Synonymous codon usage bias is correlative to intron number and shows disequilibrium among exons in plants

Abstract

Background: Evidence has been assembled to suggest synonymous codon usage bias (SCUB) has close relationship with intron. However, the relationship (if any) between SCUB and intron number as well as exon position is at present rather unclear.

Results: To explore this relationship, the sequences of a set of genes containing between zero and nine introns was extracted from the published genome sequences of three algal species, one moss, one fern and six angiosperms (three monocotyledonous species and three dicotyledonous species). In the algal genomes, the frequency of synonymous codons of the form NNG/NNC (codons with G and C at the third position) was positively related to intron number, but that of NNA/NNT was inversely correlated; the opposite was the case in the land plant genomes. The frequency of NNC/NNG was higher and that of NNA/NNT lower in two terminal exons than in the interstitial exons in the land plant genes, but the rule showed to be opposite in the algal genes. SCUB patterns in the interstitial and two terminal exons mirror the different evolutionary relationships between these plant species, while the first exon shows the highest level of conservation is therefore concluded to be the one which experiences the heaviest selection pressure. The phenomenon of SCUB may also be related to DNA methylation induced conversion of CG to AT.

Conclusions: These data provide some evidence of linkage between SCUB, the evolution of introns and DNA methylation, which brings about a new perspective for understanding how genomic variation is created during plant evolution.

Figure 1

The ratio between the frequencies of NNA/Ts to NNC/Gs is correlated differently with intron number between (A) E. fasciculatus and (B) O. sativa. NNA/Ts: frequency of SCs having an A and a T in the third position, NNC/Gs: frequency of SCs having a C and a G in the third position. The equivalent ratios for the other genomes investigated are shown in Additional file 3: Figure S2.

Figure 2

SC frequency based on the third nucleotide is correlated differently with intron number in algal (A) and land plant (B-E) genomes. NNA, NNT, NNC and NNG are defined as the ratio of total SCs with A, T, C and G at the third position, respectively to 59 SCs. The equivalent relationships in the other genomes investigated are shown in Additional file 4: Figure S3.

Figure 3

The Mean of NNA/Ts to NNC/Gs ratios within SCs differs between (A) E. fasciculatus and (B) O. sativa. NNA/Ts: frequency of SCs having an A and a T in the third position of an amino acid, NNC/Gs: frequency of SCs having a C and a G in the third position of an amino acid. The mean is defined as the average of NNA/Ts to NNC/Gs ratios of 18 amino acids with SCs. 2EG-10EG: gene sequences arranged into between two and ten exons. The equivalent ratios in the other genomes investigated are shown in Additional file 5: Figure S4.

Figure 4

SC frequency based on the third nucleotide shows disequilibrium among exons and differs between algal (A) and land plant (B-E) genomes. NNA, NNT, NNC and NNG are defined as the ratio of total SCs with A, T, C and G at the third position, respectively to 59 SCs. The equivalent ratios in the other genomes investigated are shown in Figure S5.

Figure 5

The association between the DNA methylation induced conversion of C to T and SCUB based on intron number. A, C: SC frequencies based on the second-third nucleotide combinations (NNN); B, D, SC frequencies based on the third-next codon’s first nucleotide combinations (NN|N). The equivalent associations in the other genomes investigated are shown in Additional file 7: Figure S6 and Additional file 8: Figure S7.

Figure 6

The association between the DNA methylation induced conversion of C to T and SCUB based on exon position. A, C: SC frequencies based on the second-third nucleotide combinations (NNN); B, D, SC frequencies based on the third-next codon’s first nucleotide combinations (NN|N). The equivalent associations in the other genomes investigated are shown in Additional file 9: Figure S8 and Additional file 10: Figure S9.

Figure 7

Cluster and PC analysis of SCUB frequencies based on intron number and exon position in algal and land plant genomes. (A, B) Cluster analysis, (C, D) Principal component analysis. PC1, PC2: coefficients associated with the first two extracted principal components. A, C: SC frequency in genes comprising from zero to nine introns. B, D: SC frequency in exons of gene sequences arranged into between two and ten exons.