Age-Related Changes in the Composition of Gut Bifidobacterium Species

Abstract

Bifidobacteria are one of the major components in human microbiota that are suggested to function in maintaining human health. The colonization and cell number of Bifidobacterium species in human intestine vary with ageing. However, sequential changes of Bifidobacterium species ranging from newborns to centenarians remain unresolved. Here, we investigated the gut compositional changes of Bifidobacterium species over a wide range of ages. Faecal samples of 441 healthy Japanese subjects between the ages of 0 and 104 years were analysed using real-time PCR with species-specific primers. B. longum group was widely detected from newborns to centenarians, with the highest detection rate. B. breve was detected in approximately 70% of children under 3 years old. B. adolescentis and B. catenulatum groups were predominant after weaning. B. bifidum was detected at almost all ages. The detection rate of B. dentium was higher in the elderly than in other ages. B. animalis ssp. lactis was detected in 11.4% of the subjects and their ages were restricted. B. gallinarum goup was detected in only nine subjects, while B. minimum and B. mongoliense were undetected at any age. The presence of certain Bifidobacterium groups was associated with significantly higher numbers of other Bifidobacterium species/subspecies. Inter-species correlations were found among each species, exception for B. animalis ssp. lactis. These results revealed the patterns and transition points with respect to compositional changes of Bifidobacterium species that occur with ageing, and the findings indicate that there may be symbiotic associations between some of these species in the gut microbiota.

Fig. 1

Distribution of seven Bifidobacterium species in the faeces of healthy Japanese subjects aged 0–104 years. Cell numbers were determined as the log10 of cells per gram wet weight of faecal samples. The box plots show the interquartile range (IQR) of the cell numbers shown in each section. Open circles and asterisks indicate outliers between 1.5- and 3.0-fold IQR and over 3.0-fold IQR, respectively. The age of each group is as shown in Table 1. The detection limits for each bacterial species were determined using real-time PCR to be 10⁶/g wet weight of faeces

Fig. 2

Detection rate of each bifidobacteria species in each segmented age group. ADO, B. adolescentis group; LAC, B. animalis spp. lactis; BIF, B. bifidum; BRE, B. breve; CAT, B. catenulatum group; DEN, B. dentium; GAL, B. gallinarum group;LON, B. longum group. The age of each group is as shown in Table 1

Fig. 3

Number of Bifidobacterium species. a Distribution of the numbers of Bifidobacterium species that were found in individuals in each age group. The age of each group is as shown in Table 1. b Differences in species numbers between subjects with or without the presence of identical species (i.e. the number of species excluding the identical species). ADO, B. adolescentis group; LAC, B. animalis spp. lactis; BIF, B. bifidum; BRE, B. breve; CAT, B. catenulatum group; DEN, B. dentium; GAL, B. gallinarum group; LON, B. longum group

Fig. 4

Possible ecological compatibility among Bifidobacterium species. The solid and dashed lines indicate positive and negative correlations, respectively. The widths of the connected lines indicate the abundance of the coefficient value. ADO, B. adolescentis group; LAC, B. animalis spp. lactis; BIF, B. bifidum; BRE, B. breve; CAT, B. catenulatum group; DEN, B. dentium; GAL, B. gallinarum group; LON, B. longum group. Online Resource 1. Distribution of B. longum ssp. longum in the faeces of healthy Japanese subjects aged 0–104 years. The age of each group is as shown in Table 1