Introduction

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease characterized as articular synovitis, progressive cartilage destruction and bone erosion. Although the aetiology of RA remains unclear, the aberrant B-cells activation and cytokine-mediated pathway are central to the pathogenesis of RA.

Syndecan-4, one of the members of transmembrane heparan sulphate proteoglycans (HSPGs), is the receptor of various cytokines and chemokines that can regulate a series of signal transductions [1]. It could be expressed in several cells under inflammation and involved in regulation of inflammation, angiogenesis, focal adhesion and cell migration [2]. Syndecan-4 ectodomains are constitutively shed and replaced under physiological conditions to maintain balance. However, in response to certain stimuli, such as extracellular cytokines, metalloprotease (MMP) and oxidative stress, syndecan-4 shedding increases dramatically, and the soluble syndecan-4 can promote inflammation and the synthesis of new syndecan-4 [3].

Syndecan-4 could up-regulate the levels of MMP with thrombospondin motifs (ADAMTS)‐5 to promote the inflammation and cartilage degeneration [4]. It has been shown previously that the loss of syndecan-4 protects mice from cartilage damage in animal models of osteoarthritis (OA) [5] and RA [6]. Syndecan-4 deficient mice were resistant to the induction of collagen-induced arthritis (CIA) due to loss of the migration ability of B cells [7]. The tumour necrosis factor transgenic (hTNFtg) mouse model, an established animal model of RA, demonstrated that syndecan-4 was prominently involved in synovial fibroblasts activation and its attachment to the cartilage in the early stage of arthritis [6]. Intra-articular injection of anti-syndecan-4 antibodies into wild-type mice prevented cartilage damage [5]. Syndecan-4 blocking antibody reduced interleukin-1 receptor 1 (IL1R1) surface presentation, leading to reduced RA symptoms in the hTNFtg RA mouse model [8].

Syndecan-4 can be highly expressed by the fibroblast-like synoviocyte (FLS) of RA, as well as interact with receptor protein tyrosine phosphatase sigma (RPTPσ), a highly-expressed phosphatase in arthritic FLS which can regulate the RA FLS migration [9]. Godmann et al. reported that syndecan-4 was highly expressed in RA synovium, and messenger RNA (mRNA) level was also upregulated in RA‐FLS [8]. Syndecan-4 modulates the response of RA‐FLS to the inflammatory cytokine IL-1 [8]. A most recent study demonstrated that highly expressed syndecan-4 led to the initiation of inflammation and decreased apoptosis in RA‐FLS, thereby illustrating the important role of syndecan-4 in the pathogenesis of RA [10].

Despite evidence implicating the relevant role of syndecan-4 in the pathogenesis of RA, the syndecan-4 expression in serum and synovial fluid (SF) of RA patients has not been reported. The correlation of syndencan-4 with clinical features of RA, including disease activity and autoantibody, also remains unclear. In the present study, we aimed to describe the expression of syndecan-4 in serum and SF of RA patients, and reveal the correlation between syndecan-4 and disease activity in RA.

Materials and methods

Patients and samples

The syndecan-4 concentrations in sera were tested in 60 RA patients, and 20 OA patients and 20 healthy controls by enzyme linked immunosorbant assay (ELISA). The OA patients and healthy controls were matched with the RA patients by age and gender. The correlation of syndecan-4 concentration in sera and disease activity of RA patients were analyzed.

SF samples were collected from 25 RA patients and 25 OA patients. The correlation of syndecan-4 concentration in sera and disease activity of RA patients were also analyzed.

Syndecan-4 expression in synovial tissues of 5 RA patients and 5 OA patients were tested by immunohistochemistry. All the synovial tissues were derived from patients undergoing knee replacement. These RA patients and OA patients were also matched by age and gender.

All the above RA patients fulfilled the 2010 American College of Rheumatology/The European League Against Rheumatism (ACR/EULAR) classification criteria. All the enrolled OA patients fulfilled the OA classification criteria [11, 12]. Patients with comorbidities that may increase serum syndecan-4 levels., including pneumonia, malignancies, heart failure, dilated cardiomyopathy, atopic dermatitis, haemodialysis, inflammatory arthritis other than RA, connective tissue diseases and infections were excluded.

None of the patients have received biological disease modifying anti-rheumatic drugs or target therapy. Disease activity of RA patients were evaluated by disease activity score based on 28 joint count-C reactive protein (DAS28-CRP), DAS28-erythrocyte sedimentation rate (DAS28-ESR), simplified disease activity index (SDAI) and clinical disease activity index (CDAI).

This trial was approved by the Peking University Ethics Committee.

All the participants gave written informed consent.

Syndecan-4 concentrations in sera and synovial fluid

Blood and SF were collected into tubes free of anticoagulants, centrifuged 3000 rpm for 10 min to remove cellular debris and stored at − 80 °C until use. Syndecan-4 concentrations in sera and SF were measured by the human syndecan-4 ELISA kit (R&D systems, the USA), following the manufacturer’s protocols. Goat anti human syndecan-4 capture antibody was diluted to a terminal concentration of 0.8 ug/mL using PBS, coated onto the wells of a polystyrene microtiter plate and incubated at room temperature overnight. Test sera without dilution were added to the microwells and incubated at room temperature for 2 h. After washed for 3 times, a detective antibody with a terminal concentration of 100 ng/mL was then added and incubated at room temperature for 2 h. Streptavidin-HRP was then added after washed for 3 times. After incubation for 20 min, substrate solution was added to the microwells. The reaction was terminated by the addition of stop solution and absorbance was measured at 450 nm and 540 nm. Finally, sample concentration of syndecan-4 was determined by extrapolation in a standard curve prepared from seven geometric syndecan-4 standards.

Immunohistochemistry

The protocol of previous study was followed [13] with minor modifications by anti-syndecan-4 antibodies. Briefly, 10 mm thick serial cryostat sections of synovia were dried for 1 h and stored at liquid nitrogen (− 196 ℃) until required. Prior to immunohistochemical analysis, slides were left to equilibrate to room temperature for 30 min, fixed in acetone (4 ℃) for 10 min, air dried, and then sections were rehydrated in phosphate buffered saline (PBS) for 5 min. All primary antibodies were used at 5 mg/ml since initial experiments showed that this concentration gave the optimum specific staining over the range of 2.5–10 mg/ml. Antibody binding was detected using DAB staining kit (Vector Labs, Burlingame, California, USA). Sections were counterstained with Mayer Haematoxylin and mounted. In control experiments, isotype matched control mouse Ig or rabbit Ig were added instead of primary antibodies and stained with the DAB kit as above.

Synovial fibroblast culture

Synovial tissues were obtained from RA and OA patients during joint replacement surgery of knees. According to Kellgren Lawrence grade (K–L grade) of osteoarthritis, all these ten patients were graded as 4. Tissues were minced and treated with 4 mg/ml type I collagenase (Gibco, USA) in serum free Dulbecco’s modified Eagle’s medium (DMEM; Gibco, USA) at 37 °C for 1 h. Primary cultured synovial cell lines were established and maintained in DMEM including 10% heat-inactivated fetal bovine serum (FBS; Gibco, USA), 1% penicillin–streptomycin (100 U/ml; Gibco, USA), and 1% L-glutamine (2 mM; Gibco, USA), in a humidified incubator at 37 °C in the presence of 5% CO2. Harvested cells were continuously cultured to obtain synovial fibroblasts, and cells of passages 3–6 were used in the next experiments.

Immunofluorescence

Cultured synovial fibroblasts of passages 3–6 were used for immunofluorescence. Cells were treated with rabbit anti-human syndecan-4, 4 °C overnight. Washed with PBS and incubated with goat anti-rabbit IgG–Alexa Fluor ®488 (Abcam, UK), room temperature for 1 h. The biotinylated secondary antibody was detected by streptavidin-Alexa 594 (1:1000 Molecular Probes Code S-11227; Molecular Probes, Eugene, Oregon, USA).

Statistical analysis

All statistical analyses were performed using SPSS 20.0 (IBM Company, the USA) and all charts were performed by GraphPad Prism 5.0 (CA Company, the USA). The results of immunohistochemistry and immunofluorescence were analyzed by Image-Pro Plus 6.0 software (Media Cybernetics Company, the USA). Quantitative variables were expressed as mean ± SD or median (min–max) for normal distribution or non normal distribution data, respectively. Statistical analysis was performed by the χ2 test (frequencies), nonparametric Mann–Whitney U test (non normal distribution data) and Student’s t test (normal distribution data), as appropriate. Correlation analysis was performed by Spearman correlation analysis. P value less than 0.05 was considered significant.

Results

Demographic and clinical features of RA patients

Among the 60 RA patients, there were 12 (20%) males and 48 (80%) females with mean age of 54.8 ± 11.6 years and median disease duration of 6.3 (0.5–34) years. 37 (61.7%) were rheumatoid factor (RF) positive and 53 (88.3%) were anti-cyclic citrullinated protein (anti-CCP) positive. The baseline demographic and clinic features of 60 RA patients, 20 OA patients and 20 healthy controls were demonstrated in Table 1. Generally, the age and gender were comparable among the three groups.

Table 1 The baseline demographic and clinic features of 60 RA patients
Full size table

The expression of syndecan-4 in serum, synovial fluid and synovial tissue

The median syndecan-4 concentration in sera of 60 RA patients, 20 OA patients and 20 healthy controls were 509.3 (345.2–769.8) pg/mL, 363.6 ± 137.6 pg/mL and 204.4 ± 59.1 pg/mL, respectively. The serum syndecan-4 concentration was significantly higher in RA than in OA patients and healthy controls (p < 0.001, p < 0.001, respectively). It was also higher in OA patients than in healthy controls (p < 0.001). (Fig. 1).

Fig. 1
figure 1

The comparison of syndencan-4 concentrations among serum from 60 RA patients, 20 OA patients and 20 healthy controls. RA: rheumatoid arthritis; OA: osteoarthritis; HC: healthy control

Full size image

Syndecan-4 concentration was significantly higher in sera than in SF of RA patients [509.3 (345.2–769.8) pg/mL vs 94.8 (23.5–280.3) pg/mL, p < 0.001] (Additional file 1: Figure S1).

Syndecan-4 concentrations was also significantly higher in sera than in SF of OA patients [363.6 ± 137.6 pg/mL vs 51.0 (27.2–163.6) pg/mL, p < 0.001] (Additional file 1: Figure S1).

Syndecan-4 concentration in SF was slightly higher in RA patients compared to OA patients, however without statistical difference (p = 0.497).

Syndecan-4 was detected by immunohistochemistry in the lining and sublining layers of all the 5 RA and 5 OA patents (Fig. 2). The mean area, mean diater, mean density and mean integrated optical density (IOD) were comparable between RA and OA patients (Additional file 1: Table S1).

Fig. 2
figure 2

The comparison of syndecan-4 expression in synovial tissue between 5 RA patients and 5 OA patients. RA: rheumatoid arthritis; OA: osteoarthritis

Full size image

Immunofluorescence demonstrated that Syndecan-4 was positively stained in cultured synovial fibroblasts from both RA and OA patients. The IOD was similar between RA and OA patients (10,668 ± 7570 vs. 8106 ± 12,760, p = 0.255). (Additional file 1: Figure S2).

The comparison of serum syndecan-4 concentration between sera-positive and sera-negative RA patients

Serum syndecan-4 concentration was significantly higher in RF positive patients than in RF negative patients [586.2 (390.2–1124.8) pg/mL vs 434.3 (312.0–537.5) pg/mL, (p = 0.009)]. But it was comparable between anti-CCP antibody positive and negative patients [517.4 (346.2–781.0) pg/mL vs. 500.7 (373.8–537.5) pg/mL, p = 0.542] (Additional file 1: Figure S3).

Then we defined patients with at least one of the above two antibodies positive as “sera positive”, and there were 53 sera positive and 7 sera negative patients. The result was the same as that when stratified by anti-CCP antibody.

The correlation between Syndecan-4 concentration and disease activity of RA patients

Among the 60 RA patients, the syndecan-4 concentration in sera was positively correlated with TJC, SJC, PGA, EGA, and the disease activity scores, including DAS28-ESR, DAS28-CRP, SDAI and CDAI (Table 2). The correlation coefficents (r value) were mild to moderate (Table 2). Based on any of these disease activity scores, the patients in high disease activity (HDA) had the highest serum syndecan-4 concentration, compared with those in remission, low disease activity (LDA) or moderate disease activity (MDA). (Table 3, Fig. 3) Plotted correlation graphs showed the exact Corresponding Relation of sera Syndecan-4 concentration and disease activity. (Additional file 1: Figure S4).

Table 2 The correlation of serum syndecn-4 concentration with disease activity in 60 RA patients
Full size table
Table 3 The comparison of serum syndecn-4 concentration among different disease activity group in 60 RA patients
Full size table
Fig. 3
figure 3

The comparison of serum syndecan-4 between sera-positive and sera-negative RA patients. RA: rheumatoid arthritis. RF: rheumatoid factor; anti-CCP: anti-cyclic citrullinated peptide

Full size image

The syndecan-4 concentration in SF from the 25 RA patients,was positively correlated with PGA, EGA, CRP, DAS28-ESR, DAS28-CRP and SDAI (Additional file 1: Table S2). All of the correlation coefficents (r value) were higher than 0.5.

Discussion

In the current study, we demonstrated higher serum syndecan-4 concentration in RA patients than OA patients and healthy controls. We also confirmed the correlation between syndecan-4 with RA phenotypes, including disease activity and RF. Our findings provided further evidence for the involvement of syndecan-4 in the pathogenesis of RA.

The role of syndecan-4 in articular cartilage degeneration has been proved by substantial evidences [4,5,6, 14], but its involvement in inflammation of arthritis has not been widely recognized. We found serum syndecan-4 was positively related with disease activity, TJC and SJC of RA, suggesting syndecan-4 may be involved in RA-associated inflammation. Syndecan-4 can be up-regulated by TNFα, IL-1β and lipopolysaccharide (LPS) through a functional nuclear factor (NF)-κB site in its promoter [3]. Considering the promotion effect of IL-17 on osteoclastogenesis by inducing the expression of the receptor activator of NF-κB,Th17 might have potential role in inducing syndecan-4 expression [15]. The ectodomain of syndecan-4 could shed from the membrane under oxidative stress and inflammation [16, 17]. Sheded heparan sulfate-substituted syndecan-4 ectodomains in turn increased ICAM-1, VCAM1, TNFα and IL-1β expression and NF-κB activation which all participate in the progression of RA, suggesting positive feedback regulation of inflammatory pathways [18]. TNF alpha inhibitors may control syndecan-4 [19]. It needs further investigation.

The exact pathways and mechanisms by which syndecan-4 contribute to inflammation and joint destruction in RA has emerged. Syndecan-4 was required for the action of tenascin-C which was identified as a major target of autoantibodies in RA and might regulate the pannus invasion into cartilage [20]. Inhibition of syndecan-4 suppressed tenascin-C activity and over-expression of syndecan-4 enhanced the effects of tenascin-C. Tenascin-C and syndecan-4 might work together to control fibroblast morphology and regulate matrix contraction [21]. Dimerisation of syndecan-4 is critically involved in IL-1 signal transduction and IL-1R1 surface presentation via caveolin-dependent mechanisms both in vitro and an animal model of human RA [8]. A most recent published study demonstrated syndecan-4 silencing significantly suppressed the levels of reactive oxygen species (ROS), nitric oxide (NO) and inflammation and promoted the apoptosis of RA‐FLSs. The apoptosis of RA‐fibroblast-like synoviocytes (FLSs) regulated by syndecan-4 was mainly through the intrinsic pathway in which the P53, Caspase‐3, B‐cell leukemia/lymphoma 2 (Bcl2), and Bcl2‐associated X protein (Bax) were involved. It provided evidence that Syndecan-4 mediated the inflammation and apoptosis of RA‐FLSs by mediating the ROS and NO release [10].

We found the concentration of syndecan-4 was significantly lower in SF than in serum both in RA patients and in OA patients. The underlying mechanism remains unclear. The Syndecan-4 in SF and serum may be originated from different sources. Syndecan-4 is expressed ubiquitously by a variety of cells, including mesenchymal stem cells, epithelial cells, endothelial cells, fibroblast, and ect [22]. The cartilage degradation is an important source of Syndecan-4 in SF. Unfortunately, we did not investigate the cartilage inflammation and cartilage degradation of these patients. Although the SF syndecan-4 concentration was low, it was positively correlated with RA composite disease activity, as well as TJC, SJC and CRP. The correlation coefficents (r value) of syndecan-4 in SF with disease activity was even higher than that of syndecan-4 in sera with disease activity.

We also found that the serum syndecan-4 level was significantly higher in RF positive than in RF negative RA patients. This might suggest the involvement of syndecan-4 in the humoral immunity. Endo T et al. found that syndecan-4 deficient mice had reduced numbers and migration function of B cells, causing deficient germinal center (GC) formation in draining lymph nodes, suggesting that syndecan-4 may contributed to the development of CIA by promoting GC formation [7].

Syndecan-4 is one of the receptors of a proliferation inducing ligand (APRIL), which has been proved to be a critical signal for plasmablasts survival and an important molecular in the regulation of maturation and differentiation of B cells [23]. It role of APRIL on the pathogenesis of RA had been revealed [24]. Our previous study also demonstrated that the serum APRIL was higher in RA than in OA patients, higher in sero-positive RA patients than in sero-negative ones, and more infiltration of APRIL positive cells in RA synovium than in OA patients [25]. The underlying specific mechanisms require further investigation.

We found two outliers in Fig. 3, indicating patients with remarkably high levels of syndecan-4 in the remission or low disease activity groups. This may be related with cartilage degradation. Syndecan-4 is involved in the degradation of cartilage, possibly, by upregulating the expression of MMP-3 and the activity of ADAMTS5 [14].

Interestingly, we found the serum syndecan-4 concentration was higher in OA patients than healthy controls. OA is degenerative joint disease characterized by progressive loss of cartilage matrix and cartilage clefts. Studies have demonstrated the involvement of syndecan-4 in the pathogenesis of OA [5, 14, 26]. We found syndecan-4 expression in synovium was similar in RA and OA, suggesting striking similarities in the pathogenesis of these two diseases. Whereas, Godmann et al. observed dramatically higher expression of syndecan-4 in synovium of RA than OA [8]. This may be due to different degrees of joint inflammation or bone/cartilage destruction among patients.

There were several limitations in our study. First, Syndecan-4 is involved in the degradation of cartilage, but we did not investigate the cartilage inflammation and cartilage degradation. Further research must be conducted. Second, limited samples in the anti-CCP antibody negative group. This may explain why the concentration of Syndecan-4 in the anti-CCP negative group was not significantly different from the concentration in anti-CCP-positive RA patients..

Conclusions

The serum syndecan-4 concentration is higher in RA patients than in OA patients and healthy controls, and significantly higher in RF-positive RA patients than in RF-negative ones. Serum and SF syndecan-4 concentrations were both positively correlated with the disease activity of RA patients. Although the role of syndecan-4 in RA remains elusive, it should be given more attention and prioritised.