Melissa N van Tok

Disease-specific and inflammation-independent stromal alterations in spondylarthritis synovitis

OBJECTIVE The molecular processes driving the distinct patterns of synovial inflammation and tissue remodeling in spondylarthritis (SpA) as compared to rheumatoid arthritis (RA) remain largely unknown. Therefore, we aimed to identify novel and unsuspected disease-specific pathways in SpA by a systematic and unbiased synovial gene expression analysis. METHODS Differentially expressed genes were identified by pan-genomic microarray and confirmed by quantitative polymerase chain reaction and immunohistochemical analyses of synovial tissue biopsy samples from patients with SpA (n=63), RA (n=28), and gout (n=9). The effect of inflammation on gene expression was assessed by stimulating fibroblast-like synoviocytes (FLS) with synovial fluid and by analysis of synovial tissue samples at weeks 0 and 12 of etanercept treatment. RESULTS Using very stringent statistical thresholds, microarray analysis identified 64 up-regulated transcripts in patients with SpA synovitis as compared to those with RA synovitis. Pathway analysis revealed a robust myogene signature in this gene set. The myogene signature was technically and biologically reproducible, was specific for SpA, and was independent of disease duration, treatment, and SpA subtype (nonpsoriatic versus psoriatic). Synovial tissue staining identified the myogene expressing cells as vimentin-positive, prolyl 4-hydroxylase β-positive, CD90+, and CD146+ mesenchymal cells that were significantly overrepresented in the intimal lining layer and synovial sublining of inflamed SpA synovium. Neither in vitro exposure to synovial fluid from inflamed SpA joints nor in vivo blockade of tumor necrosis factor modulated the SpA-specific myogene signature. CONCLUSION These data identify a novel and disease-specific myogene signature in SpA synovitis. The fact that this stromal alteration appeared not to be downstream of local inflammation warrants further analysis of its functional role in the pathogenesis of the disease.

Interleukin‐9 Overexpression and Th9 Polarization Characterize the Inflamed Gut, the Synovial Tissue, and the Peripheral Blood of Patients With Psoriatic Arthritis

To investigate the expression and tissue distribution of Th9‐related cytokines in patients with psoriatic arthritis (PsA).

Non-conventional forms of HLA-B27 are expressed in spondyloarthritis joints and gut tissue.

Objectives Human leukocyte antigen (HLA)-B27 (B27) is the strongest genetic factor associated with development of Ankylosing Spondylitis and other spondyloarthropathies (SpA), yet the role it plays in disease pathogenesis remains unclear. We investigated the expression of potentially pathogenic non-conventional heavy chain forms (NC) of B27 in synovial and intestinal tissues obtained from SpA patients. We also determined the presence of NC-B27 in joints, lymphoid and gastrointestinal tissue from B27 transgenic (TG1) rats with M.tuberculosis-induced SpA. Methods Expression of NC-B27 in human SpA joints and gut and in (21-3 × 283-2)F1 HLA-B27/Huβ2m rat tissue was determined by immunohistochemistry, flow cytometry and confocal microscopy analysis using HC10 and HD6 antibodies. Results Both HC10- and HD6-reactive HLA molecules were present in synovial tissue from SpA patients. Both NC-B27 and KIR3DL2, a ligand for NC-B27, were expressed in inflamed terminal ileal tissues in patients with early SpA. Infiltrating cells in inflamed joint tissues isolated from B27 TG1 rats expressed high levels of NC-B27. NC-B27 were also expressed in joint-resident cells from ankle and tail joints of B27 TG1 rats prior to clinical arthritis. The expression of NC-B27 on B27 TG1 rat CD11b/c+, CD8α+, cells from spleens and LNs increased with animal age and disease progression. Conclusions Non-conventional HLA class 1 molecules are expressed on resident and infiltrating cells in both synovial and GI tissues in human SpA. NC-B27 expression in joints and lymphoid tissues from B27 TG1 rats prior to the onset of arthritis is consistent with the hypothesis that they play a pathogenic role in SpA.

A2.15 Relative Overexpression of Transmembrane Versus Soluble TNF in Human and Experimental Spondyloarthritis

Background Macrophages and their pro-inflammatory cytokines, including TNF, are pivotal mediators of chronic synovitis in rheumatoid arthritis (RA) as well as spondyloarthritis (SpA). Despite similar levels of synovial macrophage infiltration and similar clinical responses to TNF blockade in both diseases, SpA is characterised by a more pronounced infiltration with alternatively activated CD163+ macrophages and ongoing osteoproliferation. This study aimed to investigate whether these differences were related to a differential expression and/or function of TNF between both diseases. Methods Expression of transmembrane TNF (tmTNF) and soluble TNF (sTNF) was measured in IFN-γ, IL-4 or IL-10 polarised macrophages obtained from healthy donors. Expression of TNF and its receptors was measured in synovial fluid (SF) and synovial tissue biopsies (ST) of actively inflamed knee joints of SpA and RA patients. Mice transgenically overexpressing tmTNF (TgA86) were evaluated for spondylitis and arthritis. Results In vitro polarisation with IL-10 specifically induced the expression of CD163 on macrophages, mimicking the phenotype in SpA synovitis. IL-10 and IL-4 polarised macrophages secreted less TNF than classically, IFN-γ-polarised macrophages (p < 0.05). In contrast, IL-4 polarised macrophages expressed more tmTNF compared to the IFN-γ polarised macrophages (p < 0.05), with a similar trend for IL-10 polarised macrophages. In line with these in vitro data, the sTNF SF levels were significantly lower in SpA compared to RA (p = 0.01) despite similar TNF mRNA levels in ST. This was not related to higher expression of TNF receptors as both TNFR1 and TNFR2 were similarly expressed in ST, both at protein and mRNA levels. On the contrary, the SF levels of sTNFR1 and sTNFR2, which are both cleaved from the cell membrane by the same enzyme as tmTNF, were even decreased in SpA versus RA. Investigating the potential pathophysiological role of relative overexpression of tmTNF versus sTNF in vivo, clinical analysis revealed that tmTNF transgenic mice developed arthritis, resulting in deformation and loss of grip strength, and spondylitis as evidenced by crinkled tails with a 100% incidence. Axial and peripheral joint inflammation was confirmed by histology. In contrast to mouse strains overexpressing sTNF, tmTNF tg mice did not develop systemic disease and weight loss and showed clear signs of osteoproliferation on histology. Conclusions tmTNF is relatively overexpressed by CD163+ alternatively polarised macrophages in SpA synovitis and leads to an axial and peripheral SpA phenotype in transgenic mice.

Innate Immune Activation Can Trigger Experimental Spondyloarthritis in HLA-B27/Huβ2m Transgenic Rats

Spondyloarthritis (SpA) does not display the typical features of auto-immune disease. Despite the strong association with MHC class I, CD8+ T cells are not required for disease induction in the HLA-B27/Huβ2m transgenic rats. We used Lewis HLA-B27/Huβ2m transgenic rats [21-3 × 283-2]F1, HLA-B7/Huβ2m transgenic rats [120-4 × 283-2]F1, and wild-type rats to test our hypothesis that SpA may be primarily driven by the innate immune response. In vitro, splenocytes were stimulated with heat-inactivated Mycobacterium tuberculosis and cytokine expression and production was measured. In vivo, male and female rats were immunized with 30, 60, or 90 µg of heat-inactivated M. tuberculosis and clinically monitored for spondylitis and arthritis development. After validation of the model, we tested whether prophylactic and therapeutic TNF targeting affected spondylitis and arthritis. In vitro stimulation with heat-inactivated M. tuberculosis strongly induced gene expression of pro-inflammatory cytokines such as TNF, IL-6, IL-1α, and IL-1β, in the HLA-B27 transgenic rats compared with controls. In vivo immunization induced an increased spondylitis and arthritis incidence and an accelerated and synchronized onset of spondylitis and arthritis in HLA-B27 transgenic males and females. Moreover, immunization overcame the protective effect of orchiectomy. Prophylactic TNF targeting resulted in delayed spondylitis and arthritis development and reduced arthritis severity, whereas therapeutic TNF blockade did not affect spondylitis and arthritis severity. Collectively, these data indicate that innate immune activation plays a role in the initiation of HLA-B27-associated disease and allowed to establish a useful in vivo model to study the cellular and molecular mechanisms of disease initiation and progression.

Insulin-Like Growth Factor I Does Not Drive New Bone Formation in Experimental Arthritis

Introduction Insulin like growth factor (IGF)-I can act on a variety of cells involved in cartilage and bone repair, yet IGF-I has not been studied extensively in the context of inflammatory arthritis. The objective of this study was to investigate whether IGF-I overexpression in the osteoblast lineage could lead to increased reparative or pathological bone formation in rheumatoid arthritis and/or spondyloarthritis respectively. Methods Mice overexpressing IGF-I in the osteoblast lineage (Ob-IGF-I+/-) line 324–7 were studied during collagen induced arthritis and in the DBA/1 aging model for ankylosing enthesitis. Mice were scored clinically and peripheral joints were analysed histologically for the presence of hypertrophic chondrocytes and osteocalcin positive osteoblasts. Results 90–100% of the mice developed CIA with no differences between the Ob-IGF-I+/- and non-transgenic littermates. Histological analysis revealed similar levels of hypertrophic chondrocytes and osteocalcin positive osteoblasts in the ankle joints. In the DBA/1 aging model for ankylosing enthesitis 60% of the mice in both groups had a clinical score 1<. Severity was similar between both groups. Histological analysis revealed the presence of hypertrophic chondrocytes and osteocalcin positive osteoblasts in the toes in equal levels. Conclusion Overexpression of IGF-I in the osteoblast lineage does not contribute to an increase in repair of erosions or syndesmophyte formation in mouse models for destructive and remodeling arthritis.

The role of Bob1 in rheumatoid arthritis: potential implications for autoimmunity

Background Rheumatoid arthritis (RA) is a prototypic autoimmune disease characterized by a prominent humoral autoimmunity. Of particular relevance is the local production of autoantibodies such as rheumatoid factor and anti-citrullinated protein antibodies in the inflamed synovial tissue. The mechanisms underlying break of B cell tolerance and local autoantibody production remains poorly understood. Objectives To identify cellular and molecular pathways implicated in RA-specific humoral autoimmunity. Methods Synovial tissue samples were obtained by arthroscopy from untreated individuals with RA (n=33) and inflammation matched SpA controls (n=58). Gene expression profiling was performed on tissue samples of patients with established arthritis using 44K Whole Genome Human microarrays (Agilent). Top differentially expressed genes were validated on three independent cohorts by Taqman based RT-qPCR and immunohistochemistry. Collagen-induced arthritis (CIA) and Experimental autoimmune encephalomyelitis (EAE) experiments were conducted using Bob1 knockout mice and their littermate controls. Results Microarray screening for genes differentially expressed in the inflamed synovium, the key target of the disease process in RA, revealed a prominent and disease-specific B cell/plasma cell signature with the B cell-specific transcriptional co-activator Bob1 and its transcriptional target BCMA among the most upregulated genes. Validation by RT-qPCR on two independent cohorts representing early and established arthritis confirmed microarray data and demonstrated elevated expression of Bob1 and BCMA not only in established RA, but also at the early phase of the disease. Quantitative evaluation of immunohistochemical stainings of synovial tissue with monoclonal antibody for Bob1 revealed significant increase in Bob1 positive cells in RA synovium (p<0.01). Next we determined whether lack of functional Bob1 modifies disease onset or severity in CIA. Interestingly, the results showed that Bob1-/- mice were fully resistant to CIA induction compared to their wild-type littermates. This remarkable protection from CIA is explained by failure to produce pathogenic anti-collagen autoantibodies in the absence of Bob1. In contrast, Bob1-/- mice were susceptible to MOG protein induced EAE and incidence and severity of clinical disease were not altered in these mice comparing to wild-type littermates, suggesting that absence of Bob1 does not impact on antigen-presentation/costimulatory capacity of B cells. Conclusions The specific increase in Bob1 expressing cells in RA synovitis and the resistance of Bob1-defecient mice to development of CIA indicate that Bob1/BCMA axis may contribute to humoral autoimmunity in RA. The relationship between an aberrant Bob1 expression and the break of peripheral tolerance in RA is currently under investigation. Disclosure of Interest None Declared

The Initiation, but Not the Persistence, of Experimental Spondyloarthritis Is Dependent on Interleukin-23 Signaling

IL-17A is a central driver of spondyloarthritis (SpA), its production was originally proposed to be IL-23 dependent. Emerging preclinical and clinical evidence suggests, however, that IL-17A and IL-23 have a partially overlapping but distinct biology. We aimed to assess the extent to which IL-17A-driven pathology is IL-23 dependent in experimental SpA. Experimental SpA was induced in HLA-B27/Huβ2m transgenic rats, followed by prophylactic or therapeutic treatment with an anti-IL23R antibody or vehicle control. Spondylitis and arthritis were scored clinically and hind limb swelling was measured. Draining lymph node cytokine expression levels were analyzed directly ex vivo, and IL-17A protein was measured upon restimulation with PMA/ionomycin. Prophylactic treatment with anti-IL23R completely protected against the development of both spondylitis and arthritis, while vehicle-treated controls did develop spondylitis and arthritis. In a therapeutic study, anti-IL23R treatment failed to reduce the incidence or decrease the severity of experimental SpA. Mechanistically, expression of downstream effector cytokines, including IL-17A and IL-22, was significantly suppressed in anti-IL23R versus vehicle-treated rats in the prophylactic experiments. Accordingly, the production of IL-17A upon restimulation was reduced. In contrast, there was no difference in IL-17A and IL-22 expression after therapeutic anti-IL23R treatment. Targeting the IL-23 axis during the initiation phase of experimental SpA—but not in established disease—inhibits IL-17A expression and suppresses disease, suggesting the existence of IL-23-independent IL-17A production. Whether IL-17A can be produced independent of IL-23 in human SpA remains to be established.

The Transcriptional Coactivator Bob1 Is Associated With Pathologic B Cell Responses in Autoimmune Tissue Inflammation

The molecular mechanisms steering abnormal B cell responses in autoimmune diseases remain poorly understood. We undertook this study to identify molecular switches controlling pathologic B cell responses in rheumatoid arthritis (RA).

08.22 The role of transmembrane rather than soluble tnf in spondyloarthritis

Background TNF plays a key role in immune-mediated inflammatory diseases including rheumatoid arthritis (RA) and spondyloarthritis (SpA). Here we aimed to investigate how TNF can lead to completely different disease phenotypes such as destructive peripheral polysynovitis in RA versus axial and peripheral remodelling arthritis in SpA. Materials and methods We assessed expression of TNF, TNF-R, and TACE (ADAM17) in synovial fluid, synovial tissue, and synovial fibroblasts from SpA and RA. tmTNFtg mice (TgA86)1 were clinically scored for development of peripheral and axial disease, and sacrificed at the end of the experiments for radiologic and histologic assessment. Mechanistic studies included bone marrow chimaera experiments and crossing with TNF-R1 or TNF-R2 knock-out animals. Results Arthritis was characterised by lower levels of sTNF and higher levels of tmTNF in SpA versus RA. This misbalance was related to decreased TACE activity in SpA versus RA FLS, as further confirmed by lower levels of other soluble molecules cleaved by TACE (including sTNF-RI, sTNF-RII, and sCD163) in the inflamed SpA joint. Assessing whether tmTNF has a functional role in SpA pathology, mice selectively over-expressing the transmembrane form of TNF spontaneously developed a deforming arthritis and spondylitis, starting at 4 weeks of age and reaching a 100% incidence. Histology revealed peripheral and axial synovitis, enthesitis, and osteitis, as well as inflammation of the connective tissue located at the junction of the annulus fibrosus with the vertebral bone. tmTNFtg mice did not develop extra-articular inflammation. Structural phenotyping by histology and radiology revealed mild destructive features in combination with foci of hypertrophic chondrocytes and axial and peripheral new bone formation leading to bridging of tail vertebra over time. Mechanistic experiments revealed that this SpA-like phenotype was mediated by tmTNF expression on stromal cells but not on hematopoietic cells and required the expression of TNF-R1. Conclusions Collectively, these data suggest that tmTNF expressed by stromal cells is responsible for the key pathological features of SpA. References 1.Alexopoulou L, et al. Eur J Immunol1997;27(10):2588–92.