Background Although in vivo studies have demonstrated that periodontitis aggravates experimental arthritis, there are no animal models that mimic the co-occurrence of these diseases.

Objectives To investigate the arthritogenic effect of lipopolysaccharide (LPS) in a mouse model of periodontal disease.

Methods Periodontitis was induced in CD1 mice by injection of 0.01 or 0.05 µg of LPS in 5 µl of PBS every 48 hour into the vestibular gingiva of the second molar on the left maxilla. Untreated mice or injected with LPS at the tail were used as controls. Mice (n=10 per condition) were monitored daily and arthritis was estimated by conventional visual scoring method (scale 0–5) and recording the paw swelling with a calliper. 2 weeks after the 9th injection mice were sacrificed to collect blood, maxilla and paw samples. The left maxilla was analysed by microCT and the alveolar bone loss was assessed measuring the distance between the cementum-enamel junction (CEJ) and the alveolar bone crest (ABC) of each molar. Ultrasound (US) was performed to measure the ankle joint space. Periodontal and paw tissues were processed for histological analysis. Inflammation, vascular proliferation and bone resorption were scored (0–3) in maxilla. Inflammation, pannus formation, cartilage and bone destruction were scored (0–5) in ankle joints. CXCL1, IL-1β, IL-6 and TNF serum levels were determined by ELISA.

Results Ankle swelling and inflammation were noted after the 5th periodontal injection of 0.05 µg of LPS, picked at day 18 and continued for the next 15 days with paw swelling and score higher than those of untreated mice (at the sacrifice p<0.001). 0.01 µg of LPS did not induce paw changes. Therefore, the subsequent assessments were conducted only in mice injected with 0.05 µg of LPS. The CEJ-ABC distance was greater in the inoculated (0.29±0.08 mm) than in the control (0.17±0.05 mm) mice (p<0.001). Histological analysis showed that LPS induced a mild vascular proliferation (score 0.8±0.42) in periodontal tissue and a substantial alveolar bone resorption (score 1.8±0.42), but not inflammation. US revealed the presence of effusion and a 1.5-fold higher joint space in the ankle of mice with periodontitis than in controls (p<0.05). Leukocyte infiltration (score 2.36±1.56) and synovial proliferation (score 2.09±1.54) were observed after histology in ankle joints of mice injected orally. The same sections had slight cartilage (score 1.32±1.21) and bone destruction (score 0.68±0.72). Animals that received LPS tail injection did not show any clinical and histological signs of arthritis. CXCL1 and TNF were higher in arthritic mice (CXCL1:2226.87±264.38 pg/ml; TNF:24.55±7.0 pg/ml), than in controls (CXCL1:445.97±92.09 pg/ml; TNF:3.22±1.04 pg/ml). Although there was no statistical difference, IL-1β and IL-6 were highest in LPS-mice (IL-1β:79.49±11.99 pg/ml; IL-6:196.02±40.62 pg/ml).

Conclusions This study shows that experimental arthritis and periodontal disease can co-occur after LPS oral injection in mice. Our model may be useful to improve the understanding of the mechanisms underlying the link between periodontitis and arthritis.

Disclosure of Interest None declared