Oral pathobiont induces systemic inflammation and metabolic changes associated with alteration of gut microbiota

Abstract

Periodontitis has been implicated as a risk factor for metabolic disorders such as type 2 diabetes, atherosclerotic vascular diseases, and non-alcoholic fatty liver disease. Although bacteremias from dental plaque and/or elevated circulating inflammatory cytokines emanating from the inflamed gingiva are suspected mechanisms linking periodontitis and these diseases, direct evidence is lacking. We hypothesize that disturbances of the gut microbiota by swallowed bacteria induce a metabolic endotoxemia leading metabolic disorders. To investigate this hypothesis, changes in the gut microbiota, insulin and glucose intolerance, and levels of tissue inflammation were analysed in mice after oral administration of Porphyromonas gingivalis, a representative periodontopathogens. Pyrosequencing revealed that the population belonging to Bacteroidales was significantly elevated in P. gingivalis-administered mice which coincided with increases in insulin resistance and systemic inflammation. In P. gingivalis-administered mice blood endotoxin levels tended to be higher, whereas gene expression of tight junction proteins in the ileum was significantly decreased. These results provide a new paradigm for the interrelationship between periodontitis and systemic diseases.

Figures

Figure 1. Systemic and local inflammation in P. gingivalis-administered and sham-administered mice (N = 8 in each group).

(a) Serum level of IL-6. All data are means ± SD. Significant differences were observed between the P. gingivalis-administered groups and the sham-administered group (***p < 0.001, Mann-Whitney U-test). (b) Histological findings of gingival tissues of P. gingivalis-administered and sham-administered mice. Sections of the periodontium around the disto-buccal root of the first molar were H-E stained. Right panels are magnified views of the boxed areas. No difference of gingival inflammation was observed between P. gingivalis-administered and sham-administered mice.

Figure 2. P. gingivalis induced insulin resistance.

(a) Intraperitoneal glucose tolerance test. Blood glucose levels were determined at the indicated times after intraperitoneal load of glucose (1 g/kg). (b) Area under blood concentration curve (AUC) of the Fig. 2a. (c) Intraperitoneal insulin tolerance test (ITT). Blood glucose levels were determined at the indicated times after intraperitoneal load of insulin (0.5 unit/kg). (N = 6 in each group). (d) Insulin sensitivity was assessed by inverse AUC of ITT. All data are means ± SD. (*p

Figure 3. Relationship between P. gingivalis administration and inflammatory changes of epididymal adipose tissue and liver.

(a) Epididymal adipose tissues were fixed 10% formalin, sectioned, and stained with a rat anti-mouse F4/80 primary antibody (Abd Serotec). Arrow heads indicate F4/80 positive macrophages. (b) Oil red O staining of the liver tissue are shown. Increased number of lipid containing hepatocytes (arrow heads) are seen in P. gingivalis-administered mice. (c) P. gingivalis administration increased hepatic triglyceride content. (N = 8 in each group). All data are means ± SD. (**p < 0.01, Mann-Whitney U-test).

Figure 4. Effect of oral administration of P. gingivalis on the gene and protein expressions in epididymal adipose tissue.

(a) Comparison of relative gene expression levels in the adipose tissue between the P. gingivalis-administered and the sham-administered mice (N = 8 in each group). The relative quantity of experimental mRNA was normalized to the relative quantity of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. (b) CTRP9 contents in the epididymal adipose tissue were determined by ELISA of tissue lysates. All data are means ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001, Mann-Whitney U-test).

Figure 5. Effect of oral administration of P. gingivalis on the gene and protein expressions in liver.

(a) Comparison of relative gene expression levels in the liver tissue between the P. gingivalis-administered and the sham-administered mice (N = 8 in each group). The relative quantity of experimental mRNA was normalized to the relative quantity of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. (b) TNF-α levels in the liver tissue were determined by ELISA of tissue lysates. All data are means ± SD. (*p < 0.05; **p < 0.01; ***p < 0.001, Mann-Whitney U-test).

Figure 6. Comparison of the gut microbiota between P. gingivalis-administerd and sham-administered mice by 16S rRNA sequencing analysis.

Relative abundances of each bacterial group in Phylum (a) and OTU (b) level are indicated by boxplot. Each defined OTU obtained from 16S rRNA deep sequencing was compared to the genome database (GenomeDB) from NCBI using blastin (c). Close relative species and percent similarities are shown. Fold increases indicate the mean ratio of relative abundance of the OTU in the gut microbiota from the P. gingivalis-administered group to sham-administered group. (*p < 0.05, **p < 0.01, Mann-Whitney U-test).

Figure 7. Effect of oral administration of P. gingivalis on the serum endotoxin levels.

(a) Serum endotoxin (LPS) concentration (EU/ml) were determined after 10 cycles of P. gingivalis administration or sham administration (N = 8 in each group). All data are means ± SD. (b) Serum endotoxin (LPS) concentration (EU/ml) were determined at the indicated times after single administration of P. gingivalis (N = 6 in each group). All data are means ± SD. (*p < 0.05, one-way ANOVA).

Figure 8. Comparison of relative gene expression levels in the small intestine (a) and colon tissue (b) between the sham-administered group and the P. gingivalis-administered group (N = 8 in each group).

The relative quantity of experimental mRNA was normalized to the relative quantity of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. All data are means ± SD. (*p