Angiogenesis in rheumatoid arthritis is fostered directly by toll-like receptor 5 ligation and indirectly through interleukin-17 induction

Abstract

Objective: To examine the impact of Toll-like receptor 5 (TLR-5) on endothelial cell function in rheumatoid arthritis (RA) and vascularization in collagen-induced arthritis (CIA).

Methods: Endothelial cell migration and tube formation assays were used to demonstrate the direct role of TLR-5 ligation in angiogenesis. Mice with CIA were treated with the TLR-5 agonist flagellin to document the effect of TLR-5 ligation in RA pathology. Vascularization in CIA was determined by immunohistochemical analysis and determination of cytokine levels in ankle joints. Spleen Th17 cells and joint interleukin-17 (IL-17) were quantified by fluorescence-activated cell sorting analysis and enzyme-linked immunosorbent assay. The development of Th17 cells induced by TLR-5 ligation was validated in RA peripheral blood mononuclear cells.

Results: Ligation of TLR-5 to endogenous ligands expressed in RA synovial fluid contributed to endothelial cell infiltration and tube formation. Furthermore, treatment with flagellin after the onset of CIA exacerbated joint inflammation; in contrast, inflammation in control mice remained at a plateau phase. We showed that TLR-5-enhanced disease severity was attributable to Th17 cell differentiation and joint vascularization in CIA. Examination of the underlying mechanism using RA peripheral blood mononuclear cells documented that ligation of TLR-5 in myeloid cells and production of Th17-promoting cytokines were necessary for Th17 cell polarization. Additionally, we demonstrated that blockade of the IL-17 cascade markedly reduced endothelial cell migration activated by flagellin-conditioned medium, suggesting that TLR-5 ligation can mediate RA angiogenesis either directly by attracting endothelial cells or indirectly by fostering Th17 cell development.

Conclusion: Our data demonstrate a novel role for TLR-5 in RA angiogenesis; thus, TLR-5 may be a promising new target for RA treatment.

Figure 1. TLR5 ligation promotes endothelial cell migration and tube formation

A. Flagellin induced endothelial chemotaxis was performed in a Boyden chamber with varying concentrations, n=3. Endothelial cells were preincubated by anti-TLR5 (10 µg/ml) for 1h prior to examining endothelial migration (B) or tube formation (C) in response to RA synovial fluid (SF) (1:20 dilution) (n=12 RA SFs) or flagellin (10 ng/ml) (n=3). D. Photomicrographs taken of representative wells treated with PBS, FGF (20 ng/ml), RA SF+IgG, RA SF+anti-TLR5, flagellin+IgG or flagellin+anti-TLR5. E. Cells were stimulated with flagellin (10 ng/ml) for 0–65 minutes, and the cell lysates were probed for pAKT, pERK, p-p38 or degradation of IκB, n=3. F. Cells were treated with DMSO or 1 or 5µM chemical inhibitors for PI3K (LY294002) or ERK (PD98059), p38 (SB203580) and BAY (11-7082) for 2h prior to performing endothelial chemotaxis in response to flagellin, n=3. Values are the mean ± SE. * represents p <0.05.

Figure 2. CIA disease severity and vascularization are markedly enhanced by post onset treatment of TLR5 agonist

A. Ankles harvested from PBS control or CIA induced mice were stained with anti-TLR5 antibody. B. TLR5 positive immunostaining was scored on a 0–5 scale on synovial tissue lining, sublining macrophages (mac) and endothelial cells (endo) (n=5–7) (original magnification × 200). C. Changes in joint circumference was recorded in CIA mice that were treated with PBS or flagellin (20 µg) i.p., n=10 mice. D. Demonstrates the representative H&E staining in control (PBS) and flagellin (Flag) treated CIA mice (original magnification × 200). E. Effect of flagellin post onset treatment on CIA ankle joint inflammation, lining thickness and bone erosion was scored on a 0–5 scale, n=6–8. F. Hemoglobin levels are quantified in CIA ankles harvested from control and flagellin treatment groups on day 57 and results are demonstrated as hemoglobin (g/dl). G. CIA synovial tissue treated with control (PBS) or flagellin (Flag) was harvested on day 57 and was immunostained with anti-VWF antibody (original magnification × 200) and (H) endothelial staining was quantified on a 0–5 scale, n=6–9. VWF+ staining in the CIA synovial tissues is demonstrated by arrows. Values are the mean ± SE. * represents p <0.05.

Figure 3. IL-17 and its associated factors are elevated in flagellin compared to control treatment in CIA mice

Changes in joint IL-17 (A), IL-6 (C), IL-1β (D), CCL20 (E) VEGF (F) protein levels in CIA mice treated with control or flagellin were quantified by ELISA, n=6–8. B. Total number of TH-17 positive cells per spleen was determined by Flow cytometry analysis in CIA mice treated with control or flagellin, n=5. Values are mean ± SE. * indicates p<0.05.

Figure 4. Like in CIA, TLR5 ligation drives TH-17 cell differentiation in normal peripheral blood mononuclear cells

Normal peripheral blood mononuclear cells were either untreated or treated with a mixture of IL-6, IL-1β, IL-23 (10 ng/ml) and TGF-β (5 ng/ml), LPS (100 ng/ml) or flagellin (10 and 100 ng/ml; InvivoGen) for 5 days. Cell condition media was harvested and cells were thereafter treated with 4h of PMA (50 ng/ml), ionomycin (750 ng/ml) and Brefeldin A (3 µg/ml) to quantify TH-17 cells or IL-17 and its related cytokines, n=3. A. Percent TH-17 cells was determined in normal peripheral blood mononuclear cells differentially treated as detailed above. B. Demonstrates representative Flow cytometry histogram of the data presented in A. Protein levels of IL-17 (C), IL-6 (D) and IL-1β (E) was determined in the condition media of the normal peripheral blood mononuclear cells treated as mentioned above employing ELISA, n=3. Values are mean ± SE or data obtained from each individual experiment. * indicates p<0.05.

Figure 5. Myeloid TLR5 ligation and their cell to cell interaction with CD4+ T cells promote TH-17 cell differentiation

A. RA peripheral blood mononuclear cells were treated similar to normal cells (except for IL-6 and IL-1β treatment) as detailed in Fig. 4 and % TH-17 cells was determined by FACS analysis, n=3. B. Demonstrates representative FACS histogram of the data presented in A. RA CD4+ T cells and CD16+ monocytes were negatively selected and were either cultured alone or together. Subsequently cells were treated with PBS or flagellin (100 ng/ml) for 5 days and % TH-17 cells (C–D) or concentration of IL-17 (E) in supernatants was determined by FACS analysis (n-3) or ELISA (n=6). IL-17 protein levels were not determined (N.D.) in RA myeloid cells alone (E). Values are mean ± SE or data obtained from each experiment. * indicates p<0.05.

Figure 6. Myeloid cell production of IL-6 and IL-1β is responsible for TLR5 mediated TH-17 cell differentiation and blockade of IL-17 pathway reduces flagellin mediated endothelial chemotaxis

Peripheral blood mononuclear cells were either untreated or treated with (10 µg/ml) IgG control, anti-IL-6, anti-IL-1β or anti-IL-6 plus anti-IL-1β antibodies prior to adding flagellin (100 ng/ml) for 5 days. PBS and LPS (100 ng/ml) treatment were used at negative and positive controls. Concentration of IL-17 in supernatants (A) and % TH-17 cells (B) was determined by ELISA or FACS analysis, n=9. C. Migration of endothelial cells untreated (IgG control) or treated with antibodies to IL-17RA or IL-17RC (10 µg/ml) was examined in response to IL-17 (2.5 ng/ml) or flagellin (100 ng/ml) activated peripheral blood mononuclear cell supernatants, n=3. Endothelial cell chemotaxis was also determined in response to flagellin condition media or IL-17 (2.5 ng/ml) immunoneutralized with anti-IL-17 or IgG control (10µg/ml). Values are mean ± SE or data obtained from each experiment. * indicates p<0.05.