Dr Chul-Soo Cho, Department of Internal Medicine, Division of Rheumatology, St Mary's Hospital, School of Medicine, Catholic University of Korea, 62 Youido-dong, Youngdeungpo-ku, Seoul, 150–713, Korea. E-mail: firstname.lastname@example.org
Interleukin (IL)-4 has been demonstrated to have anti-inflammatory and anti-tumour activity. Because aberrant angiogenesis is a significant pathogenic component of tumour growth and chronic inflammation, we investigated the effect of IL-4 on the production of vascular endothelial growth factor (VEGF) by synovial fibroblasts derived from patients with rheumatoid arthritis (RA). Fibroblast-like synoviocytes (FLS) were prepared from synovial tissues of RA and incubated with different concentrations of IL-4 in the presence or absence of transforming growth factor (TGF)-β. VEGF level was measured by enzyme-linked immunosorbent assay and semiquantitative reverse transcription–polymerase chain reaction. Treatment of FLS with IL-4 alone caused a dose-dependent increase in VEGF levels. In contrast, IL-4 exhibited the inhibitory effect on VEGF production when FLS were stimulated with TGF-β. Combined treatment of IL-4 and IL-10 inhibited TGF-β-induced VEGF production in an additive fashion. TGF-β increased the induction of cyclooxygenase-2 mRNA, which was inhibited significantly by the treatment of IL-4. NS-398, a COX-2 inhibitor, inhibited TGF-β-induced VEGF production in a dose-dependent manner. Furthermore, exogenous addition of prostaglandin E2 (PGE2) restored IL-4 inhibition on TGF-β induced VEGF production. Collectively, our results suggest that IL-4 have an anti-angiogenic effect, especially in the inflammatory milieu of RA by inhibiting the VEGF production in synovial fibroblasts.
Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by hyperplasia of synovial lining cells, infiltration of mononuclear cells and abundant new vessel formation in the synovium . The inflamed synovial tissues known as pannus invade adjacent cartilages and bone aggressively, and lead eventually to joint destruction in RA patients. Rheumatoid synovial tissues exhibit similar characteristics of tumour cells in uncontrolled growth, invasiveness and exuberant angiogenesis . New vessel formation relieves hypoxia by satisfying the substantial metabolic requirement of growing tissues. On the other hand, angiogenesis may exacerbate synovitis by facilitating growth of the pannus and increasing the recruitment of inflammatory cells into synovium .
Vascular endothelial growth factor (VEGF) is a heparin-binding dimeric glycoprotein that acts on endothelial cells to promote proliferation and induce vascular permeability as well [4,5]. VEGF was detected at high levels in serum and synovial fluid of RA patients compared to those of osteoarthritis or normal controls [6,7]. Serum VEGF levels in patients with RA were correlated with levels of erythrocyte sedimentation rate (ESR) and C-reactive protein [8,9]. Cultured synovial cells produce VEGF in response to hypoxia [10,11] and a variety of stimuli such as prostaglandin E2 (PGE2), tumour necrosis factor (TNF)-α, interleukin (IL)-1, IL-6, transforming growth factor (TGF)-β[10–14] and CD40 ligand . Recently, we demonstrated that subcutaneous injection of anti-VEGF peptides, dRK6, ameliorated the disease severity of collagen-induced arthritis in mice , indicating that the inhibition of VEGF would be a potential therapeutic strategy in RA.
IL-4 is a multi-functional cytokine secreted by T lymphocytes, basophils and mast cells. IL-4 exhibits anti-inflammatory properties by inhibiting the induction of proinflammatory mediators, such as TNF-α and IL-1, from monocytes [17–19]. It can also suppress the production of IL-6, leukaemia inhibitory factor (LIF) and granulocyte–macrophage colony-stimulating factor (GM-CSF) PGE2 by RA synovial fibroblasts, while up-regulating IL-1 receptor antagonists and soluble TNF receptor production [20–24]. Moreover, Volpert et al. reported that IL-4 had an anti-angiogenic activity by blocking corneal neovascularization induced by basic fibroblast growth factor . However, the effect of IL-4 on VEGF production in rheumatoid synovial fibroblasts has not been reported. Thus, we investigated whether IL-4 could modulate the basal and stimulated production of VEGF from rheumatoid synovial cells.
Materials and methods
Isolation and cultures of fibroblast-like synoviocytes
Cultured synovial cells used in this study were obtained from RA patients who underwent total joint replacement surgery or synovectomy. The synovial tissues were minced into 2–3-mm pieces and incubated for 4 h with 4 mg/ml collagenase (type I; Worthington Biochemical, Freehold, NJ, USA) in Dulbecco's modified Eagle's medium (DMEM) at 37°C in 5% CO2. After digestion, the dissociated cells were then centrifuged at 500 g, resuspended in DMEM supplemented with 10% fetal calf serum (FCS; Life Technologies, Grand Island, NY, USA), 2 mM glutamine, penicillin (100 unit/ml) and streptomycin (100 µg/ml) and plated in 75-cm2 flasks. After overnight incubation, non-adherent cells were removed. Cultures were kept at 37°C in 5% CO2 and culture medium was replaced every 3 days. When cells reached near confluence they were harvested by trypsinization, and one-third of them were resuspended in culture flasks. A homogeneous population of fibroblast-like synoviocytes (FLS) at passages 4–8 was used for this experiment.
Effects of IL-4, TGF-β and PGE2 on VEGF production by synoviocytes
FLS were seeded in 24-well plates at 6 × 104 cells per well in DMEM supplemented with 5% FCS and cultured at 37°C for 24 h. Cells were washed with serum-free DMEM medium and then incubated in serum-free DMEM supplemented with insulin–transferrin–selenium-A (ITSA; Life Technologies) for 48 h until the medium was replaced with fresh DMEM/ITSA. Subsequently, IL-4 (R&D systems, Minneapolis, MN, USA), TGF-β (10 ng/ml; PeproTech, London, UK) or IL-1β (10 ng/ml; Endogen, Woburn, MA, USA) was added to the wells, and the cells were incubated for 24 h. To assess the effect of IL-4 on VEGF production stimulated by TGF-β, FLS was incubated with various concentrations of IL-4 (0·1–50 ng/ml) in the presence of TGF-β (10 ng/ml). In selected wells, to determine whether PGE2 is involved in the IL-4 effect on TGF-β-induced VEGF production, varying concentrations of PGE2 (0·1–10 µM; Sigma-Aldrich, St Louis, MO, USA) was added to the wells at the beginning of culture. After 24 h of incubation (unless otherwise stated), cell-free media were collected and stored at − 20°C until assayed. All cultures were set up in triplicate and results are expressed as mean ± s.d.
Determination of VEGF concentration
VEGF in culture supernatants was measured by sandwich enzyme-linked immunosorbent assay (ELISA), as described previously . Ninety-six-well microtitre plates were coated with 100 µl per well of 0·4 µg/ml goat anti-human VEGF165 (R&D systems) buffered with 50 mM of sodium carbonate, pH 9·6. After incubation overnight at 4°C, the plates were blocked with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) for 1 h at room temperature. Human recombinant VEGF165 (R&D Systems) or test samples were added to the wells and then incubated for 2 h at room temperature. The plates were incubated with 0·2 µg/ml biotinylated goat anti-human VEGF165 (R&D Systems) at room temperature for 2 h. Peroxidase-labelled extravidin (Sigma-Aldrich), diluted 1 : 1000, was added to react with the plates at room temperature for 1 h. Colour reaction was induced by the addition of 3,3′,5,5′-tetramethylbenzidine (TMB)/H202 substrate solution and was stopped 30 min later by the addition of 1 M phosphoric acid. An automated microplate reader was used to measure the optical density at a wavelength of 450 nm. Between each of these steps, the plates were washed four times with PBS containing 0·1% Tween 20. Human recombinant VEGF165 diluted in culture medium was used as a calibration standard, ranging from 10 to 2000 pg/ml.
Reverse transcription–polymerase chain reaction (RT–PCR) of VEGF and cyclooxygenase (COX)-2 messenger RNA (mRNA)
FLS were cultured with various concentrations of IL-4 in the presence or absence of TGF-β (10 ng/ml). After 6–12 h of culture, RNA was extracted using RNAzol B (Biotecx, Houston, TX, USA) according to the manufacturer's instructions. Five micrograms of total RNA was then reverse-transcribed using an RT system kit (Life Technologies). PCR amplification of cDNA aliquots was performed by adding 2·5 mM dNTPs, 2·5 units Taq DNA polymerase (Takara Shuzo, Shiga, Japan) and 0·25 µM each of sense and anti-sense primers. The reaction was performed in PCR buffer (1·5 mM MgCl2, 50 mM KCl, 10 mM Tris HCl, pH 8·3) in a total volume of 25 µl. The following primers were used: VEGF sense (5′-TCTTGGGTGCATTGGAGCCTC-3′) and anti-sense (5′-AGCTCATCTCTCCTATGTGC-3′), cyclooxygenase-2 (COX-2) sense (5′-GCAGTTGTTCCAGACAA GCA-3′) and anti-sense (5′-CAGGATACAGCTCCACAGCA-3′) and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) sense (5′-CGATGCTGGGCGTGAGTAC-3′) and anti-sense (5′-CGTTCAGTCCAGGGATGA CC-3′). Cycling conditions were as follows: 45 s of denaturation at 94°C for VEGF, COX-2 and GAPDH; 1 min of annealing at 55°C for COX-2 and GAPDH and at 62°C for VEGF, followed by 30 s of elongation at 72°C. The PCR rounds were repeated for 25 cycles for VEGF and GAPDH, and 27 cycles for COX-2. PCR products were run on a 1·5% agarose gel and stained with ethidium bromide. Results are expressed as a ratio of VEGF and COX-2 PCR product to GAPDH product.
Data are expressed as the mean ± standard deviation (s.d.). The results were analysed using a non-parametric Mann–Whitney U-test. P-values less than 0·05 were considered significant.
Stimulatory effect of IL-4 on VEGF production in unstimulated synovial fibroblasts
To analyse the effect of IL-4 on VEGF production, FLS were incubated with varying concentrations of IL-4. TGF-β and IL-1β were included as positive control for induction of VEGF. After 24 h of culture, supernatants were collected for measurement of VEGF protein. As shown in Fig. 1, basal levels of VEGF were detectable from FLS incubated in medium alone (45 ± 8 pg/ml). TGF-β (10 ng/ml) and IL-1β (10 ng/ml) strongly increased the production of VEGF by 4·5-fold and 2·2-fold over the media alone (the latter data not shown), respectively, as reported previously . Addition of increasing concentrations of IL-4 (0·1–50 ng/ml) increased the VEGF production in a dose-dependent fashion. The production of VEGF by 50 ng/ml of IL-4 (the maximal concentration used in the study) was twofold over the unstimulated FLS.
Inhibitory effect of IL-4 on TGF-β induced VEGF production in synovial fibroblasts
In the rheumatoid synovial joints, resident synoviocytes are exposed to several inflammatory mediators, some of which have a potent angiogenic activity [10–14]. Thus, it is clinically relevant to investigate the effect of IL-4 on the VEGF production induced by TGF-β, which is found at a high level in the RA joints. As shown in Fig. 2, the addition of IL-4 resulted in the down-regulatory effect on VEGF production in TGF-β-stimulated fibroblasts by contrast to unstimulated fibroblasts. This effect of IL-4 (1–50 ng/ml) was dose-dependent. The inhibitory effect of IL-4 on VEGF production was also noted when FLS was stimulated with 10 ng/ml of IL-1β (38% inhibition at 50 ng/ml of IL-4). We also tested whether IL-10, another T helper type 2 (Th2) anti-inflammatory cytokine, could regulate TGF-β-induced VEGF production by FLS. The results showed that IL-10 (10 ng/ml) also inhibited the VEGF production induced by TGF-β, which was comparable to the same concentration of IL-4. Moreover, simultaneous treatment of IL-4 with IL-10 resulted in an additive effect on the suppression of VEGF production, compared to each cytokine alone (Fig. 3).
Effect of IL-4 on VEGF mRNA expression in FLS
To determine whether IL-4 modulates VEGF expression at the mRNA level, FLS were cultured with IL-4 (1–50 ng/ml) in the presence or absence of TGF-β (10 ng/ml) and the levels of VEGF mRNA were measured by semiquantitative RT–PCR analysis, as described in Materials and methods. Unstimulated FLS exhibited very low levels of VEGF mRNA expression. Addition of IL-4 alone increased VEGF expression in a dose-dependent manner. In contrast, as observed in ELISA data, IL-4 inhibited VEGF mRNA expression induced by TGF-β in a dose-dependent manner (Fig. 4). These findings indicate that modulation of VEGF by IL-4 is regulated predominantly at mRNA level.
IL-4 inhibition on VEGF production is mediated by down-regulation of PGE2
PGE2 is known to be a potent inducer for VEGF production. It has been suggested that IL-4 inhibits PGE2 synthesis in several cells, including rheumatoid synovial cells, peritoneal macrophages and periodontal ligament fibroblasts [26–28]. Thus, it is possible that the decrease in VEGF production by IL-4 is mediated by the IL-4 regulation of PGE2. To address this assumption, we measured the expression of COX-2, a key enzyme for PGE2 generation, in FLS stimulated with TGF-β. As shown in Fig. 5a, unstimulated FLS showed a weak expression of COX-2 mRNA, which increased significantly following stimulation of FLS with TGF-β. The addition of IL-4 resulted in the suppression of TGF-β-induced COX-2 mRNA expression in a dose-dependent manner (Fig. 5a). NS-398, a selective COX-2 inhibitor, exhibited a similar inhibitory action on VEGF production to that observed in IL-4 (Fig. 5b). Moreover, exogenous addition of PGE2 to the FLS cultures cancelled the IL-4 inhibition of VEGF production almost completely (Fig. 6), suggesting that PGE2 mediates the inhibitory effect of IL-4 on TGF-β-induced VEGF production.
In the present study, we observed that IL-4 induced VEGF production in unstimulated synovial fibroblasts. Our data confirm previous reports showing that IL-4 plays an important role in angiogenesis by inducing VEGF release from several cells types [29–31]. IL-4 has a mitogenic action for microvascular endothelial cells by enhancing tube formation in a dose-dependent fashion [32,33]. Moreover, IL-4 stimulates angiogenesis in the rat cornea by up-regulating the VCAM-1 release in vascular endothelial cells . However, most of these results are in vitro observations and/or were carried out in the absence of VEGF or basic fibroblast growth factor (bFGF). Therefore, it is unclear whether IL-4 promotes angiogenesis in the in vivo conditions with high VEGF and FGF, such as RA joints.
In contrast to unstimulated FLS, IL-4 exhibited an inhibitory effect on VEGF production when FLS were stimulated with TGF-β (Fig. 2), suggesting that IL-4 plays an anti-angiogenic role in certain physiological conditions. The anti-angiogenic property of IL-4 was also demonstrated in several cancer models with high VEGF expression levels [35–38], in which angiogenesis could be stimulated by tumour hypoxia and a resultant increase in angiogenic factors (VEGF, bFGF, etc.). IL-4-expressing tumours had a reduced level of vascularization compared with controls . Furthermore, retroviral delivery of IL-4 into rat C6 cell gliomas leads to the rapid inhibition of tumour growth in situ. Considering that the suppression of tumour growth and metastasis is linked directly to a decrease in neovascularity, the significant regression of lung metastasis by IL-4 in a murine renal cell carcinoma model further supports the anti-angiogenic role of IL-4 .
It has been well demonstrated that IL-4 could scarcely be detected in synovial fluids and synovium of RA patients, while TGF-β, a well-known potent inducer of VEGF, is present in high amounts . The intra-articulator injection of the IL-4 gene resulted in a significant reduction of paw swelling, as well as decreased evidence of bone destruction and cartilage in the adjuvant- and collagen-induced arthritis model [42,43]. Recently, Hass et al. demonstrated that IL-4 exhibited the anti-arthritic effect in the rat adjuvant-induced arthritis model by inhibition of angiogenesis . In the present study, we also observed that the inhibitory effect of IL-4 was not restricted to the VEGF production as TGF-β-induced IL-6 production was down-regulated significantly by IL-4 treatment in a dose-dependent manner (data not shown). Considering that IL-6 is potent inducer of VEGF production in synovial fibroblasts, IL-4 may also inhibit the VEGF production indirectly via suppression of IL-6 synthesis. In this context, it can be postulated that the absence of an IL-4 inhibitory effect on VEGF production in RA joints contributes to the increased neoangiogenesis responsible for chronic inflammatory process.
The importance of COX-2 in the regulation of VEGF has been described in several reports. For example, overexpression of COX-2 has been linked to increased VEGF expression in colon cancer cells . Up-regulation of VEGF by cobalt chloride-stimulated hypoxia was mediated by induction of COX-2 in human prostate cancer cell line . Moreover, inhibition of COX-2 activity by genetic ablation decreased tumour and stromal VEGF levels with marked attenuation of angiogenesis and tumour growth . In our results, TGF-β increased the COX-2 (Fig. 5a) and PGE2 production by FLS (data not shown), which is consistent with a previous report . The induction of COX-2 mRNA expression by TGF-β was suppressed dose-dependently by adding IL-4 as well as NS398, a selective COX-2 inhibitor. As well as inhibition by IL-4 on VEGF production, NS-398, which mitigates endogenous eicosanoid production, suppressed TGF-β-induced VEGF expression in a dose-dependent manner (Fig. 5b). Finally, IL-4 inhibition of VEGF production was recovered by the addition of exogenous PGE2 (Fig. 6). Collectively, these findings indicate that inhibitory effect of IL-4 on VEGF production is mediated by suppression of endogenous COX-2 and PGE2.
In summary, we observed the opposing effect of IL-4 on VEGF production in stimulated versus unstimulated synovial fibroblasts; IL-4 increases spontaneous VEGF production while inhibiting TGF-β-stimulated VEGF production by synovial fibroblasts. IL-4 inhibition of TGF-β-stimulated VEGF production is mediated through the suppression of COX-2 and PGE2. Our findings suggest that the beneficial effect of IL-4 in RA could be explained, in part, by inhibition of VEGF in the inflammatory milieu.
This work was supported by a grant from the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (0405-DB01-0104–0006).