result in human diseases, such as Cystic Fibrosis. Also, there are many silent mutations that are associated with human diseases, while the mechanisms are not clear. It is of great biological significance to study how codon usage bias is correlated with translational selection in human genes.  

To examine how translational selection differentially operates in human genes, scientists classified genes into three groups, expression-invariable genes (EIG) that are expressed in all the 10 tissues at relatively consistent expression levels, expression-variable genes (EVG) that are expressed in all 10 tissues at diverse expression levels, and tissue-specific (TS) genes that are expressed only in one tissue. They found that EIGs are under stronger translational selection. First, EIGs show more obvious positive correlation between CUB and gene expression level, and this is not contributed by high expression level only. Second, high CUB genes from EIGs are extremely biased in using optimal codons. 

This study indicates that translational selection is likely a fundamental mechanism to be universally operative in a diversity of species and will inspire new insights into the evolution of translational selection from unicellular to multicellular organisms. 

The work has been published online in Biology Direct on July 10, 2014 and is freely available at http://www.biologydirect.com/content/9/1/17.  

Relationship between CUB and gene expression level(Image by ZHANG Zhang's lab)