Chronic oral infection with Porphyromonas gingivalis accelerates atheroma formation by shifting the lipid profile

Abstract

Background: Recent studies have suggested that periodontal disease increases the risk of atherothrombotic disease. Atherosclerosis has been characterized as a chronic inflammatory response to cholesterol deposition in the arteries. Although several studies have suggested that certain periodontopathic bacteria accelerate atherogenesis in apolipoprotein E-deficient mice, the mechanistic link between cholesterol accumulation and periodontal infection-induced inflammation is largely unknown.

Methodology/principal findings: We orally infected C57BL/6 and C57BL/6.KOR-Apoe(shl) (B6.Apoeshl) mice with Porphyromonas gingivalis, which is a representative periodontopathic bacterium, and evaluated atherogenesis, gene expression in the aorta and liver and systemic inflammatory and lipid profiles in the blood. Furthermore, the effect of lipopolysaccharide (LPS) from P. gingivalis on cholesterol transport and the related gene expression was examined in peritoneal macrophages. Alveolar bone resorption and elevation of systemic inflammatory responses were induced in both strains. Despite early changes in the expression of key genes involved in cholesterol turnover, such as liver X receptor and ATP-binding cassette A1, serum lipid profiles did not change with short-term infection. Long-term infection was associated with a reduction in serum high-density lipoprotein (HDL) cholesterol but not with the development of atherosclerotic lesions in wild-type mice. In B6.Apoeshl mice, long-term infection resulted in the elevation of very low-density lipoprotein (VLDL), LDL and total cholesterols in addition to the reduction of HDL cholesterol. This shift in the lipid profile was concomitant with a significant increase in atherosclerotic lesions. Stimulation with P. gingivalis LPS induced the change of cholesterol transport via targeting the expression of LDL receptor-related genes and resulted in the disturbance of regulatory mechanisms of the cholesterol level in macrophages.

Conclusions/significance: Periodontal infection itself does not cause atherosclerosis, but it accelerates it by inducing systemic inflammation and deteriorating lipid metabolism, particularly when underlying hyperlidemia or susceptibility to hyperlipidemia exists, and it may contribute to the development of coronary heart disease.

Figure 1. Determination of alveolar bone loss (A–D).

Three-dimensional (A and B) and two-dimensional (C and D) microcomputed tomography images of representative samples from sham-infected mice (A and C) and infected mice (B and D) are presented. The infected group (A) showed an increase in the distance from the cement-enamel junction (CEJ) to the alveolar bone crest compared with the sham-infected group (B). Two-dimensional micro-CT sections for the infected group (C) but not the sham-infected group (D) showed alveolar bone loss at the bifurcation. Numerical analysis of alveolar bone loss between the control group and the infected group (E; N = 10 in each group) was performed. The distance between the cement-enamel junction and alveolar bone crest in each tooth root was determined from three-dimensional micro-CT images using image visualization software. Box plots present medians and the 25th and 75th percentiles as boxes and the 10th and 90th percentiles as whiskers. Significant differences were observed between the infected group and the sham-infected group (* P<0.01; ** P<0.001, Mann-Whitney U-test).

Figure 2. Effects of oral infection with P. gingivalis on serum levels of interleukin (IL)-6 (A), serum amyloid A (SAA; B), and anti-P. gingivalis antibody (C; N = 5 in each group).

All experiments were performed in triplicate wells for each condition and repeated at least twice. Representative data are shown. Box plots present medians and the 25th and 75th percentiles as boxes and the 10th and 90th percentiles as whiskers. Significant differences were observed between the infected group and the sham-infected group (* P<0.05; ** P<0.01, Mann-Whitney U-test).

Figure 3. Effects of oral infection with P. gingivalis on aortic atherosclerosis in wild-type and B6.Apoeshl mice.

(A) Representative aortas from wild-type and B6.Apoeshl mice are depicted. (B) Aortic atherosclerosis expressed as a percentage of the total area (N = 8 in each group). Box plots present medians and the 25th and 75th percentiles as boxes and the 10th and 90th percentiles as whiskers. Significant differences were observed between the infected group and the sham-infected group (* P<0.01, Mann-Whitney U-test). (C) Representative aortic sinus cross sections from wild-type and B6.Apoeshl mice. Original magnifications, 40×.

Figure 4. Comparison of relative gene expression levels in the aorta between the control group and the infected group or between the short-term and long-term infected groups (N = 5 in each group).

The relative quantity of experimental mRNA was normalized to the relative quantity of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA. The box plots present medians and the 25th and 75th percentiles as boxes and the 10th and 90th percentiles as whiskers. (* P<0.01, Mann-Whitney U-test)

Figure 5. Comparison of the relative gene expression levels in the liver between the control group and the infected groups or between the short-term and long-term infected groups (N = 5 in each group).

The relative quantity of mRNA was normalized to the relative quantity of GAPDH mRNA. The box plots present medians and the 25th and 75th percentiles as boxes and the 10th and 90th percentiles as whiskers. (* P<0.05; ** P<0.01, Mann-Whitney U-test)

Figure 6. Induction of IRF3 phosphorylation in RAW 264.7 macrophages by TLR2 and TLR4 ligands.

Cells were stimulated with 1.0 µg/ml of P. gingivalis LPS, 0.1 µg/ml of E. coli LPS or 0.1 µg/ml of Pam3CSK4 for 12 hrs. Phosphorylation of IRF3 cellular extracts was analyzed by western blotting. Blot is a representative of three independent experiments.

Figure 7. Lipopolysaccharide from P. gingivalis and E. coli inhibit cholesterol efflux and enhance cholesterol uptake.

(A) Peritoneal macrophages from wild-type and B6.Apoeshl mice were loaded with BODIPY-cholesterol and treated with vehicle, 1 µg/ml P. gingivalis LPS or 0.1 µg/ml E. coli LPS in the presence or absence of 1 µM GW3965. Data are presented as apoA1-dependent cholesterol efflux. (B) Cholesterol uptake was estimated as the total of cellular and effluxed cholesterol. The results are shown as the mean ± S.D. of three independent experiments. There were significant differences found between the wild-type and B6.Apoeshl macrophages (unpaired t-test; § P<0.05) and between control and LPS-treated macrophages in the absence (paired t-test; * P<0.05) or presence of GW3965 (paired t-test; † P<0.05).

Figure 8. Lipopolysaccharide from P. gingivalis and E. coli alter the LDLR expression in macrophages.

Peritoneal macrophages from wild-type and B6.Apoeshl mice were cultured in the sterol depletion medium for 8 hours and the effect of LPS was evaluated in the presence or absence of GW3965 following 18 hours of incubation. (A) Total RNA was extracted from the cells and gene expression of LDLR and Idol was analyzed by real-time PCR. Data are expressed as mRNA expression relative to the expression without stimulation. Results are shown as mean ± S.D. of three independent experiments. There were significant differences between the wild-type and B6.Apoeshl macrophages (unpaired t-test; § P<0.05) and between control and LPS treated macrophages in the presence (paired t-test; † P<0.01) or absence of GW3965 (paired t-test; * P<0.05). (B) Cell lysates were separated by SDS-PAGE and immunoblotted with anti-LDLR and anti-GAPDH antibodies. Results are representative of those in three independent experiments. The result of densitometric analysis of western blotting shown as mean ± S. E. of three independent experiments. Significant difference between sterol-depletion control and the different LPS stimulations are indicated (paired t-test; * P<0.05).