Niche-specialist microbial communities in major oral habitats
Distinct core species in each oral habitat. A total of 989 OTUs (971 thousand sequences in 20 samples) were defined, including 908 OTUs belonging to 651 genera, and 81 OTUs of unclassified genera (Fig. S1A). Alpha- diversity analysis revealed that gingiva and palatal samples were richer (median chao1 index > 250), more diverse (median simpson index < 0.4) and evener (median simpsoneven index > 0.3) than those of tongue and buccal samples respectively (Fig. 1A-C and Fig. S1A, C-D). The coordination of the Bray-curtis PCoA plot (beta-diversity measure) showed that bacterial communities from tongue and buccal samples clustered together, while those from palate and gingival samples formed another two clusters (Fig. 1D). Gingiva and palatal samples harbored a higher relative abundance of Lactobacillus, Staphylococcus, and Streptococcus, while tongue and buccal samples harbored the highest relative abundances of Streptococcus (Fig. 1E-F and Fig. S1B). In particular, the Corynebacterium, Rodentibacter and Stenotrophomonas were proportionally enriched in buccal samples, the Sporosarcina and Atopostipes were proportionally enriched in gingiva, while the Neomicrococcus, Staphylococcaceae and Methylobacterium were enriched in palate (Fig. 1G). Among 989 OTUs identified in different oral sites, 25 were overabundant in gingiva, 57 were overabundant in palate, 21 were overabundant in buccal tissue, and 6 of these taxa were overabundant in tongue (Fig. 1H). We also performed compact visualizations of microbial metagenomes using Graphical phylogenetic analysis (GraPhlAn) of each sample (Fig. S1E-H). All these results suggest the hypothesis that the oral microbiota is site-specialist, and each oral habitat has its distinctive core species[7].
Potential interactions and niche-sharing among oral taxa in different sites. We argue that most microorganisms restricted to the habitat result from the co-evolution of microbes in the unique site leading to highly specific taxon-taxon interplays. Therefore, the network analysis was performed at the top 50 OUT to provide details supporting this hypothesis. Among 216 links identified, there showed negative correlations in gingival and palatal samples (Fig. 2A, D) while a predominance of positive correlations in tongue and buccal samples (Fig. 2B-C). Specifically, fewer links and a lower level of centralization were found in palate (Fig. 2D), whereas more links and a higher level of centralization were found in buccal (Fig. 2C) than in other sample groups. Network analysis was also performed at the genus level (Fig. S2). The results confirm the modular organization of the microbiota, and further accentuate initial findings from the analysis at the level of OTUs.
The functional role in the distinct communities. Not surprisingly, through COG and KEGG pathways analysis (Fig. 2E and Fig. S2E), the potential functional pathways were differently enriched in each oral area. The relative abundance of several metabolic pathway-related OTUs as well as energy production and conversion-related OTUs increased in palate samples. In gingiva, there were enriched OTUs associated with cell functional activities like cell wall/membrane/envelope biogenesis, cytoskeleton, intracellular trafficking, secretion, and vesicular transport, cell motility and so on. While in tongue and buccal samples, it was the relative abundance of defense mechanisms-related, replication, recombination and repair-related, and cell cycle control, cell division, chromosome partitioning-related OTUs developed. The findings firstly suggest that the functional role that the specific members play in each oral mucosa presents differently and provides a strong microbiological barrier to support mucosal homeostasis.
The distribution of Th cells and γδT cells in different oral sites. Flow cytometry analysis showed that the proportion of Th1 cells in tongue (38.18%, [27.03%-49.34%]) and buccal (18.62%, [12.99%-24.24%] samples was higher than those in gingiva (7.79%, [4.92%-10.67%]) and palate (7.19%, [3.62%-10.76%], P < 0.01, Fig. 3A). The proportion of γδT cells in buccal (25.82%, [22.1%-29.54%]) and gingiva (20.42%, [18.31%-22.53%]) samples was higher than those in palate (14.18%, [11.69%-16.67%]) and tongue (9.38%, [5.38%-13.37%], P < 0.01, Fig. 3E). Meanwhile, the proportion of Th2, Th17, Treg cells showed no significance in each oral site (Fig. 3B-D).
Correlation analysis between immune cells and the oral microbiomes. Th1 cells revealed a negative correlation with some opportunity infectious bacterial such as Acinetobacter, Massilia, Aeromonas, Micrococcus and Enterobacter; and some strictly anaerobic or facultative anaerobic bacterial such as genus Staphylococcus, Lactobacillus, Sporosarcina, Atopostipes, Faecalibaculum and Bifidobacterium while those positively correlated with Treg cells (Fig. 4A). Th1 cells showed a negative correlation with the diversity of oral microbiota (R2 = 0.58, P < 0.05, Fig. 4B-C). From the recent findings, there is no significant correlation between Th2, Th17, γδT cells and oral communities in different oral habitats (Fig. 4A-B, D-G). These works arise the potential correlation between the site-specialist microbiota and the distribution of mucosal immune cells, till their causal relationship need more further study.
Age-related microbiome communities in the major life stages
Distinctive core species in each life stage. A total of 638 OTUs (709 thousand sequences in 10 samples) were defined, including 451 OTUs belonging to 265 genera, and 187 OTUs of unclassified genera (Fig. S3A). Our findings revealed that the community richness was higher in old mice samples than that in adult samples (P < 0.05, Fig. 5A, Fig. S3A-F), while community diversity and evenness showed no significant difference (P < 0.05, Fig. 5B-C). The coordination of the Bray-curtis PCoA plot (beta-diversity measure) performed two clustering groups (Fig. 5D). GraPhlAn was used to visualize the microbial metagenomes of each sample (Fig. 5E-F). From bar graph representing the relative community composition at the genus level (Fig. 5G, Fig. S3G), genus Streptococcus was the most general microbiome in oral cavity. LEfSe analysis as well as the richness heatmap discovered that the Actinocrinis puniceicyclus, Actinospica, Actinospicaceae and Lactobacillus were proportionally enriched in old samples, while the Jeotgalicoccus, Bosea and Sphingomonas were significantly enriched in adult samples (Fig. 5H-I, Fig. S3I). These data suggests that species abundance, rather than the overall species diversity, largely contributes to the observed differences in salivary microbiota between the adult and old groups.
Taxa-taxa interactions in different periods. Among 139 links identified, there showed negative correlations in old samples while a half of positive correlations in adult samples (Fig. 6A-B). Specifically, fewer links and a lower level of centralization were found in the old (Fig. 6B), whereas more links and a higher level of centralization were found in the adult (Fig. 6A). In addition, adult samples contained several health-associated taxa (Bifidobacterium, Bradyrhizobium, Lactobacillus, Turicibacter, and Corynebacterium; Fig. 6A and Fig. S4A), while disease-associated taxa (Pseudomonas, Enterobacteriaceae, Enterobacter, Staphylococcus, Bacillaceae, Geobacillus, and unclassified genera) were found in old samples (Fig. 6B and Fig. S4B). Network analysis was also performed at the genus level (Fig S4A-B). The analysis at the genus level also readily identified the interactions among specific taxa across the sample groups, such Lactobacillus, Caldalkalibacillus_thermarum, and Bifidobacterium.
The functional role of the oral microbiota in the phases. Differences in predicted COG and KEGG functional pathways were observed in the saliva microbiota of adult and old mice (Fig. 6C and Fig. S4C). Remarkably, increase in metabolic pathways were observed in adult mice, while cell period pathways like cell cycle control, cell division, chromosome partitioning, replication, recombination and repair, were increased in old mice. Taken together, these results suggest that changes of the oral microbiota during ageing and the interference with signaling pathways contribute to the disorders in barrier function and disease susceptibility.
The distribution of Th cells and γδT cells in different stages. The findings revealed that the proportion of Th1 cells in adult groups (28.62%, [21.45%-35.8%]) was higher than those in old groups (11.59%, [9.82%-13.37%], P < 0.01, Fig. 7A, E), while the proportion of Th2 (31.3%, [16.16%-46.44%]), Th17 (27.06%, [15.76%-38.36%]) and Treg (29.74%, [15.71%-43.77%]) T cells in old samples was higher than those in adult samples (P < 0.01, Fig. 7B-D, F-H). Meanwhile, the proportion of γδT cells showed no significance between two groups (Fig. S5A-B).
Correlation analysis between immune cells and the oral microbiomes. The correlations of the relative abundance of bacteria with immune cell subsets, including Th1, Th2, Th17, Treg and γδT cells, were shown in a spearman correlation heatmap (Fig. 8A). At the genus level, the abundance of old-enriched bacterial positively correlated with the proportion of Th2 (Fig. 8D, R2 = 0.11, P < 0.05), Th17 (Fig. 8E, R2 = 0.16, P < 0.05), Treg (Fig. 8F, R2 = 0.16, P < 0.05) and negatively correlated with the proportion of Th1 (Fig. 8B, R2 = 0.19, P < 0.05). In contrast, Sphingomonas, Burkholderia−Caballeronia−Paraburkholderia and Staphylococcus which enriched in adult mice samples, negatively correlated with the proportion of Th2 (R2 = 0.11, P < 0.05), Th17 (R2 = 0.16, P < 0.05), and Treg (R2 = 0.16, P < 0.05) and positively correlated with the proportion of Th1 (R2 = 0.19, P < 0.05).