Anticitrullinated protein antibodies facilitate migration of synovial tissue-derived fibroblasts

Figures

• 

  Figure 1. Increased mobility and adherence of synovial fibroblasts in presence of polyclonal ACPAs. Cell mobility was analysed during a period of 6 hours in the presence of 1 µg/mL polyclonal ACPA IgGs (ACPA) or non-ACPA control IgGs (IgG) or without antibody treatment (-) in both non-starved (A) and starved (B) fibroblast cultures. The graphs represent mean±SD values obtained from 10 independent experiments, using cells of 10 different patients and 6 replicates for each treatment. Dot line indicate migration index of non-treated FLS. Image-based evaluation of cell migration in starved fibroblast cultures were performed using phase contrast microscopy with 10x magnification (C). Real-time cell migration was measured in the presence of 1 µg/mL ACPA or control IgG or without any treatment in non-starved (D) and starved (E) fibroblast cultures using IncuCyte. The graphs represent mean±SD values obtained by using cells from 10 individuals patients and 6 replicates in 10 independent experiments. A titration experiment was performed on starved synovial fibroblasts with presence of ACPA (F) and anti-PR3 (G) with indicated concentrations. (H) Cell mobility were tested on starved FLS with presence…

• 

  Figure 2. Starvation-induced protein citrullination is essential for synovial fibroblast migration induced by ACPAs. Confocal microscopy images visualise PAD-2 (green) and PAD-4 (red) expressions as well as polyclonal ACPA IgGs (ACPA) and non-ACPA control IgG (IgG) binding patterns in FLS cultures following serum withdrawal (starved) or in non-starved FLS, in the presence of 10% fetal bovine serum (A). An original magnification of 400x was used for all images obtained with confocal microscopy and nuclei are shown in blue. Starved FLS cultures were also stained with ACPAs (red colour) pre-incubated overnight with citrullinated (cit) or native fibrinogen with ratio of protein:antibody (10:1) (cit-Fib or Arg-Fib, respectively) or in the absence of fibrinogen (B). To block ACPA-induced migration, FLS migration was compared in the presence of 1 µg/mL polyclonal ACPAs and control IgG or with both antibodies pre-incubated with citrullinated or native fibrinogen in the indicated ratio overnight. Migration index was calculated from three independent experiments, using cells of three individual patients and six replicates for each treatment, mean±SD values are shown (C). Similar blocking experiments…

• 

  Figure 3. Role of PI3K in the ACPA-induce synovial fibroblast migration. Starved FLS were exposed to 1 µg/mL polyclonal ACPA IgG (ACPA) or non-ACPA control IgG (IgG) or to TNF 10 ng/mL. Immunoblot analysis of Akt (Thr308 or Ser473 residues), ERK1/2, JNK and p38 phosphorylation are shown, whereas GAPDH protein was detected as control. The presented data are representative for at least three independent experiments for all tested proteins (A). Change of Akt phosphorylation at the Thr308 residue is shown in response to 1 µg/mL ACPA or IgG treatments (B). The data were calculated following densitometry analysis in five independent experiments, levels were normalised to baseline intensities, mean±SD values are shown. FLS were pretreated with 200 nM of the PI3K inhibitor wortmannin (C) or 2 µg/mL of the PTEN inhibitor SF-1670 (D) for 2 hours prior to the analysis of cell migration in the presence of 1 µg/mL ACPA or control IgG. DMSO (0.01%) was used as solvent control. Graphs represent fold-change in cell mobility, mean±SD values were calculated from three independent experiments, *P<0.05. ACPA, anticitrullinated protein/peptide antibody; DMSO, dimethyl sulfoxide; FLS, fibroblast-like…

• 

  Figure 4. Collaboration of ACPA-mediated signals with the inflammatory cytokines IL-8 and TNF in inducing FLS migration. CXCR1 and CXCR2 expressions in FLS cultures were analysed with immunohistochemistry by light microscopy with the original magnification of 250x (left panel) and flow cytometry (right panel) (A). Migration of non-starved (B) and starved FLS (C) were assessed after exposing the cells to suboptimal (0.5 µg/mL) dose of ACPA or IgG in the presence or absence of 10 ng/mL recombinant human IL-8. Mean±SD values were calculated from three independent experiments, using cells of five individual patients and three replicates, *p<0.05. Starved FLS mobility was analysed by combining high dose (10 ng/mL) of IL-8 with increasing ACPA concentrations (D) or suboptimal concentration of ACPA with increasing concentrations of IL-8 (E). On similar grounds, increasing concentrations of TNF were combined with ACPA applied at a suboptimal dose (F). Mean±SD values were calculated based on at three independent experiments, using cells of three different patients and six sample replicates, *p<0.05 (G). Confocal microscopy images show increased PAD-4 (red colour) and PAD-2 (green colour) expression…

• 

  Figure 5. The monoclonal ACPAs 1325:01B09 and 1325:04C03 display selective binding and stimulatory capacities on FLS and osteoclasts. The effects of eight individual ACPA clones were tested in concentration of 1 µg/mL on fibroblast migration (A) and osteoclast differentiation (B) assays. The graphs represent migration index expressed as fold-difference normalised to untreated controls using data from three independent experiments, using cells of three different patients, all performed six replicate samples, *p<0.05. Osteoclast counts expressed in fold-difference normalised to untreated controls using data from two to five independent experiments, performed at least three replicates, *p<0.05. IHC stainings show binding capacity of individual ACPA clones, 1325:01B09 and 1325:04C03 to FLS and osteoclasts (C). Confocal microscopy images illustrate the binding of the ACPA clones (1325:01B09, 1325:04C03) and control (1362:01E02) monoclonal antibodies (red colour) to starved FLS (D). To block monoclonal ACPA binding, starved FLS were also stained with 1325:01B09 (red colour) pre-incubated with citrullinated (cit) or native fibrinogen (cit-Fib or Arg-Fib, respectively) or in the absence of…

• 

  Figure 6. The monoclonal ACPA clones 1325:01B09 and 1325:04C03 bind to synovial targets in patients with RA but not in healthy controls. The immunohistochemistry images illustrate binding patterns of the monoclonal ACPA clones 1325:01B09 and 1325:04C03 to synovial tissues obtained from healthy donors (A) or patients with RA (B). Mouse IgG1 and the non-citrulline-specific 1362:01E02 (E02) monoclonal antibody were used as controls. CD55 and podoplanin (PDPN) stainings highlight FLS-rich areas of the synovial membrane. Immunohistochemistry stainings were analysed with light microscopy using 250x original magnification. (C) Confocal microscopy images illustrate a partial colocalisation of the ACPA clone 1325:01B09 with CD55 (C) and PDPN (D), highlighted by closed arrows. Stainings with the monoclonal ACPA 1325:04C03 and the control 1362:01E02 antibodies are also shown. Moreover, partially colocalisation of 1325:01B09 F(ab’)2 with PDPN were highlighted by arrows (E). An original magnification of 400x was used for confocal microscopy. ACPA, anticitrullinated protein/peptide antibody; DAPI, 4,6-diamidino-2-phenylindole; FLS, fibroblast-like synoviocytes; RA, rheumatoid arthritis.