A novel diindolylmethane analog, 1,1-bis(3'-indolyl)-1-(p-chlorophenyl) methane, inhibits the tumor necrosis factor-induced inflammatory response in primary murine synovial fibroblasts through a Nurr1-dependent mechanism

Abstract

The progression of rheumatoid arthritis involves the thickening of the synovial lining due to the proliferation of fibroblast-like synoviocytes (FLS) and infiltration by inflammatory cells. Tumor necrosis factor alpha (TNFα) is a pro-inflammatory cytokine involved in progression of the disease. Under rheumatoid conditions, FLS express the tumor necrosis factor (TNF)-recognition complex (TNFR1, TNFR2, VCAM-1 and ICAM-1), which induces local macrophage activation and leads to downstream nuclear factor κB (NF-κB) signaling. The NF-κB-regulated inflammatory gene, cyclooxygenase (COX), increases synthesis of prostaglandins that contribute to the propagation of inflammatory damage within the joint. Because the nuclear orphan receptor, NR4A2 (Nurr1), can negatively regulate NF-κB-dependent inflammatory gene expression in macrophages, we postulated that activation of this receptor by the Nurr1 ligand 1,1-bis(3'-indolyl)-1-(p-chlorophenyl) methane (C-DIM12) would modulate inflammatory gene expression in synovial fibroblasts by inhibiting NF-κB. Treatment with C-DIM12 suppressed TNFα-induced expression of adhesion molecules and NF-κB regulated genes in primary synovial fibroblasts including vascular adhesion molecule 1 (VCAM-1), PGE2 and COX-2. Immunofluorescence studies indicated that C-DIM12 did not prevent translocation of p65 and stabilized nuclear localization of Nurr1 in synovial fibroblasts. Knockdown of Nurr1 expression by RNA interference prevented the inhibitory effects of C-DIM12 on inflammatory gene expression, indicating that the anti-inflammatory effects of this compound are Nurr1-dependent. Collectively, these data suggest that this receptor may be a viable therapeutic target in RA.

Keywords: C-DIM12; NF-κB; Nurr1; Rheumatoid arthritis; Synovial fibroblast; TNFα.

Figure 1. Immunophenotyping of Primary Synovial Fibroblasts

Isolated synovial fibroblasts from WT mice were subjected to FACS analysis on passage 3 after seeding to verify purity. (a) negative control for dual stain cells (b) Density plot showing expression of CD90.2 on the majority of the isolated cells (>90 %) whereas very few Mac-1 positive myeloid cells can be detected (<1%). (c) FS Lin vs VCAM-1 PE scatter plot showing cluster positive for VCAM-1 expression (d) Overlaid histogram showing VCAM-1 (~96%) (e) FS Lin vs ICAM-1 PE scatter plot showing cluster positive for ICAM-1 expression (f) ICAM-1 expression (87%) (blue shaded area) on SFs or unstained control (red shaded area) (g) representative image for Vimentin expression in synovial fibroblasts.

Figure 2. TNFα induced VCAM-1 and ICAM-1 expression in synovial fibroblasts

(a) TNFα dose-response of VCAM-1 expression was determined to be most optimal at 12 h (b) Primary synovial fibroblasts were cultured in a dose dependent manner with TNFα (10 ng/ml) for 12hr. VCAM-1 surface expression was determined by FACS, as described in Methods. TNFα-induced (c) VCAM-1 and (d) I-CAM1 expression is decreased upon C-DIM12 treatment. Data are expressed as mean ± S.E.M (n=4). Statistical significance is compared with saline control. *P < 0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 3. C-DIM12 Decreases Inflammatory PGE₂ and COX-2 Expression in Synovial Fibroblasts

(a) PGE2 dose response determined after 24 hrs of TNFα (10ng/ml) stimulation (b) Upon TNFα (10ng/ml) exposure, PGE2 expression is increased and effectively suppressed upon 10μM C-DIM12 treatment. (c) Overlaid histogram showing COX-2 expression in stimulated synovial fibroblast fluorescence intensity shifted to the right increase COX- and followed by treatment of C-DIM12 (10 μM) decrease back to basal condition. (d) Synovial fibroblasts were stimulated with TNFα (10ng/ml) treated with 10μM CDIM12 analyzed for COX-2 expression. Data are expressed as mean ± S.E.M (n=4). Statistical significance is compared with saline control. *P < 0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 4. C-DIM12 decrease IκBα and p65 NF-κB Expression in Synovial Fibroblasts

(a and c) The dose response of IκBα and NF-κB expression was determined 30 mins after TNFα (10 ng/ml). (b and d) synovial fibroblasts were stimulated with TNFα (10ng/ml) treated with 10μM CDIM12 analyzed by flow cytometry. Data are expressed as mean ± S.E.M (n=4). Statistical significance is compared with saline control. *P < 0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 5. C-DIM12 does not prevent p65 translocation and promotes Nurr1 sequestration in the nuclei of TNFα-induced primary synovial fibroblast cells

Primary synovial fibroblasts were treated with 10μm C-DIM12 for 1 hour followed by saline or 10 ng/ml of TNFα for 30 mins and fixed for immunofluorescence to examine p65 and Nurr1 translocation with DAPI(blue), phalloidan (green), p65 (red). (a–b) Representative images and quantification of p65 nuclear expression was quantified by mean fluorescence intensity encompassing the nuclei (DAPI boundary; background subtracted), demonstrating increased nuclear p65 upon TNFα with or without C-DIM12(c–d). Representative images and quantification of Nurr1 expression following TNFα and C-DIM12 treatment showed increased Nurr1 protein translocation from the cytoplasm to the nucleus. Data are expressed as mean ± S.E.M. (n=3). *P < 0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 6. C-DIM12-dependent inhibition of inflammatory gene expression requires Nurr1

(a) Synovial fibroblasts were treated with concentrations of siRNA (denoted RNAi) or scrambled RNAi (control RNAi) demonstrating 1.2ug to be optimal. Synovial fibroblasts were transfected with Nurr1 RNAi or control RNAi for 24 hours followed by saline or 10 ng/ml TNFα with or without C-DIM12 for 24 hours and assessed for Nurr1 expression via flow cytometry. (b) Nurr1 was effectively knocked-down at a protein level, even in the presence of TNFα. (c–f) IL-6, MCP-1, IL-10 and IFN-γ protein expression upon knock-down and treatment was assessed via flow cytometry demonstrating a loss of C-DIM12-mediated suppression upon Nurr1 knock-down. Data are expressed as mean ± S.E.M (n=4). Statistical significance is compared with saline control. *P < 0.05, **P<0.01, ***P<0.001, ****P<0.0001.