Active invasion of oral and aortic tissues by Porphyromonas gingivalis in mice causally links periodontitis and atherosclerosis

Abstract

Atherosclerotic vascular disease is a leading cause of myocardial infarction and cerebrovascular accident, and independent associations with periodontal disease (PD) are reported. PD is caused by polymicrobial infections and aggressive immune responses. Genomic DNA of Porphyromonas gingivalis, the best-studied bacterial pathogen associated with severe PD, is detected within atherosclerotic plaque. We examined causal relationships between chronic P. gingivalis oral infection, PD, and atherosclerosis in hyperlipidemic ApoEnull mice. ApoEnull mice (n = 24) were orally infected with P. gingivalis for 12 and 24 weeks. PD was assessed by standard clinical measurements while the aorta was examined for atherosclerotic lesions and inflammatory markers by array. Systemic inflammatory markers serum amyloid A, nitric oxide, and oxidized low-density lipoprotein were analyzed. P. gingivalis infection elicited specific antibodies and alveolar bone loss. Fluorescent in situ hybridization detected viable P. gingivalis within oral epithelium and aorta, and genomic DNA was detected within systemic organs. Aortic plaque area was significantly increased in P. gingivalis-infected mice at 24 weeks (P<0.01). Aortic RNA and protein arrays indicated a strong Th2 response. Chronic oral infection with P. gingivalis results in a specific immune response, significant increases in oral bone resorption, aortic inflammation, viable bacteria in oral epithelium and aorta, and plaque development.

Figure 1. Chronic oral P. gingivalis infection and active tissue invasion induced significant alveolar bone resorption.

(A) Infection and oral microbial sampling schedule. (B) P. gingivalis infected mice (n = 6) had significantly higher levels of alveolar bone resorption relative to control mice (n = 6) at both 12- and 24 weeks of infection (***P<0.001). (C) Mandible lingual view of 12 and 24 week infected and control mice showing outlined area of bone resorption between cemento-enamel junction and alveolar bone crest. The area of bone resorption is across molar 1. (D) Gingival tissue histology. Mild epithelial hyperplasia and minimal inflammation are seen in both 12- (i)- and 24 (iii)-week P. gingivalis-infected mice relative to sham-infected age-matched controls (ii and iv, respectively). EH – epithelial hyperplasia, CEJ – cemento-enamel junction, ABC – alveolar bone crest, D – dentin, CT – connective tissue. (E) FISH demonstrates viable P. gingivalis invasion of gingival connective tissue in a 12 week-infected mouse (i) and of gingival epithelium in a 24 week-infected mouse (ii). White arrowheads indicate fluorescently labeled bacteria. Scale bar represents 10 µm.

Figure 2. P. gingivalis-specific humoral immune response.

Serum IgG and IgM titers were determined by ELISA. Both 12- and 24 week-infected mice (n = 6) had statistically significantly higher levels of IgG relative to controls (n = 6). Only 24 week-infected mice had significantly higher levels of IgM relative to controls. (*P<0.05, **P<0.01).

Figure 3. P. gingivalis infection-induced plaque growth in the aorta of mice at 12 and 24 weeks.

FISH demonstrated viable bacteria in the aortic vessel concomitant with plaque growth. Hematoxylin and eosin stained aortic cross sections (A) i–12 weeks after P. gingivalis infection with demonstrated plaque in the mouse aorta; ii–24 week P. gingivalis-infected mouse ascending aorta with plaque containing cholesterol crystals; iii–12 week control aorta at the level of the aortic valve showing a small plaque. Arrowheads delineate intimal plaque margins, elongated arrows indicate cholesterol crystals. Scale bar represents 100 µm. Bar graphs of morphometric analysis of mean aortic plaque area (B), intimal thickness (C), and intimal/medial thickness ratios (D). (N = 12 for all graphs). Plaque area of 12 and 24 week-infected mice was significantly greater than controls, (P<0.05). (E), FISH illustrating P. gingivalis (red) in 12 week-infected mouse (i) aortic vessel, (ii) adventitia (perinuclear localization of P. gingivalis can be seen the inset), and (iii) a cluster of P. gingivalis in an endothelial cell. (iv) P. gingivalis in the endothelial cell of a 24-week infected mouse, (v) brightfield view of iv, (vi) overlay of iv and v. L – aortic lumen. Mouse cell nuclei are pseudo-colored blue. White arrows indicate labeled bacterial cells. Scale bar is 10 micrometers.

Figure 4. Macrophage invasion into the aortic arch intimal layer at 12 and 24 weeks of infection.

Macrophage cell counts are significantly increased at 12 weeks follow up. (A), Immunohistochemical staining of aortic arterial cross sections demonstrate F4/80⁺ macrophage cells in 12-week-infected mouse and 12-week control mouse. White arrow heads indicate positively-stained cells. (B), Macrophage cell counts for each mouse represent counts in three high-power fields (100X) in the 12 and 24 week groups (***P<0.001). N = 6 for all groups.

Figure 5. Systemic risk factors for atherosclerosis are elevated in P. gingivalis infection.

(A, B), Graphical representation of Table 3 (n = 6). CM – chylomicron, VLDL – very low-density lipoprotein, LDL – low-density lipoprotein, HDL – high-density lipoprotein, Pg – P. gingivalis, Con – control. C, Level of oxidized LDL in serum of 24 week-infected mice is significantly higher than in controls (**P<0.01; n = 6). (D) Infected mice had statistically significantly elevated levels of SAA relative to controls. *P<0.05. E, Infection with P. gingivalis significantly reduced serum nitric oxide relative to uninfected control mice (*P<0.05; n = 6).