Cyr61 is involved in neutrophil infiltration in joints by inducing IL-8 production by fibroblast-like synoviocytes in rheumatoid arthritis

Male, DBA/1 J mice, six- to eight-weeks old, were purchased from the Shanghai Laboratory Animal Center, Chinese Academy of Science. Mice were maintained under pathogen-free conditions. All experiments were performed in accordance with guidelines and approved by the Animal Care and Use Committee of Shanghai Jiaotong University School of Medicine (2013028).

A total of 46 RA patients (6 men and 40 women, 30- to 84-years old, mean and SD 57 ± 12 years) were included in the study. The disease duration of the RA patients was 16 ± 9 years. The diagnosis of RA was based on the revised criteria of the American College of Rheumatology [30]. The control subjects were 27 osteoarthritis (OA) patients fulfilling the diagnostic criteria of OA proposed by Altman [31]. Clinical characteristics of RA and OA patients are given in Additional file 1: Table S1. ST were obtained from patients and FLS were cultured and identified as reported previously [28]. SF and cell culture supernatants were collected as reported previously [28]. The study was approved by the Institutional Medical Ethics Review Board of the Shanghai Jiaotong University School of Medicine and informed consent was obtained from each of the individuals.

To prepare single-cell suspensions from SF and ST, SF specimens were centrifuged at 500 g for 10 minutes, and cells were collected, counted and resuspended in phosphate-buffered saline (PBS) for flow cytometric analysis; ST specimens were minced into small pieces and incubated for two hours with 1 mg/ml type I collagenase (Sigma-Aldrich, Bornem, Belgium) in (D)MEM at 37°C, then cells were collected by filtering the suspension through nylon mesh and immediately used for flow cytometric analysis. For surface markers staining, fluorescence conjugated CD3, CD11b, CD14, CD15, CD16 and CD19 (eBiosciences, San Diego, CA, USA) antibodies were used. Flow cytometry was performed using a FACS Calibur cytometer (BD Biosciences, San Jose, CA, USA) and analyzed using Cellquest software (BD Biosciences).

Total RNA was extracted from cells and real-time PCR was performed as previously reported [28]. Briefly, total RNA was extracted from specimens using a Tripure isolation reagent (Roche Diagnostics, Indianapolis, IN, USA), according to the manufacturer’s instructions. The RNA quality and quantity were evaluated by a NanoDrop ND-1000 Spectrophotometer (NanoDrop, Wilmington, DE, USA). The integrity of RNA was appraised with gel analysis for the intact 28S and 18S ribosomal RNA. Messenger RNA (mRNA) was converted to cDNA using a RevertAidTM First Strand cDNA Synthesis Kit (Thermo Scientific, Glen Burnie, MD, USA) according to the manufacturer’s instructions. Real-time PCR was performed using SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. The primers used in this study are shown in Additional file 1: Table S2.

Cyr61, IL-1β and TNF-α small interfering RNA (siRNA, Additional file 1: Table S3) were designed and synthesized at Shanghai GenePharma (Shanghai, China) and gene knockdowns were performed as previously reported [28, 29]. In brief, FLS were cultured in 24-well plates. A transfection mixture of siRNA oligonucleotides and Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) in serum-free medium was added to medium-aspirated cells for four hours. Then, the medium was replaced with complete (D)MEM containing 10% fetal bovine serum for an additional 24 hour incubation.

Special inhibitors of the NF-κB and mitogen-activated protein kinase (MAPK) signaling pathways were purchased from Sigma-Aldrich and used to analyze Cyr6-induced IL-8 production. Briefly, 4 μM pyrrolidine dithiocarbamate (PDTC; an inhibitor of NF-κB activation), 10 μM SB203580 (an inhibitor of p38 MAPK), 1 μM PD98059 (an inhibitor of ERK1/2), or 20 μM SP600125 (an inhibitor of JNK) was added to the cell culture together with 5 μg/ml Cyr61 at the same time; then expression of IL-8 was determined using real-time PCR and the concentration of IL-8 in the supernatant was evaluated by ELISA. The activations of AKT, JNK, ERK1/2 and NF-κB were analyzed using western blotting with specific antibodies.

The concentration of IL-8 in the cell culture supernatant and SF was determined by a sandwich ELISA (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instructions. The level of Cyr61 was measured by ELISA as described previously [28].

Protein immune blotting was performed as described previously [28]. In brief, tissue or cell lysates were separated by SDS–PAGE electrophoresis and then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore Corporation, Bedford, MA, USA) at 100 v for 90 minutes. The phosphorylation of AKT, JNK, ERK1/2 and NF-κB and the expression of MIP-2 were analyzed using specific antibodies (Cell Signaling Technology Inc, Beverly, MA, USA). After washing with PBS, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit immunoglobulin G (IgG) at room temperature for one hour followed by washing with PBS. The target proteins were examined with an ECL system (Millipore Corporation, Bedford, MA, USA) and visualized with autoradiography film.

NF-κB nuclear translocation in FLS was studied with a confocal laser scanning fluorescence microscopy (LSM510; Zeiss, Jena, Germany) technique as described before [29]. In brief, FLS grown on glass coverslips were stimulated with 5 ug/ml Cyr61 for 30 minutes and fixed with acetone. The fixed cells were stained overnight with anti-NF-κB p65 antibody (Cell Signaling Technology Inc) and incubated for one additional hour with a PE-labeled secondary antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA). After washing, cells were incubated for three minutes with 0.25 mg/ml of 4,6-diamidino-2-phenylindole and examined using an LSM 510 confocal fluorescence microscope.

Neutrophils were isolated from peripheral blood of healthy donors according to the manufacturer’s instructions. In brief, venous blood was drawn and neutrophils were isolated immediately by Polymorphprep (Axis-Shield PoC AS, Oslo, Norway). After lysis of the erythrocytes, the neutrophils were harvested, washed twice with physiological saline and resuspended in RPMI 1640 medium supplemented with 10% fetal bovine serum at a cell concentration of 10⁶/ml. The purified cells consisted of a more than 95% pure population of viable neutrophils, as assessed by morphology and the trypan blue exclusion test.

Chemotaxis was assessed using 24-transwell Boyden chambers of 3 μm pore size (Corning Costar, Cambridge, MA, USA) for neutrophils. FLS were plated in 24-well plates and stimulated with 5 μg/ml Cyr61 (PeproTech, Rocky Hill, NJ, USA) for 48 hours. Then, the culture supernatant was harvested and pre-incubated with anti-IL-8 mAb, (5 μg/mL, PeproTech) or control mAb (5 μg/ml, PeproTech) for one hour. Then, the treated supernatant was added to the lower chambers, while neutrophils were added to the top chambers for incubation for another 90 minutes at 37°C in a humidified atmosphere with 5% carbon dioxide. The filters were fixed with ethanol and stained with crystal violet. The chemotactic response was then determined by evaluating the number of cells that had migrated through the entire thickness of the filter. Triplicate chambers were used in each experiment and five fields were examined in each filter. The results were expressed as the chemotactic index, being the number of cells that migrated towards the sample divided by the number of cells that migrated towards the control medium.

The 181 bp IL-8 promoter sequences (-135 to +46) were PCR amplified from human genomic DNA using the following primers, IL-8 (WT)-Forward: 5′-GTGAGATCTG AAGTGTGATGACTCAGG-3′, which contains an artificial BglII site, and IL-8 (WT)-Reverse: 5′-GTGAAGCTTGAAGCTTGTGTGCTCTGC-3′, which contains an artificial HindIII site [32]. The PCR product was then digested with BglII/HindIII and inserted into the corresponding restriction sites of the luciferase reporter plasmid pGL3-Basic (Promega, Fitchburg, WI, USA) to generate IL-8 (WT) Luc. To generate the IL-8 (mAP-1) Luc, IL-8 (mNF-κB) Luc and IL-8 (mC/EBP) Luc vector that contains the same IL-8 promoter sequences but with mutation that distorts the AP-1, NF-κB and C/EBP consensus, the forward primers (Additional file 1: Table S4) were used together with IL-8 (WT)-Reverse [33]. The PCR products were again digested with BglII/HindIII and ligated into pGL3-Basic.

Human skin fibroblasts (HSFs) were cultured in (D)MEM supplemented with 10% fetal bovine serum. For transient transfections, cells were grown to 70% to 80% confluence in 24-well dishes and maintained serum-free prior to transfection; then, cells were transfected with IL-8WT, IL-8mAP-1, IL-8mC/EBP or IL-8mNF-κB along with pRL-TK using the liposome–mediated method with Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer’s instructions. After a 24-hour incubation period, cells were treated with Cyr61 (5 μg/mL) for an additional two hours, at which time luciferase activity was measured using a Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions.

For chromatin immunoprecipitation (ChIP) assay, FLS cells, either with or without Cyr61 protein (5 μg/mL) stimulation, were cross-linked by formaldehyde fixation. Following cellular and nuclear lysis, isolated chromatin was sheared by sonication and subsequently incubated overnight at 4°C with antibodies against c-Jun, NF-κB p65 (Cell Signaling Technology Inc, Danvers, MA, USA), C/EBPβ (Santa Cruz Biotechnology), or control rabbit IgG (PeproTech). Immunocomplexes were subjected to cross-link reversal, extracted and precipitated as described in the protocol according to the manufacturer’s instructions. The eluted DNA and the aliquots of chromatin prior to immunoprecipitation (input) were subjected to semiquantitative PCR. The PCR primers were used for amplifying IL-8 promoters (-125 to +11) with the following sequences: -125 forward, 5′-ACTCAGGTTTGCCC TGAGGGGA-3′ and +11 reverse, 5′-TGCCTTATGGAGTGCTCCGGTG -3′. The PCR conditions were as follows: one cycle at 95°C for five minutes; 34 cycles at 95°C for 30 seconds, 65°C for 30 seconds, and 72°C for one minute; one cycle at 72°C for five minutes. PCR products were separated by 2% agarose gel containing ethidium bromide. Densitometry was used to quantify the PCR results, and all results were normalized by respective input values.

CIA was induced as described previously [34]. Briefly, male DBA/1 J mice were injected intradermally with 150 μg of chicken type II collagen (Chondrex, Redmond, WA, USA) in 0.05 M acetic acid emulsified in Freund’s complete adjuvant. Booster injections were administered on day 21 with a total of 75 μg collagen II in Freund’s incomplete adjuvant. Joint inflammation was evaluated on a scale of 1 to 4 [35], with a maximum clinical score of 16 per mouse. Mice were treated with control IgG1 or anti-Cyr61 mAb 093G9 generated in our laboratory (200 μg/mouse) twice a week when the score reached 2.

The joints were removed from sacrificed CIA mice and fixed in 10% phosphate-buffered formalin, decalcified in 10% ethylenediaminetetraacetic acid (EDTA), embedded in paraffin, stained with H & E and examined by light microscopy according to standard protocols.

Slides were deparaffinized through a series of xylene baths and rehydrated through graded alcohols. The sections were then immersed in methanol containing 0.3% hydrogen peroxide for 20 minutes to block endogenous peroxidase activity and incubated in 2.5% blocking serum to reduce nonspecific binding. Sections were incubated overnight at 4°C with anti-human CD15 mAb or anti-mouse Gr-1 mAb (eBiosciences); mouse IgM or rat IgG was used as negative control in the study. Slides were then incubated in anti-mouse IgM HRP or anti-rat IgG HRP. Vector NovaRED substrate (Vector Labs, Burlinghame, CA, USA) was used as the peroxidase substrate and slides were counter-stained with a hematoxylin solution. Stained sections were dehydrated and then mounted by light microscopy.

All experiments were performed in triplicate. The difference among groups was determined by analysis of variance (ANOVA) and comparison between two groups was analyzed by the t-test using the GraphPad Prism 4.0 (GraphPad Software, In c., San Diego, CA, USA). A value of P <0.05 was considered statistically significant.

As large numbers of studies have identified a variety of different cells involved in the pathogenesis of RA [2], we first investigated the profile of infiltrating inflammatory cells in SF from RA and OA patients. The results showed that there were large numbers of leukocytes, including a population of CD11b⁺CD15⁺, CD3⁺, CD19⁺ and CD14⁺CD16⁺ cells in RA SF. In contrast, few to no cells were detectable in SF from OA patients (Figure 1A). Further analysis showed that 25% to 65% of the examined cells were CD11b⁺CD15⁺ neutrophils, which were significantly higher than other infiltrating cells (Figure 1B), suggesting that neutrophils might be the dominant type of infiltrating inflammatory cells in SF of RA patients. Next, we examined the occurrence and distribution of neutrophils in ST of RA patients. The results showed there were large numbers of CD15⁺ neutrophils in RA ST (Figure 1C). To further identify the population of infiltrating cells, we employed flow cytometry analysis and the results revealed remarkable leukocyte infiltration, including the population of CD11b⁺CD15⁺, CD3⁺, CD19⁺ and CD14⁺CD16⁺ cells in RA ST (Figure 1D and Additional file 1: Figure S1). CD11b⁺CD15⁺neutrophils were the dominant infiltrating inflammatory cells in ST of RA patients. These results demonstrate that inflamed joints of RA patients are heavily infiltrated with neutrophils, which is consistent with previous reports [36].

IL-8 is one of the most neutrophil chemoattractant molecules and plays a very important role in pathogenesis mediated by neutrophils in RA [5, 37]. A number of cells are known to produce IL-8, including macrophages and fibroblasts [20]. Given that we have previously shown that Cyr61 induces IL-6 production in FLS, which further drive Th17 differentiation and enhance inflammation of RA [28, 29], we further explored whether Cyr61 may also stimulate IL-8 production by FLS. We set up an in vitro cell culture system using FLS isolated from RA patients and analyzed the levels of IL-8 and Cyr61 in SF obtained from RA and OA patients. The results showed that the levels of IL-8 and Cyr61 were higher in RA SF than in OA SF (Figure 2A), consistent with other reports [15, 16, 28]. After confirming that RA SF contained higher levels of IL-8 and Cyr61, we next tested the potential effect of Cyr61 on the expression of IL-8 by FLS of RA patients. The results showed that Cyr61 significantly stimulated IL-8 mRNA expression in FLS (Figure 2B and 2C). Consistent with these observations, ELISA showed that the concentration of IL-8 in FLS culture supernatant was significantly increased upon addition of exogenous Cyr61 (Figure 2D). To further examine the autocrine role of Cyr61 in the regulation of IL-8 expression by FLS, we performed specific siRNA to knockdown Cyr61 expression in FLS [28]. The results showed that IL-8 mRNA expression was remarkably reduced in Cyr61-knockdown FLS. The reduction of IL-8 production by FLS upon Cyr61 knockdown was also confirmed by ELISA measurement of IL-8 protein levels in culture supernatant (Figure 2E). Next, we treated FLS with an anti-Cyr61 mAb (named 093G9) and results showed that 093G9 could block the effect of Cyr61 on IL-8 production by FLS (Figure 2F). These data indicate that Cyr61 can induce IL-8 expression by FLS of RA patients, which may partly contribute to the higher concentration of IL-8 observed in RA SF.

Since previous studies have shown that IL-8 is induced by either IL-1β or TNF-α in RA FLS [19‐21], we asked whether IL-1β or TNF-α contributes to IL-8 production stimulated by Cyr61 in FLS. We first found that the levels of IL-1β and TNF-α were remarkably higher in RA SF (Figure 3A), which is consistent with our previous reports [29, 38]. Next, we assessed the role of IL-1β and TNF-α in IL-8 production in FLS induced by Cyr61. We employed a specific siRNA to knockdown IL-1β or TNF-α expression in FLS and examined IL-8 production induced by Cyr61. The results showed that, compared with control siRNA treatment (siNC), IL-8 mRNA expression was not reduced in IL-1β or TNF-α-knockdown FLS when exogenous Cyr61 was added (Figure 3B and 3C). These results suggest that IL-1β and TNF-α are not likely to take part in the process of IL-8 production from Cyr61-induced FLS; in other words, Cyr61-promoted IL-8 production in RA FLS in an IL-1β and TNF-α independent pathway.

Next, we further explored whether Cyr61-induced IL-8 secretion by FLS has functional activity for inducing neutrophil migration. Using a chemotaxis assay, we found that the supernatant from Cyr61-treated FLS significantly augmented the chemotaxis of neutrophils isolated from peripheral blood of healthy individuals (Figure 4A and 4B). We further preincubated Cyr61-treated supernatant with specific IL-8 neutralizing (or control) Ab, and the results showed that addition of a neutralizing Ab against IL-8 in the supernatant indeed reduced the number of migrating neutrophils (Figure 4B), indicating that Cyr61-induced IL-8 secretion by FLS has neutrophil chemoattractant activity. This allowed us to show that Cyr61 induces IL-8 production which further leads to promoting neutrophil migration. Taken together, our data demonstrate that, in addition to IL-1β and TNF-α, Cyr61 is a newly identified inducer of IL-8 in RA FLS, suggesting that Cyr61 might be involved in the inflammatory and tissue damage induced by infiltrating neutrophils.

As we found that Cyr61-stimulated FLS promoted IL-8 expression in vitro, we asked whether Cyr61 indeed plays a role in IL-8 expression and relevant inflammation in vivo. We established a CIA mice model and treated them with 093G9 (Cyr61 neutralization Ab) [39]. Our previous reports revealed that the inflammatory score was significantly decreased and leukocyte infiltration and synovial hyperplasia in joints were ameliorated in 093G9-treated CIA mice [29]; similar results were obtained in this study and are shown in Figure 5A and Additional file 1: Figure S3. Moreover, microscopy showed that infiltrating neutrophils were obviously decreased in the mice treated with 093G9 (Figure 5A). Next, we analyzed the expression of MIP-2 (a counterpart of human IL-8) in joints of the mice with CIA and found that blocking Cyr61 also down-regulated the expression of MIP-2 (Figure 5B and 5C). This suggests that blocking Cyr61 activity reduced neutrophil migration into target tissues of CIA mice by down-regulating MIP-2 activity, resulting in the amelioration of joint inflammation and erosion.

Since the results showed that Cyr61 directly induced IL-8 production in FLS, we probed the downstream signaling pathway(s) using known inhibitors of several pathways, including PDTC (inhibitor of NF-κB activation), SP600125 (inhibitor of JNK), PD98059 (inhibitor of ERK1/2) and SB203580 (inhibitor of p38 MAPK). The results showed that Cyr61-stimulated IL-8 mRNA and protein expression in FLS were markedly decreased in the presence of the JNK, ERK1/2 and NF-κB inhibitors. In contrast, inhibition of p38 MAPK activities showed no effect on Cyr61-induced IL-8 production (Figure 6A). Further analysis showed that Cyr61 treatment led to a dramatic increase in the phosphorylation level of the JNK, ERK1/2 and NF-κB p65 subunit in FLS (Figure 6B and 6C) and enhanced NF-κB nuclear translocation as shown by laser scanning confocal immunofluorescence microcopy (Additional file 1: Figure S4). Previous studies in breast cancer cells suggested that Cyr61 could induce NF-κB activation via the PI3K/AKT pathway [23, 40, 41]. We, thus, checked whether this pathway was also activated in FLS upon Cyr61 stimulation. Indeed, we found that the phosphorylated (activated) form of AKT was strongly enhanced in response to Cyr61 treatment in FLS (Figure 6C). Based on these results, we suggest that Cyr61 induced IL-8 production in FLS depends on AKT, NF-κB, JNK and ERK1/2 signaling pathways.

Studies have shown that IL-8 expression is regulated by a sequence spanning nucleotides -1 to -133 of the upstream DNA flanking the IL-8 gene, and that this region contains response elements for AP-1, C/EBP and NF-κB and is essential and sufficient for IL-8 expression [37, 42]. To further identify the molecular mechanism responsible for Cyr61 induced IL-8 expression, we first constructed an IL-8 promoter (IL-8WT, which includes response elements for AP-1, C/EBP and NF-κB) with upstream of the luciferase gene. Further we transfected IL-8WT promoter into human skin fibroblasts (HSFs) and then treated the HSFs with Cyr61. The results showed that Cyr61 increased the IL-8WT promoter activity about seven-fold (Figure 7A), suggesting that Cyr61 is able to activate IL-8 promoter and enhance IL-8 gene expression. To further analyze the role of three known transcription factors (AP-1, C/EBP, and NF-κB) in the control of IL-8 gene expression in response to Cyr61 stimulation, we transfected the IL-8 promoter with mutations (IL-8mAP-1, IL-8mC/EBP and IL-8mNF-κB, which inhibit the binding site for AP-1, C/EBP and NF-κB, respectively) into HSFs, then treated HSFs with 5 μg/ml Cyr61 for two hours. The results showed that, upon stimulation with Cyr61, luciferase activity was not significantly increased compared with untreated HSFs (Figure 7A). These data indicate that AP-1, C/EBP and NF-κB binding motifs are essential for the Cyr61-induced IL-8 gene expression in RA FLS. To detect the in vitro binding of c-Jun, C/EBPβ and p65 to the IL-8 promoter following Cyr61 challenge, we performed a ChIP assay and examined the binding of c-Jun, C/EBPβ and p65 to the IL-8 promoter in FLS stimulated with exogenous Cyr61. The results showed that the levels of transcription factors (c-Jun, C/EBPβ and p65) binding to the IL-8 promoter in FLS were increased significantly compared with controls (Figure 7B). In contrast, treatment with 093G9 resulted in reducing binding of c-Jun, C/EBPβ and p65 to the IL-8 promoter region significantly (Figure 7B). Together, these results indicate that Cyr61 induced c-Jun, C/EBPβ and p65 binding to the corresponding response elements in the IL-8 promoter and increased the transcriptional activity of the IL-8 promoter. Based on these findings, we suggest that Cyr61 induced IL-8 production in FLS via AKT, JNK and ERK1/2 dependent AP-1, C/EBP and NF- κ B activation.

Activation of MAPK and NF-κB pathways have been shown to contribute to IL-8 expression [37, 42, 55, 56], but the role of MAPK and the NF-κB pathway for the Cyr61-induced IL-8 production in FLS remains to be determined. To address the signaling pathway of Cyr61 promoting IL-8 production in FLS, we evaluated the profile of AKT/NF-κB, a well-known Cyr61/integrins pathway [57], and three well-defined MAPK pathways (JNK, ERK and p38). As expected, AKT/NF-κB pathways contributed to Cyr61-induced IL-8 production in FLS. However, the analysis of MAPK pathways indicated that JNK and ERK pathways were involved in the Cyr61-induced IL-8 production in FLS. Interestingly, the p38 pathway was not found to contribute to the Cyr61-induced IL-8 production in FLS. Previous observations suggested that, in the IL-1β or TNF-α induced IL-8 production, the p38 MAPK pathway contributes to IL-8 gene expression by stabilizing mRNAs in RA FLS [42, 58]. Our study first shows that the p38 MAPK pathway was not involved in the Cyr61-induced IL-8 production in FLS; in other words, signaling cascades of Cyr61-induced IL-8 production are different from signaling pathways of IL-1β or TNF-α-induced IL-8 production. Considering the role of the p38 MAPK pathway in post transcriptional regulation of IL-8 production, how to stabilize the mRNAs of IL-8 in Cyr61-stimulated FLS is under investigation.

Based on the results of Cyr61-induced IL-8 production in FLS via JNK, ERK and NF-κB activation, we examined the transcriptional mechanisms regulated by Cyr61. Although it is well known that the core IL-8 promoter contains binding sites for AP-1, C/EBP and NF-κB, the different binding activity of AP-1, C/EBP and NF-κB on the IL-8 promoter has been attributed to different IL-8 production in the cells [37, 42]. We performed promoter-reporters and ChIP analysis for testing regulatory elements of the IL-8 promoter in Cyr61-treated FLS. The results showed that AP-1/c-Jun, C/EBPβ and NF-κB binding to the IL-8 promoter were all necessary for Cyr61-induced IL-8 expression in RA FLS. Earlier studies have documented that transcription factors involved in IL-8 gene transcription interact to facilitate the formation of an enhanceosome-like structure that favors the induction of the IL-8 promoter [42]. In our studies, we found that Cyr61 enhanced AP-1, C/EBPβ and NF-κB binding to the IL-8 promoter simultaneously, suggesting that signaling pathways mediated by Cyr61 provoke an interaction among these transcription factors and may contribute to the formation of an enhanceosome-like structure for IL-8 production in RA FLS, even though the p38 MAPK pathway was not active in Cyr61-induced IL-8 production in RA FLS. Based on these results, we propose that Cyr61 is able to induce IL-8 production similar to pro-inflammatory cytokines, by which Cyr61 enhances neutrophil infiltration in joints with RA.

Although a recent study showed that hypoxia might induce Cyr61 and IL-8 secretion in nasal polyp fibroblasts, no direct evidence demonstrated that Cyr61 induces IL-8 production under an inflammatory environment via an IL-1β/TNF-α independent pathway [59]. Considering that Cyr61 expression may be up-regulated as a protective response to hypoxia in vivo[60], it would be interesting to investigate whether hypoxia can enhance Cyr61-induced IL-8 production by RA FLS.