TLRs and innate immunity in spondyloarthritis: Animal model and pathogenesis

Axial spondyloarthritis including ankylosing spondylitis (AS) is an immune-mediated disease mainly affecting the sacroiliac (SI) joints. Its pathophysiological origin is unknown, several factors have been speculated to be responsible for induction of AS: A close genetic association between AS and the expression of the class I major histocompatibility molecule HLA- B27 is well established. Recently two other genetic factors have been shown to influence AS, polymorphisms in the metalloprotease ARTS-1 which is involved in MHC class I loading and the receptor for interleukin 23 (IL23R). IL23 is a key component in the induction of TH17 cells, which play a crucial role in the control of autoimmunity. Besides genetic factors, bacterial infections have been a subject of AS research. The connection is based on data collected in HLA-B27 transgenic animals, which only develop disease when kept in non-sterile conditions, and data from patients infected with Klebsiella. Both, transgenic animals in research facilities and patients, however, are not only exposed to bacteria, but also encounter viruses regularly. This project will further study and compare the possible role of bacterial substances (TLR2 and TLR4 ligands) and viral nucleic acids (TLR3, TLR7 and TLR9 ligands) for induction and chronicity of AS-disease. We will use two parallel strategies, one in vivo using model mice lacking the specific TLR and the other using material from patients with AS in an in vitro system: (1) We will induce spondyloarthritis in mice using established protocols and follow disease development using MRI and at the endpoint detect bony changes utilizing CT scan technology. We will also use histology to determine the presence of macrophages and NK cells in the ankylosing joints. Using TLR knockout mice will allow us to determine which trigger is necessary for disease development. (2) In parallel we will examine possible effects of TLR-mediated activation on the maturation of monocyte derived dendritic cells and activation of blood derived macrophages by synovial fluid from patients extracted during routine clinical examinations. Blocking TLRs with specific antibodies or digesting nucleic acids will show if there is a TLR stimulus chronically present in the affected joint. If so, we will be able to determine the nature of the stimulus. Additionally, we will measure several soluble markers of activation of the innate immune system in the patient samples. After completion of this project with the proposed experiments we will know more about innate immune mechanisms of SpA/AS-disease development and perpetuation in vivo and in vitro. Most importantly the results could help redefining treatment strategies.

During this FWF-project, a mouse model for spondyloarthritis (SpA) was established in Innsbruck, which can be used for long-term studies. The new aspect is that the sacroiliac joints were examined both with magnetic resonance imaging (MR) for the detection of inflammation as well as with micro-computerized tomography (CT) for the detection of bony changes typical for ankylosing spondylitis. Our results show that the lack of toll-like-receptors (TLRs) may indeed influence the course of SpA-disease. The SpA-disease is the most frequent immune-mediated rheumatic disease, and every third patient develops typical bony changes of the sacroiliac joints and the vertebral column. Patients often report infections as a possible trigger of SpA-activity, but this phenomenon has not been investigated in detail so far. Our earlier studies showed the unusual presence of TLRs on senescent T-cells, leading to immune cell activation upon binding of bacterial products. Others showed that HLA-B27+/+ mice do not develop SpA in a germ-free environment. A precondition for this project was to improve imaging of the murine sacroiliac joints using MR and micro-CT. The second aim was then to examine the role of TLRs in SpA. The results of the first part of the project showed a different presentation of SpA in mice with a Th1-skewed immune response compared to mice with a Th2-skewed immune response both in the MR and the profile of serum biomarkers. Specifically, Th1-prone mice without the specific TLR4 expression showed more inflammation in the sacroiliac joints than wildtype mice, although bony changes were comparable within all groups of mice. We conclude that this new Th1-prone mouse model is usefuly to examine SpA stage I (with inflammation but without bony changes), but not for SpA stages II and III with bony changes. Therefore, we then established a murine SpA model in cooperation with Prof. Dr. T. Glant in Chicago. Indeed, this model showed both inflammation by using MRs and bony changes typical for ankylosing spondylitis by using micro-CTs. In accordance with these changes, the profiles of serum biomarkers changed during the disease course. The influence of different TLRs in this mouse model could not be answered so far. For the future, this mouse model can be used for further research on the pathogenesis of SpA. Further studies will allow the development of new therapeutic approaches as well as the evaluation of new clinical decision tools for the optimal use of established treatments in different SpA-patients.