Characterisation of fibroblast-like synoviocytes from a murine model of joint inflammation

Abstract

Introduction: Fibroblast-like synoviocytes (FLS) play a central role in defining the stromal environment in inflammatory joint diseases. Despite a growing use of FLS isolated from murine inflammatory models, a detailed characterisation of these cells has not been performed.

Methods: In this study, FLS were isolated from inflamed joints of mice expressing both the T cell receptor transgene KRN and the MHC class II molecule Ag7 (K/BxN mice) and their purity in culture determined by immunofluorescence and real-time reverse transcription polymerase chain reaction (real-time RT-PCR). Basal expression of proinflammatory genes was determined by real-time RT-PCR. Secreted interleukin 6 (IL-6) was measured by enzyme-linked immunosorbent assay (ELISA), and its regulation by tumor necrosis factor-alpha (TNF-α and corticosterone (the major glucocorticoid in rodents) measured relative to other mesenchymal cell populations.

Results: Purity of FLS culture was identified by positive expression of fibronectin, prolyl 4-hydroxylase, cluster of differentiation 90.2 (CD90.2) and 248 (CD248) in greater than 98% of the population. Cultured FLS were able to migrate and invade through matrigel, a process enhanced in the presence of TNF-α. FLS isolated from K/BxN mice possessed significantly greater basal expression of the inflammatory markers IL-6, chemokine ligand 2 (CCL-2) and vascular cell adhesion molecule 1 (VCAM-1) when compared to FLS isolated from non-inflamed tissue (IL-6, 3.6 fold; CCL-2, 11.2 fold; VCAM-1, 9 fold; P<0.05). This elevated expression was abrogated in the presence of corticosterone at 100 nmol/l. TNF-α significantly increased expression of all inflammatory markers to a much greater degree in K/BxN FLS relative to other mesenchymal cell lines (K/BxN; IL-6, 40.8 fold; CCL-2, 1343.2 fold; VCAM-1, 17.8 fold; ICAM-1, 13.8 fold; P<0.05), with secreted IL-6 mirroring these results (K/BxN; con, 169±29.7 versus TNF-α, 923±378.8 pg/ml/1×10⁵ cells; P<0.05). Dose response experiments confirmed effective concentrations between 10 and 100 nmol/l for corticosterone and 1 and 10 ng/ml for TNF-α, whilst inflammatory gene expression in FLS was shown to be stable between passages four and seven.

Conclusions: This study has established a well characterised set of key inflammatory genes for in vitro FLS culture, isolated from K/BxN mice and non-inflamed wild-type controls. Their response to both pro- and anti-inflammatory signalling has been assessed and shown to strongly resemble that which is seen in human FLS culture. Additionally, this study provides guidelines for the effective characterisation, duration and treatment of murine FLS culture.

Figure 1

Validation of FLS culture. (a) Confluent monolayer of K/BxN (fibroblast-like synoviocytes) FLS as observed in vitro at 10× magnification. (b) mRNA expression of 18S, P4H, CD248, osteocalcin (OCN) and CD68 determined by standard RT-PCR at 35 cycles in one non-inflamed control FLS wild-type (WT), three K/BxN FLS lines (FLS 1 to 3), primary calvarial osteoblasts (OBs) and the macrophage cell line RAW 264.7 (MΦ). Expression of CD248 (blue), CD90.2 (magenta) and fibronectin (red) in non-inflamed FLS (c, e, g) and K/BxN FLS (d, f, h), determined by confocal fluorescence immunohistochemistry. (i) Invasion of FLS across matrigel-coated transwell inserts relative to untreated non-inflamed control FLS in the presence or absence of TNF-α (ng/ml). (j) ΔCt mRNA expression of cadherin-11 normalised for the housekeeping gene 18s. Data presented are from three individual WT control FLS and three K/BxN FLS after loading 25 ng of mRNA for RT-PCR. For Figure 1 c-h, results shown are representative of three separate K/BxN FLS lines. Figure 1b has been cropped, to allow presentation of multiple gels.

Figure 2

Synovial CD248 and fibronectin expression in vivo. Decalcified paraffin-embedded joint sections were stained for CD248 and fibronectin and counterstained with Gill's haematoxylin. Staining was examined in non-inflamed control (a, d) and inflamed K/BxN (b, e) joints. (c, f) demonstrates the inflamed synovium of K/BxN joints at increased magnification. Black arrows denote positive staining of CD248 and fibronectin respectively. Bars = 100 μm.

Figure 3

Inflammatory gene regulation by corticosteone and TNF-α in FLS. Fold change in mRNA expression of inflammatory genes in (fibroblast-like synoviocytes) FLS, determined by real-time RT-PCR. Expression of mRNA was measured at 16 hr for IL-6, chemokine ligand 2 (CCL-2), vascular cell adhesion molecule 1 (VCAM-1) and intercellular adhesion molecule 1 (ICAM-1) following pretreatment with either corticosterone (a-d) (100 nmol/l) or TNFα (e-h) (10 ng/ml) for 24 hr. Data were normalized for levels of the housekeeping gene 18S rRNA and presented as fold change in expression (± standard error) relative to untreated wild-type (WT) control FLS. *P < 0.05, **P < 0.001 versus respective untreated control; #P > 0.05 versus untreated WT control FLS. Results shown are the combined duplicates of three separate FLS lines and two WT control FLS lines.

Figure 4

Regulation of IL-6 in FLS. Dose response analysis of IL-6 mRNA expression, determined by RT real-time PCR in (fibroblast-like synoviocytes) FLS isolated from K/BxN mice following treatment with (a) corticosterone (0, 1, 10, 100, 500, 1000 nmol/l) or (b) TNFα (0, 0.1, 1, 5, 10, 25 ng/ml). (c) Fold change in IL-6 mRNA expression in wild-type (WT) control FLS, C2C12, MC3T3-E1 and K/BxN FLS, determined by real-time RT-PCR. All mRNA data were normalized for levels of the housekeeping gene 18S rRNA and presented as fold change in expression (± standard error) relative to either untreated control or untreated WT con FLS. (d) IL-6 secretion into culture media (pg/ml/100000 cells, ± standard error) in WT con FLS, C2C12, MC3T3-E1 and K/BxN FLS, determined by specific ELISA. Both mRNA and conditioned media were collected at 16 hr following treatment with either control, TNFα (10 ng/ml) or corticosterone (100 nmol/l). *P < 0.05, **P < 0.001 versus respective untreated controls; #P < 0.05 versus untreated WT control FLS. Dose response experiments are the combined duplicates of two K/BxN FLS. All other data are the combined duplicates of three separate FLS lines, two WT control FLS lines, two C2C12 repeats and two MC3T3-E1 repeats.

Figure 5

Regulation of inflammatory markers over prolonged culture. Fold change in mRNA expression of inflammatory genes in fibroblast-like synoviocytes (FLS), determined by real-time RT-PCR between passages 2 and 8 (P2 to P8). Expression of mRNA was measured at 16 hr for (a) IL-6, (b) chemokine ligand 2 (CCL-2), (c) vascular cell adhesion molecule 1 (VCAM-1) and (d) intercellular adhesion molecule 1 (ICAM-1). For each gene product data were normalized for levels of the housekeeping gene 18S rRNA and are presented as fold change in expression (± standard error) relative to the P2 control. *P < 0.05 versus passage 2. #P < 0.05 versus passage 3. Results shown are the combined duplicates of two K/BxN FLS.