No abstract

The human gut microbiota established during infancy has persistent effects on health. In vitro studies have suggested that human milk oligosaccharides (HMOs) in breast milk promote the formation of a bifidobacteria-rich microbiota in infant guts; however, the underlying molecular mechanism remains elusive. Here, we characterized two functionally distinct but overlapping fucosyllactose transporters (FL transporter-1 and -2) from Bifidobacterium longum subspecies infantis. Fecal DNA and HMO consumption analyses, combined with deposited metagenome data mining, revealed that FL transporter-2 is primarily associated with the bifidobacteria-rich microbiota formation in breast-fed infant guts. Structural analyses of the solute-binding protein (SBP) of FL transporter-2 complexed with 2′-fucosyllactose and 3-fucosyllactose, together with phylogenetic analysis of SBP homologs of both FL transporters, highlight a unique adaptation strategy of Bifidobacterium to HMOs, in which the gain-of-function mutations enable FL transporter-2 to efficiently capture major fucosylated HMOs. Our results provide a molecular insight into HMO-mediated symbiosis and coevolution between bifidobacteria and humans.

Recently, we discovered that bile acid, a main component of bile, is a host factor that regulates the composition of the cecal microbiota in rats. Because bile secretion increases on a high-fat diet and bile acids generally have strong antimicrobial activity, we speculated that bile acids would be a determinant of the gut microbiota in response to a high-fat diet. The observed changes in the rat cecal microbiota triggered by cholic acid (the most abundant bile acid in human biliary bile) administration resemble those found in animals fed high-fat diets. Here, we discuss the rationale for this hypothesis by evaluating reported diet-induced gut microbiota alterations based on the postulate that bile acids worked as an underlying determinant. The identification of host factors determining the gut microbiota greatly contributes to understanding the causal relationships between changes in the gut microbiota and disease development, which remain to be elucidated.

Escherichia coli, including the closely related genus Shigella, is a highly diverse species in terms of genome structure. Comparative genomic hybridization (CGH) microarray analysis was used to compare the gene content of E. coli K-12 with the gene contents of pathogenic strains. Missing genes in a pathogen were detected on a microarray slide spotted with 4,071 open reading frames (ORFs) of W3110, a commonly used wild-type K-12 strain. For 22 strains subjected to the CGH microarray analyses 1,424 ORFs were found to be absent in at least one strain. The common backbone of the E. coli genome was estimated to contain about 2,800 ORFs. The mosaic distribution of absent regions indicated that the genomes of pathogenic strains were highly diversified becasue of insertions and deletions. Prophages, cell envelope genes, transporter genes, and regulator genes in the K-12 genome often were not present in pathogens. The gene contents of the strains tested were recognized as a matrix for a neighbor-joining analysis. The phylogenic tree obtained was consistent with the results of previous studies. However, unique relationships between enteroinvasive strains and Shigella, uropathogenic, and some enteropathogenic strains were suggested by the results of this study. The data demonstrated that the CGH microarray technique is useful not only for genomic comparisons but also for phylogenic analysis of E. coli at the strain level.Escherichia coli strains are divided into several disease phenotypes, including nonpathogenic E. coli, enterohemorragic E. coli (EHEC), enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), and urinary tract infectious or uropathogenic E. coli (UPEC) (6). In addition, the strains belonging to the genus Shigella are recognized as E. coil strains because they have many genetic and phenotypic similarities to E. coli (27). Following determination of the complete genome sequence of nonpathogenic E. coli K-12 laboratory strain MG1655 in 1997 (2), the genomes of the two pathogenic strains with the same serotype (O157:H7) were also sequenced (7, 23). A genomic comparison of strain MG1655 and the O157:H7 strains revealed that each genome contained specific genomic regions designated strain-specific islands or loops (7, 23). Recently, the genome sequence of UPEC strain CFT073 was reported to also have unique chromosome regions that are not present in other E. coli genomes that have been sequenced (35). Similar results were obtained in a genomic study of the bacillary dysentery-causing strain Shigella flexneri 2a (12, 34). These results revealed the extensive divergence of the genomes of E. coli strains and indicated that the unique phenotype of each strain should reflect its genomic content. Although whole-genome sequencing is definitely a powerful method for genetics, it is still laborious and expensive. Recently, comparative genomic hybridization (CGH) has been used to facilitate comparisons of unsequenced bacterial genomes in order to look for characteristic genes or chro...