Background: During trained immunity, monocytes and macrophages undergo a functional and transcriptional reprogramming toward activation, which is induced by a priming stimulus and results in enhanced responsiveness to subsequent triggers. Monocytes from patients with rheumatoid arthritis (RA) display features consistent with a trained immunity phenotype. Citrullinated proteins as citrullinated vimentin (c-vimentin), which function as damage-associated patterns in RA, may be implicated in the process of trained immunity. Objectives: We aimed to investigate if c-vimentin induces trained immunity in vitro in healthy individuals. Methods: Monocytes were isolated from the peripheral blood (EDTA blood, n=22; buffy coats, n=6) from healthy donors by Ficoll-paque centrifugation and negative selection using CD3/CD19/CD56 magnetic beads. The cells were stimulated with c-vimentin (0.1 μg/ml) for 24h and re-stimulated 5 days later with the lipopolysaccharide of E.coli (LPS) (10 ng/ml). Protein as well as lactate release were estimated in cell culture supernatants at day 6 by ELISA. RT-PCR and/or Western Blotting were applied to measure mRNA and/or protein expression. The Ligand-receptor glycocapture technology LRC-TRiCEPS was used to identify candidate cell surface targets of c-vimentin. The methylation of histone H3 at lysine 4 (H3K4) was examined by chromatin immunoprecipitation. Results: Priming with citrullinated vimentin induced training in human monocytes, as suggested by the significantly increased levels of secreted interleukin-6 (IL-6), upon restimulation with LPS (1.29-fold increase, n=22, p<0.001). Likewise, the release of chemokines CXCL1 and CCL20/Macrophage Inflammatory Protein 3a was significantly increased (1.81-fold and 2.32-fold increase, respectively, n=14, both p<0.001). LRC-TRiCEPS enabled the identification of STING cell surface receptor for the ligand c-vimentin. Indeed, c-vimentin induced activation of TBK1, which is implicated in the STING signaling pathway, by phosphorylation, while STING inhibition with the covalent small molecule H151 (2μM) abolished this effect. Besides, H151 inhibited trained immunity by decreasing IL-6 release and expression (1.61-fold and 1.93-fold decrease, respectively, n=5). Trained monocytes also displayed high lactate production (primed vs. unprimed cells, n=9, p=0.004), reflecting a shift in metabolism with an increase in glycolysis. By inhibiting the metabolic pathway of glycolysis by 2-deoxyglucose (11mM), the induction of trained immunity could be counteracted (5.32-fold decrease in IL-6 release, n=7, p=0.016). Finally, c-vimentin induced H3K4 methylation with increased levels of this mark in the promoter of the IL-6 gene. By modulating the function of epigenetic enzymes with methylthioadenosine (1mM), which specifically inhibits histone methyltransferase, trained immunity was reversed (8.43-fold decrease in IL-6 release, n=6, p=0.031). Conclusion: Citrullinated vimentin induces epigenetic modifications and metabolic changes in monocytes, probably through a STING and TBK1-dependent activation, resulting in enhanced cytokine and chemokine production upon restimulation. Inhibition of the STING signaling pathway may be a novel therapeutic target for myeloid activation in RA. Disclosure of Interests: None declared

Fibroblasts are key effector cells in tissue remodeling. They remain persistently activated in fibrotic diseases, resulting in progressive deposition of extracellular matrix. Although fibroblast activation may be initiated by external factors, prolonged activation can induce an "autonomous," self-maintaining profibrotic phenotype in fibroblasts. Accumulating evidence suggests that epigenetic alterations play a central role in establishing this persistently activated pathologic phenotype of fibroblasts. We demonstrated that in fibrotic skin of patients with systemic sclerosis (SSc), a prototypical idiopathic fibrotic disease, TGF-beta induced the expression of DNA methyltransferase 3A (DNMT3A) and DNMT1 in fibroblasts in a SMAD-dependent manner to silence the expression of suppressor of cytokine signaling 3 (SOCS3) by promoter hypermethylation. Downregulation of SOCS3 facilitated activation of STAT3 to promote fibroblast-to-myofibroblast transition, collagen release, and fibrosis in vitro and in vivo. Reestablishment of the epigenetic control of STAT3 signaling by genetic or pharmacological inactivation of DNMT3A reversed the activated phenotype of SSc fibroblasts in tissue culture, inhibited TGF-beta-dependent fibroblast activation, and ameliorated experimental fibrosis in murine models. These findings identify a pathway of epigenetic imprinting of fibroblasts in fibrotic disease with translational implications for the development of targeted therapies in fibrotic diseases.

Background. Synovial hyperplasia is a hallmark of rheumatoid arthritis (RA). This might be associated with an imbalance of growth-promoting and apoptotic pathways, among them fibroblast growth factors (FGFs) and their receptors (FGFRs) . The aim was to investigate differences in FGFRs and to explore the factors that might explain the differences between RA and osteoarthritis (OA) synovial fibroblasts. Methods. To assess FGFRs expression, immunohistochemistry, flow cytometry and RT-qPCR were performed. The cells were treated with TNFa, FGF2, a demethylation agent, PKA-mimics or inhibitors, hypoxia-mimics or proteasome inhibitors. Proliferation was measured using the CCK8 assay and 20S proteasome activity by a fluorescent assay. Results. In RA, FGFR3 protein was decreased in the synovial lining layer (p<0.005) and in cultured RA synovial fibroblasts (RASF) (n=10, p < 0.01). Transcription was unchanged and DNA demethylation decrease its expression. Exposure to TNFa or FGF2 had no effect on FGFR3. PKA-modulation and hypoxia-mimics induced transient changes only. Most interesting, in RASF, the proteasome inhibitor MG-132 restored FGFR3 expression (p<0.001) to levels measured in normal or OA synovial fibroblasts. MG-132 abolished the enhanced proliferative response of RASF to FGF2 (p<0.005). Increased 20S proteasome activity correlated (r=- 0.63, p<0.05) with decreased expression of FGFR3. Conclusions. The expression of FGFR3 is reduced in RA partially due to an increased degradation in proteasomes. This leads to an imbalance in the FGF-related signal pathways and may contribute to synovial hyperplasia. Proteasome inhibitors could represent a novel therapeutic strategy in RA, particularly to prevent synovial hyperplasia.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Background: Tissue fibrosis caused by a pathological activation of fibroblasts is a major hallmark of systemic sclerosis (SSc). Epigenetic gene silencing of anti-fibrotic genes is thought to play a central role to establish the persistently activated phenotype of fibroblasts independent of external stimuli such as TGFβ, which has been identified as key-mediator of fibroblast activation. Objectives: The aims of the present study were to investigate whether the aberrant activation of JAK2-STAT3 signaling in fibrosis might be caused by epigenetic silencing of SOCS expression and whether re-establishment of the endogenous, SOCS-dependent control of JAK / STAT signaling may prevent aberrant fibroblast activation and ameliorate tissue fibrosis. Methods: The methylation status of SOCS3 in fibroblasts was evaluated by methylation-specific PCR and MeDIP assays. 5-aza-2-deoxycytidine (5-aza) and siRNA was used to inhibit DNA methyltransferases (DNMTs) in vitro and in vivo . Knockdown and overexpression experiments served to analyze the mechanism of action in cultured fibroblasts. Fibroblast-specific knockout mice were additionally used to analyze the role of SOCS3 and DNMTs in vivo . Results: Chronically increased levels of TGFβ reduced the expression of SOCS3 in normal fibroblasts to a level also found in SSc fibroblasts. Consistently, the expression of SOCS3 was severely downregulated in skin of SSc patients compared to healthy individuals with only minor differences between limited and diffuse cutaneous SSc. Methylation analyses demonstrated a prominent promoter hypermethylation of SOCS3 in SSc fibroblasts and in normal fibroblasts exposed to persistently high levels of TGFβ. Increased DNMT activity and a time-dependent induction of DNMT3A and DNMT1 expression upon chronic exposure to TGFβ resulted in promoter hypermethylation of SOCS3. Knockdown of SOCS3 induced an SSc-like phenotype in normal dermal fibroblasts with increased activation of JAK2-STAT3 signaling, enhanced expression of myofibroblast markers, increased collagen release, and aggravated experimental tissue fibrosis with increased activation of JAK2-STAT3 signaling. This effect was mimicked by overexpression of mutant JAK2 with mutations in the SOCS3 binding motif. Vice versa, forced overexpression of SOCS3 reduced TGFβ-mediated fibroblast activation and ameliorated the endogenous activation of SSc fibroblasts. Pharmacological inhibition or selective knockdown of DNMTs restored the normal expression of SOCS3, reduced fibroblast activation and collagen release, blocked STAT3-responsive transcription, and exerted potent antifibrotic effects in bleomycin- and TBRI act -induced dermal fibrosis. In addition, treatment with 5-aza or knockdown of either DNMT1 or DNMT3A induced regression of established fibrosis. Conclusion: These findings identify a novel pathway of epigenetic imprinting of fibroblasts in fibrotic disease with translational implications for the development of new targeted therapies in fibrotic diseases. We demonstrate that the chronic activation of TGFβ signaling in fibrotic diseases perturbs the epigenetic control of STAT signaling by DNMT-induced silencing of SOCS3 expression. Our data might thus strengthen the scientific rational for targeting DNA methylation in fibrotic diseases. Disclosure of Interests: Clara Dees: None declared, Sebastian Poetter: None declared, Yun Zhang: None declared, Christina Bergmann: None declared, xiang zhou: None declared, Markus Luber: None declared, Emmanuel Karouzakis: None declared, Andreas Ramming Grant/research support from: Pfizer, Novartis, Consultant of: Boehringer Ingelheim, Novartis, Gilead, Pfizer, Speakers bureau: Boehringer Ingelheim, Roche, Janssen, Oliver Distler Grant/research support from: Grants/Research support from Actelion, Bayer, Boehringer Ingelheim, Competitive Drug Development International Ltd. and Mitsubishi Tanabe; he also holds the issued Patent on mir-29 for the treatment of systemic sclerosis (US8247389, EP2331143)., Consultant of: Consultancy fees from Actelion, Acceleron Pharma, AnaMar, Bayer, Baecon Discovery, Blade Therapeutics, Boehringer, CSL Behring, Catenion, ChemomAb, Curzion Pharmaceuticals, Ergonex, Galapagos NV, GSK, Glenmark Pharmaceuticals, Inventiva, Italfarmaco, iQvia, medac, Medscape, Mitsubishi Tanabe Pharma, MSD, Roche, Sanofi and UCB, Speakers bureau: Speaker fees from Actelion, Bayer, Boehringer Ingelheim, Medscape, Pfizer and Roche, Georg Schett Speakers bureau: AbbVie, BMS, Celgene, Janssen, Eli Lilly, Novartis, Roche and UCB, Jörg Distler Grant/research support from: Boehringer Ingelheim, Consultant of: Boehringer Ingelheim, Paid instructor for: Boehringer Ingelheim, Speakers bureau: Boehringer Ingelheim

Rheumatoid arthritis (RA) is a progressive, destructive autoimmune arthritis. Break of tolerance and formation of autoantibodies occur years before arthritis. Adaptive immunity is initiated in lymphoid tissue where lymph node stromal cells (LNSCs) play a crucial role in shaping the immune response and maintaining peripheral tolerance. Here we performed the first epigenomic characterization of LNSCs during health and early RA, by analyzing their transcriptome and DNA methylome in LNSCs isolated from lymph node needle biopsies obtained from healthy controls (HC), autoantibody positive RA-risk individuals and patients with established RA. Of interest, LNSCs from RA-risk individuals and RA patients revealed a common significantly differential expressed gene signature compared with HC LNSCs. Pathway analysis of this common signature showed, among others, significant enrichment of pathways affecting the extracellular matrix (ECM), cholesterol biosynthesis and immune system. In a gel contraction assay LNSCs from RA-risk individuals and RA patients showed impaired collagen contraction compared to healthy LNSCs. In RA LNSCs a significant enrichment was observed for genes involved in cytokine signaling, hemostasis and packaging of telomere ends. In contrast, in RA-risk LNSCs pathways in cancer (cell cycle related genes) were differentially expressed compared with HC, which could be validated in vitro using a proliferation assay, which indicated a slower proliferation rate. DNA methylation analyses revealed common and specific differentially methylated CpG sites (DMS) in LNSC from RA patients and RA-risk individuals compared with HC. Intriguingly, shared DMS were all associated with antigen processing and presentation. This data point toward alterations in cytoskeleton and antigen-processing and presentation in LNSC from RA-risk individuals and RA patients. Further studies are required to investigate the consequence of this LNSC abnormality on LNSC-mediated immunomodulation.

Background Persistent activation of fibroblast with excessive release of extracellular matrix is a hallmark of systemic sclerosis (SSc). Fibroblasts can either acquire a “pro-fibrotic” phenotype with excessive matrix production or a “pro-inflammatory” phenotype with releasing of matrix-degrading enzymes and subsequent tissue destruction. Despite these well-characterized phenotypic differences, the molecular mechanisms that drive polarization of fibroblasts into these two functionally opposing phenotypes remain enigmatic. Objectives We aimed to evaluate the transcriptional network that promotes the extracellular matrix-producing fibrotic fibroblast fate. Methods We investigated the transcriptional network that induces the profibrotic phenotype of fibroblasts by in silico and immunofluorescence analyses of human fibrotic skin, lung, liver and kidney, and performed functional assays to address the fibrogenic potential of fibroblasts in vitro and in several mouse models of systemic sclerosis. The heterocyclic diamidine DB1976 was used as new therapeutic compound to induce regression of fibrosis. Results We identified the ETS transcription factor PU.1 as molecular checkpoint for acquisition of a “pro-fibrotic” phenotype of fibroblasts. Our data demonstrate that expression of PU.1 is effectively silenced in fibroblasts during tissue homeostasis. When the epigenetic control of PU.1 is lost and PU.1 expression is induced, fibroblasts differentiate into a fibrotic phenotype that includes the transcription of numerous pro-fibrotic mediators. PU.1 polarized resting fibroblasts and even repolarized extracellular matrix-degrading inflammatory fibroblasts to an extracellular matrix-producing fibrotic phenotype. PU.1 is associated with a network of pro-fibrotic factors including members of the TEAD–HIPPO, canonical TGF-β–SMAD and AP1 signaling pathways. Other transcription factors with fibrotic abilities, such as SNAI2 and myocyte enhancer factor (MEF) 2, bind in close vicinity to PU.1-binding sites within the genome and contribute to the recruitment of the transcription machinery that drives the switch towards the fibrotic phenotype. PU.1 has a major coordinating role within this complex network of transcription factors in fibroblasts, as the inactivation of PU.1 alone is sufficient to prevent fibrotic polarization in vitro and in vivo. Finally, we investigated pharmacological targeting of PU.1 as a potential strategy to prevent uncontrolled fibrotic tissue remodelling. DB1976 showed anti-fibrotic effects in vivo in various fibrosis models and across several organs. Treatment with DB1976 not only prevented bleomycin-mediated skin fibrosis, but also induced regression of pre-established fibrosis, and was well tolerated. Conclusion These findings suggest that PU.1 inhibition may represent a novel and effective therapeutic approach to treat a wide range of fibrotic diseases. Inactivation of PU.1 effectively reverted the fibrotic phenotype of fibroblasts to a resting state and induced the regression of tissue fibrosis: Disclosure of Interests: Thomas Wohlfahrt: None declared, Simon Rauber: None declared, Markus Luber: None declared, Alina Soare: None declared, Stefanie Weber: None declared, Alexandru-Emil Matei: None declared, Chih-Wei Chen: None declared, Emmanuel Karouzakis: None declared, Hans Kiener: None declared, Elena Pachera: None declared, Clara Dees: None declared, Alexander Kreuter: None declared, Astrid Juengel: None declared, Steffen Gay: None declared, Oliver Distler Grant/research support from: Prof. Distler received research funding from Actelion, Bayer, Boehringer Ingelheim and Mitsubishi Tanabe to investigate potential treatments of scleroderma and its complications, Consultant for: Prof. Distler has/had consultancy relationship within the last 3 years with Actelion, AnaMar, Bayer, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, Italfarmaco, iQvia, Lilly, medac, MedImmune, Mitsubishi Tanabe Pharma, Pharmacyclics, Novartis, Pfizer, Sanofi, Serodapharm and UCB in the area of potential treatments of scleroderma and its complications. In addition, he had/has consultancy relationship within the last 3 years with A. Menarini, Amgen, Abbvie, GSK, Mepha, MSD, Pfizer and UCB in the field of arthritides and related disorders, Georg Schett: None declared, Jörg Distler: None declared, Andreas Ramming Grant/research support from: Novartis

Objectives: Most DAMPs in inflammatory diseases are TLR2- and TLR4-ligands and according to the current concept, repeated stimuli would result in tolerance. Aims of the study were to verify this assumption, to investigate whether epigenetic effectors are involved and to explore the situation in rheumatoid arthritis (RA). Methods: A trained immunity (TI) and tolerance protocol was established using peripheral blood monocytes from healthy donors, β-glucan and lipopolysaccharide (LPS). The training or tolerance capacities of RA-relevant DAMPs were tested. Results: β-Glucan-, oS100A4-, HMBG1-, and HSP90-pretreated monocytes showed increased IL-6 responses to LPS re-stimulation. β-Glucan, oS100A and tenascin C induced training of monocytes to release more TNFα. In comparison to β-glucan, most DAMPs tested induced less TI, with exception of oS100A4. Monocytes exposed to oS100A4 showed increased IL-1β, IL-6, and TNFα in response to LPS, in spite that both stimulate TLR4. RNASEq upon β-glucan or oS100A4 revealed similar changes in chemokines/cytokines and epigenetic effectors; 17 epigenetic effectors correlated with chemokine/cytokine gene expression; PRDM8 was associated with more chemokine and cytokine transcripts. Knockdown of PRDM8 abolished TI induced by oS100A4. In RA, plasma S100A4 correlated with increased CSF2, and increased PRDM8 transcription in RA monocytes was associated with increased plasma CCL5 and IL-6, as well as therapy-resistance. Conclusion: Bypass of tolerance by DAMPs might be a phenomenon as important as TI, since it could explain how chronic inflammation can be maintained in spite of an environment with multiple TLR2/TLR4-ligands. In RA monocytes, a PRDM8-dependent TI mechanism could be responsible for sustained chemokine/cytokines levels.

Activation of macrophages and overexpression of TNFα is associated with the pathogenesis of chronic inflammatory diseases. However, the mechanisms leading to TNFα overexpression are still unknown. 5-methylocytosine (5-mC) is an epigenetic modification that is associated with silenced genes. Recent studies showed that it is converted to 5-hydroxylmethylocytosine (5-hmC) and reactivates gene expression through the action of the family of Ten-Eleven-Translocation (TET1-3) enzymes. In this study, we show that 5-hmC levels are increased globally and specifically in the TNFα promoter during the differentiation of monocytes to macrophages. In addition, the levels of 5-hmC are increased upon LPS stimulation of macrophages. Furthermore, CRIPSR stable knockout of TET1 decreases the expression of TNFα and other pro-inflammatory cytokines. In conclusion, we showed that TET1 contributes to the activation of macrophages possibly through regulation of 5-hydroxymethylation in the promoter of pro-inflammatory cytokine genes. The TET1 enzyme could be a promising therapeutic target to inhibit the persistent inflammation caused by macrophages in chronic inflammatory diseases.

Background Trained immunity is a process of innate immune memory in which a primary stimulus such as beta-glucan can enhance the response of monocytes to secondary stimuli. The concept that specific damage associated molecular patterns (DAMPs) in rheumatoid arthritis (RA) could cause trained immunity which is involved in the disease pathogenesis has not been investigated so far. The oligomeric form of S100A4 (oS100A4) is a potent inducer of proinflammatory cytokines which is found in the plasma of patients with rheumatoid arthritis (RA). Objectives Aims are to investigate whether oS100A4 induces trained immunity in monocytes and characterize the molecular pathways involved in this process. Methods Monocytes were isolated from peripheral blood of healthy donors using anti-CD14 magnetic beads. To induce training, monocytes were stimulated with 2 µg/ml of oS100A4 and 1 µg/ml β-glucan for 24 hours (n=8). We searched for differential gene expression by RNA sequencing in order to identify factors that play a role in the initial stages of trained immunity. On day 4, LPS (10 ng/ml) was added. After 24 hours, IL-6 and TNFalpha were measured in cell culture supernatants by ELISA. The training protocol was repeated in monocytes transfected with PRDM8 siRNA using Lipofectamine (n=4). In addition, plasma levels of S100A4, CCL5 and IL-6 were measured in a cohort of RA patients (n=36) and healthy controls (n=18) by ELISA and PRDM8 transcripts in RA peripheral blood monocytes were quantified by RT-PCR. Results Monocytes primed with oS100A4 showed increased releases of IL-6 and TNFalpha in response to a subsequent LPS stimulation. RNA-Seq revealed the differential expression of 902 genes upon oS100A4 and 667 upon beta-glucan (mean and median > 2 fold, p<0.01). Among the differential genes, 601 were upregulated in S100A4 and 447 in beta-glucan stimulated cells. Upregulated genes included chemokine/cytokine and epigenetic factors. When we compared the upregulated genes from oS100A4 and beta-glucan stimulated cells, 83% of chemokines/cytokines and 50% epigenetic factors were identical. Interestingly, the histone methyltransferase PRDM8 was found to be a major regulator of pro-inflammatory mediators by both stimuli. siRNA knockdown of PRDM8 abolished the training effect of oS100A4 by decreasing the LPS induced release of IL-6 and TNFalpha (p<0.01). Furthermore, we analyzed a cohort of monocytes taken from RA patients. Higher PRDM8 transcription in RA monocytes was associated with increased plasma levels of CCL5 and IL-6 (r = 0.52 and 0.55, p < 0.01). RA patients in remission versus active patients showed significantly lower PRDM8 transcripts (p < 0.05). Conclusion Oligomeric S100A4 induced trained immunity in monocytes similarly to beta-glucan. PRDM8 histone methyltransferase is involved in this process that appears to be activated in monocytes of RA patients. Disclosure of Interests Emmanuel Karouzakis: None declared, Agnieszka Pajak: None declared, Niels Riksen: None declared, Leo Joosten: None declared, Mihai Netea: None declared, Esther Lutgens: None declared, Eric Stroes: None declared, Adrian Ciurea Consultant for: AbbVie, Celgene, Janssen-Cilag, MSD, Eli Lilly, Novartis, Pfizer, UCB, Speakers bureau: Abbvie, Celgene, Janssen-Cilag, MSD, Eli Lilly, Novartis, Pfizer, UCB, Oliver Distler Grant/research support from: Prof. Distler received research funding from Actelion, Bayer, Boehringer Ingelheim and Mitsubishi Tanabe to investigate potential treatments of scleroderma and its complications, Consultant for: Prof. Distler has/had consultancy relationship within the last 3 years with Actelion, AnaMar, Bayer, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, Italfarmaco, iQvia, Lilly, medac, MedImmune, Mitsubishi Tanabe Pharma, Pharmacyclics, Novartis, Pfizer, Sanofi, Serodapharm and UCB in the area of potential treatments of scleroderma and its complications. In addition, he had/has consultancy relationship within the last 3 years with A. Menarini, Amgen, Abbvie, GSK, Mepha, MSD, Pfizer and UCB in the field of arthritides and related disorders, Mariam Grigorian: None declared, Michel Neidhart: None declared

Fibroblasts are polymorphic cells with pleiotropic roles in organ morphogenesis, tissue homeostasis and immune responses. In fibrotic diseases, fibroblasts synthesize abundant amounts of extracellular matrix, which induces scarring and organ failure. By contrast, a hallmark feature of fibroblasts in arthritis is degradation of the extracellular matrix because of the release of metalloproteinases and degrading enzymes, and subsequent tissue destruction. The mechanisms that drive these functionally opposing pro-fibrotic and pro-inflammatory phenotypes of fibroblasts remain unknown. Here we identify the transcription factor PU.1 as an essential regulator of the pro-fibrotic gene expression program. The interplay between transcriptional and post-transcriptional mechanisms that normally control the expression of PU.1 expression is perturbed in various fibrotic diseases, resulting in the upregulation of PU.1, induction of fibrosis-associated gene sets and a phenotypic switch in extracellular matrix-producing pro-fibrotic fibroblasts. By contrast, pharmacological and genetic inactivation of PU.1 disrupts the fibrotic network and enables reprogramming of fibrotic fibroblasts into resting fibroblasts, leading to regression of fibrosis in several organs.

Background Rheumatoid arthritis (RA) and other types of inflammatory arthritis follow a characteristic anatomical pattern of joint involvement. We have recently shown that local synovial stromal cells, specifically synovial fibroblasts, exhibit joint-specific transcriptomes and functions. In particular, hand SF exhibited prominent proliferative and chemotactic activities. Density of stroma and leukocyte infiltration were increased in hand synovium. Objectives To explore the role of hand/feet-specific lncRNA HOTTIP in shaping the function of hand synovial fibroblasts in arthritis. Methods We studied transcriptomes and epigenomes of hand, shoulder and knee SF from patients with RA or osteoarthritis and from knees of non-arthritic subjects using RNA-sequencing, Illumina HiSeq 2000 n=21), histone ChIP-sequencing (Illumina HiSeq 2500, n=7) and Infinium HumanMethylation450 BeadChip (n=12). qPCR was used to confirm RNA-sequencing data in a larger cohort of SF from different joints. We silenced the lncRNA HOTTIP in hand SF using LNA GapmeRs, followed by RNA-sequencing, qPCR, protein-protein interaction analysis of RNA-sequencing data (STRING), and in vitro assays for proliferation (BrdU assay) and apoptosis (Annexin V/PI staining). Results Genome-wide DNA methylation patterns and histone marks at actively transcribed DNA regions (H3K27ac) and enhancers (H3K4me1) defined joint-specific origin of SF. SF from hands and feet specifically expressed the lncRNA HOTTIP. This distal-specific HOTTIP expression coincided with the enrichment of H3K4me3 and H3K27ac and a decrease in repressive marks (H3K27me3, DNA methylation) at the HOTTIP promoter in hand SF. In contrast, the HOTTIP promoter displayed scarce activating, but abundant repressive epigenetic marks in shoulder and knee SF. Silencing of HOTTIP in hand SF altered the expression of 447 protein-coding genes (log ratio >|2|, FDR<0.05). These genes were strongly enriched in the mitotic cell cycle protein interaction network (n=48 genes, p=3.3x10–7). Several of the enriched mitotic cell cycle genes, including NCAPG, CENPO, ZWILCH and BUB1 were confirmed as downregulated by HOTTIP silencing in a larger cohort of hand SF (n=6). The basal expression of 36 out of the 48 enriched cell cycle genes correlated with the basal HOTTIP expression in hand SF (n=6, RNA-sequencing, R>|0.06|). We further measured these correlations in a larger cohort of hand SF (n=21) for a subset of the 36 genes using qPCR. Among the measured genes, TADA 3 and CDC27 were confirmed to correlate with HOTTIP expression in hand SF (R=0.5, p<0.05). Silencing of HOTTIP for 24 hour, 48 hour and 72 hour decreased the incorporation of BrdU into DNA of hand SF as measured by BrdU proliferation assay (p<0.05, n=3). Apoptosis of hand SF increased at 48 hour of HOTTIP silencing (p<0.05, n=3). Conclusions The lncRNA HOTTIP, which is specifically expressed in hand joints via epigenetic mechanisms, is a master regulator of mitotic cell cycle genes and proliferation in hand SF. Distal-specific expression of HOTTIP might imprint hand SF with enhanced proliferative potential, thereby shaping the location-specific joint pathology, e.g. prominent synovial hyperplasia and increased severity of hand arthritis in RA. Disclosure of Interest M. Frank Bertoncelj Grant/research support from: euroTEAM, BTCure, IRR, Promedica, Georg und Berta Schwyzer Winiker Grant, T. Masterson: None declared, E. Karouzakis Grant/research support from: BTCure, GSK, C. Kolling: None declared, A. Filer: None declared, C. Buckley: None declared, S. Gay Grant/research support from: euroTEAM, BTCure, GSK, IRR, O. Distler Grant/research support from: Abbvie, Actelion, Bayer, BiogenIdec, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone, Lilly, medac, MedImmune, Mepha, MSD, Mitsubishi Tanabe Pharma, Novartis, Pfizer, Pharmacyclics, Sanofi, Sinoxa and UCB, Consultant for: Abbvie, Actelion, Bayer, BiogenIdec, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone, Lilly, medac, MedImmune, Mepha, MSD, Mitsubishi Tanabe Pharma, Novartis, Pfizer, Pharmacyclics, Sanofi, Sinoxa and UCB, C. Ospelt Grant/research support from: euroTEAM, BTCure, CABMM, IRR, Promedica

Background Synovial fibroblasts (SF) promote chronic joint inflammation and joint destruction in rheumatoid arthritis (RA). We have shown recently that SF from different joints exhibit profound differences in their transcriptomes, epigenomes and functions, which creates a unique microenvironment in each joint. This might influence the susceptibility of distinct joints to develop RA or lead to joint-specific differences in the disease severity or therapeutic response. Objectives To analyse differences in the JAK-STAT pathway in SF from different joints. Methods SF were isolated from knee, shoulder and hand joints of RA and osteoarthritis patients undergoing joint replacement surgery and from knee synovial biopsies of non-arthritic subjects with arthralgia. Transcriptomes and epigenomes of SF were determined by RNA-seq, Illumina HiSeq 2000 n=21), ChIP-seq (Illumina HiSeq 2500, n=7) and Infinium HumanMethylation450 BeadChip (n=12). We used MetaCore (Thomson Reuters) for the pathway enrichment analysis of RNA-seq data. SF were stimulated with IL-6/soluble IL-6 receptor (IL-6/sIL-6R, 50 ng/ml each). The amount of STATs and phospho-STAT3 (p-STAT3) was measured by Western blot with normalisation to α-tubulin. Results The JAK-STAT pathway was enriched in knee SF versus hand and shoulder SF (FDR<0.05). JAK1 (normalised reads – mean ±SD: 10673±2084) and STAT1 (15520±2678) were the top expressed Janus kinase and STAT mRNAs in SF, respectively, whereas the expression of JAK3 (40±26) and STAT4 (164±91) mRNAs was low. Looking into joint-specific differences, STAT1 mRNA was higher in knee SF and shoulder SF compared with hand SF (p<0.05, FDR<0.15). Accordingly, STAT1 protein was increased in knee SF (STAT1/α-tubulin ratio: 0.83±0.02, p=0.02, n=4) and shoulder SF (1.02±0.02, p=0.001, n=5) versus hand SF (0.57±0.02, n=3). JAK1, STAT2 and STAT5B mRNAs were higher in knee compared with hand SF (p<0.05, FDR<0.05) and STAT2 and STAT6 mRNAs were higher in knee versus shoulder SF (p<0.05, FDR<0.05). TYK2 mRNA was high in hand SF compared with shoulder and knee SF (p<0.05, FDR<0.05). SF from different joints exhibited comparable DNA methylation at the promoters of these genes. Activating histone marks H3K4me3 and/or H3K27ac were enriched at the promoters of JAK1, STAT1, STAT2 and STAT5B in knee versus hand SF. This indicated that the abundance of activating histone marks at gene promoters might shape joint-specific expression of a subset of Janus kinase and STAT genes. Stimulation of SF with IL-6/sIL-6R increased the phosphorylation of STAT3 in knee (p-STAT3/α-tubulin ratio 1.8±1.0, p=0.03, n=5) and shoulder SF (1.8±0.7, p=0.03, n=6) compared with hand SF (0.9±0.5, n=6). The basal amount of STAT3 protein and the ratio pSTAT3/STAT3 was higher in knee SF (STAT3/α-tubulin ratio: 0.8±0.1; p-STAT3/STAT3: 2.2±0.8, n=3) and shoulder SF (0.6±0.4; 2.1±1, n=4) versus hand SF (0.3±0.02; 1.0±0.4, n=2). Conclusions Here we show substantial quantitative and qualitative differences in the JAK-STAT signalling pathway in SF from different joints. Knee SF, in particular, exhibit increased expression of Janus kinase and STAT genes and enhanced JAK-STAT signalling upon stimulation with IL-6/sIL-6R. This suggests that RA in different joints might not be equally sensitive to Janus kinase inhibitors or blockade of IL-6. This has important implications in clinical practice and drug discovery in RA. Disclosure of Interest T. Masterson: None declared, K. Klein: None declared, E. Karouzakis Grant/research support from: BTCure, GSK, O. Distler Grant/research support from: Abbvie, Actelion, Bayer, BiogenIdec, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone, Lilly, medac, MedImmune, Mepha, MSD, Mitsubishi Tanabe Pharma, Novartis, Pfizer, Pharmacyclics, Sanofi, Sinoxa and UCB, Consultant for: Abbvie, Actelion, Bayer, BiogenIdec, Boehringer Ingelheim, ChemomAb, espeRare foundation, Genentech/Roche, GSK, Inventiva, iQone, Lilly, medac, MedImmune, Mepha, MSD, Mitsubishi Tanabe Pharma, Novartis, Pfizer, Pharmacyclics, Sanofi, Sinoxa and UCB, C. Ospelt Grant/research support from: euroTEAM, BTCure, CABMM, IRR, Promedica, M. Frank Bertoncelj Grant/research support from: AbbVie Rheumatology grant 2017 euroTEAM, BTCure, IRR, Promedica, Georg und Berta Schwyzer Winiker Grant

Introduction Inflammatory arthritis, such as rheumatoid arthritis (RA) and spondyloarthropaties, follows a characteristic anatomical pattern of joint involvement. We suggest that the local cell types, systemic triggers and site-specific exogenous factors that activate these local cells synergistically contribute to the site-specific occurrence of arthritis. Objectives To explore the role of the local stromal cells – synovial fibroblasts (SF), in defining the joint-specific synovial biology, relevant for RA. Methods We studied transcriptomes, epigenomes and functions of hand, shoulder and knee SF from patients with RA or osteoarthritis and from knees of nonarthritic subjects with arthralgia by using RNA-sequencing (Illumina HiSeq 2000, n=21), histone ChIP-sequencing (Illumina HiSeq 2500, n=7), Infinium HumanMethylation450 BeadChip (n=12) and in vitro assays for proliferation, adhesion and chemotaxis. We silenced the long noncoding RNA HOTTIP in hand SF using LNA GapmeR, followed by RNA-sequencing (n=2), protein-protein interaction analysis (STRING) and qPCR confirmation of the HOTTIP target genes. Paraffin embedded RA synovial tissues (n=48) from different joints were scored by the Krenn synovitis score for leukocyte infiltration, synovial lining and density of synovial stroma. Results The transcriptomes, the global DNA methylation patterns and the histone marks, which mark the actively transcribed DNA regions (H3K27ac) and enhancers (H3K4me1), defined the joint-specific origin of SF. Hand SF showed prominent proliferative, chemotactic and matrix-degrading properties. Hand synovial tissues exhibited increased density of stroma and leukocyte infiltration. The homeobox (HOX) genes that regulate hand embryogenesis (HOXA13, HOTTIP) were the top differentially expressed genes with hand-specific expression in SF. This hand-specific expression pattern coincided with the specific enrichment of the activating histone marks H3K4me3 and H3K27ac and the absence of repressive H3K27me3 and DNA methylation at the HOTTIP and HOXA13 promoters in hand SF. In contrast, shoulder and knee SF displayed abundant H3K27me3 and DNA methylation, but scarce H3K4me3 and H3K27ac at the HOTTIP and HOXA13 promoters. Silencing of HOTTIP in hand SF altered the expression of 447 mRNA genes with a log ratio >|2| (FDR<0.05). The HOTTIP regulated genes were strongly enriched in the mitotic cell cycle protein interaction network (n=48 genes, p=3.29×10–7). Several of these genes were confirmed as downregulated by HOTTIP silencing in a larger cohort of hand SF (n=6, p<0.05). Besides, the basal expression of the enriched cell cycle genes, including CDC27 and TADA3, significantly correlated with the basal HOTTIP expression in hand SF (n=21). Conclusions The lncRNA HOTTIP, which is specifically expressed in hands via epigenetic mechanisms regulates the mitotic cell cycle genes. This might imprint hand SF with an enhanced proliferative potential and might promote the synovial hyperplasia in hand joints, thereby increasing the severity of hand RA. Acknowledgements IRR-IRF, Promedica, Georg und Berta Schwyzer Winiker Stiftung, CABMM. Disclosure of interest M. Frank Bertoncelj Grant/research support from: IRR-IRF, Promedica, Georg und Berta Schwyzer Winiker Stiftung, CABMM, E. Karouzakis: None declared, T. Masterson: None declared, C. Kolling: None declared, O. Distler: None declared, C. Ospelt Grant/research support from: IRR-IRF, Promedica, Georg und Berta Schwyzer Winiker Stiftung, CABMM

DNA methylation is an important epigenetic modification that is known to be altered in rheumatoid arthritis synovial fibroblasts (RASF). Here, we compared the status of promoter DNA methylation of SF from patients with very early RA with SF from patients with resolving arthritis, fully established RA and from non-arthritic patients. DNA was hybridized to Infinium Human methylation 450k and 850k arrays and differential methylated genes and pathways were identified. We could identify a significant number of CpG sites that differed between the SF of different disease stages, showing that epigenetic changes in SF occur early in RA development. Principal component analysis confirmed that the different groups of SF were separated according to their DNA methylation state. Furthermore, pathway analysis showed that important functional pathways were altered in both very early and late RASF. By focusing our analysis on CpG sites in CpG islands within promoters, we identified genes that have significant hypermethylated promoters in very early RASF. Our data show that changes in DNA methylation differ in RASF compared to other forms of arthritis and occur at a very early, clinically yet unspecific stage of disease. The identified differential methylated genes might become valuable prognostic biomarkers for RA development.

Introduction Rheumatoid arthritis (RA) is an autoimmune disease with unknown etiopathogenesis where systemic autoimmunity precedes clinical disease onset. Adaptive immunity is initiated in lymphoid tissue where lymph node stromal cells (LNSC) play a crucial role in shaping the immune response and maintaining peripheral tolerance. Objectives We developed an experimental model for studying the functional capacities of human LNSC during the earliest phases of RA and compared their cellular and molecular characteristics to LNSC from healthy volunteers. Methods ACPA +RA patients (n=24), ACPA +RA risk individuals (n=23) and seronegative healthy controls (n=14;HC) underwent ultrasound-guided inguinal lymph node biopsy. Human LNSCs were isolated and expanded in vitro for cellular (flow cytometry), molecular (methylome, transcriptome and microRNA) and functional analyses. Results Key LN chemokines CCL19, CCL21 and CXCL13 were induced in LNSCs upon stimulation with TNFα and lymphotoxin α1β2, but to a lesser extent in LNSCs from RA patients. RNA sequencing was performed on LNSC of HC (n=5), ACPA +RA risk individuals (n=6) and ACPA +RA patients (n=4). Of interest, LNSC from ACPA +RA risk individuals and ACPA +RA patients revealed a common significantly differential expressed gene signature compared with HC LNSC. Pathway analysis of this common signature showed, among others, significant enrichment of pathways affecting actin cytoskeleton, focal adhesion and cell junction. Accordingly, in a gel contraction assay LNSC from ACPA +RA risk individuals and RA patients showed impaired collagen contraction compared to healthy LNSC. In RA LNSC a significant enrichment was observed for genes involved in TGFb signalling while in RA-risk LNSC cell cycle genes were differentially expressed compared with HC. DNA methylation analyses revealed common differentially methylated CpG sites (DMS) in LNSC from ACPA +RA patients (n=5) and ACPA +RA risk individuals (n=3) compared with HC (n=4). These DMS were significantly hypomethylated and associated with antigen processing and presentation (HLA-DRB1). Conclusions This data point towards alterations in the cytoskeleton and antigen-processing and presentation in LNSC from ACPA+RA risk individuals and RA patients. Further studies are required to investigate the influence of this LNSC abnormality on immune responses. Disclosure of interest C. Ospelt: None declared, E. Karouzakis: None declared, J. Hähnlein: None declared, H. Semmelink: None declared, R. Gay: None declared, P.-P. Tak Employee of: Currently: Senior Vice President R and D Pipeline, Global Development Leader and Chief Immunology Officer, GSK, D. Gerlag Employee of: Currently: Head Clinical Unit Cambridge at GSK, S. Gay: None declared, L. Van Baarsen: None declared

Background Lymph node stromal cells (LNSC) play a crucial role in shaping the immune response and maintaining peripheral tolerance. We developed an experimental model for studying the functional capacities of human LNSC during the earliest phases of RA and compared their cellular and molecular characteristics to LNSC from healthy volunteers. Methods ACPA+ RA patients (n=24), ACPA+ RA-risk individuals (n=23) and seronegative healthy controls (n=14;HC) underwent ultrasound-guided inguinal lymph node biopsy. Human LNSCs were isolated and expanded in vitro for cellular (flow cytometry), molecular (methylome, transcriptome and microRNA) and functional (contraction) analyses. Results RNA sequencing was performed on LNSC of HC (n=5), ACPA+ RA-risk individuals (n=6) and ACPA+ RA patients (n=4). Of interest, LNSC from ACPA+ RA-risk individuals and ACPA+ RA patients were more similar to each other compared with HC. Pathway analysis of commonly increased genes in RA (-risk) LNSC showed, among others, significant enrichment of pathways affecting actin cytoskeleton, focal adhesion and cell junction. DNA methylation (Illumina HumanMethylation450 array) analyses revealed 459 differentially methylated CpG sites (DMS) in LNSC from ACPA+ RA patients (n=5) versus HC (n=4), 504 DMS between ACPA+ RA-risk individuals (n=3) versus HC and 665 DMS when comparing RA patients with RA-risk individuals (delta β-value >0.1, p<0.05). 34 DMS were different in both RA and RA-risk LNSC compared to healthy LNSC. 80% of these DMS were significantly hypomethylated and associated with antigen processing and presentation (HLA-DRB1), immune response and regulation of actin cytoskeleton. Accordingly, in a gel contraction assay LNSC from ACPA+ RA-risk individuals and RA patients showed impaired collagen contraction compared to healthy LNSC. Healthy LNSC (n=5) covered 26.5% +/-2.5 of the well, while RA-risk (n=4) and RA (n=5) LNSC only covered 33.9% +/-5.9 and 30.6% +/-6.5. Conclusions This data point towards alterations in the cytoskeleton and antigen-processing and presentation in LNSC from ACPA+ RA-risk individuals and RA patients. Further studies are required to investigate the influence of this LNSC abnormality on immune responses. Disclosure of Interest C. Ospelt: None declared, E. Karouzakis: None declared, J. Hähnlein: None declared, J. Semmelink: None declared, R. Gay: None declared, P. Tak Employee of: GSK, D. Gerlag Employee of: GSK, S. Gay: None declared, L. van Baarsen: None declared

A number of human diseases, such as arthritis and atherosclerosis, include characteristic pathology in specific anatomical locations. Here we show transcriptomic differences in synovial fibroblasts from different joint locations and that HOX gene signatures reflect the joint-specific origins of mouse and human synovial fibroblasts and synovial tissues. Alongside DNA methylation and histone modifications, bromodomain and extra-terminal reader proteins regulate joint-specific HOX gene expression. Anatomical transcriptional diversity translates into joint-specific synovial fibroblast phenotypes with distinct adhesive, proliferative, chemotactic and matrix-degrading characteristics and differential responsiveness to TNF, creating a unique microenvironment in each joint. These findings indicate that local stroma might control positional disease patterns not only in arthritis but in any disease with a prominent stromal component.