Background Several potential mechanisms are proposed in Reactive arthritis, including auto-immunity by molecular mimicry and persistent inflammation in response to intra-cellular live Chlamydia. Progress in this field has been limited by a lack of suitable animal models. ZAP-70W163C mutant BALB/c mice (known as SKG) develop a spondyloarthropathy syndrome after microbial beta-glucan injection.
Objectives We studied whether C. muridarum genital infection triggered reactive arthritis in SKG mice, to explore the relationship between host immunity, bacterial clearance and joint inflammation.
Methods SKG and BALB/c mice were genitally infected with mouse-adapted C. muridarum (5x102-106 IFUs). Conjunctivis and arthritis were assessed weekly for 12 weeks post infection (wpi) and skin and joint sections were scored after sacrifice. Bacterial load was quantified in genital swabs. Chlamydial MOMP antigen-specific cytokine production was assessed in splenocytes 1 wpi; cytokines were measured in serum, skin and joint explants at 12 wpi. Regulatory T cells (Treg) were depleted from FoxP3-DTR BALB/c or SKG mice. C. muridarum genomic ompA DNA was quantified by PCR in spleens, iliac lymph nodes and joints. Non parametric tests assessed statistical significance.
Results After vaginal infection with C. muridarum, female SKG mice developed asymmetrical arthritis starting 5 wpi, with enthesitis, synovitis, sacro-iliitis and dermatitis in SKG mice but no joint or skin inflammation in BALB/c mice. Spondyloarthritis was milder after genital infection of male mice, but the incidence of conjunctivitis was 40%. Live bacteria were required, as inoculation of UV-inactivated C. muridarum did not induce reactive arthritis in SKG mice. C. muridarum load 1 wpi correlated with paw swelling in female SKG mice 7 wpi. Furthermore, treatment with rifampicin and doxycycline significantly decreased paw swelling either when administered after the genital infection or after arthritis onset. MOMP-specific splenocyte proliferation was not significantly different in infected BALB/c and SKG mice 1 wpi. However, C. muridarum load was significantly increased in SKG genital swabs. MOMP-specific IFN-γ and IL-17A production was significantly decreased, and TNF production significantly increased in infected SKG relative to infected BALB/c mice. IL-6 was significantly increased in skin and joint explants from SKG relative to BALB/c mice 12 wpi. Treg depletion induced severe arthritis, conjunctivitis and dermatitis within 7 days of infection in SKG but not BALB/c mice, accompanied by increased MOMP-specific splenocyte IFN-γ, TNF and IL-6 production 1 wpi. We detected no Chlamydial genomic DNA in extra-genital tissues of either mouse strain. Autoantibodies were not detected in arthritic mice.
Conclusions The ZAP-70W163C mutation predisposes mice to reactive arthritis after genital C. muridarum infection, which was not associated with autoantibodies or persistent live Chlamydia in joints. Control of bacterial load by the SKG host immune response is reduced. Reactive arthritis likely results from exuberant inflammatory response to infection in SKG mice, as factors reducing inflammation including Treg, and reduced bacterial burden through UV inactivation or antibiotic treatment, reduce arthritis severity.
Disclosure of Interest None Declared