Epigenetic suppression of synovial inflammation and osteoclast differentiation in rheumatoid arthritis by I-BET762

Abstract

The BRD and extra-terminal domain (BET) family of proteins was inhibited using I-BET762, a small molecule inhibitor, which mimics interaction with acetylated lysin residue in BET protein. We explored the roles of the bromodomain (BRD) of chromatin adapators in regulating synovial inflammation in rheumatoid arthritis (RA). The expression levels of bromodomain proteins, c-Myc, and the components of the MAPK and NF-κB signaling pathways were detected by western blot. The BRD3, BRD4, and c-Myc expression were suppressed by I-BET762-treated RA-FLS. I-BET762 inhibited the phosphorylation of p38 MAPK and NF-κB signaling pathways. Moreover, secretion levels of pro-inflammatory mediators (IL-6 and IL-8), chemokine (CXCL-10), and MMP-1 and -3 significantly decreased after I-BET762 treatment. Cell migration and invasion were also reduced in response to I-BET762 treatment. I-BET762 treatment concurrently inhibited p38 MAPK and NF-κB, thereby suppressing the inflammatory properties of RA-FLS. In osteoclastogenesis, TRAP-positive multi-nucleated cells were reduced by I-BET762 in a dose-dependent manner. The actin ring formation of RANKL-induced osteoclast was inhibited in the presence of I-BET762. I-BET762 down-regulated pro-inflammatory, matrix-degrading, chemo-attractive properties in RA-FLS and, bone resorption in osteoclast. These data suggest that I-BET762 inhibit synovial inflammation and bone resorption in RA. In vivo animal experiment may be needed for confirmation.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-36645-5.

Introduction

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by persistent joint inflammation and destruction that is mediated by infiltrating immune cells and tissue-resident cells in joints^1,2. Pannus, a hyperplastic synovial tissue with dense inflammatory infiltrates, invades adjacent bone and cartilage and contributes directly to joint destruction^3,4. At the pannus-cartilage interface, activated immune cells and fibroblast-like synoviocytes (FLS) secrete a large amount of pro-inflammatory cytokines and chemokines, including tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IL-8, and CXCL-10, as well as tissue-degrading matrix metalloproteinase (MMP) such as MMP-1, MMP-3, MMP-13^5,6. Given that pannus is the pathologic hallmark of RA, pannus consist of activated macrophages that secrete tumor necrosis factor (TNF), and numerous activated FLS that responds to paracrine TNF-α⁷. In response to inflammatory cytokine, particularly TNF-α, rheumatoid arthritis fibroblast-like synoviocytes (RA-FLS) undergo phenotypical changes, resembling cancer-cell-like features, and they invade and destroy the adjacent cartilage^2,5,6,8,9. Recently, BET protein inhibitors have been shown to inhibit the growth of various cancers, particularly blood cancers, including leukaemia, lymphoma, and myeloma¹⁰. According to previous reports, chromatin changes are triggered by Myc activation, leading to increased histone acetylation (H3 and H4), methylation (H3K4me3), and DNA accessibility^8,9. Although epigenetic modifications have been extensively studied in the oncology field, they have limited influence on autoimmune diseases¹¹. Hence, our study may have a potential of novel targeted treatment for autoimmune diseases. Increasing evidence indicates that osteoclasts play an important role in local bone erosion in RA. Cartilage destruction not only occurs via inflammatory cytokine-induced MMP expression but also via regulation of osteoclastogenesis¹². The transformation of normal synoviocytes into aggressive RA-FLS is closely associated with the epigenetic alteration that leads to the activation of pro-inflammatory signaling including mitogen-activated protein kinase (MAPK), nuclear factor-kappa B (NF-κB), and activator protein-1 (AP-1)^13–17. Persistent H3K27 acetylation and increased chromatin accessibility have been implicated in the sustained expression of TNF-inducible genes in RA-FLS, highlighting epigenetic regulation in the disease pathogenesis^3,7,9. Epigenetic modifications can function as a bridge between the environment and the genome through chromatic modification that can influence gene activity and expression without changing the DNA sequence^18,19. Although epigenomic profiling in autoimmunity is still nascent, but intriguing data from candidate genes and unbiased approaches show that an altered epigenomic landscape contributes to the biology in RA²⁰.

The bromodomain (BRD) and Extra-Terminal domain (BET) family of proteins, including BRD2, BRD3, BRD4, and BRDT, are chromatin ‘readers’ that bind to acetylated histone lysine residues and couple these positive histone markers to the transcriptional machinery, thereby promoting gene expression by disrupting chromatin complexes essential for mRNA transcription, elongation, and splicing^8,21. Interestingly, BRD inhibitors through epigenetic modification ameliorate the production of inflammatory cytokines of macrophages, the pro-inflammatory and invasive function of FLS, and osteoclastogenesis^5,8,22–24. In a previous study, the persistent H3K27 acetylation with increased chromatin accessibility was associated with a constant expression of TNF-inducible genes in RA-FLS²⁵. Especially, BRD4 is also a critical mediator of transcriptional elongation, functioning to recruit the P-TEFb to phosphorylate RNA polymerase II^10,22,26. Hence, targeting the regulatory binding proteins of acetylated histones is a potential treatment option for RA through epigenetic modification^3,7,26. I-BET762, developed by GlaxoSmithKline, as a novel pan-BET inhibitor with potent suppression of BRD2, BRD3, and BRD4^8,21. Interestingly, BET bromodomain inhibitors have been implicated in the inhibition of macrophage inflammatory cytokine secretion, the differentiation of monocytes into bone-resorbing osteoclasts^27–31, and the pro-inflammatory and invasive function of FLS, which all contribute to joint inflammation and destruction^9,10. These BET inhibitors had promising effects in animal models of arthritis and RA-FLS. Indeed, BET inhibition by using I-BET151 suppressed the pro-inflammatory response of macrophages, osteoclastogenesis, and attenuated RA-FLS invasiveness^15,32. Another BET inhibitor, JQ1, was reported to decrease the proliferation of RA-FLS, and their production of pro-inflammatory cytokines and MMPs in vitro and attenuate synovial inflammation and joint destruction in a collagen-induced arthritis animal model¹⁹. I-BET762 strongly inhibits the expression of a variety of cytokines and chemokines in mouse bone marrow-derived macrophages stimulated with LPS^12–16. The anti-inflammatory effect of I-BET762 also caused long-lasting suppression of pro-inflammatory functions in Th1 and Th17 cells. The triazolobenzodiazepine I-BET762 (GSK525762A) had phase I trials (NCT01587703) targeting NMC (NUT (nuclear protein in testis) midline carcinoma). One of the I-BET762, Molibresib, also targeted hematological malignancies in phase II clinical trials (NCT01943851). After the initial successful inhibition by I-BET762, it developed the potential of pharmacological immunoregulatory anti-cancer activities in various preclinical cancer models^33,34. In addition, it strongly suppresses the inflammatory response of bone marrow-derived macrophages, Type 1 T helper (Th1), CD4⁺ T helper, and Th17 cells^{8,22–24,35,36}. Notably, BET inhibitor might suppress the c-Myc, an oncogenic transcription factor known to modulate cellular proliferation and transformation in RA-FLS^37,38. c-Myc protein interacts with BRD, and the complex leads to c-Myc degradation by BET inhibition³⁷. Therefore, altering epigenetic modifications, especially, targeting BRD4 has been postulated as a novel target to treat RA. The BET inhibitors ameliorated joint inflammation and destruction in an animal model of collagen-induced arthritis^4,20. Inflammatory signaling such as MAPK, NF-κB plays a significant role in the pathogenesis of several immune-mediated diseases, including rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE)^25,36. Here, we aimed to investigate the impact of pan-BET inhibition on pro-inflammatory and tissue-destructive response in RA-FLS and osteoclast.

Methods

Reagents

I-BET762 (GSK525762), a pan-BET inhibitor, was obtained from Sigma-Aldrich (St. Louis, MO). This chemical was dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) to generate a 10 mM stock solution and stored at −20 °C.

Ethical statement

Synovial biopsy tissues were obtained from patients with RA undergoing joint surgery. RA was diagnosed according to the American College of Rheumatology (ACR) criteria for RA³⁹. Written informed consent was obtained from all patients. The study was approved by the Institutional Review Boards of Seoul National University Hospital (1504–086-665). All methods were performed in accordance with the Declaration of Helsinki.

FLS isolation and culture

Synovial specimens from knees joints of RA patients were cut into small pieces using surgical forceps and a scalpel^36,40, followed by incubation with 1 mg/mL type II collagenase (Worthington Biochemical, Lakewood, NJ) in serum-free Dulbecco’s modified Eagle medium (DMEM) (Welgene, Gyengsangbuk-do, Republic of Korea). After 2 ~ 3 h, the incubated synovial tissue fragments were washed and filtered through a 40-μm cell strainer. 10% heat- inactivated fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 °C and 5% CO₂. Dissociated synovial cells, RA-FLS between passages four and seven were used for experiments.

Cell viability assay

RA-FLS were plated at a density of 2 × 10³ cells per well in 96 well plates in serum-free DMEM. I-BET762 was administered at various concentration (0.1, 0.5, 1, 2, 5, and 10 μM) with or without TNF-α (10 ng/ml) for 24, 48, and 72 h. Cell viability and proliferation were assessed using Cell-counting kit 8 (CCK-8) solution (CCK-8, Dojindo, Kumamoto, Japan) and the absorbance was measured at 450 nm using a microplate reader (Epoch; BioTek Instruments, Inc, Winooski. VT).

Quantitative real-time PCR

Total RNA was extracted using the RNeasy Kit (Qiagen AG, Hombrechtikon, Switzerland). cDNA synthesis was performed using the cDNA synthesis kit (Bioneer, Daejeon, Republic of Korea). SYBR master mix was used to amplify cDNA samples in a QuantStudio 5 (Thermo Fisher Scientific) according to the manufacturer’s protocol.

Enzyme-linked immunosorbent assay (ELISA)

RA-FLS were seeded at a density of 2 × 10⁴ cells per well in 24 well plates. Cells were treated with DMSO 0.01% (v/v), 5 μM of I-BET762 for 1 h, followed by stimulation with TNF-α (10 ng/ml) for 48 h. Production of IL-6, IL-8, and CXC chemokine ligand (CXCL)-10 was measured using Duoset ELISA kit (R&D System, Minneapolis, MN) according to the manufacturer’s instruction.

Western blot

RA-FLS lysates were separated by electrophoresis and transferred to PVDF membranes (Thermo Fisher Scientific, Van Allen Way Carlsbad, CA92008, USA). After blocking, the membranes were incubated with the primary antibodies at 4 °C overnight and then horseradish peroxidase-conjugated secondary antibodies for 2 h at 25 °C in the dark. Membrane-bound secondary antibodies were detected by chemiluminescence substrate (Elpis Biotech, Daejeon, Republic of Korea). To quantify the MAPK signaling in the cytoplasm and NF-κB nuclear translocation, cytoplasmic and nuclear fractions were obtained according to the manufacturer’s instruction (Thermo Fisher Scientific, Meridian Rd., Rockford, USA).

Wound healing assay

RA-FLS were seeded into 24 well plates at 70 ~ 80% confluence and serum-starved (1% FBS/DMEM) overnight. The monolayer cells were scratched and incubated with I-BET762 for 1 h, followed by stimulation with TNF-α (10 ng/ml) for 24 h. The cells were fixed and stained using 0.1% crystal violet staining (Sigma-Aldrich, St Louis, MO) for 10 min. Images were captured at 0 and 24 h after wounding using a phase-contrast microscope. The wound area was calculated using the Image J (NIH, version 1.8).

Transwell invasion assay

Chemotaxis analysis was performed using transwell chambers with 8-μm pores (Corning, NY). RA-FLS (4 × 10⁴ cells per chamber) were pre-treated with I-BET762 and plated in the upper chambers of transwell insert of Matrigel as extracellular matrix (ECM) materials (Corning, NY). After 1 h, DMEM containing 5% FBS ad TNF-α (10 ng/ml) was added to the lower chamber as a chemoattractant. After 72 h, migrated cells were stained with 1% crystal violet. Invaded cells were counted in five representative microscopic fields. Images were captured using Leica Application suite software (version 4.12.0) and analyzed with Image J.

Osteoclast differentiation

Peripheral blood mononuclear cell (PBMC) from healthy donors were isolated by Ficoll-Hypaque (Eurobio, Courtaboeuf, France) density-gradient centrifugation. The cells were differentiated into osteoclast with 20 ng/ml macrophage-colony-stimulating factor (M-CSF) and 50 ng/ml receptor activator of NF-κB ligand (RANKL) in α-Minimun Essential Medium (α-MEM) supplemented with 10% FBS and 1% penicillin/streptomycin for 14 days. The medium was replaced every 2 days.

TRAP staining

The cells were fixed with 10% formalin neutral buffer at room temperature for 5 min after 14 days of culture. Cells were stained using tartrate-resistant acid phosphatase (TRAP) staining kit (Cosmo Biotech.co., Tokyo, Japan) according to the manufacturer’s instruction²⁷. The images of TRAP staining were obtained using microscopy (DFC295, Leica, Wetzlar, Germany) and its software LAS V3.8 (Leica).

Actin ring formation assay

To evaluate morphology of actin ring and multi-nucleus, the cells were stained with Phalloidin-TRIC (Sigma-Aldrich; St Louis, MO) and DAPI. The samples were washed with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde (PFA) and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in PBS for 5 min. Cells were washed again in PBS and the cell nuclei were counter stained and mounted with ProLong Gold® (Cell signlaing Technology, Beverly, MA) with DAPI (4,6-diamidin-2-phenylindol). The samples were analyzed using an inverted fluorescence microscope (EVOS FL Cell Imagin system; Life Technologies, Darmastadt, Germany).

Gelatin zymography assay

MMP-9 activity was detected by the gelatin zymography assay using culture supernatant of osteoclast after 14 days. Media from an HT-1080 fibrosarcoma cell line was used as a positive control for MMP-9 activity. Culture supernatants were loaded onto a precast sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) containing 0.1% gelatin. After electrophoresis, the gels were renatured for 30 min at room temperature and then replaced with fresh 1 × zymogram developing buffer (50 nM Tris HCL, pH 8.0, 5 mM CaCl₂, and 0.02% NaN₃) and incubated at 37 ℃. The gels were washed in 2.5% Triton X-100 for 30 min with gentle shaking, followed by 30 min wash in developing buffer. Gels were incubated at 37 ℃ for 24 h in Tris–HCl buffers stained with 0.25% Coomassie blue R-250, and destained in a solution (10% methanol, 10% acetic acid). Proteolysis was detected as a white zone in a blue field. MMP activity was captured using the digital camera and analyzed by Image J software. The activity of MMP-9 was shown as a ratio (%) of MMP-9 activity compared to vehicle.

Statistical analysis

Data are presented as mean ± standard error of the mean (SEM). Statistical analysis was performed using the Mann–Whitney test. p < 0.05 was considered statistically significant. All analyses were performed using GraphPad Prism sofrtware version 8.0 (Graphpad Software Inc., San Diego, CA).

Results

The effect of I-BET762 on the expression of BET bromodomain proteins in RA-FLS in response to TNF-α

First, we obtained synovial biopsy tissues from Seoul National University Hospital. To isolate high RA-FLS yields, synovial biopsy tissues were dissociated by enzymatic digestion combined with mechanical disruption as already described method. Next, RA-FLS proliferate in culture when exposed to TNF- α, which is produced by immune cells present in the inflamed joint^2,7. Cell viability of RA-FLS treated with I-BET762 in the presence of TNF-α remained unchanged, indicating no cytotoxic effects (Supplementary Fig. 1a). I-BET762 at concentrations ranging from 0.1 to 10 μM did not affect RA-FLS cell proliferation at 48, 72, and 96 h (Supplementary Fig. 1b). Blockade of the BET family protein inhibits epigenetic interactions between bromodomains (BRD) and acetylated histones. Histone acetylation is modulated by pro-inflammatory cytokines present in the joints of patients with RA. It is caused by chronic exposure to TNF, triggering the activation of fibroblast-like synoviocytes (FLS)⁹. We investigated the impact of I-BET762 on the expression of BET bromodomain proteins in RA-FLS upon TNF-α treatment. We observed that TNF-α alone increased both mRNA and protein expression of BRD2, BRD3, and BRD4 in RA-FLS. However, in the presence of I-BET762, mRNA levels of BRD2 remained unaltered, while those of BRD3 and BRD4 significantly decreased in RA-FLS (Fig. 1a). This reduction was confirmed at the protein level for BRD3 and BRD4 but not for BRD2 (Fig. 1b and Supplementary Fig. 5a). Since BET inhibitors prevent BRD proteins from binding to acetylated H3 and H4 of euchromatin, it is possible that they also block the binding of c-Myc protein to its acetylated target genes^14,17,38,41. BRD4 and c-Myc promote gene transcription by suppressing the promoter pausing of RNA polymerase, in part by increasing the accessibility of DNA to binding by the BET coactivators⁸. BRD4 and c-Myc function as positive regulatory components of P-TEFb which is a multi-protein complex essential for transcription regulation. To better understand the epigenetic regulation of FLS in the pathogenesis of RA, we investigated c-Myc expression in transcriptional and translational alteration in I-BET762-treated RA-FLS. Both mRNA and protein expression of c-Myc were downregulated by I-BET762 treatment (Fig. 1c,d). Taken together, these alterations of epigenomic landscape may influence the biology of FLS in RA by I-BET762 treatment.

Fig. 1.

I-BET762 modulates expression of BRD2, BRD3, and BRD4, and c-Myc in rheumatoid arthritis synovial fibroblasts. (A) mRNA expression levels of BRD2, BRD3, BRD4, and (C) c-Myc were assessed in RA-FLS following treatment with I-BET762 using quantitative real-time PCR. (B) Protein levels of BRD2, BRD3, BRD4 (n = 4), and (D) c-Myc in RA-FLS were analyzed by western blotting after treatment with I-BET762. Data are represented as the mean ± SEM (n = 3). *p < 0.05, **p < 0.01, and ***p < 0.001. BRD = bromodomain (BRD) and extra-terminal domain (BET) family of proteins, RA-FLS = rheumatoid arthritis fibroblast-like synoviocytes. TNF-α = tumor necrosis factor-α.

I-BET762 suppresses the production of pro-inflammatory mediators in RA-FLS

As reported previously, pro-inflammatory cytokines are sustained by TNF-α induced histone acetylation in RA synovial tissues^3,42–44. In our study, TNF-α stimulation increased the production of IL-6 and IL-8 in RA-FLS. However, I-BET762 treatment significantly downregulated the mRNA expression of IL-6 and IL-8 in RA-FLS at 24 h after TNF-α stimulation (greater than two-fold, p < 0.005, p < 0.01, respectively) (Fig. 2a,b). Additionally, I-BET762 dose-dependently reduced the protein expressions of IL-6, IL-8, and CXCL-10 (p < 0.005, p < 0.01, p < 0.01 respectively) (Fig. 2c–e). Interestingly, mRNA levels of TGF-β and MMP-2 did not change. On the other hand, IL-10 and IL-4 levels tended to increase in RA-FLS (Supplementary Fig. 2). These data suggest that diminished cytokines and chemokines secretory activity are associated with inhibition of the function of acetylated histones by I-BET762 treatment in RA-FLS.

Fig. 2.

I-BET762 suppresses expression of inflammatory cytokines and chemokines in stimulated RA-FLS. RA-FLS were treated with TNF-α in the presence of I-BET762. (A) Relative mRNA expressions of IL-6 and (B) IL-8 were quantified in RA-FLS (n = 3) using quantitative real-time PCR (qPCR). GAPDH was utilized as an internal control for normalization. (C) Concentrations of IL-6, (D) IL-8, and (E) CXCL-10 in the culture supernatants of RA-FLS were measured by ELISA after 48 h of treatment with I-BET762. Data are represented as mean ± SEM, *p < 0.05, **p < 0.01, and ***p < 0.001 compared to the vehicle-treated group. BRD = bromodomain (BRD) and extra-terminal domain (BET) family of proteins, RA-FLS = rheumatoid arthritis fibroblast-like synoviocytes. TNF-α = tumor necrosis factor-α.

I-BET762 suppresses migration and invasion of RA-FLS

FLS from the inflamed joints of patients with advanced RA shows increased expression of adhesion molecules^2,20,36,45. we tested whether adhesion molecules are modulated in I-BET762 treated RA-FLS. I-BET762 attenuated vascular cell adhesion molecule 1(VCAM-1) expression in TNF-α-stimulated RA-FLS (p < 0.005), while intercellular adhesion molecule 1 (ICAM-1) expression remained unaffected (Fig. 3a–c, and Supplementary Fig. 6a). MMPs also promote the disassembly of the type II collagen network, causing biomechanical dysfunction in R^18,19. Above all, MMP-1 and MMP-3 may be able to degrade all the important structural proteins in the extracellular matrix of cartilage. The serum concentrations of MMP-1 and MMP-3 correlate with disease activity and predict the progression of joint destruction in RA^23,25,36. The aggressive phenotype of RA-FLS is characterized by the upregulated expression of MMPs^1,2. MMPs are involved in cartilage degradation, indicating that the cells mediate bone and cartilage damage by collage degradation^32,46. In line with this, we evaluated the effect of I-BET762 on the expression of MMPs in this cell type. The TNF-α stimulation upregulated the mRNA and protein expression of MMP-1 and MMP-3 in RA-FLS (Fig. 3d–g). Pre -treatment with I-BET762 showed significantly decreased the mRNA (Fig. 3d,e) and protein expression of MMP-1 and MMP-3 by 2 -to tenfold (MMP-1; p < 0.001, MMP-3; p < 0.005 respectively) (Fig. 3f,g), while MMP-2 gene expression was not affected (Supplementary Fig. 2d). These data suggest that I-BET762 modulate the tissue-degradation, aggressive behaviour and adhesive ability of RA-FLS. To gain better insight into the regulation of cellular mobility during a TNF-α response in FLS, we determined whether BET bromodomain proteins play a crucial role in the aggressive phenotype of RA-FLS. In wound healing migration assays, I-BET762 inhibited the migration of RA-FLS compared to vehicle control (p < 0.001, Fig. 4a). Importantly, when adding TNF-α to this model, FLS from RA patients increased invasion, whereas I-BET762 treated FLS significantly decreased the invasion and cell-to-cell interaction (p < 0.0001, Fig. 4b). In this result, we observed that the invasive capacity of RA-FLS was significantly inhibited with the same concentration of I-BET762 when compared to the vehicle control using Matrigel basement membrane matrix invasion system. Collectively, I-BET762 effectively suppressed the pro-inflammatory properties of RA-FLS, leading to reduced production of inflammatory cytokines, chemokines, and tissue-destructive enzymes with invasion capacity.

Fig. 3.

I-BET762 suppresses expression matrix metalloproteinase of RA-FLS. (A) Protein expression of ICAM-1 and VCAM-1 in RA-FLS after I-BET762 treatment was assessed by western blot. (B) ICAM-1 expression remained unchanged, while (C) VCAM-1 expression was significantly decreased after I-BET762. Relative mRNA expression levels of (D) MMP-1 and (E) MMP-3 were quantified using qPCR in RA-FLS (n = 3) treated with I-BET762. (F) Concentration of secreted MMP-1 (n = 3) and (G) MMP-3 (n = 4) in culture supernatants of RA-FLS treated with 0.01% DMSO or 5 μM of I-BET762 for 1 h prior to TNF-α (10 ng/mL) stimulation for 48 h were measured by ELISA. Data are represented as the mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001. ICAM-1 = intercellular adhesion molecule-1, VCAM-1 = vascular cell adhesion molecule-1.

Fig. 4.

I-BET762 suppresses migration and invasion of RA-FLS. (A) RA-FLS (n = 5) were stimulated with TNF-α (10 ng/mL) for 48 h in the presence or absence of 5 μM of I-BET762. Migration was calculated by comparing the number of cells that migrated into the wounded areas after I-BET762 treatment with that of control cells. (B) Invasion was assessed using a trans-well invasion assay. RA-FLS (n = 5) were stimulated with TNF-α (10 ng/mL) for 72 h in the presence of 5 μM of I-BET762. Cell images were captured using a Leica upright bright field microscope at 10X magnification. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 vs vehicle control group.

I-BET762 inhibits MAPK and NF-κB signaling pathways in RA-FLS

Next, we assessed the effects of I-BET762 on TNF-α- induced MAPK activation in RA-FLS to gain further insights into mechanism that underlie the sustained inflammatory response of RA-FLS. RA-FLS were pre-treated with I-BET762 and then stimulated with TNF-α. Extracellular signal-regulated kinase (ERK), c-Jun NH2-terminal kinase (JNK), and p38 levels were analyzed in a time-dependent manner (Fig. 5a). I-BET762 did not affect the phosphorylation of ERK or JNK (Fig. 5b,c), but significantly accelerated p38 dephosphorylation at 60 min, with no detectable effect at earlier time points (Fig. 5d).

Fig. 5.

I-BET762 modulates MAPK and NF-κB signaling pathways in RA-FLS. (A) RA-FLS were stimulated with TNF-α in the presence of I-BET762. Phosphorylation of extracellular signal-regulated kinase (p-ERK), c-Jun N-terminal kinase (p-JNK), and p38 (p-p38) were analyzed at 5, 15, 30, or 60 min after activation. β-actin served as the loading control. Representative western blots of MAPK pathway were shown. Densitometry analysis of (B) p-ERK, (C) p-JNK and (D) p-p38 expression normalized to total expression was shown. (E) Phosphorylated NF-κB p65 was assessed by western blotting (n = 3). Histone H3 was used as a loading control. *p < 0.05.

The total protein levels of ERK, JNK, and p38 were not influenced by the I-BET762 treatment (Supplementary Fig. 7a). These data suggests that p38 signaling was activated in inflammation status in RA-FLS, but, I-BET762 treatment accelerated the dephosphorylation of p38.

A sustained TNF-induced NF-κB (RelA/p65) signaling in FLS³ maintains open chromatin and histone acetylation, which contributes to sustained transcription. We investigated whether BET inhibitors could affect the phosphorylation of NF-κB (RelA/p65) in the nucleus after stimulation with TNF-α. I-BET762 depleted the nuclear translocation of p65 in RA-FLS (Fig. 5e, and Supplementary Fig. 7b). Collectively, I-BET762 treatment significantly inhibited p38 phosphorylation and the nuclear translocation of p65 NF-κB in the signal pathway.

I-BET762 dose-dependently inhibits osteoclast differentiation of peripheral blood mononuclear cells

In line with this result of RA-FLS with BET inhibition, we next sought the role of I-BET762 in osteoclastogenesis. We investigated the effect of I-BET762 on the modification of gene expression and cellular functions during osteoclast differentiation in response to M-CSF (20 ng/ml) and RANKL (50 ng/ml) in PBMC. Human PBMC was cultured for osteoclast for 14 days in various concentration of I-BET762 (100, 250, 500, 1000 nM). There was no change in cell viability, as assessed by CCK-8 assays for day 1, day 7, and day 14 (Supplementary Fig. 3a). The concentration of 1 μM I-BET762 was chosen in further studies in osteoclastogenesis. Thus, we moved forward to test whether I-BET762 affected osteoclast formation in response to M-CSF and RANKL. First, we verified that RANKL-induced osteoclast formation was inhibited by I-BET762 treatment in a time-dependent manner (Supplementary Fig. 4a). Next, human PBMC was pre-treated with M-CSF (20 ng/ml) and RANKL (50 ng/ml) for 4 h and then subjected to 1 μM I-BET762. Further quantitative analyses of osteoclast transcriptional gene expression via qRT-PCR analysis revealed that cathepsin K gene expression significantly decreased relative to vehicle control group (Supplementary Fig. 3b) but the nuclear factor of activated T cells, c1 (NFATc1) gene expression did not change in osteoclast. To assess the epigenetic modification effect of the bromodomain proteins in osteoclast formation from human PBMC, we analyzed the RANKL-induced multinucleated osteoclast by TRAP staining. We found that the number and size of osteoclasts increased in control, while I-BET762 treatment decreased osteoclast formation into multi-nucleated TRAP-positive cells in a dose-dependent manner (Fig. 6a). Collectively, these data suggest that I-BET762 suppressed osteoclast formation at gene and protein levels, ruling out a possible effect on the cytotoxic effects during osteoclastogenesis.

Fig. 6.

I-BET762 inhibits osteoclast differentiation of peripheral blood mononuclear cells. (A) Evaluation of the differentiation of peripheral blood mononuclear cell (PBMC) to osteoclasts using bright field microscopy and TRAP staining after 14 days of cell culture with I-BET 762. TRAP-positive, multinucleated (three or more than three nuclei) cells were counted in triplicate. The number of osteoclasts was decreased after treatment with I-BET 762 in a dose-dependent manner. Data are shown as mean ± SEM from four independent donors. * p < 0.05. (B) The cells were cultured for 14 days in various concentrations of I-BET762 (100, 250, 500, 1000 nM). Multinucleated osteoclasts were visualized by staining F-actin with fluorescent-labelled phalloidine (in red) and the nucleus stained with DAPI (in blue). I-BET762 decreased actin ring formation in a dose-dependent manner. The white arrowhead indicates actin rings. (C) MMP-9 was detected as transparent bands on the blue background of a Coomassie blue-stained gel zymography demonstrating the activity of MMP-9 in osteoclast after 14 days of culture. I-BET762 decreased MMP-9 activity in a dose-dependent manner. The gels are representative of three different experiments (n = 3). Their staining densities were evaluated by densitometric analysis. Data are presented as means ± SEM (n = 3). I-BET762 treated osteoclast as compared with vehicle group, * p < 0.05.

I-BET762 suppresses actin ring formation and expression of MMP-9 in osteoclast

To further elucidate whether epigenetic regulatory proteins contribute to the activated phenotype of osteoclast differentiation, we stained with the fluorescent labelled phalloidin and DAPI to visualize F-actin and the nucleus (Fig. 6b). According to previous reports, the actin ring formation in RANKL-induced osteoclast is required for osteoclast bone resorption^32,47. Actin ring formation plays an important role in organizing the bone matrix during osteoclast differentiation. Actin cytoskeletons in osteoclasts are dot-like actin accumulated structures localized in the cell periphery that has a ring-like structure called the actin ring. Our result show that the arrows indicate completely encircled actin cytoskeletal morphology, including filopodia, podosomes, actin belts, and actin ring in the control group. Blockade of BET proteins after I-BET762 treatment in human osteoclasts significantly disrupted podosome belts and fractionated actin rings in a dose-dependent manner, implicating the I-BET762 affected actin modulation against mature osteoclast differentiation in vitro. These data confirm an inhibitory effect on osteoclast formation from human PBMC by I-BET762 treatment.

Recent studies reported that MMP-dependent collagenolysis of the bone extracellular matrix (ECM) increased osteoclast-mediated bone resorption in vitro. Osteoclast-mediated bone resorption is controlled by a network of secreted and membrane-tethered metalloproteinase. Among the MMPs, the gelatinase subfamily, consisting of MMP-2 (gelatinase A) and MMP-9 (gelatinase B), is important in collagen degradation⁴⁸. To evaluate the regulation of gelatinase in osteoclast, the gelatin zymography assay was performed to quantify the activity of bone resorption in the culture supernatants of osteoclast (Fig. 6c, and Supplementary Fig. 8). RANKL-induced osteoclast promoted a significant increase of MMP-9, the secreted proteinase activity. 1 μM I-BET762 treatment significantly diminished MMP-9 activity in a dose-dependent manner upon RANKL induction. These data suggest that targeting the MMP-9 proteinase retard bone resorption in human osteoclasts. Taken together, I-BET762 may play an important role not only in the differentiation of TRAP (+) mononuclear cells but also in the formation of the actin ring and MMP activity.

Discussion

Pan-BRD inhibitor, I-BET762, suppressed the production of pro-inflammatory cytokines chemokines, and MMPs in RA-FLS following stimulation with TNF-α. It was initially designed as a novel drug candidate to attenuate inflammation by dissociating the BET family of proteins from the enhancer region of inflammatory genes. Furthermore, I-BET762 inhibited the migration and invasion of activated RA-FLS in our study. Like BRD protein, which binds to euchromatin through acetylated H3 and H4, c-Myc also could bind to DNA in regions of high acetylation. It is possible that BET inhibitor blocks not only the transcription of the c-Myc gene but also the binding of c-Myc protein to pro-inflammatory genes. In addition, BRD4 and c-Myc promote gene transcription by RNA polymerase II, in part by increasing the accessibility of DNA to binding by the BET coactivators⁴². It is a rationale for targeting BET bromodomains to inhibit c-Myc-dependent transcription. Pan-BET inhibitors led to the displacement of BRD proteins from these enhancer regions, resulting in transcriptional repression. BRD3, BRD4, and c-Myc expression were significantly downregulated by I-BET762 (Fig. 1). These results suggest that suppression of BET proteins recruitment as well as the c-Myc expression by I-BET762 treatment, define the selectivity of I-BET762. Inflamed synovium is characterized by the production of large amounts of inflammatory cytokines by activated immune cells, leading to the transformation of synoviocytes into cancer-cell-like RA-FLS. This process is associated with epigenetic modifications, including persistent H3K27 acetylation, which increases chromatin accessibility and sustains the expression of TNF-inducible genes in RA-FLS^7,49. Thus, targeting epigenetic regulatory proteins such as BET presents a potential therapeutic approach to reverse the pro-inflammatory response of RA-FLS and immune cells, thereby reducing articular inflammation. Our study demonstrates that I-BET762 effectively suppresses the production of key inflammatory cytokines and MMPs in RA-FLS by TNF-α stimulation. Similar to previous reports, it reduced cell mobility and invasion, partly achieved by downregulating VCAM-1 expression (Fig. 3c), an essential molecule for cell mobility^7,13,49. Especially, the BRD4 of BET proteins binds the acetylated lysin-310 of p65, regulating the transcriptional activity of NF-κ^21,23,28. MAPKs, including ERK, JNK, and p38 play critical roles in regulating various cellular functions, including cell growth, differentiation, and survival, in response to inflammatory stimuli^16,45,50. These pathways are highly activated in RA-FLS. Numerous reports have been shown that theses inflammatory signaling are highly activated in immune mediated disease. Using the genetic and chemical approaches, p38 inhibitors have shown efficacy in humans, but concerns about drug safety, such as hepatotoxicity, limit their clinical use^15,51. According to our data, I-BET762 promote the dephosphorylation of p38, but not that of ERK and JNK, suggesting that the inhibitory effect of I-BET762 may be mediated, in part, by suppressing p38 activation in a time-dependent manner. Our data also showed that blocking of BET proteins contributed to regulate NF-κB activation in the nucleus and suppress MAPK/p38 pathway in cytoplasm (Fig. 5). The c-Myc, a proto-oncogenic transcription factor, not only promote cell growth and proliferation but inhibits apoptosis in RA-FLS. Its upregulation may be linked to NF-κB signaling in RA-FLS. Furthermore, c-Myc and NFATc1 also promote osteoclastogenesis, and bone resorption, which may be decreased by I-BET151 treatment^32,46. In particular, a novel BRD4 inhibitor was reported to suppress osteoclastogenesis and ovariectomized osteoporosis by blocking RANKL-mediated MAPK and NF-κB pathways^52,53. We showed that the osteoclastogenesis and actin ring formation repressed by I-BET762 treatment (Fig. 6).

Our study supports the therapeutic potential of epigenetic inhibitor in RA, however, several limitations should be acknowledged. First, the limited sample size of patient- derived data restricts the ability to draw definitive conclusions. Second, the therapeutic effects of I-BET762 were not validated in an in vivo RA model, limiting the assessment of its efficacy under physiological conditions. Although in vitro experiments were conducted to investigate cellular and molecular mechanisms, such models do not fully recapitulate the complex physiological and immunological environment of living organisms, and thus the findings may not directly translate to in vivo settings. Future studies should incorporate appropriate animal models, such as the collagen-induced arthritis (CIA) model, to further evaluate the therapeutic potential of I-BET762. Finally, comprehensive preclinical studies are essential to assess the feasibility, safety, and translational relevance of candidate therapies prior to clinical application.

Although further studies are necessary to confirm the efficacy and safety of I-BET762, our results suggest that targeting epigenetic regulation with I-BET762 may offer a potential therapeutic strategy for RA. In conclusion, I-BET762 suggests inhibition of the pro-inflammatory response in RA-FLS and osteoclastogenesis through p38 MAPK and NF-κB pathway. This study supports the potential therapeutic value of I-BET762 in the treatment of RA.

Author contributions

Y.W.S conceived the overall study design. R.H.K. carried out the experimental data, performed the analysis, and wrote the manuscript with support from Y.W.S. S.U.C. supervised the project and all authors commented on the manuscript. Y.W.S and R.H.K authors contributed to the final version of the paper.

Data availability

All data generated or analysed during this work are included in this published article and the original raw western films are available in the supplementary information.

Declarations

Competing interests

The authors declare no competing interests.