Abstract Codon usage bias has been documented in a wide diversity of species. The relative contributions of mutation and various forms of natural selection on codon usage bias are different. Plant specific WRKY transcription factors play important roles in transcriptional regulation and signal transduction. However, systematical analysis on codon usage bias of Medicago truncatula WRKY (MtWRKY) genes has not been reported. In the present study, a systematic examination of codon usage for MtWRKY genes was carried out. The results showed that all MtWRKY genes were below the high effective number of codons(ENC) standard curve, suggesting that other factors independent of nucleotide composition had affected codon usage bias. Neutrality plots (GC12 vs. GC3) were used to analyze the relationships among the 3 codon positions. There was a significant positive correlation (r=0.34, P＜0.01) between GC12 and GC3 codons of MtWRKY genes, indicating that GC mutational bias led to similar GC content in all codon positions. Moreover, GC content in MtWRKY genes showed a wide range of GC3s values (0.2~0.5), indicating that mutational pressure was the main factor in shaping codon usage. The CT contents were higher than that of GA on the 3rd position of codons according to parity rule 2 analysis. GC or AT were used disproportionately, with C and T used more frequently than G and A in the 3rd position of codon in MtWRKY genes. This result indicated that natural selection contributed to MtWRKY codon usage bias, but mutational bias was the major influence on codon usage. Correspondence between frequency of optimal codons (Fop) and GC content analysis showed that there was a significant positive correction between Fop and GC content of exon (r=0.57, P＜0.01) and weak positive correction between Fop and GC content of intron (r=0.09, P＞0.05). There was a significant positive correction between Fop and length of exon (r=0.28, P＜0.05), and a negative correction between Fop and length of intron (r=-0.01, P＞0.05). These results indicated that mutation contributed to codon usage bias of exon, while natural selection was the main factor that shaped the intron sequences. This study identified 4 optimal codons, on which the 3rd position exclusively used G or C. All together, these results provide important information for codon optimization on transgenic studies of WRKY genes in the future.
