Multiple mechanisms might be responsible for generating the observed

diversity in 5S rRNA genes in a genome. In organisms containing multiple rRNA genes, the homogeneity of primary structures is believed to be maintained through gene conversion by homologous recombination (Hashimoto et al., 2003), as a form of concerted evolution (Abdulkarim & Hughes, 1996). Although the observed homogeneity of 5S rRNA genes in the majority of species analyzed could be attributed to the effect of homologous recombination, the recombination appeared to be compartmentalized or ineffective in some genomes. The observed high degree of diversity in the primary structures of the 5S rRNA genes in partial or split rRNA gene operons and the rrnC operon in T. tengcongensis suggested that these rRNA genes have been excluded Cell Cycle inhibitor from participation in concerted evolution. Such compartmentalization was also present in B. subtilis that has two similarity groups of rRNA genes appeared to have evolved

independently, as evidenced by their relation to different 5S rRNA genes-rrn23S spacers. Despite the lack of sequence homogeneity, secondary structures of these genes were well conserved, most likely due to the life and death driving force of ribosomal constraints. Compared with whole 16S and 23S rRNA genes, 5S rRNA genes are a less ideal taxonomical marker for use in analyses of complex microbiomes. The main reason is the widespread intragenomic 5S rRNA gene diversity. Approximately, 12.3% (96 of 779) Etoposide mw of the unique

species analyzed had > 3% intragenomic variation of 5S rRNA genes, compared to only about 1% of species with similar degree of variation in 16S and 23S rRNA genes (Coenye & Vandamme, 2003; Acinas et al., 2004; Pei et al., 2010). This high degree of diversity most often occurs between a standalone 5S rRNA Vildagliptin gene (orphan or split) and a 5S rRNA gene in a complete rRNA operon. The lack of standalone 16S or 23 S rRNA genes appears to be the main reason for the lower intragenomic diversity among 16S or 23S rRNA genes. Orphan 5S rRNA genes are sometimes overlooked by a whole-genome annotation program because of their small size. Compared with rrnDB (Lee et al., 2009), a publically accessible database that collects existing data on structure RNA genes from whole-genome sequencing projects, 11 genomes listed in Table 1 that had additional 5S rRNA genes in our study are not listed in rrnDB. The additional 5S rRNA genes would have been invisible if blast search of 5S rRNA genes against the whole genomes were not performed. Nevertheless, in 26 of the 52 genomes listed in Table 1, correct records of the orphan 5S rRNA genes can be found in rrnDB. The remaining 15 of the 52 genomes have no entries in rrnDB. Divergent evolution between paralogous 5S rRNA genes in a genome may corrupt the record of evolutionary history and obscure the true identity of an organism.