S Herman

CTLA-4 directly inhibits osteoclast formation

CTLA-4 is a regulator of co-stimulation and inhibits the activation of T cells through interfering with the interaction of CD80/86 on antigen-presenting cells with CD28 on T cells. CTLA-4 binds to the surface of antigen-presenting cells, such as dendritic cells and monocytes through CD80/86. Monocytes can differentiate in osteoclasts, the primary bone resorbing cells. Herein, we investigated whether the binding of CTLA-4 affects the differentiation of monocytes into osteoclasts in vitro and vivo. We show that CTLA-4 dose-dependently inhibits RANKL- as well as tumour necrosis factor (TNF)-mediated osteoclastogenesis in vitro without the presence of T cells. Furthermore, CTLA-4 was effective in inhibiting TNF-induced osteoclast formation in a non-T cell dependent TNF-induced model of arthritis as well as the formation of inflammatory bone erosion in vivo. These data suggest that CTLA-4 is an anti-osteoclastogenic molecule that directly binds osteoclast precursor cells and inhibits their differentiation. These findings are an attractive explanation for the anti-erosive effect of abatacept, a CTLA-4 immunoglobulin fusion protein used for the treatment of rheumatoid arthritis.

Nucleic acid-stimulated antigen-presenting cells trigger T cells to induce disease in a rat transfer model of inflammatory arthritis

Autoimmune responses to heterogeneous nuclear ribonucleproteins (hnRNP) occur in many systemic autoimmune diseases, particularly in patients with rheumatoid arthritis (RA) and systemic lupus erythematosus. In RA, humoral and/or cellular autoimmunity to hnRNP-A2/B1 is the most prominent anti-nuclear reactivity, being detectable in more than 50% of patients. However, its pathogenic role has not been fully elucidated yet. Here, we report that splenocytes from rats with pristane-induced arthritis transfer disease after in vitro restimulation with hnRNP-A/B antigens. Remarkably, disease transfer can be blocked by nuclease treatment of hnRNPs and is also achieved with splenocytes stimulated with hnRNP-A/B associated DNA or RNA oligonucleotides (ON) alone. Induction of proinflammatory cytokines in splenocytes stimulated with hnRNP-A/Bs or ONs involves Toll-like receptors (TLR) 7 and 9 but not TLR3. Furthermore, although T cells are the main mediators of disease transfer they require restimulation with TLR-activated antigen-presenting cells such as macrophages in order to become arthritogenic. Thus, the autoantigenic properties of hnRNPs appear to be mediated by their associated nucleic acids binding to TLR7 and 9. Our data explain the specific selection of hnRNP-A2/B1 as autoantigen in RA and reveal the requirement of interaction between innate and adaptive immunity to initiate and drive inflammation in autoimmune arthritis.

Cell death and cytokine production induced by autoimmunogenic hydrocarbon oils

Hydrocarbon oils such as pristane or hexadecane induce arthritis and lupus in rodents sharing clinical and pathological features with the human diseases rheumatoid arthritis and systemic lupus erythematosus, respectively. In pristane-induced lupus in the mouse induction of apoptosis and augmentation of type-I Interferon signalling by pristane have been suggested to contribute to pathology, whereas in pristane-induced arthritis (PIA) in the rat the pathological mechanisms are still elusive. Here we show that pristane induces cell death in rat and human cells. Increased numbers of apoptotic cells were found in draining lymph nodes of pristane-injected rats and increased percentages of apoptotic and necrotic cells were observed in peripheral blood. In addition, neutrophil extracellular trap formation was triggered by pristane and hexadecane in neutrophils. Because levels of interleukin (IL)-1β were elevated in sera of pristane-injected rats, with levels mirroring the course of PIA, we examined the effect of pristane at single cell level in vitro, using rat splenocytes and the human monocytic cell line THP-1. Pristane and other hydrocarbon oils induced IL-1β secretion in THP-1 cells as well as in rat splenocytes. The potassium channel inhibitor glibenclamide partly inhibited IL-1β induction, suggesting involvement of the inflammasome. Elevated levels of IL-1α were also found in supernatants of cells treated with pristane and hexadecane. In conclusion, autoimmunogenic hydrocarbon oils induce various forms of cell death in rat and human cells. The higher serum IL-1β levels in pristane-injected animals might be caused by both inflammasome-dependent and -independent mechanisms, such as passive release from dying-cells and probably extracellular maturation of pro-IL-1β.

Regulatory T cells form stable and long-lasting cell cluster with myeloid dendritic cells (DC)

Regulatory T cells (Treg) with the capacity to suppress T-cell proliferation exert various effects on T cell function. In addition, Treg have been shown to modulate the phenotype and function of antigen-presenting cells (APC) including dendritic cells (DC), B cells and monocytes/macrophages. However, the specific mechanism(s) of how Treg affect APC have not been entirely identified so far. In this study, we analyzed the interaction of human Treg and effector T cells (Teff) with peripheral blood myeloid and monocyte-derived dendritic cells in vitro. A strong tendency for cell cluster formation between Treg and DC was observed, which was dependent on the adhesion molecules ICAM-1, LFA-3 and ICAM-3. In addition, Treg were found to express higher levels of LFA-1, LFA-2, LFA-3 and ICAM-3 both before and after activation with anti-CD3 antibodies. Using in vitro live cell imaging, we were further able to show that Treg-DC cell clusters, in contrast to Teff-DC clusters, were stable and long lasting. Co-cultures of DC with Treg diminished the up-regulation of activation induced costimulatory molecule expression on DC, and further reduced the production of tumor necrosis factor alpha and stimulated the production of IL-4. In summary, our data indicate that Treg-DC cluster formation might enable Treg to modulate phenotypic and functional characteristics of DC and help to constrain Teff activation.

Inhibition of Inflammation and Bone Erosion by RNA Interference-Mediated Silencing of Heterogeneous Nuclear RNP A2/B1 in Two Experimental Models of Rheumatoid Arthritis.

The nuclear protein heterogeneous nuclear RNP A2/B1 (hnRNP A2/B1) is involved in posttranscriptional regulation of gene expression. It is constitutively expressed in lymphoid organs and highly up‐regulated in the synovial tissue of patients with rheumatoid arthritis (RA), who may also generate autoantibodies to this protein. This study was undertaken to investigate the potential involvement of hnRNP A2/B1 in the pathogenesis of autoimmune arthritis, by silencing hnRNP A2/B1 expression in 2 animal models of RA.

1.58 rheumatoid factor binding is influenced by the N-Glycans of their IGG targets

Background and Objectives In autoimmune conditions, like rheumatoid arthritis (RA), the own immune system attacks the body and causes inflammation. One of the consequences of inflammation during autoimmune diseases is a changed glycosylation profile of IgG antibodies. RA is not only characterised by an altered N-glycan structure bound to the Fc portion of IgG molecules but also by circulating rheumatoid factors (RF) that recognizes the Fc portion of IgG as antigen. The aim of our work was to investigate the glycosylation of immune complexes circulating in the blood of patients with seropositive and seronegative RA. Material and Methods Sera from 70 patients with RA (RF positive n = 43 and RF negative n = 27) and 8 healthy donors (control) have been used in the analysis. (I) Random serum IgG-complexes were captured with F(ab’)2 fragments of anti-human-IgG and analysed by ELISA for their interaction with anti-human-IgG, anti-human-IgM, AAL lectin (fucose-specific) and LCA lectin (manose-specific). (II) The IgG complexes eluted from the ELISA-plates were analysed by Western Blots for the presence of IgG, IgM and AAL as well as LCA binding. (III) IgG affinity purified with AAL-, SNA- and jacalin-agarose columns was tested by ELISA as target for IgM rheumatoid factor. Results (I) The accessibility of fucose residues for the AAL lectin of circulating IgG complexes was much higher in patients with RA when compared to healthy donors. This reactivity was independent of the presence of IgM-RF. The accessibility of mannose residues for the LCA lectin of circulating IgG complexes was much higher in patients with RA when compared to healthy donors. This reactivity was only to be observed in the presence of IgM-RF. (II) Western Blot confirmed an increased presence of IgM in IgG complexes from seropositive patients with RA. The fucosylation of IgG did not differ much between all groups. (III) No difference was detected in IgM-RF affinity towards positive and negative fractions of AAL- and SNA-reactive IgG. However, IgM-RF displayed a reduced binding to the minor IgG fraction isolated with jacalin-affinity chromatography. Conclusion In not-denatured IgG circulating in patients with RA the glycan shows a lower accessibility for the fucose-specific AAL lectin, independent of the presence of IgM-RF. In contrast, the increased binding of the LCA lectin is due to the presence of IgM complexed to IgG. IgM-RF did not discriminate between IgG reactive with AAL and SNA lectin. AAL – Aleuria aurantia lectin (α-1,6 core fucose) LCA – Lens culimaris agglutinin (α-mannose) SNA - Sambucus nigra agglutinin (α-2,6-sialic acid)

A7.15 in vitro silencing of HNRNP-A2/B1 in synovial fibroblasts reveals involvement in regulation of several signal transduction pathways

Background and objectives The heterogeneous nuclear ribonucleoprotein (hnRNP) A2/B1 is involved in post-transcriptional regulation of gene expression. It has been shown to be highly upregulated in synovial tissue of patients with rheumatoid arthritis (RA). In addition, autoantibodies and T cells directed against hnRNP-A2/B1 can be found in RA patients. Recently, it was shown that silencing of hnRNP-A2/B1 in two animal models of RA, namely Collagen-induced arthritis and K/BxN serum transfer arthritis, led to reduction of arthritis severity.1 To further elucidate the role of hnRNP-A2/B1 in RA, we sought of analysing the signalling pathways affected by silencing of hnRNP-A2/B1 in human fibroblast-like synoviocytes (FLS). Materials and methods siRNA-mediated silencing of hnRNP-A2/B1 in FLS was achieved by lipofectamine-based transfection. After three days, successful reduction of hnRNP-A2/B1 expression was analysed by real-time quantitative polymerase chain reaction (RT-qPCR). The role of hnRNP-A2/B1 in FLS was investigated by activating cells with TNF-α. Proteome Profiler Arrays were used to analyse cytokine production and phosphorylation of various signal transduction molecules. Interleukin (IL) -6 and IL-8 secretion was assayed using enzyme-linked immunosorption assay (ELISA). Results Silencing of hnRNP-A2/B1 led to a reduction of phosphorylation of AKT and mammalian target of rapamycin (mTOR) in TNF-α stimulated cells and a slight reduction in phosphorylation of p70 S6 kinase, which is a downstream signalling component of mTOR. Moreover, down-regulation of hnRNP-A2/B1 led to reduced levels of phosphorylated MAPK14 (p38α), and a reduction of MSK2 phosphoryl ation. Analysis of supernatants revealed reduced levels of CCL5 (RANTES), CXCL10 (IP-10), CCL20 (MIP-3α) and Serpine E1, but increased levels of ICAM-1. In addition, reduced secretion of Dickkopf-1 (DKK-1) or IGFBP-3 could be detected in silenced cells. Surprisingly, secretion of IL-6 and IL-8 was slightly increased in hnRNP-A2/B1 silenced FLS. Conclusions hnRNP-A2/B1 seems to play an important role in regulation of several signalling pathways, mainly the mTOR pathway, which is involved in translation and cell growth. Further analyses will be needed to fully understand the role of hnRNP-A2/B1 in signalling pathways operative in FLS and other inflammatory cell types involved in the pathogenesis of RA. Reference Herman S, et al. Inhibition of Inflammation and Bone Erosion by RNA Interference-Mediated Silencing of Heterogeneous Nuclear RNP A2/B1 in Two Experimental Models of Rheumatoid Arthritis. Arthritis Rheumatol 2005;67:2536–46

A6.38 Toll-like receptor 7 and 9 in the pathogenesis of inflammatory autoimmune arthritis

Background and objectives There is evidence that release and insufficient removal of endogenous nucleic acids may be involved in triggering harmful autoimmune reactions important in the initiation of systemic autoimmune diseases including rheumatoid arthritis (RA). Nucleic acid sensing molecules, such as the endosomal Toll-like receptors (TLRs) 7 and 9, have been linked to pathogenetic autoimmune processes, particularly in systemic lupus erythematosus, but their role in RA is less clear. We aimed to study the role of TLR7 and TLR9 in the pathogenesis of inflammatory arthritis by antagonising or stimulating them in rats with pristane-induced arthritis (PIA). Materials and methods Arthritis was induced in Dark Agouti rats with the mineral oil pristane. Antagonists or agonists, respectively, for TLR7 and TLR9, a non-inhibitory control sequence or PBS as placebo were applied every other day. Treatment was started before disease induction. Arthritis was scored using established scoring systems, inflammation and bone erosion were quantified by histological analysis. Serum cytokine levels were measured by ELISA. Results While the control sequence showed no effect on arthritis development and severity, the TLR9 antagonist reduced arthritis severity significantly in PIA. In contrast, a slight aggravation of disease severity was observed in animals treated with the TLR7 antagonist. Inhibition of TLR9 led to strongly reduced bone erosion, whereas it appeared moderately aggravated in animals treated with the TLR7 inhibitor. Furthermore, IL-6 serum levels were reduced in animals treated with the TLR9 antagonist. However, these effects were only seen when the inhibitor was applied before disease onset. When treatment with the antagonists was started at disease-onset neither disease severity nor bone erosion were affected. Treatment with agonists for TLR9 or TLR7 showed no significant effect on disease severity in animals treated with the TLR9 agonist. In contrast, disease was significantly aggravated in animals treated with the TLR7 agonist and this effect was more pronounced than that observed in previous experiments with the TLR7 antagonist. Conclusions Inhibition of TLR9 in rats with PIA significantly reduced inflammation and bone erosion whereas stimulation of TLR7 aggravated disease severity. Therefore, these results suggest different roles for TLR7 and TLR9 in the T cell-dependent initiation phase of PIA and thus an important involvement of the DNA (CpG) recognising TLR9 and the RNA recognising TLR7 in the initiation of autoimmune arthritis which needs to be further elucidated in ongoing and future experiments.

THU0102 The Involvement of Toll-Like Receptor 9 in the Pathogenesis of Erosive Arthritis

Background Toll-like receptors (TLR) such as TLR7 and TLR9 that are activated by nucleic acids are known to be involved in the pathogenesis of systemic lupus erythematosus (SLE) while their role in the initiation and perpetuation of rheumatoid arthritis (RA) is unknown. To address this issue we applied oligonucleotides specifically antagonizing one or both TLRs in two different disease models: (i) pristane-induced arthritis (PIA) in DA rats which is driven by autoimmune processes induced by subcutaneous injection of the mineral oil pristane and (ii) the KRN serum transfer model in C57Bl/6 mice reflecting the late effector phase of erosive arthritis. In addition, involvement of TLR9 was also investigated in TLR9 knock-out mice. Objectives To investigate the role of TLR9 in the pathogenesis of autoimmune arthritis both in the early and in the effector disease phase. Methods Arthritis was induced in rats with the mineral oil pristane, in C57Bl/6 and TLR9 -/- mice by injection of KRN serum. Immunoregulatory oligodeoxynucleotide (ODN) sequences (IRS) antagonizing TLR7 or TLR9 were applied either subcutaneously (PIA) or intra-peritoneally (KRN). A non-inhibitory ODN was used as control and PBS served as placebo. Arthritis was scored using established scoring systems, inflammation and bone erosion were quantified by histology. Serum and cell culture cytokine levels were measured by ELISA. Results In the PIA model the TLR7 inhibitor and a TLR7/9 dual inhibitor showed no effect on arthritis development and severity. In contrast, arthritis severity was significantly reduced (p<0.05) by the TLR9 antagonist and bone erosion was almost completely abolished (p<0.01). Furthermore, IL-6 serum levels were significantly lower than in mice treated with the TLR7 antagonist or a control oligonucleotide. However, these effects were only seen when the inhibitor was applied before disease onset. In line with this observation, neither inhibitor affected arthritis onset and severity in the serum transfer model which is independent of the adaptive immune system and arthritis appeared to be even aggravated and osteoclastogenesis enhanced in mice lacking a functional TLR9 gene. Conclusions TLR9 inhibition significantly reduced inflammation and bone erosion in PIA, but not in the KRN serum transfer model. These results suggest an important involvement of the DNA (CpG) recognizing TLR9 in the initiation of autoimmune arthritis whereas in the later disease phases TLR9 may even play a beneficial role similar to what has been found in models of SLE and other models for autoimmune diseases such as diabetes. Antagonizing TLR9 in human RA may therefore show beneficial effects only in the very earliest phase of the disease while its activation may have therapeutic effects in established disease. Disclosure of Interest None Declared

A2.11 Involvement of the Nucleic Acid Recognising Toll-Like Receptors TLR7 and TLR9 in the Pathogenesis of Erosive Arthritis

Background There is substantial evidence that abundant release and/or insufficient removal of endogenous nucleic acids can trigger autoimmune reactions via activation of nucleic acid recognising Toll-like receptors (TLR) such as TLR7 and TLR9, which may lead to the development of SLE and other systemic autoimmune diseases. However, in RA the involvement of these TLRs is less well understood. Interestingly, in rats with pristane-induced arthritis (PIA) disease can be transferred by T cells together with antigen-presenting-cells pre-activated with TLR7 or TLR9 agonists (Hoffmann et al, J Autoimmunity 2011; 36: 288–300). Objective We aimed to study the role of TLR7 and TLR9 in the pathogenesis of inflammatory erosive arthritis by antagonising them in PIA and the KRN serum transfer model. Methods Arthritis was induced in rats with the mineral oil pristane, and in C57Bl/6 mice by injection of KRN serum. Immunoregulatory oligodeoxynucleotide (ODN) sequences (IRS) antagonising TLR7 or TLR9 were applied either subcutaneously (PIA) or intra-peritoneally (KRN). A non-inhibitory ODN was used as control and PBS served as placebo. Arthritis was scored using established scoring systems, inflammation and bone erosion were quantitatively analysed by histology. Serum and cell culture cytokine levels were measured by ELISA. Results While the TLR7 inhibitor and the control ODN showed no effect on arthritis development and severity, the TLR9 antagonist reduced arthritis severity significantly in PIA. Bone erosion was almost completely abolished, whereas it was moderately aggravated in animals treated with the TLR7 inhibitor. Furthermore, IL-6 serum levels were significantly reduced in animals treated with the TLR9 antagonist. However, these beneficial effects were only observed when the inhibitor was applied before disease onset. Moreover, neither inhibitor affected arthritis onset and severity in the serum transfer model, which is independent of the adaptive immune system. Summary and Conclusions Inhibition of TLR9 significantly reduced inflammation and bone erosion in PIA but not in the KRN serum transfer model that reflects the late effector phase of erosive arthritis. Therefore, these results suggest important involvement of the DNA (CpG) recognising TLR9 in the initiation of autoimmune arthritis whereas nucleic acid binding TLRs do not seem to play a major role in the later phases of the disease. Antagonizing TLR9 in human RA may only act beneficial in the earliest phase of the disease.