Robustness of gut microbiota of healthy adults in response to probiotic intervention revealed by high-throughput pyrosequencing

Abstract

Probiotics are live microorganisms that potentially confer beneficial outcomes to host by modulating gut microbiota in the intestine. The aim of this study was to comprehensively investigate effects of probiotics on human intestinal microbiota using 454 pyrosequencing of bacterial 16S ribosomal RNA genes with an improved quantitative accuracy for evaluation of the bacterial composition. We obtained 158 faecal samples from 18 healthy adult Japanese who were subjected to intervention with 6 commercially available probiotics containing either Bifidobacterium or Lactobacillus strains. We then analysed and compared bacterial composition of the faecal samples collected before, during, and after probiotic intervention by Operational taxonomic units (OTUs) and UniFrac distances. The results showed no significant changes in the overall structure of gut microbiota in the samples with and without probiotic administration regardless of groups and types of the probiotics used. We noticed that 32 OTUs (2.7% of all analysed OTUs) assigned to the indigenous species showed a significant increase or decrease of ≥10-fold or a quantity difference in >150 reads on probiotic administration. Such OTUs were found to be individual specific and tend to be unevenly distributed in the subjects. These data, thus, suggest robustness of the gut microbiota composition in healthy adults on probiotic administration.

Keywords: 16S ribosomal RNA gene; gut microbiota; probiotics; pyrosequencing.

Figure 1.

Assessment of the quantitative accuracy of the analysis of the bacterial composition of two mock communities by various methods. PCA analysis of the data was obtained from various methods using mock01 (a) and mock02 (b). Closed circle: expected, open circle: duplicate qPCR, closed square: pyrosequencing of 16S V1-2 region using 27Fmod, open square: pyrosequencing of 16S V1-2 region using 27F, closed triangle: pyrosequencing of 16S V5-6 region, open diamond: Sanger sequencing of nearly full-length 16S clone.

Figure 2.

Change in OTU number in faecal microbiota with and without probiotic administration. (a) Individual, (b) group, (c) type of probiotics. Black bar indicates Pro(−) samples. Grey bar indicates Pro(+) samples. The error bars represent standard deviation.

Figure 3.

Average UniFrac distance within Pro (−) and Pro(+) and between Pro(−) and Pro(+) for each group, type of probitotics, and all subjects. Average UniFrac distance between any pair of the three distances for type of probiotics and all subjects (a and b), and each group (c and d). Open circle, closed circle, and closed square indicate average UniFrac distance within Pro (−), within Pro (+), and between Pro(−) and Pro(+) samples, respectively.

Figure 4.

Average UniFrac distance within Pro(−) and Pro(+) and between Pro(−) and Pro(+) for each subject. Open circles, closed circles, and closed squares indicate average UniFrac distance within Pro(−), within Pro(+), and between Pro(−) and Pro(+) samples, respectively. Closed triangles indicate average UniFrac distance between samples (S00) of 18 unrelated individuals.

Figure 5.

OTUs showing ≥2-fold change and their difference in quantity between the Pro(−) and Pro(+) samples. The x-axis represents the scale of fold change between the Pro(+) and Pro(−) samples. The y-axis represents the difference (number of reads) in quantity between the Pro(+) and Pro(−) samples. Closed and open circles indicate the administrated probiotic and indigenous species, respectively.

Figure 6.

Distribution of 32 OTUs showing a significant change in 18 subjects. The y-axis indicates the number of OTUs showing significant change between the Pro(−) and Pro(+) samples in each subject (see Supplementary Tables S9 and S10). Open and closed bars indicate increased and decreased OTUs, respectively.