Effects of long-term intake of a yogurt fermented with Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131 on mice

Effects of long-term intake of a yogurt fermented with Lactobacillus delbrueckii subsp.... Abstract The gut is an extremely complicated ecosystem where micro-organisms, nutrients and host cells interact vigorously. Although the function of the intestine and its barrier system weakens with age, some probiotics can potentially prevent age-related intestinal dysfunction. Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131, which are the constituents of LB81 yogurt, are representative probiotics. However, it is unclear whether their long-term intake has a beneficial influence on systemic function. Here, we examined the gut microbiome, fecal metabolites and gene expression profiles of various organs in mice. Although age-related alterations were apparent in them, long-term LB81 yogurt intake led to an increased Bacteroidetes to Firmicutes ratio and elevated abundance of the bacterial family S24-7 (Bacteroidetes), which is known to be associated with butyrate and propanoate production. According to our fecal metabolite analysis to detect enrichment, long-term LB81 yogurt intake altered the intestinal metabolic pathways associated with propanoate and butanoate in the mice. Gene ontology analysis also revealed that long-term LB81 yogurt intake influenced many physiological functions related to the defense response. The profiles of various genes associated with antimicrobial peptides-, tight junctions-, adherens junctions- and mucus-associated intestinal barrier functions were also drastically altered in the LB81 yogurt-fed mice. Thus, long-term intake of LB81 yogurt has the potential to maintain systemic homeostasis, such as the gut barrier function, by controlling the intestinal microbiome and its metabolites. aging, intestinal epithelial barrier, metabolome, microbiome, probiotics Introduction The prevention of age-related chronic disorders has become a major focus of current medical research (1, 2). Because age-related deterioration of mechanical, physiological and psychological functions is mostly inevitable, the incidence of age-related diseases such as atherosclerosis, arthritis, cancer, osteoporosis, type 2 diabetes, hypertension and Alzheimer’s has increased in the world (1, 3). Fascinatingly, several lines of evidence have shown that interactions between intestinal micro-organisms and the host are crucial in the context of health and disease (4–7). Dysbiosis, for example, which is defined as an imbalance in the repertoire of the intestinal microbiota, is correlated with aberrant immune responses such as abnormal inflammatory cytokine production (8). Intestinal dysbiosis has the potential to alter gut metabolites, and is more frequent in older individuals than in young people (9). Dysbiosis in the elderly is thought to be responsible for a defective intestinal barrier with an associated altered secretion of the antimicrobial peptides (AMPs) produced by intestinal epithelial cells, thereby resulting in the development of age-related diseases (10). Notably, it has been shown that probiotic bacteria can have a direct effect on the functioning of the intestinal epithelial barrier (11). Previous studies have revealed that age-related changes occur in gut microbes (12), intestinal metabolites (13) and gene expression in various organs (14). However, the current lack of multi-omics approaches for investigating the systemic biological changes associated with aging mean that the effects of aging on the body are not fully understood. It has been shown recently that the administration of probiotic-containing products such as yogurt can improve the gut environment and reduce the risk of inflammatory disorders such as autoimmune diseases and allergies (15). Among the many probiotic types, LB81 yogurt, which contains Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131, is a traditionally consumed product. However, it remains unclear whether long-term intake of LB81 yogurt is beneficial to health. Here, we first determined the influence of aging on the fecal microbiome, fecal metabolites and gene expression in various organs in mice. As other studies (12–14) have found, each of them are altered by aging. We also evaluated the effect of long-term LB81 yogurt intake on the survival rates, body weights, fecal microbiome, fecal metabolites and gene expression in various organs in mice. When LB81 yogurt was used to supplement a normal mouse diet (from 8 to 25 months of age), the mice did not show any alterations in their gut bacterial alpha diversity index compared with the control mice; however, the increased ratio of Bacteroidetes to Firmicutes and the fecal metabolites associated with propanoate metabolism and butanoate metabolism were significantly linked with the long-term intake of this yogurt. On the basis of our gene ontology (GO) analysis results, long-term intake of LB81 yogurt modulated many different systemic functions, and the expression patterns of the intestinal barrier-associated genes were drastically altered. These findings suggest that long-term LB81 yogurt intake can help to maintain gut homeostasis and possibly prevent aging-associated diseases. Methods Mice The 8-month-old ICR male mice obtained from Japan SLC (Hamamatsu, Japan) were fed on a normal diet (AIN-93) (Oriental Yeast, Tokyo, Japan) (n = 39, hereafter referred to as the control mice) or the same diet containing 1–2% LB81 yogurt (n = 40, hereafter referred to as the LB81 yogurt-intake mice) until they were 28 months of age. The survival rates and body weight changes were monitored. The 2-month-old ICR male mice obtained from Japan SLC and were fed on a normal diet (AIN-93) (Oriental Yeast) (n = 6, hereafter referred to as the control mice) until 4 months of age. Fecal samples were longitudinally collected from the control mice (n = 22) and the LB81 yogurt-intake mice (n = 25) at 10, 15, 21 and 25 months of age. Each mouse was reared in a single cage under specific pathogen-free conditions at the Experimental Animal Facility of Meiji Co., Ltd (Japan). The experiments were performed according to the guidelines of the Animal Research Committee of Meiji Co., Ltd and the Institute of Medical Science (University of Tokyo, Japan). 16S rRNA gene analysis Bacterial DNA extraction from each fecal sample was performed as described previously with some modifications (16). The 16S rRNA V3–V4 region was PCR-amplified and purified as described previously (17). For each sample, an equal amount of each DNA amplicon library was mixed and sequenced on the MiSeq instrument (Illumina, San Diego, CA, USA) using a MiSeq v3 Reagent kit with 20% PhiX (Illumina). The 16S rRNA gene analysis was performed using QIIME (version 1.9.1) (18), as described previously (17). Briefly, the raw sequence data were subjected to quality filtering and paired-end read merging. The insufficient sequence depth below the 15000 threshold meant that three of the sequenced samples (two from the control mice, one from the LB81 yogurt-intake mice) were removed from the following analysis. The processed sequence reads were assigned to operational taxonomic units (OTUs) with a 97% identity threshold using the Greengenes database (version 13.8) (18). Faith’s phylogenetic alpha diversity estimate and principal coordinate analysis (PCoA) of the weighted UniFrac distance matrices were performed using QIIME. To characterize the changes occurring in family-level microbial abundance during aging, the relative abundance at each time point was standardized by Z-scores using the mean relative abundance at 10 months of age. Gut microbial families were clustered based on the standardized profile using the R function Kmeans in the amap package, with K = 4 and distance metric ‘correlation’. Metabolome analysis To extract the ionic metabolites, ~50 mg of each fecal sample was dissolved in Milli-Q water (1:9, w/v ratio). After centrifugation, 20 µl of an internal standard solution (H3304-1002, Human Metabolome Technologies, Inc., HMT, Yamagata, Japan) was added to 80 µl of the supernatant. To remove the macromolecules, the solution was centrifugally filtered through a Millipore 5-kDa cutoff filter (UltrafreeMC-PLHCC, HMT) (9100 × g, 4°C, 60 min) and subsequently analyzed by capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS). Metabolomic measurements were performed by a facility service at HMT, as described previously (19–21). The peaks detected were annotated against the putative metabolites from the HMT metabolite database linked to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (release 54.1) (22, 23). For this analysis, we used the annotated KEGG compound identity (ID) and the relative area value for the peak area of each metabolite was normalized against the sample volume. The relative area values below the detection limit were set to 2−52 as in the HMT reference. Metabolites with multiple KEGG compound ID annotations were excluded from the following analysis, except for the amino acids, where we used the KEGG compound IDs of the l-form amino acids. The aging effect was analyzed using the metabolites whose relative areas showed more than a 2-fold change at any of 10, 15, 21 and 25 month time points compared with any of the other time points. The relative area values were Z-score standardized using the relative area values at 10 months of age, and then clustered using the R function Kmeans in the amap package, with K = 5 and distance metric ‘correlation’. The KEGG pathway enrichments for the metabolites in each metabolite cluster were examined using MBROLE 2.0 (24). The effect of the probiotics was examined using the relative area values of the control and the LB81 yogurt-intake mice at 25 months of age. Metabolites whose relative area values changed >2-fold between the control and the LB81 yogurt-intake group were examined using MBROLE 2.0. Microarray analysis Jejunum, ileum, distal colon, liver and spleen tissues were collected from the control mice at 4 (n = 6) and 28 (n = 9) months of age, and the LB81 yogurt-intake mice at 28 months of age (n = 10). Equal amounts of total RNA, previously extracted from the tissues of each mouse, were pooled and mixed for each group. The RNA mixtures (100 ng) were individually amplified using the GeneChip 3′IVT Express kit and then hybridized to an Affymetrix Mouse Genome 430 2.0 chip (GPL1261; 45101 probes) (n = 1 per group). After hybridization, the chip was washed, stained on the Affymetrix Fluidics Station 450 and scanned using the Affymetrix GeneChip Scanner 3000 7G System, with all the operations being performed by the Affymetrix GeneChip Command Console (AGCC) Software (accession code GEO: GSE104375, GSE111578 and GSE111579). The analysis steps described below were performed separately to compare the aging effect (4 versus 28 months of age in the control mice) and to compare the effect of the probiotics (the control versus the LB81 yogurt-intake mice at 28 months of age). CEL files were processed and normalized by frozen robust multi-array analysis using the frma and mouse4302frmavecs packages from Bioconductor in the R statistical language (25). Detection calls were generated by MAS5.0 using the affy package from Bioconductor (26). The probes were annotated with the mouse4302.db package from Bioconductor and the annotation table of the Mouse Genome 430 2.0 array provided by Affymetrix (release 36). Probe sets that fulfilled any of the following criteria were excluded: (i) the detection calls were absent from both the control and the sample of interest; (ii) the probe set ID contained ‘x_’ or ‘s_’; (iii) the gene symbol was NA; and (iv) the transcript assignment grade of the probe set was lower than ‘A’. In cases where multiple probe sets had the same gene symbol annotation, one probe set was selected in the following order: (i) the probe set with the largest number of SWISS-PROT annotations; (ii) the unique probe set id (‘_at’); (iii) the probe set that targeted the longest region; and (iv) the probe set with the highest sum of the gene expression signals. To detect the differentially expressed genes, the intensity-dependent Z-score was calculated (27). The fold change of the signal between the control and the sample of interest was divided by the standard deviation obtained at each log-transformed average signal with a window size of 0.5. The GO biological process term enrichment was examined for the probe sets whose intensity-dependent Z-scores were >2 or < −2 by Fisher’s exact test using the topGO and mouse4302.db packages from Bioconductor (28). Statistical analysis Statistical analysis was performed using R statistical language. A P-value of <0.05 was defined as statistically significant. Results The aging gut microbiome To investigate age-related changes in the gut microbiome, we longitudinally collected fecal samples from 22 mice at 10, 15, 21 and 25 months of age, and performed 16S rRNA gene sequencing on these samples. Samples with insufficient sequence depths were removed, and the sequenced samples from 20 mice were used for the following analysis. Our analysis of the gut microbial compositions in the mice was based on the relative abundance of the OTUs that clustered with 97% identity scores using QIIME (18). The gut microbial compositions at the phylum level showed that the relative abundance of Actinobacteria decreased, whereas that of the Bacteroidetes increased with age (Fig. 1A). The ratio of Bacteroidetes to Firmicutes, which represent the major bacterial phyla in this study, increased with age and peaked at 21 months of age (P < 0.01) (Fig. 1B). The bacterial diversity evaluation within each sample showed that bacterial alpha diversity increased at 21 and 25 months (P < 0.01) (Fig. 1C). To assess the bacterial community across the aging stages, a PCoA for the weighted UniFrac distance matrices was performed, the results of which showed that at 21 months of age the gut microbial community occupied the largest distance from the other age groups (Fig. 1D). We also analyzed the profiles of the family-level microbial abundance by clustering them using the K-means algorithm (Fig. 1E and Supplementary Table 1). Interestingly, we identified three clusters that increased at different time points with peaks at 15 months (bacterial cluster 1), 21 months (bacterial cluster 2) and 25 months (bacterial cluster 3), whereas only one bacterial gut cluster decreased with aging (bacterial cluster 4). The most abundant bacterial families in bacterial clusters 1, 2, 3 and 4 were Erysipelotrichaceae (Firmicutes), S24-7 (Bacteroidetes), F16 (TM7) and Bifidobacteriaceae (Actinobacteria), respectively. Thus, the intestinal microbiome underwent considerable changes at around 21 months of age and its composition shifted with aging in at least four different patterns. Fig. 1. View largeDownload slide Gut microbial compositions at different age points. (A) Gut microbial compositions at the phylum level based on the relative abundance of OTUs for the fecal samples from 20 mice at 10, 15, 21 and 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance of OTUs at phylum level. The P-value was calculated using a pairwise Wilcoxon signed rank test for paired data with Holm P-value correction. (C) Bacterial alpha diversity based on Faith’s phylogenetic alpha diversity index. The P-value was calculated using a pairwise t-test for paired data with Holm P-value correction. (D) PCoA on the gut microbial community according to the weighted UniFrac distance matrices. (E) Age-related changes in gut microbial abundance were classified by the K-means algorithm on the basis of the standardized relative abundance of OTUs at the family level. Data are shown as the mean values of the standardized relative abundance for each bacteria ± SD. NS, not significant, **P < 0.01, *P < 0.05. Fig. 1. View largeDownload slide Gut microbial compositions at different age points. (A) Gut microbial compositions at the phylum level based on the relative abundance of OTUs for the fecal samples from 20 mice at 10, 15, 21 and 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance of OTUs at phylum level. The P-value was calculated using a pairwise Wilcoxon signed rank test for paired data with Holm P-value correction. (C) Bacterial alpha diversity based on Faith’s phylogenetic alpha diversity index. The P-value was calculated using a pairwise t-test for paired data with Holm P-value correction. (D) PCoA on the gut microbial community according to the weighted UniFrac distance matrices. (E) Age-related changes in gut microbial abundance were classified by the K-means algorithm on the basis of the standardized relative abundance of OTUs at the family level. Data are shown as the mean values of the standardized relative abundance for each bacteria ± SD. NS, not significant, **P < 0.01, *P < 0.05. Fecal metabolites associated with aging Some gut microbes produce a variety of intestinal metabolites that have close relationships with health and disease through the circulatory system and regulation of the intestinal mucosal barrier (29). Therefore, we investigated whether any changes in fecal metabolites occurred upon aging in the mice. Fecal samples were longitudinally collected at 10, 15, 21 and 25 months of age. Metabolites from the pooled fecal samples were measured at each time point using CE-TOFMS instrumentation (Supplementary Table 2). To characterize the changes occurring in the metabolites during the aging process, we clustered them based on the standardized relative area values using the K-means algorithm. Notably, five clusters were defined and three of them showed increased abundances at the following time points: 15 months (metabolite cluster 1), 21 months (metabolite cluster 2) and 25 months (metabolite cluster 3); however, two of them showed decreased abundances with aging (metabolite clusters 4 and 5) (Fig. 2A). To assess the metabolic pathways involved in each cluster, we tested for KEGG metabolic pathway enrichment using MBROLE 2.0 (Fig. 2B and Supplementary Table 3). In the metabolite cluster 1, central carbon metabolism such as ‘galactose metabolism’ (P = 5 × 10−3), ‘pentose and glucuronate interconversions’ (P = 6 × 10−3) and ‘glycolysis/gluconeogenesis’ (P = 2 × 10−2) was enriched. In metabolite cluster 2, metabolism related to short-chain fatty acids (SCFAs) such as ‘propanoate metabolism’ (P = 2 × 10−2) and ‘butanoate metabolism’ (P = 3 × 10−2) showed enrichment. In metabolite cluster 3, carbohydrate metabolism such as ‘glyoxylate and dicarboxylate metabolism’ (P = 2 × 10−2), ‘ascorbate and aldarate metabolism’ (P = 2 × 10−2) and ‘starch and sucrose metabolism’ (P = 4 × 10−2) was enriched. In metabolite cluster 4, which was defined as displaying a decreased profile, ‘aminoacyl-tRNA biosynthesis’ (P = 3 × 10−9) was significantly enriched with 18 amino acids being involved in this pathway (Supplementary Table 3). In metabolite cluster 5, lipid metabolism (e.g. ‘glycerophospholipid metabolism’) was enriched (P = 2 × 10−2). Thus, the metabolites involved in carbohydrate metabolism (metabolite clusters 1 and 3) and SCFA (metabolite cluster 2) increased transiently with aging, whereas the metabolites related to amino acid metabolism (metabolite cluster 4) and lipids (metabolite cluster 5) decreased with age. Fig. 2. View largeDownload slide Age-related changes in the fecal metabolites. (A) Age-related changes in the fecal metabolite amounts were classified by the K-means algorithm on the basis of the standardized relative area values. The fecal samples from 22 mice at 10, 15, 21 and 25 months of age were used. Data are shown as the mean values for the standardized relative area of each metabolite ± SD. (B) KEGG pathways in each fecal metabolite cluster that were significantly enriched on the basis of MBROLE 2.0. The bar graph shows the top six enriched KEGG pathways where uncorrected P < 0.05. Fig. 2. View largeDownload slide Age-related changes in the fecal metabolites. (A) Age-related changes in the fecal metabolite amounts were classified by the K-means algorithm on the basis of the standardized relative area values. The fecal samples from 22 mice at 10, 15, 21 and 25 months of age were used. Data are shown as the mean values for the standardized relative area of each metabolite ± SD. (B) KEGG pathways in each fecal metabolite cluster that were significantly enriched on the basis of MBROLE 2.0. The bar graph shows the top six enriched KEGG pathways where uncorrected P < 0.05. The effect of aging on the gene expression profiles of the intestines and other organs We next focused on how gene expression might change in various organs with aging. The intestinal tissues (jejunum, ileum and distal colon) and other tissues (liver and spleen) from mice at 4 months (n = 6) and 28 months (n = 9) of age were isolated, and their gene expression profiles were evaluated using cDNA microarrays. According to the GO analysis results, the GO term ‘defense response’ (P = 9 × 10−27) in the liver was the most significantly enriched among the up-regulated genes in the aged mice when compared with the results of the same analysis in the younger mice (Fig. 3A). In the liver, the genes involved in the ‘innate immune response’ (P = 2 × 10−18) and the ‘adaptive immune response’ (P = 2 × 10−10) were also associated with aging in mice. In the distal colon, the genes involved in ‘leukocyte activation’ (P = 3 × 10−16) and ‘lymphocyte activation’ (P = 7 × 10−16) were enriched in the aged mice. In the spleen, the biological terms ‘defense response’ (P = 8 × 10−12) and ‘defense response to bacterium’ (P = 2 × 10−10) were remarkably enriched in the aged mice. In contrast, no significantly enriched GO terms were identified among the up-regulated genes in the jejunum and ileum tissues from the aged mice, a finding that is consistent with a previous study (30). As for the down-regulated genes in the aged mice compared with those in the younger mice, ‘lipid metabolic process’ was commonly enriched in the jejunum (P = 5 × 10−9) and ileum (P = 2 × 10−15) (Fig. 3B). These results suggest that the biological functions of each organ are preferentially affected by age. Fig. 3. View largeDownload slide GO terms for the differentially expressed genes in the aged mice compared with the younger mice. GO term enrichment analysis was performed for the up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the aged mice (n = 9) compared with the younger mice (n = 6). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the aged mice (28 months of age) compared with the younger mice (4 months of age). Only GO terms where the number of annotated genes was below 1000 and where the P-values were <10−9 in at least one tissue were visualized. The color intensity is based on the uncorrected P-values calculated via Fisher’s exact test. Fig. 3. View largeDownload slide GO terms for the differentially expressed genes in the aged mice compared with the younger mice. GO term enrichment analysis was performed for the up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the aged mice (n = 9) compared with the younger mice (n = 6). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the aged mice (28 months of age) compared with the younger mice (4 months of age). Only GO terms where the number of annotated genes was below 1000 and where the P-values were <10−9 in at least one tissue were visualized. The color intensity is based on the uncorrected P-values calculated via Fisher’s exact test. Effect of long-term LB81 yogurt intake on survival rates and body weight changes Because the gut microbiome, fecal metabolites, and gene expression in various organs were altered with aging, we investigated the effects of the long-term intake of LB81 yogurt on aging. We fed a normal diet with (n = 40) or without (n = 39) 1–2% LB81 yogurt to mice for 20 months, and monitored their survival rates and body weight changes. No significant difference in the survival rates of the mice fed on LB81 yogurt and the control mice was noted (Fig. 4A). Additionally, the body weight of the mouse group fed on LB81 yogurt was similar to that of the control group (Fig. 4B). Thus, long-term intake of LB81 yogurt had no effect on mouse survival or body weight. Fig. 4. View largeDownload slide Physiological effects of long-term supplementation with LB81 yogurt. (A) Kaplan–Meier survival curve for the control (n = 39) and the LB81 yogurt-intake groups (n = 40). The P-value was calculated using a log-rank test. (B) Body weight changes in the control and the LB81 yogurt-intake groups during the study period. Data are shown as the mean ± SD. Control mice: gray color, LB81 yogurt-intake mice: black color. Fig. 4. View largeDownload slide Physiological effects of long-term supplementation with LB81 yogurt. (A) Kaplan–Meier survival curve for the control (n = 39) and the LB81 yogurt-intake groups (n = 40). The P-value was calculated using a log-rank test. (B) Body weight changes in the control and the LB81 yogurt-intake groups during the study period. Data are shown as the mean ± SD. Control mice: gray color, LB81 yogurt-intake mice: black color. Effect of long-term LB81 yogurt intake on the intestinal microbiome To elucidate the effect of long-term administration of LB81 yogurt on the gut microbial composition in the mice, we used pyrosequencing-based 16S rRNA gene profiling to analyze the fecal microbiota from the 25-month-old mice fed on a normal diet with (n = 24) or without (n = 20) LB81 yogurt supplementation for 17 months. The relative abundance of gut microbes at the phylum level showed that the gut microbial composition changed in the mice fed on LB81 yogurt (Fig. 5A). Of note, we found that the Bacteroidetes to Firmicutes ratio was significantly higher in the mice fed on LB81 yogurt than in the control mice (P < 0.05) (Fig. 5B). There was, however, no significant difference in bacterial alpha diversity (Faith’s phylogenetic alpha diversity) between the control and the LB81 yogurt-fed groups (Fig. 5C), and the PCoA analysis based on the weighted UniFrac distance did not separate the control group from the LB81 yogurt-treated group (Fig. 5D). Our comparison of the relative abundance of the bacterial community at the family level showed that S24-7 (Bacteroidetes) was the most abundant bacteria among the significantly increased bacteria in the LB81 yogurt-treated group as compared with the control group (P < 0.05) (Fig. 5E). In addition, the relative abundance of Streptococcaceae (Firmicutes), Deferribacteraceae (Deferribacteres) and Paraprevotellaceae (Bacteroidetes) was significantly higher in the LB81 yogurt-treated group than in the control group (P < 0.05) (Fig. 5F–H). In contrast, the relative abundance of Turicibacteraceae (Firmicutes) and Moraxellaceae (Proteobacteria) was down-regulated in the LB81 yogurt-treated group (P < 0.05) (Fig. 5I and J). Thus, long-term LB81 yogurt intake increased the ratio of Bacteroidetes to Firmicutes in the mice whose diet was supplemented with it, but overall the microbial diversity did not alter in these mice. Fig. 5. View largeDownload slide Gut microbial compositions and diversities in the control mice and the LB81 yogurt-intake mice. (A) Gut microbial compositions based on the relative abundance of OTUs at the phylum level from the control (n = 20) and the LB81 yogurt-intake group (n = 24) at 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance data from the OTUs at phylum level. The P-value was calculated using a two-sided Wilcoxon rank sum test for unpaired data. (C) Bacterial alpha diversities for the fecal microbial community using Faith’s phylogenetic alpha diversity index. The P-value was calculated using the two-sided Wilcoxon rank sum test for unpaired data. (D) PCoA on the weighted UniFrac distance matrices for the microbial community. (E–H) Top four gut microbial families whose relative abundances increased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. (I and J) Top two gut microbial families whose relative abundances decreased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. The P-value was calculated using a one-sided Wilcoxon rank sum test for unpaired data. *P < 0.05. NS, not statistically significant. Fig. 5. View largeDownload slide Gut microbial compositions and diversities in the control mice and the LB81 yogurt-intake mice. (A) Gut microbial compositions based on the relative abundance of OTUs at the phylum level from the control (n = 20) and the LB81 yogurt-intake group (n = 24) at 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance data from the OTUs at phylum level. The P-value was calculated using a two-sided Wilcoxon rank sum test for unpaired data. (C) Bacterial alpha diversities for the fecal microbial community using Faith’s phylogenetic alpha diversity index. The P-value was calculated using the two-sided Wilcoxon rank sum test for unpaired data. (D) PCoA on the weighted UniFrac distance matrices for the microbial community. (E–H) Top four gut microbial families whose relative abundances increased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. (I and J) Top two gut microbial families whose relative abundances decreased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. The P-value was calculated using a one-sided Wilcoxon rank sum test for unpaired data. *P < 0.05. NS, not statistically significant. Effect of long-term LB81 yogurt intake on fecal metabolites Because LB81 yogurt is a probiotic source that is thought to have beneficial effects on host immune defenses by modulating the concentrations of, for example, the intestinal metabolites butyrate and propanoate (31), we also analyzed fecal metabolite levels using CE-TOFMS. Our enrichment analysis of the metabolomic data revealed that two pathways, propanoate metabolism and butanoate metabolism, were significantly enhanced in the LB81 yogurt-fed group (Table 1), suggesting that long-term intake of this yogurt alters these gut-associated metabolites. Table 1. Significantly enriched KEGG pathways for the up-regulated fecal metabolites in the LB81 yogurt-intake mice (n = 25) compared with the control mice (n = 22) KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid aKEGG pathways with uncorrected P-values of <0.05 are shown in this table. bEnrichment annotations involving >1 compound are shown in the table. View Large Table 1. Significantly enriched KEGG pathways for the up-regulated fecal metabolites in the LB81 yogurt-intake mice (n = 25) compared with the control mice (n = 22) KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid aKEGG pathways with uncorrected P-values of <0.05 are shown in this table. bEnrichment annotations involving >1 compound are shown in the table. View Large Effect of long-term LB81 yogurt intake on gene expression in the intestines, liver and spleen As described above, long-term LB81 yogurt intake was found to regulate gut metabolism. Gut microbiomes and metabolites have also been shown to affect the functions of the intestine (e.g. intestinal barrier functioning), liver and spleen (29, 32, 33). Therefore, we isolated the jejunum, ileum, distal colon, liver and spleen tissues from the control (n = 9) and LB81 yogurt-treated mice (n = 10) at 28 months of age, and examined their gene expression profiles using a cDNA microarray. Compared with the control mice, among the up-regulated genes in the LB81 yogurt-fed group, we found that genes involved in the ‘defense response’ were commonly enriched in the jejunum (P = 1 × 10−10), ileum (P = 4 × 10−4), distal colon (P = 3 × 10−4), liver (P = 2 × 10−6) and spleen (P = 1 × 10−6) (Fig. 6A). In the jejunum, we found that the genes associated with ‘digestion’ (P = 6 × 10−10) were enriched among the up-regulated genes in the aged mice. In the liver, we found that the genes associated with the ‘acute inflammatory response’ (P = 2 × 10−7) were enriched. A different profile was also seen for the spleen, where the genes involved in ‘disruption by host of symbiont cells’ (P = 4 × 10−8) and its subterm ‘killing by host of symbiont cells’ (P = 4 × 10−8) were enriched. Conversely, in comparison with the control mice, among the down-regulated genes in the LB81 yogurt-intake mice, those involved in the ‘regulation of B cell proliferation’ (P = 2 × 10−11) and its subterm ‘positive regulation of B cell proliferation’ (P = 4 × 10−10) were enriched in the spleen (Fig. 6B). Collectively, these results suggest that long-term intake of LB81 yogurt alters the gene expression profiles associated with immune functions in the tissues of the jejunum, ileum, distal colon, liver and spleen from the aged mice. Fig. 6. View largeDownload slide GO terms for the differentially expressed genes in the LB81 yogurt-intake group compared with the control group. GO term enrichment analysis was performed for up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the LB81 yogurt-intake mice (n = 10) compared with the control mice (n = 9). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the LB81 yogurt-intake mice compared with the control mice. Only GO terms where the number of annotated genes is below 1000 and where the P-values were <10−6 in at least one tissue were visualized. The color intensity and circular size indicate the uncorrected P-values calculated via Fisher’s exact test. Fig. 6. View largeDownload slide GO terms for the differentially expressed genes in the LB81 yogurt-intake group compared with the control group. GO term enrichment analysis was performed for up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the LB81 yogurt-intake mice (n = 10) compared with the control mice (n = 9). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the LB81 yogurt-intake mice compared with the control mice. Only GO terms where the number of annotated genes is below 1000 and where the P-values were <10−6 in at least one tissue were visualized. The color intensity and circular size indicate the uncorrected P-values calculated via Fisher’s exact test. Effect of long-term LB81 yogurt intake on intestinal barrier functions The administration of probiotic-containing products has been shown to enhance intestinal barrier function (11). In the present study, we found that the genes involved in the ‘defense response’ were significantly enriched in the jejunum, ileum and distal colon tissues of the LB81 yogurt-fed group (Fig. 6A). Therefore, we next focused on the genes associated with AMPs, tight junctions, adherens junctions and mucus. Among the AMP genes, up-regulation of Reg3b, Reg3g, Prss22 and Ang4 expression was common in the jejunum, ileum and distal colon tissues of the LB81 yogurt-fed group (Fig. 7). Moreover, enhanced expression of Reg3a, Prss2, Pla2g2a, Defa4, Mmp7 and Ltf also occurred in the jejunum, while that of Prss12, Defb1 and Lyz2 increased in the ileum. These results suggest that long-term intake of LB81 yogurt has the potential to enhance the expression of intestinal barrier-related genes. Conversely, both increased and decreased expression of the genes associated with tight- and adherens-junctions was observed in the LB81 yogurt-fed mice. Furthermore, among the mucus-related genes, enhanced expression of Muc2, C1galt1 and Retnlb genes occurred only in the distal colon. These findings raise the possibility that the altered intestinal metabolism induced by long-term LB81 yogurt intake is associated with the maintenance of intestinal barrier functions. Fig. 7. View largeDownload slide Differentially expressed intestinal barrier-related genes in the LB81 yogurt-intake group compared with the control group. Expression of intestinal barrier-related genes in the jejunum, ileum and distal colon tissues was compared between the LB81 yogurt-intake mice (n = 10) and the control mice (n = 9). Colors indicate the intensity-dependent Z-score. Yellow and blue colors represent increased and decreased gene expression in the LB81 yogurt-intake mice compared with the control mice, respectively. Fig. 7. View largeDownload slide Differentially expressed intestinal barrier-related genes in the LB81 yogurt-intake group compared with the control group. Expression of intestinal barrier-related genes in the jejunum, ileum and distal colon tissues was compared between the LB81 yogurt-intake mice (n = 10) and the control mice (n = 9). Colors indicate the intensity-dependent Z-score. Yellow and blue colors represent increased and decreased gene expression in the LB81 yogurt-intake mice compared with the control mice, respectively. Discussion In the present study, we performed comprehensive analyses on the gut microbiome, fecal metabolites and gene expression profiles of various organs in mice. As the results of the longitudinal examination in the mice fed with LB81 yogurt revealed, long-term intake of the probiotics had various effects on the normal age-related changes that occur over time. Aging is associated with an alteration in the gut microbiome of mice and humans (12, 34). The present study showed clearly the age-related changes in the gut microbial composition, such as a decreased abundance of Actinobacteria and an increased abundance of Bacteroidetes with aging. Additionally, the gut microbial composition underwent considerable changes at around 21 months of age (Fig. 1A, B and D). These findings are consistent with those from a previous study where the gut microbial composition of middle-aged mice (~20 months of age) was found to differ from that of the older mice (~28 months of age) (12). Thus, the age of around 21 months might be a turning point in the intestinal environment during mouse aging. Although the intestinal microbiome and metabolome are influenced by aging (12, 13), alterations to them can be diverse. In the present study, the abundance of gut microbial families changed with aging in four different patterns (Fig. 1E) and the intestinal metabolite amounts were altered in five different patterns (Fig. 2A and B). Of particular note, the age-related changes in metabolite cluster 1 were similar to those of bacterial cluster 1 (Figs 1E, 2A and B). Metabolite cluster 2 also showed similar age-related changes to those in bacterial cluster 2 (Figs 1E, 2A and B). Thus, age-related changes in the gut microbiome might be linked with intestinal metabolite alterations. Therefore, further studies will be required to better understand the correlation between the gut microbiome and intestinal metabolites during aging. Here, we showed that many immune-related GO terms were significantly enriched with aging in the distal colon, liver and spleen (Fig. 3). These results imply that age-related changes to the biological functions represented by chronic low-grade inflammation with advanced age, which is known as ‘inflamm-aging’ (30, 35), are induced. A decreased Bacteroidetes to Firmicutes ratio is known to be associated with various medical conditions and diseases, such as obesity and hypertension (36, 37). As shown in Fig. 5(B), the long-term intake of LB81 yogurt results in an increased Bacteroidetes to Firmicutes ratio, suggesting that increasing this ratio by long-term LB81 yogurt intake might prevent dysbiosis-associated diseases. A diet high in saturated fat has been shown to alter the intestinal microbiome, leading to an increased incidence of obesity (38). Thus, an interesting future direction would be to determine whether long-term intake of LB81 in mice improves the altered intestinal environments introduced by high-fat diets. We found that propanoate and butanoate metabolic pathways in the mice were enhanced by long-term LB81 yogurt intake (Table 1). Because L. delbrueckii subsp. bulgaricus and Streptococcus have been shown to produce SCFA (39, 40), LB81 yogurt itself might have the potential to be associated with the production of propanoate and butanoate. Additionally, SCFAs have been shown to regulate bacterial composition and pathogenic bacteria by adjusting the pH in the intestine (41). Interestingly, the relative abundance of S24-7, which has been reported to be associated with the metabolism of propanoate and butanoate (42, 43), was significantly enhanced in the LB81 yogurt-fed group (Fig. 5E). These data indicated the possibility that long-term intake of LB81 yogurt could be associated with SCFA metabolism via modulation of commensal bacteria in addition to their direct supply of SCFAs. However, S24-7 is not a symbiotic micro-organism in humans. Therefore, it would be interesting to investigate whether the intestinal microbiome similar to S24-7 exists in humans after long-term intake of LB81 yogurt in the future. AMPs, tight-junctions, adherens-junctions and mucus all play important roles in inhibiting pathogen invasion (44, 45). Although aging contributes to intestinal barrier dysfunction (46, 47), probiotics are effective at preventing age-related intestinal dysfunction in both mice and humans (11). It is worth noting that long-term LB81 yogurt intake resulted in the increased expression of Reg3b, Reg3g, Prss22 and Ang4 in the jejunum, ileum and distal colon (Fig. 7), and enhanced expression of Muc2, C1galt1 and Retnlb in the distal colon. Thus, the present study provides evidence that long-term intake of LB81 yogurt can regulate intestinal epithelial function in mice. Our study has revealed that aging is associated with alterations in the intestinal microbiome, metabolites and gene expression of various organs in mice, and that long-term LB81 yogurt intake can support systemic functions such as intestinal barrier functioning in these mice. However, further studies are needed to determine whether probiotic yogurts can improve age-related dysfunction in humans. Supplementary Table 1 lists the bacterial families in bacterial clusters. Supplementary Table 2 lists the fecal metabolites from mice fed with the normal diet. Supplementary Table 3 shows the enriched KEGG pathways in fecal metabolite clusters. Supplementary Table 4 lists the fecal metabolites from the control mice and the LB81 yogurt-intake mice at 25 months of age. Supplementary data Supplementary data are available at International Immunology Online. Funding This work was supported by the Takeda Science Foundation (to S.U.); the Canon Foundation (to S.U.); by Grants-in-Aid for Challenging Exploratory Research (17K19543) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (to S.U.); by a grant (964930) from the Japan Agency for Medical Research and Development (AMED) (to S.U.); by Grants-in-Aid for Young Scientists (B) (16K16144) (to Y.K.) from the Japan Society for the Promotion of Science (JSPS); by Grants-in-Aid for Young Scientists (B) (16K20992) (to T.S.) from the JSPS; by Grants-in-Aid for Young Scientists (B) (16K19148) (to N.T.) from the JSPS; and a grant from Meiji Co., Ltd (to S.U.). Acknowledgements We thank S. Yin and B. Batmunkh for technical assistance, and K. Ogawa and N. Nagaya for secretarial assistance. Y.U., Y.K., K.F. and S.U. conceived and designed the study. Y.U., T.S., K.K., S.H., S.K., S.H. and Y.A. performed the experiments. Y.U. and Y.K. conducted all the data analysis. N.T., Y.O. and H.O helped with the data analysis. H.T., S.I. and G.E. provided scientific insights. Y.U., Y.K., T.S., K.F. and S.U. wrote and edited the manuscript. S.U. directed the research. Conflicts of interest statement: The authors declared no conflicts of interest. References 1 He , S. and Sharpless , N. E . 2017 . Senescence in health and disease . Cell 169 : 1000 . Google Scholar CrossRef Search ADS PubMed 2 Shlisky , J. , Bloom , D. E. , Beaudreault , A. R. , et al. 2017 . Nutritional considerations for healthy aging and reduction in age-related chronic disease . Adv. Nutr . 8 : 17 . Google Scholar CrossRef Search ADS PubMed 3 Campisi , J . 2013 . Aging, cellular senescence, and cancer . Annu. Rev. Physiol . 75 : 685 . Google Scholar CrossRef Search ADS PubMed 4 Cho , I. and Blaser , M. J . 2012 . The human microbiome: at the interface of health and disease . Nat. Rev. Genet . 13 : 260 . Google Scholar CrossRef Search ADS PubMed 5 Gilbert , J. A. , Quinn , R. A. , Debelius , J. , et al. 2016 . Microbiome-wide association studies link dynamic microbial consortia to disease . Nature 535 : 94 . Google Scholar CrossRef Search ADS PubMed 6 Mirzaei , M. K. and Maurice , C. F . 2017 . Ménage à trois in the human gut: interactions between host, bacteria and phages . Nat. Rev. Microbiol . 15 : 397 . Google Scholar CrossRef Search ADS PubMed 7 Rooks , M. G. and Garrett , W. S . 2016 . Gut microbiota, metabolites and host immunity . Nat. Rev. Immunol . 16 : 341 . Google Scholar CrossRef Search ADS PubMed 8 Schirmer , M. , Smeekens , S. P. , Vlamakis , H. , et al. 2016 . Linking the human gut microbiome to inflammatory cytokine production capacity . Cell 167 : 1125 . Google Scholar CrossRef Search ADS PubMed 9 Ardesia , M. , Villanacci , V. and Fries , W . 2015 . The aged gut in inflammatory bowel diseases . Minerva Gastroenterol. Dietol . 61 : 235 . Google Scholar PubMed 10 Nicoletti , C . 2015 . Age-associated changes of the intestinal epithelial barrier: local and systemic implications . Expert Rev. Gastroenterol. Hepatol . 9 : 1467 . Google Scholar CrossRef Search ADS PubMed 11 Madsen , K. , Cornish , A. , Soper , P. , et al. 2001 . Probiotic bacteria enhance murine and human intestinal epithelial barrier function . Gastroenterology 121 : 580 . Google Scholar CrossRef Search ADS PubMed 12 Langille , M. G. , Meehan , C. J. , Koenig , J. E. , et al. 2014 . Microbial shifts in the aging mouse gut . Microbiome 2 : 50 . Google Scholar CrossRef Search ADS PubMed 13 Calvani , R. , Brasili , E. , Praticò , G. , et al. 2014 . Fecal and urinary NMR-based metabolomics unveil an aging signature in mice . Exp. Gerontol . 49 : 5 . Google Scholar CrossRef Search ADS PubMed 14 Jonker , M. J. , Melis , J. P. , Kuiper , R. V. , et al. 2013 . Life spanning murine gene expression profiles in relation to chronological and pathological aging in multiple organs . Aging Cell 12 : 901 . Google Scholar CrossRef Search ADS PubMed 15 Borchers , A. T. , Selmi , C. , Meyers , F. J. , Keen , C. L. and Gershwin , M. E . 2009 . Probiotics and immunity . J. Gastroenterol . 44 : 26 . Google Scholar CrossRef Search ADS PubMed 16 Kim , S. W. , Suda , W. , Kim , S. , et al. 2013 . Robustness of gut microbiota of healthy adults in response to probiotic intervention revealed by high-throughput pyrosequencing . DNA Res . 20 : 241 . Google Scholar CrossRef Search ADS PubMed 17 Eri , T. , Kawahata , K. , Kanzaki , T. , et al. 2017 . Intestinal microbiota link lymphopenia to murine autoimmunity via PD-1+CXCR5-/dim B-helper T cell induction . Sci. Rep . 7 : 46037 . Google Scholar CrossRef Search ADS PubMed 18 Caporaso , J. G. , Kuczynski , J. , Stombaugh , J. , et al. 2010 . QIIME allows analysis of high-throughput community sequencing data . Nat. Methods 7 : 335 . Google Scholar CrossRef Search ADS PubMed 19 Soga , T. and Heiger , D. N . 2000 . Amino acid analysis by capillary electrophoresis electrospray ionization mass spectrometry . Anal. Chem . 72 : 1236 . Google Scholar CrossRef Search ADS PubMed 20 Soga , T. , Ueno , Y. , Naraoka , H. , Ohashi , Y. , Tomita , M. and Nishioka , T . 2002 . Simultaneous determination of anionic intermediates for Bacillus subtilis metabolic pathways by capillary electrophoresis electrospray ionization mass spectrometry . Anal. Chem . 74 : 2233 . Google Scholar CrossRef Search ADS PubMed 21 Soga , T. , Ohashi , Y. , Ueno , Y. , Naraoka , H. , Tomita , M. and Nishioka , T . 2003 . Quantitative metabolome analysis using capillary electrophoresis mass spectrometry . J. Proteome Res . 2 : 488 . Google Scholar CrossRef Search ADS PubMed 22 Sugimoto , M. , Wong , D. T. , Hirayama , A. , Soga , T. and Tomita , M . 2010 . Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral, breast and pancreatic cancer-specific profiles . Metabolomics 6 : 78 . Google Scholar CrossRef Search ADS PubMed 23 Kanehisa , M. , Furumichi , M. , Tanabe , M. , Sato , Y. and Morishima , K . 2017 . KEGG: new perspectives on genomes, pathways, diseases and drugs . Nucleic Acids Res . 45 : D353 . Google Scholar CrossRef Search ADS PubMed 24 López-Ibáñez , J. , Pazos , F. and Chagoyen , M . 2016 . MBROLE 2.0-functional enrichment of chemical compounds . Nucleic Acids Res . 44 : W201 . Google Scholar CrossRef Search ADS PubMed 25 McCall , M. N. , Bolstad , B. M. and Irizarry , R. A . 2010 . Frozen robust multiarray analysis (fRMA) . Biostatistics 11 : 242 . Google Scholar CrossRef Search ADS PubMed 26 Gautier , L. , Cope , L. , Bolstad , B. M. and Irizarry , R. A . 2004 . affy–analysis of Affymetrix GeneChip data at the probe level . Bioinformatics 20 : 307 . Google Scholar CrossRef Search ADS PubMed 27 Quackenbush , J . 2002 . Microarray data normalization and transformation . Nat. Genet . 32 : 496 . Google Scholar CrossRef Search ADS PubMed 28 Alexa , A. and Rahnenfuhrer , J . 2010 . topGO: enrichment analysis for gene ontology. R package version 2 . 29 Postler , T. S. and Ghosh , S . 2017 . Understanding the holobiont: how microbial metabolites affect human health and shape the immune system . Cell Metab . 26 : 110 . Google Scholar CrossRef Search ADS PubMed 30 Steegenga , W. T. , de Wit , N. J. , Boekschoten , M. V. , et al. 2012 . Structural, functional and molecular analysis of the effects of aging in the small intestine and colon of C57BL/6J mice . BMC Med. Genomics 5 : 38 . Google Scholar CrossRef Search ADS PubMed 31 Furusawa , Y. , Obata , Y. , Fukuda , S. , et al. 2013 . Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells . Nature 504 : 446 . Google Scholar CrossRef Search ADS PubMed 32 Lin , L. and Zhang , J . 2017 . Role of intestinal microbiota and metabolites on gut homeostasis and human diseases . BMC Immunol . 18 : 2 . Google Scholar CrossRef Search ADS PubMed 33 Thaiss , C. A. , Zmora , N. , Levy , M. and Elinav , E . 2016 . The microbiome and innate immunity . Nature 535 : 65 . Google Scholar CrossRef Search ADS PubMed 34 Odamaki , T. , Kato , K. , Sugahara , H. , et al. 2016 . Age-related changes in gut microbiota composition from newborn to centenarian: a cross-sectional study . BMC Microbiol . 16 : 90 . Google Scholar CrossRef Search ADS PubMed 35 Baylis , D. , Bartlett , D. B. , Patel , H. P. and Roberts , H. C . 2013 . Understanding how we age: insights into inflammaging . Longev. Healthspan 2 : 8 . Google Scholar CrossRef Search ADS PubMed 36 Turnbaugh , P. J. , Ley , R. E. , Mahowald , M. A. , Magrini , V. , Mardis , E. R. and Gordon , J. I . 2006 . An obesity-associated gut microbiome with increased capacity for energy harvest . Nature 444 : 1027 . Google Scholar CrossRef Search ADS PubMed 37 Yang , T. , Santisteban , M. M. , Rodriguez , V. , et al. 2015 . Gut dysbiosis is linked to hypertension . Hypertension 65 : 1331 . Google Scholar CrossRef Search ADS PubMed 38 Turnbaugh , P. J. , Bäckhed , F. , Fulton , L. and Gordon , J. I . 2008 . Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome . Cell Host Microbe 3 : 213 . Google Scholar CrossRef Search ADS PubMed 39 Corr , S. C. , Hill , C. and Gahan , C. G . 2009 . Understanding the mechanisms by which probiotics inhibit gastrointestinal pathogens . Adv. Food Nutr. Res . 56 : 1 . Google Scholar CrossRef Search ADS PubMed 40 Koh , A. , De Vadder , F. , Kovatcheva-Datchary , P. and Backhed , F. 2016 . From Dietary Fiber to Host Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites . Cell 165 : 1332 . 41 Ríos-Covián , D. , Ruas-Madiedo , P. , Margolles , A. , Gueimonde , M. , de Los Reyes-Gavilán , C. G. and Salazar , N . 2016 . Intestinal short chain fatty acids and their link with diet and human health . Front. Microbiol . 7 : 185 . Google Scholar CrossRef Search ADS PubMed 42 Meisel , M. , Mayassi , T. , Fehlner-Peach , H. , et al. 2017 . Interleukin-15 promotes intestinal dysbiosis with butyrate deficiency associated with increased susceptibility to colitis . ISME J . 11 : 15 . Google Scholar CrossRef Search ADS PubMed 43 Ormerod , K. L. , Wood , D. L. , Lachner , N. , et al. 2016 . Genomic characterization of the uncultured Bacteroidales family S24-7 inhabiting the guts of homeothermic animals . Microbiome 4 : 36 . Google Scholar CrossRef Search ADS PubMed 44 Okumura , R. and Takeda , K . 2017 . Roles of intestinal epithelial cells in the maintenance of gut homeostasis . Exp. Mol. Med . 49 : e338 . Google Scholar CrossRef Search ADS PubMed 45 Turner , J. R . 2009 . Intestinal mucosal barrier function in health and disease . Nat. Rev. Immunol . 9 : 799 . Google Scholar CrossRef Search ADS PubMed 46 Ren , W. Y. , Wu , K. F. , Li , X. , et al. 2014 . Age-related changes in small intestinal mucosa epithelium architecture and epithelial tight junction in rat models . Aging Clin. Exp. Res . 26 : 183 . Google Scholar CrossRef Search ADS PubMed 47 Mabbott , N. A. , Kobayashi , A. , Sehgal , A. , Bradford , B. M. , Pattison , M. and Donaldson , D. S . 2015 . Aging and the mucosal immune system in the intestine . Biogerontology 16 : 133 . Google Scholar CrossRef Search ADS PubMed © The Japanese Society for Immunology. 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Immunology Oxford University Press

Effects of long-term intake of a yogurt fermented with Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131 on mice

International Immunology , Volume Advance Article (7) – May 15, 2018

Loading next page...
 
/lp/ou_press/effects-of-long-term-intake-of-a-yogurt-fermented-with-lactobacillus-pR30XKGtrn
Publisher
Oxford University Press
Copyright
© The Japanese Society for Immunology. 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0953-8178
eISSN
1460-2377
D.O.I.
10.1093/intimm/dxy035
Publisher site
See Article on Publisher Site

Abstract

Abstract The gut is an extremely complicated ecosystem where micro-organisms, nutrients and host cells interact vigorously. Although the function of the intestine and its barrier system weakens with age, some probiotics can potentially prevent age-related intestinal dysfunction. Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131, which are the constituents of LB81 yogurt, are representative probiotics. However, it is unclear whether their long-term intake has a beneficial influence on systemic function. Here, we examined the gut microbiome, fecal metabolites and gene expression profiles of various organs in mice. Although age-related alterations were apparent in them, long-term LB81 yogurt intake led to an increased Bacteroidetes to Firmicutes ratio and elevated abundance of the bacterial family S24-7 (Bacteroidetes), which is known to be associated with butyrate and propanoate production. According to our fecal metabolite analysis to detect enrichment, long-term LB81 yogurt intake altered the intestinal metabolic pathways associated with propanoate and butanoate in the mice. Gene ontology analysis also revealed that long-term LB81 yogurt intake influenced many physiological functions related to the defense response. The profiles of various genes associated with antimicrobial peptides-, tight junctions-, adherens junctions- and mucus-associated intestinal barrier functions were also drastically altered in the LB81 yogurt-fed mice. Thus, long-term intake of LB81 yogurt has the potential to maintain systemic homeostasis, such as the gut barrier function, by controlling the intestinal microbiome and its metabolites. aging, intestinal epithelial barrier, metabolome, microbiome, probiotics Introduction The prevention of age-related chronic disorders has become a major focus of current medical research (1, 2). Because age-related deterioration of mechanical, physiological and psychological functions is mostly inevitable, the incidence of age-related diseases such as atherosclerosis, arthritis, cancer, osteoporosis, type 2 diabetes, hypertension and Alzheimer’s has increased in the world (1, 3). Fascinatingly, several lines of evidence have shown that interactions between intestinal micro-organisms and the host are crucial in the context of health and disease (4–7). Dysbiosis, for example, which is defined as an imbalance in the repertoire of the intestinal microbiota, is correlated with aberrant immune responses such as abnormal inflammatory cytokine production (8). Intestinal dysbiosis has the potential to alter gut metabolites, and is more frequent in older individuals than in young people (9). Dysbiosis in the elderly is thought to be responsible for a defective intestinal barrier with an associated altered secretion of the antimicrobial peptides (AMPs) produced by intestinal epithelial cells, thereby resulting in the development of age-related diseases (10). Notably, it has been shown that probiotic bacteria can have a direct effect on the functioning of the intestinal epithelial barrier (11). Previous studies have revealed that age-related changes occur in gut microbes (12), intestinal metabolites (13) and gene expression in various organs (14). However, the current lack of multi-omics approaches for investigating the systemic biological changes associated with aging mean that the effects of aging on the body are not fully understood. It has been shown recently that the administration of probiotic-containing products such as yogurt can improve the gut environment and reduce the risk of inflammatory disorders such as autoimmune diseases and allergies (15). Among the many probiotic types, LB81 yogurt, which contains Lactobacillus delbrueckii subsp. bulgaricus 2038 and Streptococcus thermophilus 1131, is a traditionally consumed product. However, it remains unclear whether long-term intake of LB81 yogurt is beneficial to health. Here, we first determined the influence of aging on the fecal microbiome, fecal metabolites and gene expression in various organs in mice. As other studies (12–14) have found, each of them are altered by aging. We also evaluated the effect of long-term LB81 yogurt intake on the survival rates, body weights, fecal microbiome, fecal metabolites and gene expression in various organs in mice. When LB81 yogurt was used to supplement a normal mouse diet (from 8 to 25 months of age), the mice did not show any alterations in their gut bacterial alpha diversity index compared with the control mice; however, the increased ratio of Bacteroidetes to Firmicutes and the fecal metabolites associated with propanoate metabolism and butanoate metabolism were significantly linked with the long-term intake of this yogurt. On the basis of our gene ontology (GO) analysis results, long-term intake of LB81 yogurt modulated many different systemic functions, and the expression patterns of the intestinal barrier-associated genes were drastically altered. These findings suggest that long-term LB81 yogurt intake can help to maintain gut homeostasis and possibly prevent aging-associated diseases. Methods Mice The 8-month-old ICR male mice obtained from Japan SLC (Hamamatsu, Japan) were fed on a normal diet (AIN-93) (Oriental Yeast, Tokyo, Japan) (n = 39, hereafter referred to as the control mice) or the same diet containing 1–2% LB81 yogurt (n = 40, hereafter referred to as the LB81 yogurt-intake mice) until they were 28 months of age. The survival rates and body weight changes were monitored. The 2-month-old ICR male mice obtained from Japan SLC and were fed on a normal diet (AIN-93) (Oriental Yeast) (n = 6, hereafter referred to as the control mice) until 4 months of age. Fecal samples were longitudinally collected from the control mice (n = 22) and the LB81 yogurt-intake mice (n = 25) at 10, 15, 21 and 25 months of age. Each mouse was reared in a single cage under specific pathogen-free conditions at the Experimental Animal Facility of Meiji Co., Ltd (Japan). The experiments were performed according to the guidelines of the Animal Research Committee of Meiji Co., Ltd and the Institute of Medical Science (University of Tokyo, Japan). 16S rRNA gene analysis Bacterial DNA extraction from each fecal sample was performed as described previously with some modifications (16). The 16S rRNA V3–V4 region was PCR-amplified and purified as described previously (17). For each sample, an equal amount of each DNA amplicon library was mixed and sequenced on the MiSeq instrument (Illumina, San Diego, CA, USA) using a MiSeq v3 Reagent kit with 20% PhiX (Illumina). The 16S rRNA gene analysis was performed using QIIME (version 1.9.1) (18), as described previously (17). Briefly, the raw sequence data were subjected to quality filtering and paired-end read merging. The insufficient sequence depth below the 15000 threshold meant that three of the sequenced samples (two from the control mice, one from the LB81 yogurt-intake mice) were removed from the following analysis. The processed sequence reads were assigned to operational taxonomic units (OTUs) with a 97% identity threshold using the Greengenes database (version 13.8) (18). Faith’s phylogenetic alpha diversity estimate and principal coordinate analysis (PCoA) of the weighted UniFrac distance matrices were performed using QIIME. To characterize the changes occurring in family-level microbial abundance during aging, the relative abundance at each time point was standardized by Z-scores using the mean relative abundance at 10 months of age. Gut microbial families were clustered based on the standardized profile using the R function Kmeans in the amap package, with K = 4 and distance metric ‘correlation’. Metabolome analysis To extract the ionic metabolites, ~50 mg of each fecal sample was dissolved in Milli-Q water (1:9, w/v ratio). After centrifugation, 20 µl of an internal standard solution (H3304-1002, Human Metabolome Technologies, Inc., HMT, Yamagata, Japan) was added to 80 µl of the supernatant. To remove the macromolecules, the solution was centrifugally filtered through a Millipore 5-kDa cutoff filter (UltrafreeMC-PLHCC, HMT) (9100 × g, 4°C, 60 min) and subsequently analyzed by capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS). Metabolomic measurements were performed by a facility service at HMT, as described previously (19–21). The peaks detected were annotated against the putative metabolites from the HMT metabolite database linked to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (release 54.1) (22, 23). For this analysis, we used the annotated KEGG compound identity (ID) and the relative area value for the peak area of each metabolite was normalized against the sample volume. The relative area values below the detection limit were set to 2−52 as in the HMT reference. Metabolites with multiple KEGG compound ID annotations were excluded from the following analysis, except for the amino acids, where we used the KEGG compound IDs of the l-form amino acids. The aging effect was analyzed using the metabolites whose relative areas showed more than a 2-fold change at any of 10, 15, 21 and 25 month time points compared with any of the other time points. The relative area values were Z-score standardized using the relative area values at 10 months of age, and then clustered using the R function Kmeans in the amap package, with K = 5 and distance metric ‘correlation’. The KEGG pathway enrichments for the metabolites in each metabolite cluster were examined using MBROLE 2.0 (24). The effect of the probiotics was examined using the relative area values of the control and the LB81 yogurt-intake mice at 25 months of age. Metabolites whose relative area values changed >2-fold between the control and the LB81 yogurt-intake group were examined using MBROLE 2.0. Microarray analysis Jejunum, ileum, distal colon, liver and spleen tissues were collected from the control mice at 4 (n = 6) and 28 (n = 9) months of age, and the LB81 yogurt-intake mice at 28 months of age (n = 10). Equal amounts of total RNA, previously extracted from the tissues of each mouse, were pooled and mixed for each group. The RNA mixtures (100 ng) were individually amplified using the GeneChip 3′IVT Express kit and then hybridized to an Affymetrix Mouse Genome 430 2.0 chip (GPL1261; 45101 probes) (n = 1 per group). After hybridization, the chip was washed, stained on the Affymetrix Fluidics Station 450 and scanned using the Affymetrix GeneChip Scanner 3000 7G System, with all the operations being performed by the Affymetrix GeneChip Command Console (AGCC) Software (accession code GEO: GSE104375, GSE111578 and GSE111579). The analysis steps described below were performed separately to compare the aging effect (4 versus 28 months of age in the control mice) and to compare the effect of the probiotics (the control versus the LB81 yogurt-intake mice at 28 months of age). CEL files were processed and normalized by frozen robust multi-array analysis using the frma and mouse4302frmavecs packages from Bioconductor in the R statistical language (25). Detection calls were generated by MAS5.0 using the affy package from Bioconductor (26). The probes were annotated with the mouse4302.db package from Bioconductor and the annotation table of the Mouse Genome 430 2.0 array provided by Affymetrix (release 36). Probe sets that fulfilled any of the following criteria were excluded: (i) the detection calls were absent from both the control and the sample of interest; (ii) the probe set ID contained ‘x_’ or ‘s_’; (iii) the gene symbol was NA; and (iv) the transcript assignment grade of the probe set was lower than ‘A’. In cases where multiple probe sets had the same gene symbol annotation, one probe set was selected in the following order: (i) the probe set with the largest number of SWISS-PROT annotations; (ii) the unique probe set id (‘_at’); (iii) the probe set that targeted the longest region; and (iv) the probe set with the highest sum of the gene expression signals. To detect the differentially expressed genes, the intensity-dependent Z-score was calculated (27). The fold change of the signal between the control and the sample of interest was divided by the standard deviation obtained at each log-transformed average signal with a window size of 0.5. The GO biological process term enrichment was examined for the probe sets whose intensity-dependent Z-scores were >2 or < −2 by Fisher’s exact test using the topGO and mouse4302.db packages from Bioconductor (28). Statistical analysis Statistical analysis was performed using R statistical language. A P-value of <0.05 was defined as statistically significant. Results The aging gut microbiome To investigate age-related changes in the gut microbiome, we longitudinally collected fecal samples from 22 mice at 10, 15, 21 and 25 months of age, and performed 16S rRNA gene sequencing on these samples. Samples with insufficient sequence depths were removed, and the sequenced samples from 20 mice were used for the following analysis. Our analysis of the gut microbial compositions in the mice was based on the relative abundance of the OTUs that clustered with 97% identity scores using QIIME (18). The gut microbial compositions at the phylum level showed that the relative abundance of Actinobacteria decreased, whereas that of the Bacteroidetes increased with age (Fig. 1A). The ratio of Bacteroidetes to Firmicutes, which represent the major bacterial phyla in this study, increased with age and peaked at 21 months of age (P < 0.01) (Fig. 1B). The bacterial diversity evaluation within each sample showed that bacterial alpha diversity increased at 21 and 25 months (P < 0.01) (Fig. 1C). To assess the bacterial community across the aging stages, a PCoA for the weighted UniFrac distance matrices was performed, the results of which showed that at 21 months of age the gut microbial community occupied the largest distance from the other age groups (Fig. 1D). We also analyzed the profiles of the family-level microbial abundance by clustering them using the K-means algorithm (Fig. 1E and Supplementary Table 1). Interestingly, we identified three clusters that increased at different time points with peaks at 15 months (bacterial cluster 1), 21 months (bacterial cluster 2) and 25 months (bacterial cluster 3), whereas only one bacterial gut cluster decreased with aging (bacterial cluster 4). The most abundant bacterial families in bacterial clusters 1, 2, 3 and 4 were Erysipelotrichaceae (Firmicutes), S24-7 (Bacteroidetes), F16 (TM7) and Bifidobacteriaceae (Actinobacteria), respectively. Thus, the intestinal microbiome underwent considerable changes at around 21 months of age and its composition shifted with aging in at least four different patterns. Fig. 1. View largeDownload slide Gut microbial compositions at different age points. (A) Gut microbial compositions at the phylum level based on the relative abundance of OTUs for the fecal samples from 20 mice at 10, 15, 21 and 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance of OTUs at phylum level. The P-value was calculated using a pairwise Wilcoxon signed rank test for paired data with Holm P-value correction. (C) Bacterial alpha diversity based on Faith’s phylogenetic alpha diversity index. The P-value was calculated using a pairwise t-test for paired data with Holm P-value correction. (D) PCoA on the gut microbial community according to the weighted UniFrac distance matrices. (E) Age-related changes in gut microbial abundance were classified by the K-means algorithm on the basis of the standardized relative abundance of OTUs at the family level. Data are shown as the mean values of the standardized relative abundance for each bacteria ± SD. NS, not significant, **P < 0.01, *P < 0.05. Fig. 1. View largeDownload slide Gut microbial compositions at different age points. (A) Gut microbial compositions at the phylum level based on the relative abundance of OTUs for the fecal samples from 20 mice at 10, 15, 21 and 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance of OTUs at phylum level. The P-value was calculated using a pairwise Wilcoxon signed rank test for paired data with Holm P-value correction. (C) Bacterial alpha diversity based on Faith’s phylogenetic alpha diversity index. The P-value was calculated using a pairwise t-test for paired data with Holm P-value correction. (D) PCoA on the gut microbial community according to the weighted UniFrac distance matrices. (E) Age-related changes in gut microbial abundance were classified by the K-means algorithm on the basis of the standardized relative abundance of OTUs at the family level. Data are shown as the mean values of the standardized relative abundance for each bacteria ± SD. NS, not significant, **P < 0.01, *P < 0.05. Fecal metabolites associated with aging Some gut microbes produce a variety of intestinal metabolites that have close relationships with health and disease through the circulatory system and regulation of the intestinal mucosal barrier (29). Therefore, we investigated whether any changes in fecal metabolites occurred upon aging in the mice. Fecal samples were longitudinally collected at 10, 15, 21 and 25 months of age. Metabolites from the pooled fecal samples were measured at each time point using CE-TOFMS instrumentation (Supplementary Table 2). To characterize the changes occurring in the metabolites during the aging process, we clustered them based on the standardized relative area values using the K-means algorithm. Notably, five clusters were defined and three of them showed increased abundances at the following time points: 15 months (metabolite cluster 1), 21 months (metabolite cluster 2) and 25 months (metabolite cluster 3); however, two of them showed decreased abundances with aging (metabolite clusters 4 and 5) (Fig. 2A). To assess the metabolic pathways involved in each cluster, we tested for KEGG metabolic pathway enrichment using MBROLE 2.0 (Fig. 2B and Supplementary Table 3). In the metabolite cluster 1, central carbon metabolism such as ‘galactose metabolism’ (P = 5 × 10−3), ‘pentose and glucuronate interconversions’ (P = 6 × 10−3) and ‘glycolysis/gluconeogenesis’ (P = 2 × 10−2) was enriched. In metabolite cluster 2, metabolism related to short-chain fatty acids (SCFAs) such as ‘propanoate metabolism’ (P = 2 × 10−2) and ‘butanoate metabolism’ (P = 3 × 10−2) showed enrichment. In metabolite cluster 3, carbohydrate metabolism such as ‘glyoxylate and dicarboxylate metabolism’ (P = 2 × 10−2), ‘ascorbate and aldarate metabolism’ (P = 2 × 10−2) and ‘starch and sucrose metabolism’ (P = 4 × 10−2) was enriched. In metabolite cluster 4, which was defined as displaying a decreased profile, ‘aminoacyl-tRNA biosynthesis’ (P = 3 × 10−9) was significantly enriched with 18 amino acids being involved in this pathway (Supplementary Table 3). In metabolite cluster 5, lipid metabolism (e.g. ‘glycerophospholipid metabolism’) was enriched (P = 2 × 10−2). Thus, the metabolites involved in carbohydrate metabolism (metabolite clusters 1 and 3) and SCFA (metabolite cluster 2) increased transiently with aging, whereas the metabolites related to amino acid metabolism (metabolite cluster 4) and lipids (metabolite cluster 5) decreased with age. Fig. 2. View largeDownload slide Age-related changes in the fecal metabolites. (A) Age-related changes in the fecal metabolite amounts were classified by the K-means algorithm on the basis of the standardized relative area values. The fecal samples from 22 mice at 10, 15, 21 and 25 months of age were used. Data are shown as the mean values for the standardized relative area of each metabolite ± SD. (B) KEGG pathways in each fecal metabolite cluster that were significantly enriched on the basis of MBROLE 2.0. The bar graph shows the top six enriched KEGG pathways where uncorrected P < 0.05. Fig. 2. View largeDownload slide Age-related changes in the fecal metabolites. (A) Age-related changes in the fecal metabolite amounts were classified by the K-means algorithm on the basis of the standardized relative area values. The fecal samples from 22 mice at 10, 15, 21 and 25 months of age were used. Data are shown as the mean values for the standardized relative area of each metabolite ± SD. (B) KEGG pathways in each fecal metabolite cluster that were significantly enriched on the basis of MBROLE 2.0. The bar graph shows the top six enriched KEGG pathways where uncorrected P < 0.05. The effect of aging on the gene expression profiles of the intestines and other organs We next focused on how gene expression might change in various organs with aging. The intestinal tissues (jejunum, ileum and distal colon) and other tissues (liver and spleen) from mice at 4 months (n = 6) and 28 months (n = 9) of age were isolated, and their gene expression profiles were evaluated using cDNA microarrays. According to the GO analysis results, the GO term ‘defense response’ (P = 9 × 10−27) in the liver was the most significantly enriched among the up-regulated genes in the aged mice when compared with the results of the same analysis in the younger mice (Fig. 3A). In the liver, the genes involved in the ‘innate immune response’ (P = 2 × 10−18) and the ‘adaptive immune response’ (P = 2 × 10−10) were also associated with aging in mice. In the distal colon, the genes involved in ‘leukocyte activation’ (P = 3 × 10−16) and ‘lymphocyte activation’ (P = 7 × 10−16) were enriched in the aged mice. In the spleen, the biological terms ‘defense response’ (P = 8 × 10−12) and ‘defense response to bacterium’ (P = 2 × 10−10) were remarkably enriched in the aged mice. In contrast, no significantly enriched GO terms were identified among the up-regulated genes in the jejunum and ileum tissues from the aged mice, a finding that is consistent with a previous study (30). As for the down-regulated genes in the aged mice compared with those in the younger mice, ‘lipid metabolic process’ was commonly enriched in the jejunum (P = 5 × 10−9) and ileum (P = 2 × 10−15) (Fig. 3B). These results suggest that the biological functions of each organ are preferentially affected by age. Fig. 3. View largeDownload slide GO terms for the differentially expressed genes in the aged mice compared with the younger mice. GO term enrichment analysis was performed for the up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the aged mice (n = 9) compared with the younger mice (n = 6). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the aged mice (28 months of age) compared with the younger mice (4 months of age). Only GO terms where the number of annotated genes was below 1000 and where the P-values were <10−9 in at least one tissue were visualized. The color intensity is based on the uncorrected P-values calculated via Fisher’s exact test. Fig. 3. View largeDownload slide GO terms for the differentially expressed genes in the aged mice compared with the younger mice. GO term enrichment analysis was performed for the up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the aged mice (n = 9) compared with the younger mice (n = 6). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the aged mice (28 months of age) compared with the younger mice (4 months of age). Only GO terms where the number of annotated genes was below 1000 and where the P-values were <10−9 in at least one tissue were visualized. The color intensity is based on the uncorrected P-values calculated via Fisher’s exact test. Effect of long-term LB81 yogurt intake on survival rates and body weight changes Because the gut microbiome, fecal metabolites, and gene expression in various organs were altered with aging, we investigated the effects of the long-term intake of LB81 yogurt on aging. We fed a normal diet with (n = 40) or without (n = 39) 1–2% LB81 yogurt to mice for 20 months, and monitored their survival rates and body weight changes. No significant difference in the survival rates of the mice fed on LB81 yogurt and the control mice was noted (Fig. 4A). Additionally, the body weight of the mouse group fed on LB81 yogurt was similar to that of the control group (Fig. 4B). Thus, long-term intake of LB81 yogurt had no effect on mouse survival or body weight. Fig. 4. View largeDownload slide Physiological effects of long-term supplementation with LB81 yogurt. (A) Kaplan–Meier survival curve for the control (n = 39) and the LB81 yogurt-intake groups (n = 40). The P-value was calculated using a log-rank test. (B) Body weight changes in the control and the LB81 yogurt-intake groups during the study period. Data are shown as the mean ± SD. Control mice: gray color, LB81 yogurt-intake mice: black color. Fig. 4. View largeDownload slide Physiological effects of long-term supplementation with LB81 yogurt. (A) Kaplan–Meier survival curve for the control (n = 39) and the LB81 yogurt-intake groups (n = 40). The P-value was calculated using a log-rank test. (B) Body weight changes in the control and the LB81 yogurt-intake groups during the study period. Data are shown as the mean ± SD. Control mice: gray color, LB81 yogurt-intake mice: black color. Effect of long-term LB81 yogurt intake on the intestinal microbiome To elucidate the effect of long-term administration of LB81 yogurt on the gut microbial composition in the mice, we used pyrosequencing-based 16S rRNA gene profiling to analyze the fecal microbiota from the 25-month-old mice fed on a normal diet with (n = 24) or without (n = 20) LB81 yogurt supplementation for 17 months. The relative abundance of gut microbes at the phylum level showed that the gut microbial composition changed in the mice fed on LB81 yogurt (Fig. 5A). Of note, we found that the Bacteroidetes to Firmicutes ratio was significantly higher in the mice fed on LB81 yogurt than in the control mice (P < 0.05) (Fig. 5B). There was, however, no significant difference in bacterial alpha diversity (Faith’s phylogenetic alpha diversity) between the control and the LB81 yogurt-fed groups (Fig. 5C), and the PCoA analysis based on the weighted UniFrac distance did not separate the control group from the LB81 yogurt-treated group (Fig. 5D). Our comparison of the relative abundance of the bacterial community at the family level showed that S24-7 (Bacteroidetes) was the most abundant bacteria among the significantly increased bacteria in the LB81 yogurt-treated group as compared with the control group (P < 0.05) (Fig. 5E). In addition, the relative abundance of Streptococcaceae (Firmicutes), Deferribacteraceae (Deferribacteres) and Paraprevotellaceae (Bacteroidetes) was significantly higher in the LB81 yogurt-treated group than in the control group (P < 0.05) (Fig. 5F–H). In contrast, the relative abundance of Turicibacteraceae (Firmicutes) and Moraxellaceae (Proteobacteria) was down-regulated in the LB81 yogurt-treated group (P < 0.05) (Fig. 5I and J). Thus, long-term LB81 yogurt intake increased the ratio of Bacteroidetes to Firmicutes in the mice whose diet was supplemented with it, but overall the microbial diversity did not alter in these mice. Fig. 5. View largeDownload slide Gut microbial compositions and diversities in the control mice and the LB81 yogurt-intake mice. (A) Gut microbial compositions based on the relative abundance of OTUs at the phylum level from the control (n = 20) and the LB81 yogurt-intake group (n = 24) at 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance data from the OTUs at phylum level. The P-value was calculated using a two-sided Wilcoxon rank sum test for unpaired data. (C) Bacterial alpha diversities for the fecal microbial community using Faith’s phylogenetic alpha diversity index. The P-value was calculated using the two-sided Wilcoxon rank sum test for unpaired data. (D) PCoA on the weighted UniFrac distance matrices for the microbial community. (E–H) Top four gut microbial families whose relative abundances increased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. (I and J) Top two gut microbial families whose relative abundances decreased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. The P-value was calculated using a one-sided Wilcoxon rank sum test for unpaired data. *P < 0.05. NS, not statistically significant. Fig. 5. View largeDownload slide Gut microbial compositions and diversities in the control mice and the LB81 yogurt-intake mice. (A) Gut microbial compositions based on the relative abundance of OTUs at the phylum level from the control (n = 20) and the LB81 yogurt-intake group (n = 24) at 25 months of age. (B) The Bacteroidetes to Firmicutes ratio (log2) using the relative abundance data from the OTUs at phylum level. The P-value was calculated using a two-sided Wilcoxon rank sum test for unpaired data. (C) Bacterial alpha diversities for the fecal microbial community using Faith’s phylogenetic alpha diversity index. The P-value was calculated using the two-sided Wilcoxon rank sum test for unpaired data. (D) PCoA on the weighted UniFrac distance matrices for the microbial community. (E–H) Top four gut microbial families whose relative abundances increased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. (I and J) Top two gut microbial families whose relative abundances decreased significantly (uncorrected P < 0.05) in the LB81 yogurt-intake mice compared with the control mice. The P-value was calculated using a one-sided Wilcoxon rank sum test for unpaired data. *P < 0.05. NS, not statistically significant. Effect of long-term LB81 yogurt intake on fecal metabolites Because LB81 yogurt is a probiotic source that is thought to have beneficial effects on host immune defenses by modulating the concentrations of, for example, the intestinal metabolites butyrate and propanoate (31), we also analyzed fecal metabolite levels using CE-TOFMS. Our enrichment analysis of the metabolomic data revealed that two pathways, propanoate metabolism and butanoate metabolism, were significantly enhanced in the LB81 yogurt-fed group (Table 1), suggesting that long-term intake of this yogurt alters these gut-associated metabolites. Table 1. Significantly enriched KEGG pathways for the up-regulated fecal metabolites in the LB81 yogurt-intake mice (n = 25) compared with the control mice (n = 22) KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid aKEGG pathways with uncorrected P-values of <0.05 are shown in this table. bEnrichment annotations involving >1 compound are shown in the table. View Large Table 1. Significantly enriched KEGG pathways for the up-regulated fecal metabolites in the LB81 yogurt-intake mice (n = 25) compared with the control mice (n = 22) KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid KEGG pathway ID Pathway name P-valuea KEGG compound IDb Compound name map00640 Propanoate metabolism 2.72 × 10−2 C00164 Acetoacetic acid C01013 3-Hydroxypropionic acid map00650 Butanoate metabolism 3.55 × 10−2 C00164 Acetoacetic acid C00246 Butyric acid aKEGG pathways with uncorrected P-values of <0.05 are shown in this table. bEnrichment annotations involving >1 compound are shown in the table. View Large Effect of long-term LB81 yogurt intake on gene expression in the intestines, liver and spleen As described above, long-term LB81 yogurt intake was found to regulate gut metabolism. Gut microbiomes and metabolites have also been shown to affect the functions of the intestine (e.g. intestinal barrier functioning), liver and spleen (29, 32, 33). Therefore, we isolated the jejunum, ileum, distal colon, liver and spleen tissues from the control (n = 9) and LB81 yogurt-treated mice (n = 10) at 28 months of age, and examined their gene expression profiles using a cDNA microarray. Compared with the control mice, among the up-regulated genes in the LB81 yogurt-fed group, we found that genes involved in the ‘defense response’ were commonly enriched in the jejunum (P = 1 × 10−10), ileum (P = 4 × 10−4), distal colon (P = 3 × 10−4), liver (P = 2 × 10−6) and spleen (P = 1 × 10−6) (Fig. 6A). In the jejunum, we found that the genes associated with ‘digestion’ (P = 6 × 10−10) were enriched among the up-regulated genes in the aged mice. In the liver, we found that the genes associated with the ‘acute inflammatory response’ (P = 2 × 10−7) were enriched. A different profile was also seen for the spleen, where the genes involved in ‘disruption by host of symbiont cells’ (P = 4 × 10−8) and its subterm ‘killing by host of symbiont cells’ (P = 4 × 10−8) were enriched. Conversely, in comparison with the control mice, among the down-regulated genes in the LB81 yogurt-intake mice, those involved in the ‘regulation of B cell proliferation’ (P = 2 × 10−11) and its subterm ‘positive regulation of B cell proliferation’ (P = 4 × 10−10) were enriched in the spleen (Fig. 6B). Collectively, these results suggest that long-term intake of LB81 yogurt alters the gene expression profiles associated with immune functions in the tissues of the jejunum, ileum, distal colon, liver and spleen from the aged mice. Fig. 6. View largeDownload slide GO terms for the differentially expressed genes in the LB81 yogurt-intake group compared with the control group. GO term enrichment analysis was performed for up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the LB81 yogurt-intake mice (n = 10) compared with the control mice (n = 9). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the LB81 yogurt-intake mice compared with the control mice. Only GO terms where the number of annotated genes is below 1000 and where the P-values were <10−6 in at least one tissue were visualized. The color intensity and circular size indicate the uncorrected P-values calculated via Fisher’s exact test. Fig. 6. View largeDownload slide GO terms for the differentially expressed genes in the LB81 yogurt-intake group compared with the control group. GO term enrichment analysis was performed for up- and down-regulated genes (intensity-dependent Z-score > 2 or < −2) in the jejunum, ileum, distal colon, liver and spleen tissues from the LB81 yogurt-intake mice (n = 10) compared with the control mice (n = 9). Heatmap showing the significantly enriched GO terms for the up-regulated genes (A) and down-regulated genes (B) in the LB81 yogurt-intake mice compared with the control mice. Only GO terms where the number of annotated genes is below 1000 and where the P-values were <10−6 in at least one tissue were visualized. The color intensity and circular size indicate the uncorrected P-values calculated via Fisher’s exact test. Effect of long-term LB81 yogurt intake on intestinal barrier functions The administration of probiotic-containing products has been shown to enhance intestinal barrier function (11). In the present study, we found that the genes involved in the ‘defense response’ were significantly enriched in the jejunum, ileum and distal colon tissues of the LB81 yogurt-fed group (Fig. 6A). Therefore, we next focused on the genes associated with AMPs, tight junctions, adherens junctions and mucus. Among the AMP genes, up-regulation of Reg3b, Reg3g, Prss22 and Ang4 expression was common in the jejunum, ileum and distal colon tissues of the LB81 yogurt-fed group (Fig. 7). Moreover, enhanced expression of Reg3a, Prss2, Pla2g2a, Defa4, Mmp7 and Ltf also occurred in the jejunum, while that of Prss12, Defb1 and Lyz2 increased in the ileum. These results suggest that long-term intake of LB81 yogurt has the potential to enhance the expression of intestinal barrier-related genes. Conversely, both increased and decreased expression of the genes associated with tight- and adherens-junctions was observed in the LB81 yogurt-fed mice. Furthermore, among the mucus-related genes, enhanced expression of Muc2, C1galt1 and Retnlb genes occurred only in the distal colon. These findings raise the possibility that the altered intestinal metabolism induced by long-term LB81 yogurt intake is associated with the maintenance of intestinal barrier functions. Fig. 7. View largeDownload slide Differentially expressed intestinal barrier-related genes in the LB81 yogurt-intake group compared with the control group. Expression of intestinal barrier-related genes in the jejunum, ileum and distal colon tissues was compared between the LB81 yogurt-intake mice (n = 10) and the control mice (n = 9). Colors indicate the intensity-dependent Z-score. Yellow and blue colors represent increased and decreased gene expression in the LB81 yogurt-intake mice compared with the control mice, respectively. Fig. 7. View largeDownload slide Differentially expressed intestinal barrier-related genes in the LB81 yogurt-intake group compared with the control group. Expression of intestinal barrier-related genes in the jejunum, ileum and distal colon tissues was compared between the LB81 yogurt-intake mice (n = 10) and the control mice (n = 9). Colors indicate the intensity-dependent Z-score. Yellow and blue colors represent increased and decreased gene expression in the LB81 yogurt-intake mice compared with the control mice, respectively. Discussion In the present study, we performed comprehensive analyses on the gut microbiome, fecal metabolites and gene expression profiles of various organs in mice. As the results of the longitudinal examination in the mice fed with LB81 yogurt revealed, long-term intake of the probiotics had various effects on the normal age-related changes that occur over time. Aging is associated with an alteration in the gut microbiome of mice and humans (12, 34). The present study showed clearly the age-related changes in the gut microbial composition, such as a decreased abundance of Actinobacteria and an increased abundance of Bacteroidetes with aging. Additionally, the gut microbial composition underwent considerable changes at around 21 months of age (Fig. 1A, B and D). These findings are consistent with those from a previous study where the gut microbial composition of middle-aged mice (~20 months of age) was found to differ from that of the older mice (~28 months of age) (12). Thus, the age of around 21 months might be a turning point in the intestinal environment during mouse aging. Although the intestinal microbiome and metabolome are influenced by aging (12, 13), alterations to them can be diverse. In the present study, the abundance of gut microbial families changed with aging in four different patterns (Fig. 1E) and the intestinal metabolite amounts were altered in five different patterns (Fig. 2A and B). Of particular note, the age-related changes in metabolite cluster 1 were similar to those of bacterial cluster 1 (Figs 1E, 2A and B). Metabolite cluster 2 also showed similar age-related changes to those in bacterial cluster 2 (Figs 1E, 2A and B). Thus, age-related changes in the gut microbiome might be linked with intestinal metabolite alterations. Therefore, further studies will be required to better understand the correlation between the gut microbiome and intestinal metabolites during aging. Here, we showed that many immune-related GO terms were significantly enriched with aging in the distal colon, liver and spleen (Fig. 3). These results imply that age-related changes to the biological functions represented by chronic low-grade inflammation with advanced age, which is known as ‘inflamm-aging’ (30, 35), are induced. A decreased Bacteroidetes to Firmicutes ratio is known to be associated with various medical conditions and diseases, such as obesity and hypertension (36, 37). As shown in Fig. 5(B), the long-term intake of LB81 yogurt results in an increased Bacteroidetes to Firmicutes ratio, suggesting that increasing this ratio by long-term LB81 yogurt intake might prevent dysbiosis-associated diseases. A diet high in saturated fat has been shown to alter the intestinal microbiome, leading to an increased incidence of obesity (38). Thus, an interesting future direction would be to determine whether long-term intake of LB81 in mice improves the altered intestinal environments introduced by high-fat diets. We found that propanoate and butanoate metabolic pathways in the mice were enhanced by long-term LB81 yogurt intake (Table 1). Because L. delbrueckii subsp. bulgaricus and Streptococcus have been shown to produce SCFA (39, 40), LB81 yogurt itself might have the potential to be associated with the production of propanoate and butanoate. Additionally, SCFAs have been shown to regulate bacterial composition and pathogenic bacteria by adjusting the pH in the intestine (41). Interestingly, the relative abundance of S24-7, which has been reported to be associated with the metabolism of propanoate and butanoate (42, 43), was significantly enhanced in the LB81 yogurt-fed group (Fig. 5E). These data indicated the possibility that long-term intake of LB81 yogurt could be associated with SCFA metabolism via modulation of commensal bacteria in addition to their direct supply of SCFAs. However, S24-7 is not a symbiotic micro-organism in humans. Therefore, it would be interesting to investigate whether the intestinal microbiome similar to S24-7 exists in humans after long-term intake of LB81 yogurt in the future. AMPs, tight-junctions, adherens-junctions and mucus all play important roles in inhibiting pathogen invasion (44, 45). Although aging contributes to intestinal barrier dysfunction (46, 47), probiotics are effective at preventing age-related intestinal dysfunction in both mice and humans (11). It is worth noting that long-term LB81 yogurt intake resulted in the increased expression of Reg3b, Reg3g, Prss22 and Ang4 in the jejunum, ileum and distal colon (Fig. 7), and enhanced expression of Muc2, C1galt1 and Retnlb in the distal colon. Thus, the present study provides evidence that long-term intake of LB81 yogurt can regulate intestinal epithelial function in mice. Our study has revealed that aging is associated with alterations in the intestinal microbiome, metabolites and gene expression of various organs in mice, and that long-term LB81 yogurt intake can support systemic functions such as intestinal barrier functioning in these mice. However, further studies are needed to determine whether probiotic yogurts can improve age-related dysfunction in humans. Supplementary Table 1 lists the bacterial families in bacterial clusters. Supplementary Table 2 lists the fecal metabolites from mice fed with the normal diet. Supplementary Table 3 shows the enriched KEGG pathways in fecal metabolite clusters. Supplementary Table 4 lists the fecal metabolites from the control mice and the LB81 yogurt-intake mice at 25 months of age. Supplementary data Supplementary data are available at International Immunology Online. Funding This work was supported by the Takeda Science Foundation (to S.U.); the Canon Foundation (to S.U.); by Grants-in-Aid for Challenging Exploratory Research (17K19543) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (to S.U.); by a grant (964930) from the Japan Agency for Medical Research and Development (AMED) (to S.U.); by Grants-in-Aid for Young Scientists (B) (16K16144) (to Y.K.) from the Japan Society for the Promotion of Science (JSPS); by Grants-in-Aid for Young Scientists (B) (16K20992) (to T.S.) from the JSPS; by Grants-in-Aid for Young Scientists (B) (16K19148) (to N.T.) from the JSPS; and a grant from Meiji Co., Ltd (to S.U.). Acknowledgements We thank S. Yin and B. Batmunkh for technical assistance, and K. Ogawa and N. Nagaya for secretarial assistance. Y.U., Y.K., K.F. and S.U. conceived and designed the study. Y.U., T.S., K.K., S.H., S.K., S.H. and Y.A. performed the experiments. Y.U. and Y.K. conducted all the data analysis. N.T., Y.O. and H.O helped with the data analysis. H.T., S.I. and G.E. provided scientific insights. Y.U., Y.K., T.S., K.F. and S.U. wrote and edited the manuscript. S.U. directed the research. Conflicts of interest statement: The authors declared no conflicts of interest. References 1 He , S. and Sharpless , N. E . 2017 . Senescence in health and disease . Cell 169 : 1000 . Google Scholar CrossRef Search ADS PubMed 2 Shlisky , J. , Bloom , D. E. , Beaudreault , A. R. , et al. 2017 . Nutritional considerations for healthy aging and reduction in age-related chronic disease . Adv. Nutr . 8 : 17 . Google Scholar CrossRef Search ADS PubMed 3 Campisi , J . 2013 . Aging, cellular senescence, and cancer . Annu. Rev. Physiol . 75 : 685 . Google Scholar CrossRef Search ADS PubMed 4 Cho , I. and Blaser , M. J . 2012 . The human microbiome: at the interface of health and disease . Nat. Rev. Genet . 13 : 260 . Google Scholar CrossRef Search ADS PubMed 5 Gilbert , J. A. , Quinn , R. A. , Debelius , J. , et al. 2016 . Microbiome-wide association studies link dynamic microbial consortia to disease . Nature 535 : 94 . Google Scholar CrossRef Search ADS PubMed 6 Mirzaei , M. K. and Maurice , C. F . 2017 . Ménage à trois in the human gut: interactions between host, bacteria and phages . Nat. Rev. Microbiol . 15 : 397 . Google Scholar CrossRef Search ADS PubMed 7 Rooks , M. G. and Garrett , W. S . 2016 . Gut microbiota, metabolites and host immunity . Nat. Rev. Immunol . 16 : 341 . Google Scholar CrossRef Search ADS PubMed 8 Schirmer , M. , Smeekens , S. P. , Vlamakis , H. , et al. 2016 . Linking the human gut microbiome to inflammatory cytokine production capacity . Cell 167 : 1125 . Google Scholar CrossRef Search ADS PubMed 9 Ardesia , M. , Villanacci , V. and Fries , W . 2015 . The aged gut in inflammatory bowel diseases . Minerva Gastroenterol. Dietol . 61 : 235 . Google Scholar PubMed 10 Nicoletti , C . 2015 . Age-associated changes of the intestinal epithelial barrier: local and systemic implications . Expert Rev. Gastroenterol. Hepatol . 9 : 1467 . Google Scholar CrossRef Search ADS PubMed 11 Madsen , K. , Cornish , A. , Soper , P. , et al. 2001 . Probiotic bacteria enhance murine and human intestinal epithelial barrier function . Gastroenterology 121 : 580 . Google Scholar CrossRef Search ADS PubMed 12 Langille , M. G. , Meehan , C. J. , Koenig , J. E. , et al. 2014 . Microbial shifts in the aging mouse gut . Microbiome 2 : 50 . Google Scholar CrossRef Search ADS PubMed 13 Calvani , R. , Brasili , E. , Praticò , G. , et al. 2014 . Fecal and urinary NMR-based metabolomics unveil an aging signature in mice . Exp. Gerontol . 49 : 5 . Google Scholar CrossRef Search ADS PubMed 14 Jonker , M. J. , Melis , J. P. , Kuiper , R. V. , et al. 2013 . Life spanning murine gene expression profiles in relation to chronological and pathological aging in multiple organs . Aging Cell 12 : 901 . Google Scholar CrossRef Search ADS PubMed 15 Borchers , A. T. , Selmi , C. , Meyers , F. J. , Keen , C. L. and Gershwin , M. E . 2009 . Probiotics and immunity . J. Gastroenterol . 44 : 26 . Google Scholar CrossRef Search ADS PubMed 16 Kim , S. W. , Suda , W. , Kim , S. , et al. 2013 . Robustness of gut microbiota of healthy adults in response to probiotic intervention revealed by high-throughput pyrosequencing . DNA Res . 20 : 241 . Google Scholar CrossRef Search ADS PubMed 17 Eri , T. , Kawahata , K. , Kanzaki , T. , et al. 2017 . Intestinal microbiota link lymphopenia to murine autoimmunity via PD-1+CXCR5-/dim B-helper T cell induction . Sci. Rep . 7 : 46037 . Google Scholar CrossRef Search ADS PubMed 18 Caporaso , J. G. , Kuczynski , J. , Stombaugh , J. , et al. 2010 . QIIME allows analysis of high-throughput community sequencing data . Nat. Methods 7 : 335 . Google Scholar CrossRef Search ADS PubMed 19 Soga , T. and Heiger , D. N . 2000 . Amino acid analysis by capillary electrophoresis electrospray ionization mass spectrometry . Anal. Chem . 72 : 1236 . Google Scholar CrossRef Search ADS PubMed 20 Soga , T. , Ueno , Y. , Naraoka , H. , Ohashi , Y. , Tomita , M. and Nishioka , T . 2002 . Simultaneous determination of anionic intermediates for Bacillus subtilis metabolic pathways by capillary electrophoresis electrospray ionization mass spectrometry . Anal. Chem . 74 : 2233 . Google Scholar CrossRef Search ADS PubMed 21 Soga , T. , Ohashi , Y. , Ueno , Y. , Naraoka , H. , Tomita , M. and Nishioka , T . 2003 . Quantitative metabolome analysis using capillary electrophoresis mass spectrometry . J. Proteome Res . 2 : 488 . Google Scholar CrossRef Search ADS PubMed 22 Sugimoto , M. , Wong , D. T. , Hirayama , A. , Soga , T. and Tomita , M . 2010 . Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral, breast and pancreatic cancer-specific profiles . Metabolomics 6 : 78 . Google Scholar CrossRef Search ADS PubMed 23 Kanehisa , M. , Furumichi , M. , Tanabe , M. , Sato , Y. and Morishima , K . 2017 . KEGG: new perspectives on genomes, pathways, diseases and drugs . Nucleic Acids Res . 45 : D353 . Google Scholar CrossRef Search ADS PubMed 24 López-Ibáñez , J. , Pazos , F. and Chagoyen , M . 2016 . MBROLE 2.0-functional enrichment of chemical compounds . Nucleic Acids Res . 44 : W201 . Google Scholar CrossRef Search ADS PubMed 25 McCall , M. N. , Bolstad , B. M. and Irizarry , R. A . 2010 . Frozen robust multiarray analysis (fRMA) . Biostatistics 11 : 242 . Google Scholar CrossRef Search ADS PubMed 26 Gautier , L. , Cope , L. , Bolstad , B. M. and Irizarry , R. A . 2004 . affy–analysis of Affymetrix GeneChip data at the probe level . Bioinformatics 20 : 307 . Google Scholar CrossRef Search ADS PubMed 27 Quackenbush , J . 2002 . Microarray data normalization and transformation . Nat. Genet . 32 : 496 . Google Scholar CrossRef Search ADS PubMed 28 Alexa , A. and Rahnenfuhrer , J . 2010 . topGO: enrichment analysis for gene ontology. R package version 2 . 29 Postler , T. S. and Ghosh , S . 2017 . Understanding the holobiont: how microbial metabolites affect human health and shape the immune system . Cell Metab . 26 : 110 . Google Scholar CrossRef Search ADS PubMed 30 Steegenga , W. T. , de Wit , N. J. , Boekschoten , M. V. , et al. 2012 . Structural, functional and molecular analysis of the effects of aging in the small intestine and colon of C57BL/6J mice . BMC Med. Genomics 5 : 38 . Google Scholar CrossRef Search ADS PubMed 31 Furusawa , Y. , Obata , Y. , Fukuda , S. , et al. 2013 . Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells . Nature 504 : 446 . Google Scholar CrossRef Search ADS PubMed 32 Lin , L. and Zhang , J . 2017 . Role of intestinal microbiota and metabolites on gut homeostasis and human diseases . BMC Immunol . 18 : 2 . Google Scholar CrossRef Search ADS PubMed 33 Thaiss , C. A. , Zmora , N. , Levy , M. and Elinav , E . 2016 . The microbiome and innate immunity . Nature 535 : 65 . Google Scholar CrossRef Search ADS PubMed 34 Odamaki , T. , Kato , K. , Sugahara , H. , et al. 2016 . Age-related changes in gut microbiota composition from newborn to centenarian: a cross-sectional study . BMC Microbiol . 16 : 90 . Google Scholar CrossRef Search ADS PubMed 35 Baylis , D. , Bartlett , D. B. , Patel , H. P. and Roberts , H. C . 2013 . Understanding how we age: insights into inflammaging . Longev. Healthspan 2 : 8 . Google Scholar CrossRef Search ADS PubMed 36 Turnbaugh , P. J. , Ley , R. E. , Mahowald , M. A. , Magrini , V. , Mardis , E. R. and Gordon , J. I . 2006 . An obesity-associated gut microbiome with increased capacity for energy harvest . Nature 444 : 1027 . Google Scholar CrossRef Search ADS PubMed 37 Yang , T. , Santisteban , M. M. , Rodriguez , V. , et al. 2015 . Gut dysbiosis is linked to hypertension . Hypertension 65 : 1331 . Google Scholar CrossRef Search ADS PubMed 38 Turnbaugh , P. J. , Bäckhed , F. , Fulton , L. and Gordon , J. I . 2008 . Diet-induced obesity is linked to marked but reversible alterations in the mouse distal gut microbiome . Cell Host Microbe 3 : 213 . Google Scholar CrossRef Search ADS PubMed 39 Corr , S. C. , Hill , C. and Gahan , C. G . 2009 . Understanding the mechanisms by which probiotics inhibit gastrointestinal pathogens . Adv. Food Nutr. Res . 56 : 1 . Google Scholar CrossRef Search ADS PubMed 40 Koh , A. , De Vadder , F. , Kovatcheva-Datchary , P. and Backhed , F. 2016 . From Dietary Fiber to Host Physiology: Short-Chain Fatty Acids as Key Bacterial Metabolites . Cell 165 : 1332 . 41 Ríos-Covián , D. , Ruas-Madiedo , P. , Margolles , A. , Gueimonde , M. , de Los Reyes-Gavilán , C. G. and Salazar , N . 2016 . Intestinal short chain fatty acids and their link with diet and human health . Front. Microbiol . 7 : 185 . Google Scholar CrossRef Search ADS PubMed 42 Meisel , M. , Mayassi , T. , Fehlner-Peach , H. , et al. 2017 . Interleukin-15 promotes intestinal dysbiosis with butyrate deficiency associated with increased susceptibility to colitis . ISME J . 11 : 15 . Google Scholar CrossRef Search ADS PubMed 43 Ormerod , K. L. , Wood , D. L. , Lachner , N. , et al. 2016 . Genomic characterization of the uncultured Bacteroidales family S24-7 inhabiting the guts of homeothermic animals . Microbiome 4 : 36 . Google Scholar CrossRef Search ADS PubMed 44 Okumura , R. and Takeda , K . 2017 . Roles of intestinal epithelial cells in the maintenance of gut homeostasis . Exp. Mol. Med . 49 : e338 . Google Scholar CrossRef Search ADS PubMed 45 Turner , J. R . 2009 . Intestinal mucosal barrier function in health and disease . Nat. Rev. Immunol . 9 : 799 . Google Scholar CrossRef Search ADS PubMed 46 Ren , W. Y. , Wu , K. F. , Li , X. , et al. 2014 . Age-related changes in small intestinal mucosa epithelium architecture and epithelial tight junction in rat models . Aging Clin. Exp. Res . 26 : 183 . Google Scholar CrossRef Search ADS PubMed 47 Mabbott , N. A. , Kobayashi , A. , Sehgal , A. , Bradford , B. M. , Pattison , M. and Donaldson , D. S . 2015 . Aging and the mucosal immune system in the intestine . Biogerontology 16 : 133 . Google Scholar CrossRef Search ADS PubMed © The Japanese Society for Immunology. 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

International ImmunologyOxford University Press

Published: May 15, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off