Kober KM and Pogson GH (2013), Genome-wide patterns of codon bias are shaped b...

ECB-ART-42862

G3 (Bethesda) 2013 Jul 08;37:1069-83. doi: 10.1534/g3.113.005769.

Show Gene links Show Anatomy links

Genome-wide patterns of codon bias are shaped by natural selection in the purple sea urchin, Strongylocentrotus purpuratus.

Kober KM , Pogson GH .

???displayArticle.abstract???
Codon usage bias has been documented in a wide diversity of species, but the relative contributions of mutational bias and various forms of natural selection remain unclear. Here, we describe for the first time genome-wide patterns of codon bias at 4623 genes in the purple sea urchin, Strongylocentrotus purpuratus. Preferred codons were identified at 18 amino acids that exclusively used G or C at third positions, which contrasted with the strong AT bias of the genome (overall GC content is 36.9%). The GC content of third positions and coding regions exhibited significant correlations with the magnitude of codon bias. In contrast, the GC content of introns and flanking regions was indistinguishable from the genome-wide background, which suggested a limited contribution of mutational bias to synonymous codon usage. Five distinct clusters of genes were identified that had significantly different synonymous codon usage patterns. A significant correlation was observed between codon bias and mRNA expression supporting translational selection, but this relationship was driven by only one highly biased cluster that represented only 8.6% of all genes. In all five clusters preferred codons were evolutionarily conserved to a similar degree despite differences in their synonymous codon usage distributions and magnitude of codon bias. The third positions of preferred codons in two codon usage groups also paired significantly more often in stems than in loops of mRNA secondary structure predictions, which suggested that codon bias might also affect mRNA stability. Our results suggest that mutational bias has played a minor role in determining codon bias in S. purpuratus and that preferred codon usage may be heterogeneous across different genes and subject to different forms of natural selection.

???displayArticle.pubmedLink??? 23637123
???displayArticle.pmcLink??? PMC3704236
???displayArticle.link??? G3 (Bethesda)

Species referenced: Echinodermata
Genes referenced: ets1 LOC100887844 LOC577982

???attribute.lit??? ???displayArticles.show???

	Figure 1. Distribution of codon bias scores of (A) Nc and (B) Nc′ in S. purpuratus.
	Figure 2. Heatmap of codon usage frequencies in the five codon usage groups (CUG) in (A) Strongylocentrotus purpuratus and (B) Drosophila melanogaster. Codon usage frequencies were centered on the equal usage expectation (e.g., for fourfold degenerate codons, frequencies were centered on 1/4). Centered frequencies clustering was performed with hierarchical cluster analysis using pairwise complete-linkage by Euclidean distance for both columns (CUG) and rows (synonymous codons) using Cluster 3.0 (Eisen et al. 1998). Heatmap plots were generated with Java TreeView (Saldanha 2004). Synonymous codon labels are colored by the base composition of the third position (N3) with N3 of ‘A’ and ‘T’ nucleotides, colored pink, and ‘G’ and ‘C,’ colored green. The asterisk (*) denotes genome-wide preferred codons in S. purpuratus (see Materials and Methods).
	Figure 3. Plot of offsets β1 and β2 for a SCUMBLE model with four trends for S. purpuratus genes, colored by groups of genes clustered by codon usage distributions. Offset β1 is correlated most strongly to GC3 in all groups.
	Figure 4. Codon bias and GC composition in protein coding gene regions for different codon usage groups. Codon bias is represented as Wright’s Nc (Wright 1990). Regional GC composition is calculated from (A) GC3 content of exons, (B) GC content of exons, (C) GC content of introns, and (D) GC content of flanking regions. “Rp” denotes annotated ribosomal proteins. The genome-wide mean GC content is 36.9%.
	Figure 5. Synonymous codon usage bias and mRNA expression levels for genes in different codon usage groups in (A) S. purpuratus and (B) Drosophila melanogaster. Codon usage bias is Novembre’s Nc′. Red circles surround annotated ribosomal proteins (‘Rp’). The black arrow labeled “U” points to “ubiquitin-like/S30 ribosomal fusion protein” (SPU_005280), an example of a highly expressed gene in S. purpuratus. The black arrow labeled “E” points to Sp-Ets1/2 (SPU_002874), an example of a highly expressed, highly biased gene that does not belong to codon usage group 0.

References [+] :

Addison, Multiple gene genealogies reveal asymmetrical hybridization and introgression among strongylocentrotid sea urchins. 2009, Pubmed, Echinobase