Resistin upregulates chemokine production by fibroblast-like synoviocytes from patients with rheumatoid arthritis

Samples

Synovial tissues were obtained from patients with RA and with osteoarthritis (OA) who underwent total knee or hip replacement. FLSs were prepared from synovial tissues as described previously [12]. FLSs from RA synovial tissues were also obtained from the Japanese Collection of Research Bioresources Cell Bank. The experimental protocol was approved in advance by the Ethics Committees of Toho Medical Center Omori Hospital (M16020) and the Ethics Committees of the Faculty of Medicine, Toho University (27060, 2703024007).

Immunohistochemical assessment

Synovial tissues were fixed with freshly prepared 4% (v/v) paraformaldehyde in Tris-buffered saline. Sections (3 μm) were immersed in ethanol containing 3% (v/v) H₂O₂ for 30 min to block endogenous peroxidase activity. Sections were incubated with protein block serum-free (Agilent Technologies) for 30 min to block non-specific binding. Sections were subsequently incubated at 4 °C overnight with a rabbit anti-resistin polyclonal antibody (pAb) (Bioss Antibodies), rabbit anti-adenylyl cyclase-associated protein 1 (CAP1) monoclonal antibody (mAb) (EPR8339(B); Abcam), or isotype control (Agilent Technologies) as a primary antibody. Expression was detected using an EnVision + kit™ (Agilent Technologies) and counterstained with hematoxylin.

Non-specific binding was blocked with protein block serum-free for immunofluorescence double staining, and sections were incubated at 4 °C overnight with rabbit anti-resistin pAb or CAP1 mAb. Samples were subsequently incubated at room temperature with biotinylated anti-rabbit IgG for 40 min followed by Fluorescein Avidin D (Vector laboratories) for 20 min. Samples were incubated with mouse anti-CD68 mAb (KP1; Abcam), mouse anti-cadherin-11 mAb (16A; Acris Antibodies), or mouse anti-von Willebrand factor (vWF) mAb (F8/86; Agilent Technologies) and then with a Texas Red® horse anti-mouse IgG antibody at room temperature for 20 min. A nuclear stain was performed with 4′, 6-diamidino-2-phenylindole. Slides were examined using the BX61 (Olympus). To determine the percentages of CAP1-expressing cells, the number of CAP1-positive cells in the lining layer, and in the CD68-positive or cadherin-11-positive cells was counted under fluorescence microscope.

RNA extraction

FLSs were seeded in Roswell Park Memorial Institute (RPMI) 1640 medium containing 10% fetal bovine serum (FBS) in 10-cm dishes (1 × 10⁶ cells/dish), and were then incubated with 1000 ng/ml resistin (PeproTech) for 18 h. Total RNA was isolated from FLSs using TRIzol® (Invitrogen) according to the manufacturer’s instructions. RNA samples were digested with an RNase-free DNase set (Qiagen) to remove genomic DNA and further purified using the RNeasy kit (Qiagen). The quality of RNA samples was examined by the Agilent 2100 Bioanalyzer (Agilent Technologies) using RNA 6000 NanoChips. RNA samples with an RNA integrity number higher than 7 were used in further analyses, including a RNA-Seq and real-time polymerase chain reaction (PCR).

RNA-Seq transcriptome analysis

The RNA-Seq library was prepared using the SureSelect Strand-Specific RNA Library Prep Kit (Agilent Technologies) in accordance with the manufacturer’s instructions optimized to Illumina Multiplexed Sequencing. After purification of the amplified libraries, the DNA quality of products was assessed using the 2100 Bioanalyzer DNA 1000 Assay. Paired-end sequencing of the RNA-seq libraries was performed using an Illumina MiSeq system (Illumina; 2 × 75 bases paired-end run). FASTQ files were imported into CLC Genomics Workbench v9.01 software (CLC bio) for post-processing and data analysis. Sequences were trimmed based on the FASTQC report and mapped onto annotated human genes with support from reference human genome (hg19). Data were normalized by total reads per million and analyzed for differential gene expression empirical analysis of differential gene expression subroutine based on reads per kilobase of transcript per million reads mapped (RPKM).

Quantitative real-time PCR

Total RNA samples were reverse transcribed into cDNA using random primers and an RNA PCR Kit (AMV) Ver.3.0 (Takara Bio Inc.). CXCL1, CXCL8, and chemokine (C-C motif) ligand (CCL) 2 levels were measured using the Power SYBR Green PCR Master Mix (Applied Biosystems). The following primers were used for analyses: 5′-TGC AGG GAA TTC ACC CCA AG-3′ and 5′-CAG GGC CTC CTT CAG GAA CA-3′ for CXCL1; 5′-ACT CCA AAC CTT TCC ACC CCA-3′ and 5′-TTT CCT TGG GGT CCA GAC AGA-3′ for CXCL8; 5′-CTT CTG TGC CTG CTG CTC AT-3′ and 5′-CGG AGT TTG GGT TTG CTT GTC-3′ for CCL2; 5′-GAA GGT GAA GGT CGG AGT CA-3′ and 5′-GAG GTC AAT GAA GGG GTC AT-3′ for glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The Prism 7500 Fast Real-time PCR system (Applied Biosystems) was used for analyses, and the mRNA levels of the genes tested were represented as relative values to the expression level of GAPDH.

Reverse transcription (RT)-PCR

Total RNA was extracted with an RNeasy Mini kit (Qiagen) from the cultured FLSs of patients with RA. RT was performed using a SuperScript first-strand synthesis system for RT-PCR according to the recommendations of the manufacturer (Invitrogen) with 1 μg of total RNA from FLSs. Equal amounts of each RT product were amplified by PCR with HotStarTaq® DNA polymerase (Qiagen). The primer sequences and numbers of base pairs (bp) were as follows: for CAP1 (118 bp), 5′-AGG CAT TTG ACT CGC TGC TTG and 5′- TCG CTC CAA CTT CAA ACC TGT G; and for GAPDH (598 bp), 5′- CCA CCC ATG GCA AAT TCC ATG GCA and 5′-TCT AGA CGG CAG GTC AGG TCC ACC. After initial denaturation at 95 °C for 15 min, PCR involved amplification for 32 cycles at 95 °C for 30 s, at 56 °C for 30 s, and at 72 °C for 45 s, followed by elongation at 72 °C for 5 min. Amplified DNA fragments were resolved by electrophoresis on a 2% agarose gel, and were detected under ultraviolet light using LAS-3000 (Fujifilm) after staining the gel with ethidium bromide.

Western blot analysis

The western blotting procedure was previously described [13]. Membranes were incubated with rabbit anti-CAP1 mAb (EPR8339(B); Abcam) or rabbit anti-GAPDH pAb (Santa Cruz Biotechnology), with a dilution of 1:1000 (CAP1) and 1:100 (GAPDH), the secondary antibody (horseradish peroxidase-conjugated goat anti-rabbit antibody) was added (at a dilution of 1:2000), and an incubation was performed for 3 h using the iBind Flex Western System (Thermo Fisher Scientific). Protein bands were detected with the enhanced Novex® ECL Chemiluminescent Substrate Reagent Kit (Invitrogen) using LAS-3000 (Fujifilm).

Enzyme-linked immunosorbent assay (ELISA)

FLSs were cultured overnight in 96-well plates (2 × 10⁴ cells/well) and then incubated with recombinant human resistin (0, 10, 100, or 1000 ng/ml; PeproTech) at 37 °C for 24 h in RPMI1640 medium containing 1% FBS. Concentrations of CXCL8, CCL2, IL-1β, IL-6, and IL-32 in culture supernatants were assessed using the ELISA kit (R&D Systems), according to the instructions of the manufacturer.

Signaling pathway of resistin via CAP1

Stealth RNAi™ small interfering RNA (siRNA) targeting CAP1 and negative control siRNA were purchased from Thermo Fisher Scientific. Lipofectamine® RNAiMAX reagent (Thermo Fisher Scientific) was used to formulate transfecting siRNAs. FLSs were transfected with siRNA at 37 °C for 48 h, and cells were then treated with resistin (PeproTech) for another 24 h. CXCL8 concentrations in the culture supernatant were assessed by ELISA.

Statistical analysis

Results are expressed as the mean +/- standard error (SE). Statistical analyses were performed using StatFlex software (ver. 6; ARTEC). The production of CXCL8, CCL2, and IL-6 was analyzed by analysis of variance using Dunnett’s test. The paired t test was applied to compare CXCL8 production between control siRNA-transfected and CAP1 siRNA-transfected cells. In all analyses, p < 0.05 was considered to indicate significance.

Expression of resistin and CAP1 in synovial tissues of RA

We immunohistochemically investigated the expression of resistin and CAP1 in synovial tissues harvested from patients with RA and with OA. Resistin was strongly expressed in the synovial lining and sub-lining cells of synovial tissue in RA (Fig. 1a), while resistin expression was minimal in the synovium in OA (Fig. 1b). On the other hand, the expression of CAP1 was observed in the lining and sub-lining cells of the synovium in RA (Fig. 1c), while CAP1 was expressed in the lining cells of the synovium in OA (Fig. 1d). In fat tissue around the synovial tissue, resisitin was weakly expressed in RA and OA (Fig. 1g and h).

We performed double immunohistochemical assessment to identify resistin-expressing and CAP1-expressing cells in the synovium in RA. As shown in Fig. 2, resistin was expressed by CD68⁺ macrophages (Fig. 2a-c). However, resistin was not expressed by vWF⁺ endothelial cells (Fig. 2d-f). CAP1 was expressed by CD68⁺ macrophages (Fig. 2g-i) and also by cadherin-11⁺ FLSs (Fig. 2j-l) and vWF⁺ endothelial cells (Fig. 2m-o) in synovial tissues in RA. The percentage of CAP1-positive cells was 94% (187/200) in the lining layer. The frequency of CAP1-positive cells in CD68⁺ cells was 97% (97/100) in the lining layer and 95% (95/100) in the sub-lining layer. The percentage of CAP1-positive cells in cadherin-11⁺ cells was 95% (95/100) in the lining layer and 81% (81/100) in the sub-lining layer.

CAP1 expression in cultured FLSs established from synovial tissue in RA was evaluated with RT-PCR and western blotting. The mRNA and protein expression of CAP1 was also observed in in vitro cultured FLSs (Fig. 3).

Stimulatory effects of resistin on FLSs in RA

We also analyzed CXCL1, CXCL8, and CCL2 expression by quantitative real-time RT-PCR using three lots of FLSs from patients with RA. The expression of CXCL1, CXCL8, and CCL2 increased in all three lots of FLSs following the resistin stimulation (Fig. 4).

Chemokine production by the resistin-CAP1 pathway

We examined the protein levels of chemokine expression by resistin-stimulated FLSs in vitro. FLSs were incubated with various concentrations of resistin for 24 h. The concentrations of CXCL8 and CCL2, which were observed as upregulated chemokines by RNA-seq and real-time RT-PCR, were assessed in culture supernatants using ELISA kits. The CXCL8 level was significantly increased by the stimulation with resistin (Fig. 5a). CCL2 expression was dose-dependently increased by resistin (Fig. 5b). We also analyzed expression of IL-1β and IL-32, which were identified as upregulated cytokines by RNA-seq (Table 1), and IL-6, which was identified as a slightly upregulated cytokine by RNA-seq (fold change 1.969, 3.721, and 1.997 in each lot). IL-1β and IL-32 were not detected by ELISA with or without stimulation with resistin. The concentration of IL-6 was increased by resistin (Fig. 5c).

In order to verify the involvement of CAP1 in the resistin stimulation, siRNA for CAP1 was transfected into RA FLSs. The FLSs were pretreated with CAP1 siRNA or control siRNA. The transfection of CAP1 siRNA significantly decreased CAP1 expression from that with the transfection of control siRNA (Fig. 6a). Resistin-induced CXCL8 production by FLSs was significantly inhibited by the abrogation of CAP1 by siRNA (Fig. 6b). These results indicate that the resistin-CAP1 pathway contributes to chemokine production by RA FLSs.