The gut microbiome of healthy Japanese and its microbial and functional uniqueness

Abstract

The human gut microbiome has profound influences on the host's health largely through its interference with various intestinal functions. As recent studies have suggested diversity in the human gut microbiome among human populations, it will be interesting to analyse how gut microbiome is correlated with geographical, cultural, and traditional differences. The Japanese people are known to have several characteristic features such as eating a variety of traditional foods and exhibiting a low BMI and long life span. In this study, we analysed gut microbiomes of the Japanese by comparing the metagenomic data obtained from 106 Japanese individuals with those from 11 other nations. We found that the composition of the Japanese gut microbiome showed more abundant in the phylum Actinobacteria, in particular in the genus Bifidobacterium, than other nations. Regarding the microbial functions, those of carbohydrate metabolism were overrepresented with a concurrent decrease in those for replication and repair, and cell motility. The remarkable low prevalence of genes for methanogenesis with a significant depletion of the archaeon Methanobrevibacter smithii and enrichment of acetogenesis genes in the Japanese gut microbiome compared with others suggested a difference in the hydrogen metabolism pathway in the gut between them. It thus seems that the gut microbiome of the Japanese is considerably different from those of other populations, which cannot be simply explained by diet alone. We postulate possible existence of hitherto unknown factors contributing to the population-level diversity in human gut microbiomes.

Keywords: Japanese; gut; metagenome; microbiome.

Figure 1.

Population-level diversity in human gut microbiomes among the 12 countries. (A) The MDS plot of microbial compositions for all individuals. Each circle represents an individual microbial composition and each colour represents a country of origin. The position based on the average microbial composition for each country is displayed by the abbreviated country name. (B) Comparison of Pearson's correlation coefficients of microbial compositions in individuals within a country and between different countries. Boxes represent the inter-quartile range (IQR), and the lines inside show the median. Whiskers denote the lowest and highest values within 1.5 times the IQR. Asterisk represents P < 0.05. (C) ROC curves and AUC values from the randomForest model. Numbers in parentheses represent the AUC values of the 10 countries.

Figure 2.

Taxonomic comparison of gut microbiomes of populations from the 12 countries. Relative abundances of the four dominant phyla (A), the five genera most enriched and depleted in the JPGM (B), and M. smithii (C) in the 12 countries are shown. Vertical axes represent the relative abundance of the species calculated from the number of mapped reads to the reference genomes.

Figure 3.

Enriched and depleted functions in the JPGM. Functional categories of the KOs most enriched and depleted in the JPGM compared with those of the other 11 countries are shown. The vertical axis represents the proportion of KOs assigned to the category. Asterisks indicate adjusted P < 0.01 (Fisher's exact test).

Figure 4.

Enriched and depleted genes in the acetogenesis, methanogenesis, and dissimilatory sulfate reduction in the JPGM. Relative abundances of KOs involved in the pathways for hydrogen metabolism in acetogenesis, methanogenesis, and dissimilatory sulfate reduction among the 12 countries are shown. Asterisks indicate adjusted P < 0.01 compared with the abundances of the other 11 countries (Student's t-test).

Figure 5.

Comparison of the prevalence of NOG gene families between the JP and the other nine populations. The frequency of NOGs in the JP individuals plotted against those in the other nine countries. Each circle represents a NOG. The vertical axis represents the frequency of NOGs detected in the JP individuals. The horizontal axis represents the average frequency of NOGs detected in the individuals of the nine countries. Fifty-two NOGs significantly highly prevalent in the JP cohort compared with the others (the JP >0.7, the others <0.3, and JP/others ≥3) are coloured with blue. Three NOGs (ENOG4108MQB, ENOG4108ZIS, and ENOG4105WVE) were depicted in red.