Abstract
Ankylosing spondylitis (AS) is the archetypical disorder of Spondyloarthritis (SpA), that develops in young adults and late adolescence. AS predominantly affects the sacroiliac joint and spine and is often accompanied by extra-articular manifestations. On average, 5-10% of patients with AS develop Inflammatory Bowel Disease (IBD), and many patients exhibit low-level gut inflammation and gut dysbiosis. The human body contains sites characterized by reduced oxygen availability, such as the gut where microbes interact with immune cells, and the joints which are susceptible to mechanical injury and inflammation, as is common in AS disease. This study sought to analyse immune responses in AS patients to gut bacteria in both normoxic and hypoxic conditions, which is more reflective of in vivo conditions.
The hypothesis suggests that under hypoxic conditions, patients with AS may exhibit an abnormal immune response to gut bacteria compared to healthy controls due to altered signalling pathways.
In this study, two bacteria linked to SpA in both humans and animal models of AS were further investigated. These bacteria included a commensal from the Bacteroides fragilis group (isolate CAS10) and Campylobacter jejuni, a common enteric pathogen associated with the development of HLA-B27-associated Reactive arthritis. The study focused on examining dendritic cell (DC) activation markers, cell signalling pathways related to bacterial recognition, autophagy, and T cell effector responses.
AS patients were recruited from the Rheumatology Department of Dunedin Public Hospital, with healthy age and sex-matched controls. The monocyte-derived DC (MoDC) response to heat-killed bacteria in normoxic and hypoxic conditions was assessed by flow cytometric measurement of MoDC activation markers (major histocompatibility [MHC]-I/HLA-A, B, C, MHC-II/human leukocyte antigen [HLA]-DR, cluster differentiation [CD]86, CD80, CD40 and CD83). Cytokine secretion in supernatants was measured by Bioplex (interleukin [IL]-1β, IL-12, IL-10, interferon gamma-induced protein [IP]-10, macrophage inflammatory protein [MIP]-1α and tumour necrosis factor [TNF]). DC activation experiments identified
bimodal responses in AS to bacterial stimulation; MoDCs from AS patients with a high AS disease activity score [ASDAS] (ASDAS≥ 2.1) had muted responses to stimulation with the exception of MHC-I, which was significantly increased to Bacteroides CAS10 only in hypoxia. Secretion Th1 priming cytokine, IL-12, was generally decreased irrespective of treatment. Conversely, MoDCs from AS patients with low disease activity exhibited heightened responsiveness to stimulation with both C. jejuni and Bacteroides CAS10 compared to Healthy Controls in normoxia. However, under hypoxic conditions, this enhanced response was primarily associated with CAS10 stimulation.
Inhibitors to proteins associated with cell signalling pathways in bacterial recognition were used to identify mechanisms of DC activation; specifically, Myeloid differentiation primary response 88 (MyD88), TIR-domain-containing adapter-inducing interferon-β (TRIF) and Protein Kinase R (PKR). Inhibition of cell signalling proteins indicated that activation of AS MoDCs to C. jejuni stimulation was mediated through MyD88 and IFN signalling pathways (PKR and TRIF) in normoxia, however in hypoxia, MyD88 signalling had little role in DC activation where PKR signalling predominates with TRIF signalling. Activation of AS MoDCs by Bacteroides isolate CAS10 was through the TRIF signalling pathway in normoxic and hypoxic conditions. The autophagic response to bacterial treatment of DCs in normoxia and hypoxia was also investigated using a cationic amphiphilic tracer. This indicated that MoDCs from AS patients with high disease activity had significantly increased autophagic vesicles independent of stimulation in both normoxic and hypoxic conditions compared to Healthy Controls and AS patients with low disease activity.
The evaluation of T cell responses was conducted through coculture with bacteria-stimulated MoDCs, followed by intracellular cytokine staining and T cell proliferation analysis using Violet Proliferation dye and flow cytometry. T cell proliferation in AS was markedly reduced compared to Healthy Controls to all stimuli in normoxic conditions; however, hypoxia reduced proliferation in both Healthy Controls and AS diminishing these differences. AS patients with low disease activity produced significantly increased T helper (Th)17 responses to C. jejuni and Bacteroides isolate CAS10 in hypoxic conditions, compared to Healthy Control responses which were restrained in hypoxia. By contrast, MoDCs from AS patients with High Disease activity were unable to polarise T cells, showing negligible Th17 and Th1 responses to any stimulation including C. jejuni and Bacteroides fragilis group isolate CAS10. CD8 T cell responses from both AS patient groups were low compared to Healthy Controls indicating AS MoDCs favoured CD4+ T cell priming.
Together this work indicates a loss of tolerogenic activity to commensal organisms, such as Bacteroides, rather than a disease driven by a pathogenic bacterium. Furthermore, dissecting immune responses in AS based on disease activity may help to fully understand the spectrum of AS disease pathology; where sustained Th17 responses in hypoxic conditions may indicate dysregulation of a pathway intersecting these responses in AS patients; and defective autophagic processes may be associated with the development of a high disease burden or signal immune cell exhaustion in this chronic inflammatory disease.