Background
Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by inflammation, joint stiffness, and pain, finally leading to tissue destruction [
1]. Excessive proliferation fibroblast-like synoviocytes (FLS) play a vital role in RA progression, which promotes pannus formation, produces inflammatory mediators, and finally aggravates cartilage damage [
2]. Pannus formation is one of the driving pathologic processes which can lead to the development of joint erosion in RA. Connective tissue growth factor (CTGF) is a cysteine-rich protein secreted by FLS in RA patients, which can induce the proliferation of FLS to form pannus, attack cartilage, and exacerbate the disease [
3]. Our previous proteomic research showed that CTGF expression in FLS from RA patients was obviously higher than that in the controls [
4], and the clinical results revealed that the concentration of CTGF in serum of RA patients showed the same tendency. Besides, CTGF was a biomarker for the diagnosis of RA [
5]. Furthermore, it was also demonstrated that CTGF can prompt human umbilical vein endothelial cell (HUVEC) proliferation and migration [
4]. These results indicate that CTGF correlates well with RA disease activity. Nozawa et al. found that anti-CTGF antibody could reduce the clinical score and the content of IL-2 and matrix metalloproteinase-3 in serum in collagen-induced arthritis (CIA) mice [
6]. The findings hint that the drug which can regulate the expression of CTGF may effectively ameliorate disease progression in patients with RA.
Resolvins, one of specialized pro-resolving mediators (SPMs), are derived from omega-3 fatty acids during the resolution phase of inflammatory response. Resolvins include E resolvins (RvE) and D resolvins (RvD), and RvD comprises RvD1–6 [
7]. They have been reported to have many functions, including limiting the migration of neutrophils [
8‐
10], suppressing the production of inflammatory factors, and strengthening phagocytic ability of macrophages in acute inflammation [
11]. The effects of resolvins on chronic inflammatory disease have also attracted attention in recent years. Arnardottir et al. found that resolvin D3 (RvD3) could reduce the number of leukocytes in serum to ameliorate RA progression [
12]. Besides, Lima-Garcia et al. also discovered that resolvin D1 (RvD1) could relieve pain in adjuvant-induced arthritis in rats [
13]. These findings suggest that RvD1 is associated with RA. However, whether or not RvD1 can inhibit the expression of CTGF and pannus formation in RA progression is still unclear.
MicroRNAs (miRNAs) are a family of small non-coding RNAs that usually downregulate gene expression through targeting the 3′-UTR of mRNA [
14]. Recently, there are increasing researches revealing that miRNAs play an important role in RA progression [
15]. In FLS of RA patients, miRNA-23b can regulate arthritic inflammation by inhibiting the NF-κB signal pathway [
16]. Furthermore, SPMs can regulate the Treg/Th17 imbalance in CIA mice by upregulating miR-21 [
17]. These studies indicate that RvD1 is apt to exert its biological functions through microRNA.
In this study, we aimed to determine the effect of RvD1 on pannus formation and the expression of CTGF in RA progression, and shed light on the function of microRNA during this process.
Materials and methods
RvD1
RvD1 (C22H32O5, 7S,8R, 17S-trihydroxy-4Z, 9E, 11E, 13Z, 15E, 19Z-docosahexaenoicacid, CAS No. 872993–05-0) was purchased from Cayman Chemical Company, Ann Arbor, USA (cat. Number 10012554). The concentration and purity of the RvD1 were identified by UPLC-MS/MS. RvD1 was aliquoted into several brown glass tubes by a glass HAMILTON syringe to guarantee the activity. The tubes were then evacuated oxygen by nitrogen gas and were stored at − 80 °C to avoid repeated freezing and thawing.
Patients and samples
Samples were acquired from RA patients and healthy controls at the First Affiliated Hospital of Wenzhou Medical University from May 2016 to May 2017. RA diagnosis was in accordance with the 2010 American College of Rheumatology (ACR) criteria. The clinical information of the patients was shown in supplementary Table 1. Serum was obtained from patients with RA on the first day of clinical admission before undergoing any treatment, and the control serum was from the healthy individuals. This study was approved by the Clinical Research Ethics Committees of the First Affiliated Hospital of Wenzhou Medical University (No. 2016157). All patients participated in this study provided written informed consent.
The concentration of RvD1 determined by UPLC-MS/MS
The concentration of RvD1 was measured as described previously [
18]. Each sample was dissolved with ice-cold methanol. RvD1 levels were determined by ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS) using an UPLC I-Class system (Waters, USA) equipped with an Agilent Eclipse Plus C18 column (2.1 mm × 100 mm × 1.7 μm) paired with Sciex 6500 Q-TRAP mass spectrometer (Sciex, USA). To monitor and quantify the levels of the various LM, a multiple reaction monitoring (MRM) method was developed with signature ion fragments for RvD1. Identification and quantification of RvD1 were conducted using previously published criteria in which a minimum of 6 diagnostic ions were employed and based on peak area of MRM transitions and the linear calibration curve of RvD1 respectively.
Collagen-induced arthritis (CIA) model
DBA/1 mice (male, 7–8 weeks old) were obtained from SLAC Laboratory Animal Co. (Shanghai, China). All experiment procedures were approved by the Institutional Animal Care and Use Committee of Wenzhou Medical University. CIA model in DBA/1 mice was induced as described previously [
19]. Briefly, on day 0, mice were injected intradermally at the base of the tail with 100 μL of type II bovine collagen (2 mg/mL, Chondrex, USA) emulsified in equal volumes of complete Freund’s adjuvant (CFA, Sigma-Aldrich, USA) containing heat inactivated Bacillus Calmette-Guerin. After 21 days, mice were given a booster immunization with 100 μL of type II bovine collagen (2 mg/mL) emulsified in equal volumes of incomplete Freund’s adjuvant (IFA, Sigma-Aldrich, USA). On the booster immunization day, mice were treated with RvD1 (0, 20, and 100 ng) by tail vein injection every third day or mouse miRNA146a-5p agomir by intra-articular injection once a week until day 48. From day 21 to day 48, the mean clinical score of each mouse was evaluated according to the standardized method [
19]. The mean clinical score (0–4) was assigned as follows: 0 = no symptoms, 1 = erythema and slight swelling limited to the ankle joint and toes, 2 = erythema and slight swelling spreading from the ankle to the midfoot, 3 = erythema and severe swelling spreading from the ankle to the metatarsal joints, and 4 = ankylosing deformity with joint swelling. Mice were sacrificed on day 49. To collect the synovial fluid from mice, saline was injected into mice articular cavity, and then gathered. Serum and joint tissues were harvested for further study.
Histopathology evaluation
Samples were obtained from the knee joints of sacrificed mice. Then the specimens were fixed in 4% paraformaldehyde, decalcified in 50 nM EDTA, and embedded in paraffin. They were afterwards serially sectioned onto microscope slides at a thickness of 5 μm and then deparaffinized, rehydrated, and stained with hematoxylin and eosin (H&E) or toluidine blue.
ELISA
The concentrations of TNF-α, IL-1β, and IL-6 in mice serum and CTGF in human serum were detected using specific commercially available ELISA kits according to the manufacturer’s instructions (GenWay Biotech, USA). The concentrations of TNF-α, IL-1β, IL-6, and CTGF in supernatants from RA FLS were determined by commercially available ELISA kits (GenWay Biotech, USA). The concentration of CTGF in mice serum was detected by commercial ELISA kit according to the manufacturer’s instructions (Biorbyt, England).
Isolation and culture of RA FLS
RA FLS were isolated from synovial tissues according to the method described previously [
20]. Passage 3–5 RA FLS were used for each experiment, for these, cells were purer and had better biological functions than other passages. This study was approved by the Clinical Research Ethics Committees of the First Affiliated Hospital of Wenzhou Medical University (No. 2016157). All RA patients signed informed consent before the research started.
MTT assay
A 3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was used to assess the effects of RvD1 and miRNA-146a-5p on RA FLS proliferation. The RA FLS were seeded in a 96-well plate at the concentration of 1 × 106/ml and incubated for 24 h. Then the cells were treated with various concentrations of RvD1 (0, 20, and 100 nM) or human miRNA-146a-5p mimics. The experiments were performed in duplicate. Plates were incubated at 37 °C in 5% CO2 for different times (12 h, 24 h, 48 h, and 72 h), and MTT dissolved with dimethyl sulfoxide (DMSO; Sigma-Aldrich, USA) was added to the culture medium 4 h before the final test. The absorbance value at the 490 nm wavelength was measured using a Biotek Synergy2 spectrophotometer (Winooski, VT, USA).
Scratch migration assay
RA FLS were seeded in six-well plates at a density of 2.5 × 105 cells/well. A scratch was made along the diameter of the well using a 10-μL pipette tip (Axygen® Corning, USA) once the cells reached close to 95% confluency. The wells were gently washed 3 times with PBS to remove the detached cells. DMEM containing various concentrations of RvD1 (0, 20, 100 nM) was added to the wells, and the cells were grown for an additional 72 h. During this period, images were captured after 24 h, 48 h, and 72 h incubation by a camera under the light microscope (× 40 magnification). ImageJ 2.43 s was used to calculate the mobility ratio, and the mobility ratio was calculated using the following equation: migrated cellular area/scratched area × 100%.
Endothelial tube formation assay was performed to evaluate the angiogenic activity in vitro as described previously [
21]. Matrigel was diluted with DMEM at the ratio of 1:1, then 96-well plates were coated with Matrigel and incubated at 37 °C for about 1 h to promote gelling. Human umbilical vein endothelial cells (HUVEC) were added to each well with different concentrations of RvD1 (0, 20, 100 nM), human miRNA-146a-5p agomir, or human CTGF-specific siRNA lentivirus (Genepharma, China) and incubated in a humidified incubator at 37 °C with 5% CO
2. After 4 h, 6 h, and 8 h incubation, the plates were observed and captured by a camera under a microscope. Tube-like structures in each well were evaluated by the number of intersections among branches of the endothelial cell networks in the whole field.
CAM assay
The chick chorioallantoic membrane (CAM) assay was carried out to evaluate the angiogenic activity in vivo as described previously [
22]. Embryonated chicken eggs (~ 10 per treatment) were incubated at 37 °C and treated with RvD1 (0, 20, 100 nM), chicken miRNA-146a-5p agomir or a lentiviral vector harboring RNAi sequence targeting the CTGF gene (Gene Pharma, China) respectively, which were absorbed on sterile Whatman GB/B glass fiber filter disks (6 mm in diameter; Reeve-Angel, Clifton, NJ, USA) on day 7. The blood vessels in embryos were observed 3 days later under a stereomicroscope. ImageJ 2.43 s was used to assess the vascular and CAM areas. The percentage of angiogenic area was calculated using the following equation: vascular area / CAM area × 100%.
MiRNA microarray and MiRNA transfection assay
The microRNA expression profiling of RA FLS treated with RvD1 was determined by miRNA microarray analysis with the human miRNA array probes (Exiqon, Denmark). QRT-PCR was used to validate the upregulation of miRNA-146a-5p. RA FLS were then transfected with human miRNA146a-5p mimic/inhibitor or a non-sense strand negative control using Lipofectamine™ 3000 as a transfection reagent. Lipofectamine 3000 was mixed with 50 μl Opti-MEM. Meanwhile, miRNA146a-5p mimic/inhibitor was mixed with 50 μl Opti-MEM. The two mixtures were then mixed for 5 min and then added to the cell culture medium and left to incubate for 48–72 h at 37 °C in 5% CO2. Cells were collected for total RNA extraction.
Quantitative real-time PCR analysis
Total RNA samples in RA FLS were isolated by TRIzol Reagent according to the manufacturer’s protocol. The cDNA was synthesized by PrimeScript™RT reagent Kit with gDNA Eraser (TAKARA, Japan) and the expression of mRNA was detected with SYBR® Premix Ex Taq™ II (TAKARA, Japan) by quantitative real-time PCR (qRT-PCR). qRT-PCR of miRNA was performed with miRNA First-Strand Synthesis and SYBR qRT-PCR kit (Clontech, USA). The gene-specific primers used were listed in supplementary Table 2. The relative expression of miRNA was normalized to U6 controls while mRNA was normalized to β-actin and was calculated using the 2−ΔΔCt method.
Western blot analysis
Western blot analysis from RA FLS homogenates were performed as described previously [
23]. Equal amounts of protein per sample were separated by 10% SDS-PAGE, and subsequently transferred to nitrocellulose membrane (Pierce). The membranes were blocked for 2 h with 5% skimmed milk and then incubated in the primary antibodies overnight at 4 °C. Horseradish peroxidase-linked anti-rabbit antibodies were used as secondary antibodies. The membranes were imaged with the Image Quant LAS 4000 mini (GE Healthcare Bio-Sciences AB, Uppsala, Sweden).
Statistical analysis
SPSS software 19.0 and GraphPad Prism 7.0 were employed to analyze the data. All results were expressed as means ± SD. All of the experiments were carried out in triplicate. Normally distribution of data was determined by the Shapiro-Wilk method, and the homogeneity of variance was determined by Levene method. Student’s t test was used to analyze the difference between two sets of data that met the normal distribution and homogeneity of variance. One-way analysis of variance test was applied to analyze the differences among multigroups.
Discussion
Recently, Lima-Garcia et al. found that the aspirin-triggered RvD1 epimer had anti-hyperalgesic and suppression of pro-inflammatory effects in AIA mice through the inhibition of NF-κB activation [
13], but the specific mechanism is not yet clear. In the current study, our data revealed that the concentration of RvD1 in serum from RA patients was lower than that in healthy controls. Furthermore, an inverse correlation between the concentrations of CTGF and RvD1 in serum was detected. In addition, we also verified that RvD1 could decrease CTGF expression in RA FLS. CTGF, which is associated with several biological functions such as fibrosis, tumorigenesis, angiogenesis, and endochondral ossification [
28,
29], seems to have a close relation with RA. Nozawa et al. demonstrated that CTGF promoted the articular damage by increased osteoclastogenesis in RA patients [
30]. Ding et al. found that CTGF could promote articular damage by increased proliferation of FLS in RA [
31]. Besides, our previous proteomic research indicated that CTGF expression in FLS from RA patients was remarkably higher than that in the controls [
4], and the clinical results revealed that CTGF could be used as a biomarker for the diagnosis of RA. In this study, we demonstrated that CTGF could promote angiogenesis. Therefore, it is natural that we investigate if RvD1 has its effect on angiogenesis.
We found that RvD1 could inhibit angiogenesis by in vitro experiment study. Moreover, we demonstrated that RvD1 could inhibit pannus formation and decrease the levels of pro-inflammation cytokines and CTGF in CIA mice. Considering the role of CTGF in the process of angiogenesis, it came to us that RvD1 could downregulate the expression of CTGF to alleviate RA progression, but the mechanism is still vague.
Therefore, miRNA microarray studies were performed in RA FLS treated with RvD1. Our data revealed that RvD1 upregulated the level of miRNA-146 while synchronously downregulated the level of miRNA-155 and miRNA-181. Based on the
p value and fold changes, miR-146a-5p was selected for further study. MicroRNA-146a has been widely reported for its multiple roles in the control of the innate and adaptive immune processes and for its oncogenic role in some tumors and arthritis [
32]. Boldin et al. found that miRNA-146a-5p played a key role as a molecular brake on inflammation, myeloid cell proliferation, and oncogenic transformation [
32]. MiRNA-146a was also known as a major molecular regulator in arthritis. Mice with miRNA-146a deficiency are more likely to develop severe gouty arthritis [
33]. What is more, miRNA-146a could inhibit pathogenic bone erosion in inflammatory arthritis [
34]. Nakasa et al. affirmed that administration by intravenous injection of miR-146a could prevent joint destruction in CIA mice [
35]. We discovered that miRNA-146a could inhibit RA FLS proliferation and angiogenesis. Besides, transfection experiment revealed that over-expression of miRNA-146a-5p in RA FLS significantly decreased inflammatory mediators and CTGF levels. In vivo, we clearly demonstrated that miRNA-146a-5p could inhibit angiogenesis and decreased the level of CTGF to delay RA progression. Taking all the fact into account, we concluded that RvD1 upregulated miRNA-146 to decrease the level of CTGF thus ameliorating RA progression.
As for the mechanism of miR-146a-5p regulating the expression of CTGF, it is predicted by bioinformatics software that CTGF was not the target gene of miRNA-146a-5p, so we concluded that miRNA-146a-5p could not directly bind to CTGF mRNA to decrease CTGF expression. Previous reports and our study have demonstrated that miR-146a-5p can inhibit the producing of IL-6 by suppressing NF-κB signaling pathway. In addition, IL-6 was responsible for STAT3 upregulation [
36], and STAT3 signaling activation was required for the expression of CTGF [
37]. In our study, the results of Western blotting showed that miRNA-146a-5p inhibited STAT3 activation in the research. Consequently, we speculated that miRNA-146a-5p downregulated IL-6, thereby blocking the expression of STAT3 and finally inhibiting the expression of CTGF. In short, our hypothesis is that miR-146a-5p can downregulate CTGF expression via the IL-6/STAT3 signaling pathway.
Furthermore, miRNA microarray revealed that RvD1 decreased the level of miRNA-155 and miRNA-181 in RA FLS. MiR-155 is also one of the key miRNAs in RA pathogenesis [
38]. An increasing number of literatures have supported the concept that miRNA-155 played a key role in keeping Th1/Th2/Th17 balance in the context of autoimmunity [
38]. In CIA mice, miRNA-155 played a positive regulatory role in the development of pathogenic Th17 cells [
39]. Besides, the homeostasis of Th17/Treg was also regulated by miR-181 in RA. For example, miR-181 was reported to promote the differentiation of Th17 cells [
38]. In our study, the expressions of miRNA-155 and miRNA-181 in RA FLS were inhibited by RvD1. Despite the fact that RA FLS are not immune cells, they play a vital role in RA procession. Therefore, the effects of miRNA-155 and miRNA-181 on RA FLS are pending for further study.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.