Interleukin‐21 induces migration and invasion of fibroblast‐like synoviocytes from patients with rheumatoid arthritis

Summary

Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by synovial fibroblast hyperplasia and bone erosion. Fibroblast‐like synoviocytes (FLS) play a pivotal role in RA pathogenesis through aggressive migration and matrix invasion, and certain proinflammatory cytokines may affect synoviocyte invasion. Whether interleukin (IL)‐21 influences this process remains controversial. Here, we evaluated the potential regulatory effect of IL‐21 on the migration, invasion and matrix metalloproteinase (MMP) expression in RA‐FLS. We found that IL‐21 promoted the migration, invasion and MMP (MMP‐2, MMP‐3, MMP‐9, MMP‐13) production in RA‐FLS. Moreover, IL‐21 induced activation of the phosphoinositide 3‐kinase (PI3K), signal transducer and activator of transcription‐3 (STAT‐3) and extracellular signal‐regulated protein kinases 1 and 2 (ERK1/2) pathways, and blockage of these pathways [PI3K/protein kinase B (AKT) inhibitor LY294002, STAT‐3 inhibitor STA‐21 and ERK1/2 inhibitor PD98059] attenuated IL‐21‐induced migration and secretion of MMP‐3 and MMP‐9. In conclusion, our results suggest that IL‐21 promotes migration and invasion of RA‐FLS. Therefore, therapeutic strategies targeting IL‐21 might be effective for the treatment of RA.

Introduction

Rheumatoid arthritis (RA) is a chronic autoimmune disease of the joints characterized by synovial inflammation, hyperplasia of synovial tissues, synovial pannus and cartilage and bone destruction 1. Among the various pathological events in the synovium, fibroblast‐like synoviocytes (FLS) are reported to play a pivotal role. The FLS from RA patients (RA‐FLS) are stimulated to migrate from affected synovium to healthy synovium 2, contributing to the spread of arthritis and destruction of distant joints 3. RA‐FLS invade the extracellular matrix (ECM) directly and secrete matrix metalloproteinases (MMPs) into synovial fluid, which destroys cartilage and bone and exacerbates joint damage 4. Cytokines play a critical role in the onset and progression of RA. Several proinflammatory cytokines, such as tumour necrosis factor‐(TNF)‐α and interleukin (IL)‐6, are essential to RA development owing to their role in synoviocyte activation 5. Regulating these proinflammatory cytokines may prevent disease progression, and are candidate RA therapies.

IL‐21 is a member of the IL‐2 family of cytokines, and binds the IL‐21 receptor (IL‐21R), a heterodimer composed of IL‐21R and the common γ‐chain that is shared with receptors specific for IL‐2, IL‐4, IL‐7, IL‐9 and IL‐15 6. IL‐21 is expressed primarily by CD4⁺ T cells, including T helper type 1 (Th1), Th2, Th17 and natural killer (NK) T cells 7, 8, 9. Notably, IL‐21R is expressed highly by RA‐FLS in rheumatoid synovium tissue 10. Increased IL‐21 plasma levels are associated with enhanced disease activity and radiographic status in RA patients 11. Inhibition of IL‐21 with the IL‐21 receptor Fc fusion protein (IL‐21R.Fc) reduces inflammatory cytokine production and attenuates the progression of arthritis in collagen‐induced arthritis (CIA) animal models 12. In contrast, IL‐21R deficiency protects against severe inflammation and joint destruction in streptococcal cell wall arthritis 13.

A previous study has demonstrated that IL‐21 promotes MMP‐1, MMP‐2, MMP‐3 and MMP‐9 expression in intestinal fibroblasts 14. MMPs play important roles in the degradation of basement membranes and ECM, and are crucial for the migration and invasion of many types of cells 15, 16. However, we still know little about the effect of IL‐21 on RA‐FLS migration and invasion, which prompted us to investigate the potential effect of IL‐21. Further study of IL‐21 may help to elucidate the precise role of IL‐21 in disease, and therapies targeting IL‐21 may represent novel alternative treatments for diverse human immunological conditions.

In the present study, we explored the effects of IL‐21 on migration and invasion of RA‐FLS and observed the effect of IL‐21 on MMPs and tissue inhibitor of metalloproteinases (TIMPs) production to determine the potential mechanisms underlying the action of IL‐21 in RA‐FLS.

Materials and methods

Isolation of synovial cells and cell culture

FLS were prepared from synovectomized tissues of RA and osteoarthritis (OA) patients undergoing joint replacement surgery. Synovial tissues from five RA patients [female, aged 32, 31, 51, 49 and 60 years, disease‐modifying anti‐rheumatic drugs (DMARDs) application irregularity] and two OA patients (female, aged 67 and 69 years) were minced mechanically, washed in cold sterile phosphate‐buffered saline (PBS) and digested with 150 mg/ml of Dispase II (Roche, Reinach, Switzerland) for 4 h at 37°C, with gentle agitation. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Gibco, Life Technologies, Carlsbad, CA, USA), 100 U/ml penicillin (Gibco) and 100 µg/ml streptomycin (Gibco) in a 37°C incubator containing a 5% CO₂‐enriched atmosphere. Cells from passages 4–7 were used in all experiments. Cells at passages 4–7 were identified by flow cytometric analysis. Ethical approval was obtained from the Ethics Committee of Peking University Third Hospital.

CD4⁺ T cells and NK cells purification from human donors

Whole blood was obtained from healthy human donors from Peking University Third Hospital. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll density‐gradient centrifugation (GE Healthcare Life Sciences, Fairfield, CT, USA), according to the manufacturers' protocol. CD4⁺ T cells were purified by negative selection using the EasySep^TM human CD4⁺ T cell isolation kit (Stemcell Technologies Inc., British Columbia, Canada), as recommended by the manufacturer. NK cells were purified by negative selection using the EasySep^TM human NK cell selection kit (Stemcell). The CD4⁺ T cells and NK cells were cultured in 24‐well plates in 1 ml medium (RPMI‐1640 supplemented with 10% heat‐inactivated FBS) with 50 ng/ml phorbol myristate acetate (PMA)/ionomycin (Selleck Chemicals, Houston, TX, USA) for 18 h, then the supernatants were cultured with RA‐FLS or OA‐FLS in the lower chambers of Transwell plates.

Reverse transcription–polymerase chain reaction (RT–PCR)

Total RNA was extracted from cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions, with cDNA synthesis and RT–PCR analysis performed as described previously 17. The following primers were used for each molecule: MMP‐1: 5′‐ACTCTGGAGTAATGTCACACC‐3′ (sense) and 5′‐GTTGGTCCACCTTTCATCTTCA‐3′ (anti‐sense); MMP‐2: 5′‐CCGTCGCCCATCATCAAGTT‐3′ (sense) and 5′‐CTGTCTGGGGCAGTCCAAAG‐3′ (anti‐sense); MMP‐3: 5′‐AGTCTTCCAATCCTACTGTTGCT‐3′ (sense) and 5′‐TCCCCGTCACCTCCAATCC‐3′ (anti‐sense); MMP‐9: 5′‐GGGACGCAGACATCGTCATC‐3′ (sense) and 5′‐TCGTCATCGTCGAAATGGGC‐3′ (anti‐sense); MMP‐13: 5′‐GCTGCCTTCCTCTTCTTGA‐3′ (sense) and 5′‐TGCTGCATTCTCCTTCAGGA‐3′ (anti‐sense); TIMP‐1: 5′‐CCTTCTGCAATTCCGACCTCGTC‐3′ (sense) and 5′‐CGGGCAGGATTCAGGCTATCTGG‐3′ (anti‐sense); TIMP‐2: 5′‐GATTAGGGCCAAAGCGGTCAGTG‐3′ (sense) and 5′‐GCCTGATGCCCGTTGATGCTC‐3′ (anti‐sense); and glyceraldehyde 3‐phosphate dehydrogenase (GAPDH): 5′‐GAAGGTCGGAGTCAACGG‐3′ (sense) and 5′‐GGAAGATGGTGATGGGATT‐3′ (anti‐sense).

Enzyme‐linked immunosorbent assay (ELISA)

Cell culture supernatants were stored at −80°C until assayed. Supernatant cytokine concentrations were measured using ELISA, according to the manufacturer's protocol. MMP‐3 and MMP‐9 ELISA kits were obtained from eBioscience (San Diego, CA, USA).

Immunoblotting

After cell samples were lysed in lysis buffer, protein concentration was estimated using a bicinchoninic acid (BCA) protein assay kit (Pierce, Rockford, IL, USA). Proteins were loaded onto a 10% sodium dodecyl sulphate‐polyacrylamide gel and transferred electrophoretically to polyvinylidene fluoride membranes (PVDF; Merck Millipore, Billerica, MA, USA). After blocking with 5% bovine serum albumin (BSA) for 2 h, membranes were incubated with primary antibodies according to the supplier's protocol, followed by incubation with a horseradish peroxidase‐conjugated secondary antibody at room temperature for 1 h. Blots were developed using an enhanced chemiluminescence detection kit (Pierce). Antibodies used were as follows: anti‐p‐protein kinase B (AKT) (Ser473), anti‐AKT, anti‐phosphoinositide 3‐kinase (PI3K)p110α, anti‐p‐signal transducer and activator of transcription‐3 (STAT‐3) (Ser727), anti‐p‐STAT‐3 (Tyr705), anti‐STAT‐3, anti‐p‐extracellular signal‐regulated protein kinases 1 and 2 (ERK1/2) (T202/Y204), anti‐ERK1/2, anti‐p‐mitogen‐activated protein kinase kinase (MEK1/2) (Ser217/221), anti‐MEK1/2, anti‐TIMP‐1 and anti‐TIMP‐2 antibody from Cell Signaling Technology (Danvers, MA, USA), and anti‐MMP‐1, anti‐MMP‐2, anti‐MMP‐3, anti‐MMP‐9 and anti‐MMP‐13 from Abcam (Cambridge, UK).

Cell migration and invasion assays

Cell migration was determined using 6·5 mm Transwell chambers with 8 μm pores (Corning, NY, USA). Briefly, a total of 5 × 10⁴ cells in 100 μl DMEM were added in triplicate to the top chambers of Transwell plates, and the lower chambers were filled with DMEM containing TNF‐α (20 ng/ml) or IL‐21 (1, 10, 50, 100 ng/ml) with 10% FBS. For kinase pathways blocking experiments, PD98059 (20 μM), an ERK inhibitor, LY294002 (10 μM), an AKT inhibitor or STA21 (50 μM), a STAT‐3 inhibitor (Cell Signaling Technology) and AG490 (Selleck Chemicals, Houston, TX, USA) were used. After culturing at 37°C for 24 h, cells were fixed with 100% methanol for 30 min at room temperature. After removal of methanol and subsequent washing with PBS, the cells that invaded the matrix were stained with 0·1% crystal violet (Merck Millipore) solution for 30 min at room temperature. The non‐migrating cells were removed from the upper surface by cotton swabs. The number of cells that migrated through the membrane to the lower surface was counted in five representative microscopic fields (×100 magnification). The number of migrated cells was averaged from three ×10 field‐of‐view images and normalized to control.

Cell invasion ability was determined using Matrigel invasion chambers (BD Biosciences, Tokyo, Japan), according to the manufacturer's instructions. The upper chambers were freshly coated with Matrigel, and medium was added to the lower chamber as described above. For blocking experiments, FLS were incubated with IL‐21 (50 ng/ml) plus IL‐21R.Fc (0·01, 0·1, 1, 10 μg/ml) or immunoglobulin (Ig)G.Fc (10 μg/ml) for 48 h. Soluble recombinant human IL‐21R chimera (IL‐21R.Fc) (991‐R2) and recombinant human IgG1 chimera chimera (IgG. Fc) (110‐HG) were purchased from R&D Systems (Abingdon, UK). Cell invasion was allowed to occur for 48 h and the gel and cells on the top membrane surface were removed with cotton swabs. Cells that had penetrated to the bottom were counted.

Flow cytometric analysis

Following the required time in culture, 5 × 10⁵ FLS were harvested in DMEM, cells were washed with PBS and incubated with saturating antibody concentrations for 20 min at room temperature. Flow cytometry was performed using a fluorescence activated cell sorter (FACS)Aria II flow cytometer (Becton Dickinson, San Jose, CA, USA). Cells were gated using forward‐ versus side‐scatter to remove any dead cells and cellular debris and thus provide a uniform population of FLS. For each sample, 10 000 cells were analysed. Results were expressed as the corrected mean fluorescence intensity (MFI) following subtraction of non‐specific IgG control fluorescence. Antibodies used were as follows: vimentin‐Alexa Fluor 488 and Alexa Fluor 488 isotype were from R&D Systems; and extracellular surface markers with fluorescently conjugated antibodies specific for CD4 (BD Biosciences), CD14, CD68, CD56, intercellular adhesion molecule 1 (ICAM‐1) (eBioscience), vascular cell adhesion molecule 1 (VCAM‐1) and cadherin‐11 (R&D Biosystems). Flow cytometry data were analysed using FlowJo 2·7·4 software (TreeStar Inc., San Carlos, CA, USA).

Statistical analysis

Data are expressed as the mean ± standard error of the mean (s.e.m.). Data were compared using one‐way analysis of variance (anova), followed by Turkey's method for multiple comparison (GraphPad Prism version 6·0; GraphPad Software Inc., La Jolla, CA, USA). A P‐value less than 0·05 was considered significant in all statistical analyses.

Results

Migration induced by IL‐21 from CD4⁺ and NK cells

For CD4⁺ T and NK cell enrichment from healthy human donors, cellular purity was detected with a flow cytometer before and after sorting using the immunomagnetic isolation method. The purity of CD4⁺ T and NK cells was 98 and 88.9%, respectively (Fig. 1a). The IL‐21 concentration in the supernatant of CD4⁺ T and NK cells were up to 55 ng/ml and 11·9 ng/ml (Fig. 1b). When cultured with the CD4⁺ T and NK cell supernatant or recombinant human IL‐21 (50 ng/ml), the RA‐FLS migration was increased significantly (Fig. 1c). However, when OA‐FLS were cultured with CD4⁺ T and NK cell supernatant, the CD4⁺ T cell supernatant‐induced OA‐FLS migration was increased significantly, while NK cells and the IL‐21 group were not changed (Fig. 1d).

Figure 1.

Migration induced by interleukin (IL)−21 from CD4⁺ T and natural killer (NK) cells. CD4⁺ T and NK cell enrichment from healthy human donors by the immunomagnetic isolation method, the supernatant cultured with rheumatoid arthritis‐ fibroblast‐like synoviocytes (RA‐FLS) and osteoarthritis (OA)‐FLS to induce cell migration. (a) The purity of CD4⁺ T and NK cell enrichment by the immunomagnetic isolation method. (b) The IL‐21 concentration in the supernatant of CD4⁺ T and NK cells. (c) The migration of RA‐FLS induced by CD4⁺ T and NK cell supernatant or IL‐21 (50 ng/ml). (d) The migration of OA‐FLS induced by CD4⁺ T and NK cell supernatant or IL‐21 (50 ng/ml). All pictures were captured at ×100 magnification. Values are the mean ± standard error of the mean (s.e.m.) of three samples. *P < 0·05 versus control; ***P < 0·001 versus control. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

IL‐21 promotes migration and invasion of RA‐FLS in vitro

FLS at passages 4–7 were identified by flow cytometric analysis as a homogeneous population with the phenotype of <2·5% CD14, <1% CD68 and >98% vimentin 18 (Supporting information, Fig. S1). Cell migration and invasion are regulated in part by inflammatory cytokines, such as TNF‐α, that trigger cellular motility when applied in a uniform concentration. Transwell assay was performed to determine the role of IL‐21 in regulating the migration and invasion of RA‐FLS. RA‐FLS migration was assessed in a Transwell assay (Fig. 2a). IL‐21‐induced migrated cells were increased in a concentration‐dependent manner. We also evaluated the effects of IL‐21 on OA‐FLS migration, Transwell assay showed that IL‐21 did not affect the migration of OA‐FLS (Fig. 2b).

Figure 2.

Interleukin (IL)−21 stimulates migration of fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA) and osteoarthritis (OA). RA‐FLS and OA‐FLS migration was measured using the Transwell system. 1 × 10⁴ cells/200 μl were added to each upper Transwell chamber; the lower chamber contained 600 μl medium with or without IL‐21 (1, 10, 50 and 100 ng/ml) or tumour necrosis factor (TNF)‐α (20 ng/ml). Cells that migrated to the lower surface after 48 h were stained with 0·1% crystal violet and counted. (a) The migration of RA‐FLS induced by IL‐21. (b) The migration of OA‐FLS induced by IL‐21. All pictures were captured at ×100 magnification. Data are expressed as the mean ± standard error of the mean (s.e.m.). **P < 0·01 versus control; ***P < 0·001 versus control. Three separate experiments were performed with similar results. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

The ability to invade cartilage is a critical pathogenic behaviour of rheumatoid synoviocytes. Therefore, we also examined the role of IL‐21 in RA‐FLS invasion through Matrigel‐coated Transwell membranes. The results of the Transwell assay showed that IL‐21 increased the invasion cells of RA‐FLS (Fig. 3a,b). To confirm the effects of IL‐21 on RA‐FLS invasion, increasing concentrations of IL‐21R.Fc were added to the cultured cells. The Transwell results showed that IL‐21R.Fc at a concentration of 10 μg/ml inhibited IL‐21‐induced cell invasion (Fig. 3c,d). Hence, IL‐21 increased the migration and invasion of RA‐FLS in vitro, suggesting that IL‐21 might be an important factor for mediating the invasive capacity of RA‐FLS.

Figure 3.

Interleukin (IL)‐21 stimulates invasion of fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). RA‐FLS invasion was measured using the Transwell system. The upper compartment of the Transwell chamber was filled with 0·5 ml Matrigel before adding cells. 1 × 10⁴ cells/200 μl were added to each upper Transwell chamber. The lower chamber contained 600 μl medium with or without IL‐21 or tumour necrosis factor (TNF)‐α (20 ng/ml). Cells that migrated to the lower surface after 48 h were stained with 0·1% crystal violet and counted. (a,b) RA‐FLS invasion was measured using the Transwell system. (c,d) The cells were preincubated with IL‐21 (50 ng/ml) plus IL‐21R.Fc (0·01, 0·1, 1, 10 μg/ml) or immunoglobulin (Ig)G.Fc (10 μg/ml) for 1 h. All pictures were captured at ×100 magnification. Values are the mean ± standard error of the mean (s.e.m.) of three samples. *P < 0·05 versus control; **P < 0·01 versus control; ***P < 0·001 versus control; ^# P < 0·05 versus the IL‐21 group. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

IL‐21 promotes MMPs production and secretion by RA‐FLS

MMPs are involved in the degradation of ECM components and play an important role in joint destruction during RA. To address whether RA‐FLS respond to IL‐21, FLS were cultured in the presence of IL‐21 for 48 h, then cells were harvested for RT–PCR and immunoblot analysis and the supernatant was analysed by ELISA. As shown in Fig. 4, RT–PCR results showed specifically that IL‐21 stimulated the production of MMP‐2, MMP‐3, MMP‐9 and MMP‐13 in RA‐FLS (Fig. 4a,b), which was confirmed by immunoblot analysis (Fig. 4c,d). The activity of MMPs is down‐regulated by TIMPs, so we also evaluated the effects of IL‐21 on TIMP‐1 and TIMP‐2 production. Both RT–PCR and immunoblot analysis showed that IL‐21 did not affect the TIMP‐1 and TIMP‐2 expression in RA‐FLS.

Figure 4.

Interleukin (IL)‐21 promotes matrix metalloproteinase (MMP) synthesis and secretion by fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). RA‐FLS were stimulated with IL‐21 (0, 1, 10, 50, 100 ng/ml) for 48 h, and cells harvested for reverse transcription–polymerase chain reaction (RT–PCR) or immunoblotting. (a,b) MMP‐1, MMP‐2, MMP‐3, MMP‐9, MMP‐13, TIMP‐1 and TIMP‐2 levels were detected by RT–PCR. (c,d) Representative Western blots showing MMP‐1, MMP‐2, MMP‐3, MMP‐9, MMP‐13, tissue inhibitor of metalloproteinases (TIMP‐1) and TIMP‐2 in RA‐FLS. (e,f) RA‐FLS were cultured in a 96‐well plate for 24 h, and were then incubated for 48 h with IL‐21 (50 ng/ml) in the absence or presence of IL‐21R.Fc (0·01, 0·1, 1, 10 μg/ml) or immunoglobulin (Ig)G.Fc for 48 h, and culture supernatants were harvested for enzyme‐linked immunosorbent assay (ELISA). Values are the mean ± standard error of the mean (s.e.m.) of five samples. *P < 0·05 versus control; **P < 0·01 versus control; ***P < 0·001 versus control; ^# P < 0·05 versus the IL‐21 group.

The MMPs concentrations in the supernatant were also tested by ELISA assays, which showed that MMP‐3 and MMP‐9 levels were elevated significantly following IL‐21 treatment (Fig. 4e). Blocking the IL‐21 signal pathway using IL‐21R.Fc inhibited the production of proinflammatory cytokines in RA synovial cells 12, 19. Therefore, we used IL‐21R.Fc to investigate the effects of blocking IL‐21 on the secretion of MMPs of RA‐FLS. IL‐21R.Fc treatment inhibited MMP‐3 and MMP‐9 secretion by RA‐FLS in a concentration‐dependent manner, and 10 μg/ml IL‐21R.Fc inhibited IL‐21‐induced MMP‐3 and MMP‐9 production markedly (Fig. 4f). Together, these results suggest that IL‐21 promotes MMPs production in a concentration‐dependent manner.

IL‐21 promotes ICAM‐1 and cadherin‐11 production

Adhesion molecules ICAM‐1, VCAM‐1 and cadherin‐11 play a crucial role in the progression of cell migration and invasion; certain proinflammatory cytokines up‐regulate the expression of adhesion molecules. To study whether IL‐21 regulates vimentin, ICAM‐1, VCAM‐1 and cadherin‐11 expression, RA‐FLS were treated with IL‐21 (1, 10, 50 ng/ml) for 48 h. Stimulation of RA‐FLS with IL‐21 (50 ng/ml) resulted in a significant 11·7 and 11·0% increase in cadherin‐11 and ICAM‐1 expression, respectively (Fig. 5a,b). However, results of the flow cytometric analysis showed only a modest effect of IL‐21 on VCAM‐1 and vimentin expression (Fig. 5c,d).

Figure 5.

Interleukin (IL)‐21 stimulates intercellular adhesion molecule‐1 (ICAM‐1) and cadherin‐11 by fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). RA‐FLS were stimulated with IL‐21 (0, 1, 10, 50 ng/ml) for 48 h and the cells were harvested for flow cytometry analysis. (a) Representative fluorescence activated cell sorter (FACS) analysis of IL‐21 mediated cadherin‐11 expression on RA‐FLS. (b) Representative FACS analysis of IL‐21‐mediated ICAM‐1 expression on RA‐FLS. (c) Representative FACS analysis of IL‐21‐mediated vascular cell adhesion molecule‐1 (VCAM‐1) expression on RA‐FLS. (d) Representative FACS analysis of IL‐21‐mediated vimentin expression on RA‐FLS. Three separate experiments were performed with similar results. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

IL‐21 activates STAT‐3, ERK‐1/2 and PI3K/AKT pathways in RA‐FLS

IL‐21 activates several signalling pathways, including the Janus kinase‐1 and −3 (JAK‐1)/STAT‐3 pathway, the mitogen‐activated protein kinase pathway (MAPK) and the PI3K signalling pathway 20. MMP secretion is regulated by a number of intracellular signalling pathways, including the MAPK pathway 21, 22. To determine the intracellular mechanisms involved in IL‐21‐induced MMPs synthesis and invasion of RA‐FLS, we examined the signalling pathways that control cell migration and MMPs synthesis, such as the PI3K/AKT, STAT‐3 and ERK1/2 pathways. Immunoblot analysis showed that phosphorylation of AKT (Fig. 6a), STAT‐3 (Fig. 6c), MEK1/2 and ERK1/2 (Fig. 6e) were induced significantly by IL‐21, with maximum phosphorylation at 5–15 min after stimulation.

Figure 6.

Interleukin (IL)‐21 activates phosphoinositide 3‐kinase (PI3K), signal transducer and activator of transcription‐3 (STAT‐3) and extracellular signal‐regulated protein kinases 1 and 2 (ERK1/2) pathways in fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). RA‐FLS were stimulated with IL‐21 (50 ng/ml) for the indicated times, and cells were harvested for immunoblotting analysis. (a,b) Phosphorylation of protein kinase B (AKT) and expression of PI3K. (c,d) Phosphorylation of STAT‐3. (e,f) Phosphorylation of mitogen‐activated protein kinase kinase (MEK)1/2 and ERK1/2. Values are the mean ± standard error of the mean (s.e.m.) of four samples. *P < 0·05 versus control; **P < 0·01 versus control; ***P < 0·001 versus control.

Inhibition of STAT‐3, ERK‐1/2 and PI3K/AKT pathways decreases IL‐21‐induced RA‐FLS migration, and MMP‐3 and MMP‐9 secretion in RA‐FLS

To confirm the role of activated kinases in IL‐21‐induced migration and MMP synthesis, we pretreated RA‐FLS with specific kinase inhibitors before stimulation with IL‐21. The ERK1/2 pathway inhibitor, PD98059 (20 μM), PI3K/AKT inhibitor, LY294002 (10 μM) and STAT‐3 pathway inhibitors STA‐21 (50 μM) and AG490 (50 μM) were used at concentrations that had been established previously to not alter FLS viability (Supporting information, Fig. S2). Blocking of each kinase pathway (PI3K/AKT, ERK‐1/2 and STAT‐3 pathways) decreased IL‐21‐induced RA‐FLS migration and MMP‐3 and MMP‐9 secretion. In contrast, inhibition of the JAK‐2/STAT‐3 pathway using AG490 had no influence on IL‐21‐induced migration (Fig. 7).

Figure 7.

Inhibition of phosphoinositide 3‐kinase (PI3K), signal transducer and activator of transcription‐3 (STAT‐3) and extracellular signal‐regulated protein kinases 1 and 2 (ERK1/2) pathways decreases interleukin (IL)‐21‐induced rheumatoid arthritis‐fibroblast‐like synoviocytes (RA‐FLS) migration and matrix metalloproteinase (MMP)‐3 and MMP‐9 secretion in RA‐FLS. (a) Cells were preincubated with specific pathways inhibitors: the ERK1/2 pathway inhibitor, PD98059 (20 μM), PI3K/protein kinase B (AKT) inhibitor, LY294002 (10 μM), STAT‐3 pathway inhibitors STA‐21 (50 μM) and AG490 (50 μM) for 1 h. The lower chamber contained 600 μl medium with or without IL‐21 (1, 10, 50, 100 ng/ml) or tumour necrosis factor (TNF)‐α (20 ng/ml). After incubation for 24 h, cells that migrated to the lower surface were stained with haematoxylin and eosin and counted. (b,c) Enzyme‐linked immunosorbent assay (ELISA) analyses for MMP‐3 and MMP‐9. Cells were preincubated with specific pathway inhibitors for 1 h, then stimulated with IL‐21 (1, 10, 50, 100 ng/ml) or tumour necrosis factor (TNF)‐α (20 ng/ml) for 48 h and culture supernatants harvested for ELISA. All pictures were captured at ×100 magnification. Values are the mean ± standard error of the mean (s.e.m.) of four samples. *P < 0·05 versus control or dimethylsulphoxide (DMSO); **P < 0·01 versus control or DMSO; ***P < 0·001 versus control; ^# P < 0·05 versus the IL‐21 group; ^## P < 0·01 versus the IL‐21 group; ^### P < 0·001 versus the IL‐21 group. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Discussion

RA‐FLS constitutes the major cell population in synovial tissues and is a critical component in the development of RA. Previous studies have shown that RA‐FLS migrate and invade the cartilage and bone, contributing to pannus formation and tissue damage during RA progression 23. Therefore, it is important to identify the factors that regulate invasion. Modulation of RA‐FLS invasion may offer a novel strategy for RA therapy. It was reported previously that proinflammatory cytokines, such as IL‐6, IL‐1β and TNF‐α, result in heightened migration and invasion by RA‐FLS 24. Monoclonal antibodies against both TNF‐α and the IL‐6 receptor have shown promise in treating RA 25. IL‐21 is a recently discovered cytokine produced by activated CD4⁺ T lymphocytes; several reports have shown that IL‐21 is associated with RA development in both animal models and humans, which suggests further that IL‐21 is involved in disease progression. Our earlier studies have shown that IL‐21R is highly expressed in rheumatoid synovium tissues and by RA‐FLS, and IL‐21 induced RA‐FLS proliferation by activating the PI3K/AKT and MEK1/2/ERK1/2 pathways 26. In this study we found that IL‐21 enhanced RA‐FLS migration and invasion, and we also confirmed that the high level of IL‐21 secretion by CD4⁺ T and NK cells have a stronger role in promoting invasion. However, the culture supernatants of CD4⁺ T and NK cells contained many kinds of cytokines, such as TNF‐α and IL‐6, so recombinant human IL‐21 was used in subsequent experiments to identify the effect of IL‐21 on the migration and invasion of RA‐FLS. IL‐21 up‐regulated the production of MMPs, ICAM‐1 and cadherin‐11 by inducing the PI3K/AKT, STAT‐3 and ERK1/2 pathways in RA‐FLS cells.

Previous studies have shown that RA‐FLS‐produced MMPs are involved in RA development and progression 27. MMP‐3 plays a pivotal role in the destruction of bone and degradation of various components of cartilage, and baseline levels of MMP‐3 serve as a biomarker of progressive RA joint damage 28. Previous studies have shown that MMP‐13 promotes bone destruction in RA 28. In addition, over‐expression of MMP‐2 and MMP‐9 plays a role in RA‐FLS invasion 29, 30. MMPs are up‐regulated by proinflammatory cytokines such as TNF‐α and IL‐1β 31, indicating that proinflammatory cytokine involvement in RA tissue injury is due, in part, to promoting MMP expression. In the current study, we found that IL‐21 induced the expression of MMP‐2, MMP‐3, MMP‐9 and MMP‐13 in RA‐FLS, and this is consistent with a previous study showing that IL‐21 induces production of MMP‐3, MMP‐9 and MMP‐13 mRNA levels in RA‐FLS 32. Furthermore, we discovered that IL‐21 promotes MMP‐3 and MMP‐9 secretion by RA‐FLS, and inhibition of IL‐21 with an IL‐21R.Fc fusion protein inhibits IL‐21‐induced secretion of MMP‐3 and MMP‐9. In contrast, no considerable change in the production of TIMP‐1 and TIMP‐2 was seen in IL‐21‐stimulated RA‐FLS. The MMPs and TIMPs, and especially the disturbances of the enzyme : inhibitor ratios, are involved in degradation of the articular components during the course of RA 33. These findings are consistent with previous reports showing high MMP production without concomitant TIMP elevation in other systems.

ICAM‐1 and VCAM‐1 expressed by RA‐FLS play an important role in the sustainment of inflammation by promoting adhesion of circulating inflammatory cells to the walls of endothelial cells 34. Studies have shown that cultured RA‐FLS constitutively display surface ICAM‐1, which can be induced by a variety of cytokines, including IL‐lβ, TNF‐α and IFN‐γ. VCAM‐1 expression on FLS can be increased by the same pattern of cytokines, including IL‐4 35. ICAM‐1 and VCAM‐1 expression plays a key role in maintaining inflammation; our study suggests that IL‐21 up‐regulates the expression of ICAM‐1 and cadherin‐11 that may partly interpret the IL‐21‐induced invasion. Cadherin‐11 mediated the β‐catenin signalling involved in IL‐1β‐induced proliferation, but was not involved in TNF‐α‐induced RA‐FLS proliferation 36. In this study, we found that IL‐21 induced cadherin‐11 expression in RA‐FLS; whether cadherin‐11 mediates IL‐21‐induced proliferation and invasion needs further study.

We found that IL‐21 activates the PI3K/AKT, STAT‐3 and ERK1/2 pathways in RA‐FLS, and that activation of AKT, STAT‐3, ERK1/2 and MEK1/2 was detectable 5 min after stimulation, suggesting that stimulation is triggered directly by IL‐21. The ERK pathway was reported to participate in the inflammatory response of RA‐FLS, and the PI3K/AKT and STAT‐3 pathways are crucial for the invasion of RA‐FLS cells 37. We found that blocking these pathways with specific kinase inhibitors inhibited IL‐21‐induced RA‐FLS migration significantly, and decreased MMP‐3 and MMP‐9 secretion. Nevertheless, inhibition of STAT‐3 with AG490 did not affect IL‐21‐induced migration, which may relate to the specific inhibition of JAK‐2 activation by AG490, while IL‐21 has no activation effect on the JAK‐2/STAT‐3 pathway.

In summary, our data indicate that IL‐21 enhanced RA‐FLS migration and invasion. MMPs (MMP‐2, MMP‐3, MMP‐9 and MMP‐13) and adhesion molecules (ICAM‐1 and cadherin‐11) up‐regulation may be involved in IL‐21‐induced invasion. We validated the signalling pathways (PI3K/AKT, STAT‐3 and ERK1/2 pathways) involved in IL‐21‐induced invasion of RA‐FLS. These findings provide new insight regarding the regulation of FLS cell‐mediated joint damage of IL‐21 in RA patients. Modulation of IL‐21 may therefore offer a novel strategy to control the aggressive and destructive process of RA.

Disclosure

The authors declare no disclosures.

Author contributions

X. L. and J. Z. designed the project and supervised the work, analysed and interpreted the data and participated in writing the manuscript. R. X. coordinated and performed the experiments, collected and analysed the data and wrote the manuscript. L. S. and L. Y. helped to design the experiments and interpret the data. C. L. performed RT–PCR, Y. J. and Z. L. were involved in patient recruitment and collection of the patients' data. All authors have read, commented on and approved the final manuscript.

Supporting information

Additional Supporting information may be found in the online version of this article at the publisher's web‐site:

Fig. S1. Identification of fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). Cells at passages 4–7 were identified by flow cytometric analysis. (a) Expression of vimentin; (b) expression of CD14; (c) expression of CD68.

Fig. S2. Effect of phosphoinositide 3‐kinase (PI3K), signal transducer and activator of transcription‐3 (STAT‐3) and extracellular signal‐regulated protein kinases 1 and 2 (ERK1/2) pathway inhibitors on the activation of fibroblast‐like synoviocytes (FLS) from patients with rheumatoid arthritis (RA). RA‐FLS were preincubated with specific pathway inhibitors: the ERK1/2 pathway inhibitor, PD98059 (20 μM), PI3K/protein kinase B (AKT) inhibitor, LY294002 (10 μM), STAT‐3 pathway inhibitors STA‐21 (50 μM) and AG490 (50 μM) for 1 h; cells were then were stimulated with IL‐21 (50 ng/ml) for 5 min, and cells were harvested for immunoblotting analysis. (a) Phosphorylation of AKT; (b) phosphorylation of STAT‐3; (c) phosphorylation of ERK1/2.

Fig. S3. Interleukin (IL)‐21 in synovial fluid and serum from patients with rheumatoid arthritis (RA) and osteoarthritis (OA). (a) IL‐21 level in synovial fluid; (b) IL‐21 level in serum. Values are the mean ± standard error of the mean (s.e.m.). **P < 0·01 versus control or OA group.