A detailed comparison was made of codon usage of chloroplast genes with their host (nuclear) genes in the four angiosperm speciesOryza sativa, Zea mays, Triticum aestivum andArabidopsis thaliana. The average GC content of the entire genes, and at the three codon positions individually, was higher in nuclear than in chloroplast genes, suggesting different genomic organization and mutation pressures in nuclear and chloroplast genes. The results of Nc-plots and neutrality plots suggested that nucleotide compositional constraint had a large contribution to codon usage bias of nuclear genes inO. sativa, Z. mays, andT. aestivum, whereas natural selection was likely to be playing a large role in codon usage bias in chloroplast genomes. Correspondence analysis and chi-test showed that regardless of the genomic environment (species) of the host, the codon usage pattern of chloroplast genes differed from nuclear genes of their host species by their AU-richness. All the chloroplast genomes have predominantly A- and/or U-ending codons, whereas nuclear genomes have G-, C- or U-ending codons as their optimal codons. These findings suggest that the chloroplast genome might display particular characteristics of codon usage that are different from its host nuclear genome. However, one feature common to both chloroplast and nuclear genomes in this study was that pyrimidines were found more frequently than purines at the synonymous codon position of optimal codons.

Laccases (LACs) are versatile enzymes that catalyze oxidation of a wide range of substrates, thereby functioning in regulation of plant developmental processes and stress responses. However, with a few exceptions, the function of most LACs remains unclear in plants. In this study, we newly identified 4, 12, 22, 26, 27, 28 and 49 LAC genes for Physcomitrella patens, Amborella trichopoda, Zeamays, Ricinus communis, Vitis vinifera, Triticum aestivum and Glycine max, on the basis of exhaustive homologous sequence searches. In these plants, LACs differ greatly in sequence length and physical properties, such as molecular weight and theoretical isoelectric point (pI), but majority of them contain a signal peptide at their N-terminus. The originality of LACs could be traced back to as early as the emergence of moss. Plant LACs are clearly divided into seven distinct classes, where six ancient LACs should be present prior to the divergence of gymnosperms and angiosperms. Functional divergence analysis reveal that functional differentiation should occur among different groups of LACs because of altered selective constraints working on some critical amino acid sites (CAASs) within conserved laccase domains during evolution. Soybean and maize LACs have significantly different exon frequency (6.08 vs 4.82), and they are unevenly distributed and tend to form gene clusters on some chromosomes. Further analysis shows that the expansion of LAC gene family would be due toextensive tandem and chromosomal segmental duplications in the two plant species. Interestingly, *81.6% and 36.4% of soybean and maize LACs are potential targets of miRNAs, such as miR397a/b, miR408d, or miR528a/b etc. Both soybean and maize LACs are tissue specifically and developmental-specifically expressed, and are in response to different external abiotic and biotic stressors. These results suggest a diversity of functions of plant LAC genes, which will broaden our understanding and lay solid foundation for further investigating their biological functions in plants.