Introduction Synovial fibroblasts (SF) from different joint localizations exhibit profound differences in their expression profiles that might predispose specific joints for certain forms of arthritis.1 Homeobox (HOX) genes were shown to play important roles in limb development and might influence the site-specific development of various diseases. Objectives To analyse the expression of distally expressed HOXD genes in SF from hands and to investigate their epigenetic regulation. Methods The expression of HOXD10, HOXD11 and HOXD13 was analysed by quantitative Real-time PCR. The histone marks H3K4me1 (enhancers), H3K4me3 (transcriptional starts of transcribed genes), H3K27me3 (inactive gene promoters) and H3K27ac (active enhancers) in SF from one RA (finger) and one osteoarthritis (OA; thumb) patient were analysed by ChIP DNA sequencing (ChIPseq). Hand SF (n=7) were treated for 24 hour with the bromodomain inhibitor I-BET151 (1 µM), targeting the bromodomain and extra-terminal domain (BET) proteins. The expression of the histone-acetyltransferases CREBBP-binding protein (CBP) and p300 was silenced by transfection of antisense LNA gapmeRs in hand SF (n=4). Results HOXD10, HOXD11 and HOXD13 transcripts were significantly increased in SF from RA and OA patients in digits II-IV and wrists compared to SF from digit I (thumb). Accordingly, ChIPseq showed an increase of H3K27ac and H3K4me3 marks in the genomic region between HOXD9 and HOXD13 in SF from a RA finger compared to SF from an OA thumb, paralleled by a loss of the repressive histone marks H3K27me3 between HOXD12 and MIR10B. Treatment of SF with I-BET151 reduced the expression of HOXD10, HOXD11 and HOXD13 by 50% (±23; p<0.001), 29% (±29; p<0.05), and 27% (±27; p<0.05) respectively. Silencing of CBP reduced the expression of HOXD10 by 47% (±10; p<0.01) and HOXD11 by 22% (±19; p=0.1) but not HOXD13. Also silencing of p300 reduced the expression of HOXD10 by 47% (±24; p=0.06) and HOXD11 by 55% (±10; p<0.01) but not HOXD13. Conclusions Distally expressed HOXD genes exhibit site-specific expression between digits II-IV and the thumb. The site-specific expression is maintained by a specific set of histone modifications and regulated by epigenetic writer (CBP, p300) and reader proteins (BET proteins). This epigenetically regulated expression of HOXD genes might influence the different occurrence of RA and OA in the small joints of the digits II-IV and the thumb. Reference . Frank-Bertoncelj, et al. Nat Commun2017. Acknowledgements Swiss National Fund (PMPDP3-171315/1), Institute of Rheumatology Research (IRR), Stiftung für wissenschaftliche Forschung, Opo-Stiftung, Hartmann Müller Stiftung Disclosure of interest K. Klein Grant/research support from: Swiss National Fund (PMPDP3-171315/1), Hartmann Müller Stiftung, M. Frank-Bertoncelj: None declared, G. Lee: None declared, C. Kolling: None declared, O. Distler Grant/research support from: Stiftung für wissenschaftliche Forschung, C. Ospelt Grant/research support from: Opo Stiftung
Background The close homologues cAMP-response element binding protein (CREB) binding protein (CBP) and p300 are writers of H3K27 histone acetylation marks found in active enhancers. In addition, their bromodomains are readers of acetylated lysine residues on histone tails and are subject of drug development for inflammatory and malignant diseases. CBP and p300 are widely accepted as redundant proteins and unique functions have not been investigated yet in depth. Objectives To analyse individual functions of CBP and p300 in rheumatoid arthritis synovial fibroblasts (SF). Methods SF were treated with the pan inhibitor I-CBP (1 µM, 5 µM), targeting the bromodomains of CBP and p300, in presence and absence of TNF-α (10 ng/ml) for 24 hour. The expression of CBP and p300 was silenced by transfection of antisense LNA gapmeRs (5 nM) in SF. Knockdown was verified by Western blotting. 24 hour after transfection cells were stimulated with TNF-α (10 ng/ml) for 24 hour. The mRNA expression of potential target genes was measured by quantitative Real-time PCR, using RPLP0 as an endogenous control. The protein expression of HOXD10 after CBP and p300 silencing was verified by Western blotting. Results I-CBP dose-dependently reduced the TNF-α-induced expression of MMP1 (p<0,05), MMP3, IL6 and IL8 in SF (n=3). Antisense LNA gapmeRs targeting CBP reduced the protein expression of CBP by 68,7% (±12,9%, p<0,01, n=5) in unstimulated cells and by 89,7% (±12,9%) in presence of TNF-α. The protein expression of p300 was reduced by 55,3% (±29,8%, p<0,05, n=6) in unstimulated cells and by 62,7% (±27,9%) in presence of TNF-α after transfection of LNA gapmeRs targeting p300. Silencing of CBP in SF (n=7) reduced the TNF-α-induced expression of IL6 (p<0,05), IL8 (p<0,05), MMP3 (p<0,05), as well as the basal (p<0.001) and the TNF-α-induced expression of MMP1 (p<0,05). In contrast, silencing of p300 induced the basal expression of IL6 (p<0,05), IL8 (p<0,01), MMP1 (p<0,05), and MMP3 (p<0,05), as well as the TNF-α-induced expression of IL6 (p=0,078), IL8 (p=0,078), MMP1 (p<0,05) and MMP3 (p=0,063). Silencing of CBP in hand SF (n=4) reduced the expression of hand-specific HOX genes including HOXD10 (0,53±0,10 fold; p<0,01), HOXD11 (0,76±0,19 fold; p=0,098) and HOXA13 (0,75±0,17 fold; p=0,063), whereas HOXA9, HOXA10 and HOXA11 were not affected. Silencing of p300 reduced the expression of HOXD10 (0,65±0,24 fold; p=0,061), HOXD11 (0,45±0,10 fold; p<0,01), HOXA10 (0,70±0,14 fold; p<0,05) and HOXA13 (0,55±0,19 fold; p<0,05). The down regulation of HOXD10 after silencing of CBP and p300 in hand SF was confirmed on protein levels by Western blotting. Conclusions Our results unravel opposing functions of CBP and p300 in regulating the TNF-α-induced expression of inflammatory and matrix-degrading target genes in SF. In addition, CBP and p300 likely contribute to the maintenance of a joint-specific gene expression in SF by regulating the expression of hand-specific HOX genes. Acknowledgements Marie Heim Vögtlin grant (SNF), Stiftung für wissenschaftliche Forschung Disclosure of Interest G. Lee: None declared, C. Kolling: None declared, 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: None declared, K. Klein Grant/research support from: Marie Heim Vögtlin grant (SNF)