Article Text
Abstract
Background Systemic lupus erythematosus (SLE) is frequently featured by musculoskeletal issues as arthralgia and arthritis, which account for the most important burdens in patient-based questionnaires. In most cases, lupus arthritis appears as non-erosive in projection radiography, but presented as (minimal) erosive disease in recent MRI-based observations. Despite its high prevalence, lupus arthritis is not in the focus of intensive basic research, since it is rarely seen in murine lupus models. The mineral oil pristane can induce a systemic lupus-like disease in mice featured by typical auto-antibodies, the involvement of inner organs, and also by arthritis.
Objectives To establish pristane induced lupus (PIL) in BALB/c mice as suitable model for lupus arthritis, provide in-depth analysis of both clinical and histological features by using an established work up procedure originating from RA basic research.
Methods For disease induction BALB/c mice were injected i.p. with either 0.5ml of pristane (PIL group n=57) or PBS (healthy controls, HC, n=35) and sacrificed after 8 months. In addition, also Bl/6 mice were treated similarly (10 PIL, 10 HC). Animals were monitored for clinical signs of arthritis (paw swelling, loss of grip strength) and analyzed by histopathology techniques: H/E (overview), TRAP (osteoclasts), toluidine blue (cartilage). In order to analyze and compare disease severity, the histological features were quantified with an image analysis system (osteomeasure) and also combined to define an arthritis severity score (ASS). Specimens were also stained for granulocytes (G1), T- (anti-CD3) and B-Cells (anti-CD45) for immunohistochemistry. Blood samples were tested for anti-chromatin-abs, anti-histone-abs and rheumatoid factor.
Results After 3 months, clinical signs of arthritis developed and constantly increased; finally 75% of mice experienced at least one episode of paw swelling, while HC were free of arthritis (p<0.001). In PIL, clinical features correlated well with histological results in osteomeasure analysis: cartilage destaining, inflammatory area, erosive area, number of osteoclasts and severity score ASS (for all: r>0.7, p<0.0001).
Histological analysis showed arthritis in 58% of PIL-group, but none in HC (p<0.001). Arthritis was erosive with cartilage degradation; inflammatory infiltrates were rich in granulocytes and B cells (but low in T cells). The frequency of affected mice gradually increased over time (50% after 6 months, 58% after 8 months, 75% after 12 months, respectively), but mean disease severity did not (mean ASS 8.3 after 6 months, 7.3 after 8 months, 7.1 after 12 months, respectively). In contrast to established models of RA (TNFtg, CIA), PIL arthritis does not lead to joint destruction or ankylosis, even after 1 year of observation. Preceding the onset of arthritis, PIL mice developed anti-chromatin and anti-histone abs; RF was elevated after 8 months in both PIL and controls. Bl/6 mice do not develop arthritis; neither clinically nor in histological analysis.
Conclusions In contrast to most murine models of SLE, PIL in BALB/c mice is featured by arthritis. Despite its erosive nature, PIL arthritis does not procede to joint destruction as seen in human RA or murine RA models. PIL may provide a good tool to study lupus arthritis in knock-out or transgenic mouse models based on a BALB/c, but not on a Bl/6 background.
Disclosure of Interest None Declared