Glucosamine Sulfate Inhibits TNF-α and IFN-γ-Induced Production of ICAM-1 in Human Retinal Pigment Epithelial Cells In Vitro

Abstract

purpose. Glucosamine sulfate (GS) is a naturally occurring sugar that possesses some immunosuppressive effects in vitro and in vivo, but its mechanism is unknown. We investigated whether GS could modulate the proinflammatory cytokine-induced expression of the gene for intercellular adhesion molecule (ICAM)-1, an inflammatory protein in human retinal pigment epithelial (RPE) cells.

methods. ARPE-19 cells were used as a model to determine the effects of GS on the expression of the ICAM-1 gene upregulated by TNF-α or IFN-γ, by Western blot analysis and semiquantitative reverse transcription polymerase chain reaction (RT-PCR). The activation and nuclear translocation of the nuclear factors NF-κB and STAT1 were evaluated by immunocytochemistry, Western blot analysis, and electrophoretic mobility shift assay (EMSA).

results. Both TNF-α and IFN-γ increased the expression of ICAM-1 at the mRNA and protein levels in a time- and dose-dependent manner in ARPE-19 cells. GS effectively downregulated the TNF-α- or IFN-γ-induced expression of ICAM-1 in the protein and mRNA level in a dose-dependent manner. GS further inhibited the nuclear translocation of p65 proteins in TNF-α and phosphorylated STAT1 in IFN-γ-stimulated ARPE-19 cells.

conclusions. GS inhibits the expression of the ICAM-1 gene in ARPE-19 cell stimulated with TNF-α or IFN-γ through blockade of NF-κB subunit p65 and nuclear translocation of STAT1. This study has demonstrated a potentially important property of GS in reducing ICAM-1 mediated inflammatory mechanisms in the eye.

Intercellular adhesion molecule-1 (ICAM-1, CD54) is a transmembrane glycoprotein of 505 amino acids with a molecular mass ranging from 80 to 114 kDa, depending on the degree of glycosylation. ¹ It belongs to the immunoglobulin supergene family and contains five immunoglobulin-like domains (D1–D5) that function in cell–cell and cell–matrix adhesive interactions. These are essential for the transendothelial migration of leukocytes and the activation of T cells, in which ICAM-1 functions as a costimulatory signal (Signal-2). ²  

Retinal pigment epithelial (RPE) cells are located between the neuroretina and the choroid coat, where they act as one of the components of the outer blood–retinal barrier, participate in selective transport of metabolites between the neuroretina and choriocapillaris, and phagocytose the outer segment shed from photoreceptors. ^{3 4} They can act as antigen-presenting cells thereby participating in the immunogenic process. ⁵ ICAM-1 can be detected in the RPE cells of patients with posterior uveitis, and increased levels of ICAM-1 are considered to promote extravasation of inflammatory cells into the retina. ⁶  

Monosaccharides, especially those bearing amino groups, effectively inhibit cytotoxic T-lymphocyte function in vitro. ⁷ In addition, hexosamines inhibit natural killer cell cytotoxicity. ⁸ Of these, glucosamine sulfate (GS) is a naturally occurring sugar that possesses some immunosuppressive effects in vitro and in vivo. ^{9 10} It inhibits NF-κB activity and IL-1β bioactivity in rat chondrocytes by increasing the expression of the type II IL-1 decoy receptor. ⁹ Moreover, GS can suppress the activation of T-lymphoblasts and dendritic cells and prolong cardiac allograft survival in vivo. ¹⁰ GS also shows tumor-inhibiting activity and inhibits the biosynthesis of macromolecules including nucleic acids and proteins exclusively in tumor cell lines. ^{11 12 13 14 15 16 17}  

This study was undertaken to investigate whether GS can inhibit the upregulation of the gene for ICAM-1 in human RPE cells (ARPE-19) induced by proinflammatory cytokines, including TNF-α and IFN-γ. The mechanisms involved in the possible inhibitory effects of GS on this gene’s expression were also studied. 

Materials and Methods

Cell Culture and Treatment

ARPE-19 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and maintained in Dulbecco’s modified Eagle’s medium (DMEM)/F-12 supplemented with 4 mM l-glutamine, 10% fetal bovine serum (FBS), 100 U/mL penicillin, 100 mg/mL streptomycin, and 0.25 mg/mL amphotericin B at 37°C in 5% CO₂ in air. In each experiment, cells were grown to confluence, made quiescent for 24 hours in DMEM/F-12 medium without serum and stimulated at different times or concentrations of TNF-α and IFN-γ, as described in the figure legends. Where indicated, cells were preincubated with GS (Sigma-Aldrich, St Louis, MO), N-acetylglucosamine (N-AcG; Sigma-Aldrich) or galactosamine hydrochloride (Gal; Sigma-Aldrich) for 1 hour, and these compounds were maintained during the whole period of the experiments. 

Western Blot Analysis

Confluent cultured cells were preinoculated with GS, N-AcG, or Gal and stimulated with TNF-α and IFN-γ for the indicated periods and dosages. For measures of ICAM-1, cells were washed twice with phosphate-buffered salt solution (PBS) and detached by scraping. Cells were pelleted at 1000g, resuspended, and sonicated in cold lysis buffer (50 mM Tris-HCl [pH 7.5] 2% SDS, and 1 mM phenylmethylsulfonyl fluoride). Insoluble debris was removed by centrifugation at 16,000g at 4°C for 10 minutes. To determine IκBα, STAT1, and phosphorylated STAT1 levels, cytosolic, and nuclear fractions were obtained as described later in the article. Protein content was determined using the bicinchoninic acid method (BCA; Pierce, Rockford, IL) with bovine serum albumin (BSA) as a standard. One-dimensional SDS-PAGE (12% polyacrylamide gels) was performed. The separated proteins were transferred onto polyvinylidene difluoride (PVDF) membranes (Immobilon; Millipore Corp., Bedford, MA) which were then incubated in PBS containing 5% skimmed milk and reacted overnight at 4°C with a primary anti-human ICAM-1 rabbit polyclonal antibody (1:1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA), or IκBα (1:200 dilution; Santa Cruz Biotechnology) or with anti-STAT1 (1:200 dilution; Santa Cruz Biotechnology), anti-phosphorylated STAT1 (1:200 dilution; Santa Cruz Biotechnology) or anti-human actin antibodies (1:1000 dilution; Santa Cruz Biotechnology). After three washes in PBS, membranes were incubated for 1 hour in PBS containing a peroxidase-conjugated goat antibody raised against rabbit IgG (1:2000 dilution; Santa Cruz Biotechnology). Peroxidase activity on the membrane sheet was visualized on x-ray films by means of a standard enhanced chemiluminescence (ECL) procedure. The blots were scanned into image-analysis software (Photoshop, ver. 7.0; Adobe Systems, San Jose, CA), and the optical densities of the bands were calculated. 

RNA Isolation and Quantitative RT-PCR

Expression of the ICAM-1 gene in ARPE-19 cells was investigated at the mRNA level by RT-PCR. Total RNA from ARPE-19 cells was isolated (TRIzol Reagent; Invitrogen-Gibco, Grand Island, NY), according to the manufacturer’s instructions. Oligonucleotide primers complementary to the 5′ and 3′ ends of ICAM-1 and GAPDH cDNAs were used in RT-PCR. The sequences of the oligonucleotide primers used to amplify ICAM-1 and β-actin cDNAs were: ICAM-1 forward 5′-CCGGAAGGTGTATGAACTG-3′ and reverse 5′-CAGTTCATACACCTTCCGG-3′; and GAPDH forward 5′-GCAGGGGGGAGCCAAAAGGG-3′ and reverse 5′-TGCCAGCCCCAGCGTCAAAG-3′. Oligo(dT)_12–18 (1 mg)-primed total RNA (5 mg) was reverse-transcribed using reverse transcriptase (SuperScript III RNase H; Invitrogen-Gibco) supplied with a first-strand synthesis system for RT-PCR (Invitrogen-Gibco). PCR reactions contained 2 μL cDNA, 67 mM Tris-HCl (pH 8.8), 16 mM (NH₄)₂SO₄, 1.5 mM MgCl₂, 0.2 mM dNTPs, 10 picomoles sense primer, 10 picomoles antisense primer, and 1.25 U Taq DNA polymerase (Invitrogen-Gibco) in a 50-μL reaction volume. Annealing temperatures and MgCl₂ concentrations were optimized to create a one-peak melting curve. PCR reaction parameters were: denaturation at 94°C for 3 minutes followed by 30 cycles of denaturation at 94°C for 30 seconds, annealing at 58°C for 30 seconds, and extension at 72°C for 30 seconds. PCR amplification of β-actin was routinely used as a control to assess the integrity of the RNA and cDNA. Amplification reaction products (10 μL) were resolved on 1.2% Tris-borate-EDTA (TBE)-buffered agarose gels and visualized with ethidium bromide staining. To verify their authenticity, amplicons were excised from the gel, repurified, and subjected to DNA sequence analysis. The results of PCR were quantified on computer (ImageQuant software; Molecular Dynamics, Sunnyvale, CA). The expression level was calculated by dividing the integrated band intensity volume of the experimental sample with that of the control sample. 

Immunofluorescence Staining

ARPE-19 cells were grown in 12-well tissue culture dishes. For the study of nuclear translocation of NF-κB, quiescent cells were incubated with 10 ng/mL TNF-α for 60 minutes, with or without GS pretreatment. After incubation, cells were washed, fixed in methanol/acetone (1:1) for 1 hour at −20°C, and treated with 0.1% Triton X-100 for 10 minutes on ice. Cells were further incubated with 5% BSA in PBS for 1 hour at room temperature. Rabbit polyclonal antibodies against p50 and p65 subunits (1:200 dilution; Santa Cruz Biotechnology) were used as primary antibodies. Fluorescein isothiocyanate (FITC)-labeled goat anti-rabbit IgG was used as a secondary antibody (1:200 dilution; Santa Cruz Biotechnology). Cells were also stained with propidium iodide (2 μg/mL; Sigma-Aldrich) for the localization of nuclei. Preparations were mounted in 70% glycerol and examined by fluorescence microscopy. Images were photographed and printed at equivalent exposures. 

Preparation of Nuclear and Cytosolic Extracts

ARPE-19 cells were trypsinized, resuspended and homogenized in buffer A (10 mM HEPES [pH 7.8], 15 mM KCl, 2 mM MgCl₂, 0.1 mM EDTA [EDTA]), 1 mM dithiothreitol [DTT], and 1 mM PMSF). Nuclei and cytosolic fractions were separated by centrifugation at 1000g for 20 minutes. The cytosolic fractions (supernatant) were stored at −80°C until the analysis of STAT1 and phosphorylated STAT1. The nuclear fractions (pellets) were washed twice in buffer A and resuspended in the same buffer, with a final concentration of 0.39 M KCl. Nuclei were extracted for 1 hour at 4°C and centrifuged at 100,000g for 45 minutes. Supernatants (nuclear extracts) were dialyzed in buffer C (50 mM HEPES [pH 7.8], 50 mM KCl, 10% glycerol, 1 mM PMSF, 0.1% mM EDTA, and 1 mM DTT) and stored at −80°C until phosphorylated STAT1 analysis. Protein concentration was determined by the BCA method, as described earlier. 

Electrophoretic Mobility Shift Assay

Binding reactions were performed in a 20-μL reaction volume containing 20 mM HEPES, 1 mM MgCl₂, 50 mM KCl, 12% glycerol, 0.1 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT, 5 μg nuclear protein, and 1 μg poly[d(A-T)] (Sigma-Aldrich) as a nonspecific competitor. After 10 minutes on ice, 10 femtomoles ³²P-labeled NF-κB oligonucleotide (5′-AGTTGAGGGGACTTTCCCAGGC-3′) was added, and incubation was continued for 1 hour at room temperature. The reaction was stopped by adding gel-loading buffer. Samples were run on 4% polyacrylamide gels (38:1) in 0.25× TBE buffer at 150 V for 2.5 hours. The dried gels were exposed for autoradiography. 

Statistical Analysis

Student’s t-test was used to compare data between two groups. To compare data among three or more groups, one-way analysis of variance (ANOVA) followed by the Bonferroni test was used. Data are expressed as means ± SEM and P < 0.05 was considered statistically significant. 

Results

Effects of TNF-α and IFN-γ on ICAM-1 Expression

The ARPE-19 cells expressed low levels of ICAM-1 constitutively, as determined by Western blot analysis. Treatment for 24 hours with either TNF-α or IFN-γ increased this expression in a dose-dependent manner (Figs. 1A 2A) . Treatment with either TNF-α or IFN-γ rapidly increased the protein level of ICAM-1 in a time-dependent manner, with an early onset at 4 hours, and it reached statistically significant accumulation at 24 hours (Figs. 1B 2B) . 

Effects of GS on ICAM-1 Protein Levels Induced by TNF-α and IFN-γ

To evaluate whether GS inhibits the ICAM-1 protein production induced by either TNF-α or IFN-γ, cells were incubated for 1 hour with different doses of GS (10–1000 mg/L) before addition of either TNF-α or IFN-γ for 24 hours. GS did not affect cell viability, even at a concentration of 1000 mg/L (data not shown). GS significantly inhibited the TNF-α- or IFN-γ-induced synthesis of ICAM-1 in a dose-dependent manner (Figs. 3A 3B) . At 1000 mg/L, GS completely inhibited the production of ICAM-1 induced by 20 ng/mL TNF-α or 500 U/mL IFN-γ. 

Effects of Different Hexosamines on ICAM-1 Levels Induced by TNF-α or IFN-γ

To study whether other hexosamines also inhibit TNF-α or IFN-γ-induced ICAM-1 protein levels, we incubated ARPE-19 cells with equimolar concentrations of N-AcG or Gal (2 mM) 1 hour before TNF-α or IFN-γ stimulation for 24 hours. GS, N-AcG, and Gal all inhibited the TNF-α-induced ICAM-1 protein levels significantly, but N-AcG and Gal did not inhibit the effects of IFN-γ (Figs. 4A 4B) . 

Effects of GS on ICAM-1 mRNA Induced by TNF-α or IFN-γ

To assess whether the inhibitory effect of GS on protein levels of ICAM-1 induced by TNF-α or IFN-γ was associated with decreased steady-state mRNA levels, we examined the levels of mRNA of ICAM-1 in ARPE-19 cells by RT-PCR. Similar to our Western blot results, ICAM-1 was basally expressed at a very low level in ARPE-19 cells. Expression of ICAM-1 mRNA was significantly induced after treatment of ARPE-19 cells with either TNF-α or IFN-γ for 6 hours (Fig. 5) . However, GS cotreatment significantly inhibited these effects (Fig. 5) . 

Effect of GS on TNF-α-Induced NF-κB Activity

Because nuclear transcriptional factor NF-κB binding is essential for the activation of ICAM-1 expression induced by TNF-α, we studied whether GS would modify the NF-κB binding activity, leading decreased transcription of ICAM-1 expression in ARPE-19 cells. Stimulation of TNF-α (20 ng/mL) for 1 hour caused an increase in NF-κB binding activity shown by EMSA (Fig. 6) . GS at 1000 mg/L significantly inhibited this binding. 

Characterization of Subunits of NF-κB Involved in TNF-α-Induced NF-κB Activity

NF-κB is composed of two groups of structurally related, interacting proteins that bind DNA recognition sites as “dimmers”and whose activity is regulated by subcellular location. ¹⁸ Using an immunofluorescence assay, we examined the location of the subunits of NF-κB after stimulation by TNF-α for 1 hour, when ARPE-19 cells were preincubated with or without 1000 mg/L GS. There was diffuse cytoplasmic staining in control and GS-treated cells (Figs. 7A 7C) . Cells stimulated with TNF-α had a clear, strong nuclear staining pattern for p65 (Fig. 7E) . When preincubated with GS, TNF-α-stimulated cells did not show this, indicating the blockade of nuclear translocation of subunit p65 (Fig. 7G) . Only the nuclear staining pattern for p50 was seen in control cells. Stimulation with TNF-α did not enhance the nuclear staining pattern in cells preincubated with or without GS compared with the control (data not shown). 

Effects of GS on the Inhibitory Protein IκBα Produced by TNF-α-Induced NF-κB Activity

Cytokines such as TNF-α activate NF-κB through a kinase-mediated phosphorylation cascade followed by the subsequent degradation of IκB. ^{19 20} We studied whether the cytosol level of inhibitory protein IκBα was involved in the inhibitory effect of GS on NF-κB activity. IκBα was found to be constitutively present in control cells (Fig. 8) . Simulation with TNF-α for 45 minutes lowered the cytosol level of IκBα indicating its degradation. However, preincubation with GS restored the protein level of IκBα in cells stimulated with TNF-α. Although cytosolic IκBα degradation appears to occur after IFN-γ treatment, stimulation with IFN-γ did not significantly modify the cytosol level of IκBα. 

Effect of GS on STAT1 Phosphorylation Induced by IFN-γ

IFN-γ induction of ICAM-1 transcription occurs through the activation of Janus kinases (JAKs) and activators of transcription (STAT) signal transduction pathways. ^{21 22} The activated JAKs phosphorylate STAT, which allows the factor to translocate to the nucleus. ²³ We studied whether GS modifies the nuclear translocation of phosphorylated STAT1 in ARPE-19 cells stimulated with IFN-γ. Phosphorylated STAT1 was not detected in either control or GS-treated cells in cytosolic and nuclear cell extracts (Fig. 9) . STAT1 was also not phosphorylated in cytosolic and nuclear cell extracts by treatment with TNF-α, with or without preincubation with GS. However, STAT1 became activated, phosphorylated, and further translocated into the nucleus, as shown in both cytosolic and nuclear cell extracts after 45 minutes of treatment with 500 U/mL IFN-γ. GS (1000 mg/L) lowered the level of phosphorylated and translocated STAT1 in nuclear cell extracts and therefore increased the level of phosphorylated STAT1 in cytosolic cell extracts, indicating the partial blockade of nuclear translocation of phosphorylated STAT1 by GS in IFN-γ-activated cells. 

Discussion

In this study, human RPE cells (ARPE-19) expressed ICAM-1 constitutively. Exposure of ARPE-19 to TNF-α and IFN-γ resulted in a time- and dose-dependent increase in protein levels of ICAM-1. Moreover, GS significantly inhibited both the expression and synthesis of ICAM-1 induced by TNF-α and IFN-γ. We also found that GS inhibited NF-κB activity, prevented IκBα degradation in the cell cytoplasm, and prevented nuclear translocation of the NF-κB subunit, p65, in ARPE-19 cells stimulated with TNF-α. In addition, we demonstrated that preincubation of ARPE-19 with GS prevented the migration of activated and phosphorylated STAT1 from the cytoplasm to the nucleus. The inhibitory effect was greatest for GS compared with other hexosamines in TNF-α-induced ICAM-1 induction. Gal and N-AcG had no significant effects on the induction of ICAM-1 by IFN-γ in these cells. 

RPE cells have been implicated in the pathogenesis of both proliferative vitreoretinopathy (PVR) and certain types of posterior uveitis. In these disorders, RPE cells proliferate in the vitreous cavity and on the retinal surface and undersurface, where they may form contractile membranes and produce some inflammatory mediators to recruit and activate inflammatory cells. ^{24 25} Increased levels of proinflammatory cytokines such as TNF-α and IFN-γ, which are present in ocular tissues of patients with PVR and uveitis, increase the level of ICAM-1 protein in RPE cells. ^{26 27 28} Our findings are in accord with those in previous studies indicating that human ICAM-1 gene expression is upregulated by TNF-α or IFN-γ. ^{26 27 28} Interaction of ICAM-1 and leukocyte functional antigen (LFA)-1 is responsible for the firm adhesion of sticking leukocytes before extravasation. ²⁹  

This study is the first to demonstrate that GS effectively inhibits the TNF-α or IFN-γ-induced synthesis of ICAM-1 protein in a dose-dependent manner compared with TNF-α- or IFN-γ-treated ARPE-19 cells. At a concentration of 1000 mg/L, GS completely inhibited the protein level of ICAM-1 induced by 20 ng/mL TNF-α or 500 U/mL IFN-γ. However, at this concentration, GS did not affect cell viability. ICAM-1 is upregulated primarily at the level of gene transcription. ^{30 31} We found that this inhibitory effect of GS on protein levels of ICAM-1 induced by TNF-α or IFN-γ was associated with decreased steady state mRNA levels. GS has also been reported to inhibit the expression and synthesis of COX-2, a NF-κB-controlled protein, in human osteoarthritic chondrocytes (HOC) stimulated with IL-1β. ³²  

The NF-κB binding site upstream of the transcription start site has been shown to be particularly crucial in the induction of ICAM-1 transcription by TNF-α. ^{33 34} In the present study, we demonstrated that GS inhibited NF-κB activation and lowered the NF-κB-binding activity shown in EMSA, which was related to the downregulation of gene expression and synthesis of ICAM-1 in cells stimulated with TNF-α. NF-κB is a heterodimer mainly consisting of p65 (Rel A), and p50 and is located in the cytoplasm associated with several inhibitory molecules (IκBs). ^{35 36 37 38} Once cells are exposed to proinflammatory cytokines such as TNF-α, signal transduction leads to phosphorylation and degradation of IκB. NF-κB is released from an inhibitory signalosome, translocates to the nucleus, and induces the transcription of a variety of genes. We found that preincubation with GS specifically prevented the nuclear translocation of the NF-κB subunit p65 in ARPE-19 induced by TNF-α. By Western blot analysis, we further found that IκB levels in cytosolic fractions were markedly decreased in TNF-α-stimulated cells. Preincubation with GS prevented the TNF-α-mediated degradation of cytosolic IκBs. Taken together, these results suggest that GS exerts its effect, at least in part, by specific blockage of IκB degradation. 

IFN-γ induction of ICAM-1 transcription occurs through the activation of the JAK-STAT signal transduction pathway. ^{21 22} Activated JAKs subsequently phosphorylate STAT, which allow the factor to translocate to the nucleus. ²³ We found that GS modified the nuclear translocation of phosphorylated STAT1 in ARPE-19 stimulated with IFN-γ. STAT1 became activated, phosphorylated, and further translocated into the nucleus, as shown in both cytosolic and nuclear cell extracts after stimulation with IFN-γ. GS treatment at 1000 mg/L lowered the levels of phosphorylated and translocated STAT1 in nuclear cell extracts and, in turn, increased the level of phosphorylated STAT1 in cytosolic cell extracts, indicating partial prevention of nuclear translocation of phosphorylated STAT1 by GS in IFN-γ-activated cells. 

Our study demonstrated that although it was not statistically significant, stimulation with IFN-γ in RPE cells enhanced cytosolic IκBα degradation, indicating possible cross-talk between the TNF-α and IFN-γ pathways. TNF-α and IFN-γ synergize in the production of a large number of proinflammatory cytokines, chemokines, and expression of adhesion molecules. ³⁹ For these genes, cross-talk between TNF-α and IFN-γ occurs at multiple levels. Most of the promoters are induced synergistically by the binding of TNF-α-activated NF-κB and IFN-γ-activated STAT1 to their respective binding sites within promoters. ³⁹ Another level of cross-talk exists at the level of NF-κB activation. Although typically it does not activate NF-κB, IFN-γ enhances and prolongs TNF-α-dependent NF-κB activation through two separate mechanisms. IFN-γ synergistically enhances TNF-α-induced NF-κB nuclear translocation through a mechanism that involves the induced degradation of IkBα. ⁴⁰ The TNF receptor 1(TNFR1) interacts with STAT1, and this association inhibits TNF-α-mediated NF-κB activation. ⁴¹ IFN-γ attenuates the ability of STAT1 to interact with the TNFR1, which in turn allows for optimal NF-κB activation on TNF-α ligation in macrophages. ⁴²  

GS has become a popular nutritional supplement in relieving the symptoms of osteoarthritis (OA). ^{43 44 45 46} However, the mechanisms of the putative beneficial effects of GS on OA remain to be established. The reported therapeutic value of GS has been supported by several in vitro studies. Some have demonstrated that GS downregulates IL-1β-induced COX-2 gene expression and subsequent prostaglandin E2 synthesis in chondrocytes and decreases neutrophil function and immune activity in the synovial tissue. ^{10 32 47} GS also inhibits aggrecanase-mediated cleavage of aggrecan, the major proteoglycan in articular cartilage, and can prevent or reduce articular cartilage degradation in vitro. ^{48 49} It has also been proposed that GS may quench small bioactive molecules including NO and oxygen radicals that can damage articular cartilage. ⁵⁰ In addition to the anti-inflammatory effects of GS in the treatment of OA, the supply of GS may simply increase intracellular concentrations of UDP-N-AcG, which is essential in the formation of glycosaminoglycan, found in cartilage. ⁵¹  

Although our data reveal that GS, N-AcG, and Gal all inhibited ICAM-1 expression induced with TNF-α in ARPE-19 cells significantly, GS was the most potent. Both TNF-α and IL-1β upregulate ICAM-1 expression through the NF-κB activation pathway. However, in contrast, another study has reported that neither N-AcG nor Gal inhibits the NF-κB activation induced by IL-1β significantly. ³² This discrepancy may be due to the different cell lines used in the studies. 

In comparing the different amino sugars, our data demonstrate that hexosamines such as N-AcG and Gal, but not GS, were unable to inhibit ICAM-1 expression significantly in ARPE-19 cells stimulated with IFN-γ. N-AcG was also demonstrated to be incapable of inhibiting cartilage degradation in equine articular cartilage explant cultures. ⁴⁹  

Our results indicate that GS is a potent agent in lowering ICAM-1 gene expression and protein synthesis induced by proinflammatory cytokines including TNF-α and IFN-γ. ICAM-1/LFA-1 interaction is essential for T-cell activation as well as for migration of T-cells to target tissues. This interaction also serves as a costimulatory signal (signal-2) for T-cell activation. ⁵² Therefore, targeting of ICAM-1/LFA-1 interaction may suppress T-cell activation in autoimmune diseases and organ transplantation. Recently, GS has also been demonstrated to suppress the activation of T-cells and dendritic cells in vitro and, more important, prolong cardiac allograft survival in vivo. ¹⁰  

In conclusion, GS effectively inhibited ICAM-1 expression and synthesis, through the prevention of NF-κB activity and activated STAT1 nuclear translocation in human RPE cells stimulated with TNF-α or IFN-γ. Our study demonstrated a potentially important property of GS in reducing ICAM-1-mediated inflammatory mechanisms in the eye. 

 Supported in part by Grant TSGH-C94-63 from the Tri-Service General Hospital and by the C. Y. Chai Foundation for the Advancement of Education, Sciences, and Medicine.

 Submitted for publication July 31, 2005; revised October 13, 2005; accepted December 16, 2005.

 Disclosure: J.-T. Chen, None; J.-B. Liang, None; C.-L. Chou, None; M.-W. Chien, None; R.-C. Shyu, None; P.-I. Chou, None; D.-W. Lu, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Immunoblot analysis of the effects of TNF-α on protein levels of ICAM-1 in cultured ARPE-19 cells. (A) TNF-α increased ICAM-1 protein levels in ARPE-19 cells in a dose-dependent manner, as shown in a Western blot. ARPE-19 cells were treated at the indicated dosage for 24 hours. (B) TNF-α increased ICAM-1 protein levels in ARPE-19 cells in a time-dependent manner, as shown in a Western blot. ARPE-19 cells were treated with 20 ng/mL TNF-α at the indicated period. ICAM-1 protein levels in ARPE-19 cells were expressed as ratios to the mean value of the control group. Data are expressed as the mean ± SEM; n = 4; *P < 0.05 compared with the control.

Figure 2.

 

Immunoblot analysis of the effects of IFN-γ on levels of ICAM-1 protein in cultured ARPE-19 cells. (A) IFN-γ increased ICAM-1 protein levels in ARPE-19 cells in a dose-dependent manner as shown in a Western blot. ARPE-19 cells were treated at the indicated dosage for 24 hours. (B) IFN-γ increased ICAM-1 protein levels in ARPE-19 cells in a time-dependent manner as shown in a Western blot. ARPE-19 cells were treated with 500 U/mL IFN-γ at the indicated period. ICAM-1 protein levels in ARPE-19 cells were expressed as ratios to the mean in the control group. Data are shown as the mean ± SEM; n = 4;*P < 0.05 compared with the control.

Figure 3.

 

Effects of GS on ICAM-1 protein levels in TNF-α- or IFN-γ-stimulated ARPE-19 cells. ARPE-19 cells were preincubated with GS for 1 hour and then stimulated with either (A) TNF-α or (B) IFN-γ for 24 hours. Protein levels were quantified by Western blot analysis and are expressed as ratios to the mean of the control group. Data are shown as the mean ± SEM; n = 4; *P < 0.05 compared with TNF-α- or IFN-γ-treated groups.

Figure 4.

 

Effects of different hexosamines on ICAM-1 protein levels in TNF-α or IFN-γ-stimulated ARPE-19 cells. Cells were preincubated with hexosamines for 1 hour and then were stimulated with either (A) TNF-α or (B) IFN-γ for 24 hours. Protein levels were quantified by Western blot analysis and are expressed as ratios to the mean of the control. Data are shown as the mean ± SEM; n = 4; *P < 0.05 compared with TNF-α- or IFN-γ-treated groups. #P < 0.05 compared with N-AcG+TNF-α- and Gal+TNF-α-treated groups.

Figure 5.

 

Effects of GS on gene expression of ICAM-1 in TNF-α or IFN-γ-stimulated ARPE-19 cells. ARPE-19 cells were preincubated with GS for 1 hour and then stimulated with TNF-α or IFN-γ for 6 hours. The levels of gene expression were quantified by RT-PCR and are expressed as ratios to the average of the control group. Data are shown as the mean ± SEMs; n = 4; *P < 0.05 compared with the control. §P < 0.05 compared with TNF-α-treated cells. #P < 0.05 compared with IFN-γ-treated cells.

Figure 6.

 

Effects of GS on the NF-κB binding activity in TNF-α-stimulated ARPE-19 shown by electrophoretic mobility shift assay. Lane 1: 100× unlabeled NF-κB probe; lane 2: stimulation with TNF-α (20 ng/mL) for 1 hour; lane 3: preincubation with GS (1000 mg/L) and then stimulation with TNF-α (20 ng/mL); lane 4: preincubation with GS (100 mg/mL) and then stimulation with TNF-α (20 ng/mL) for 1 hour; lane 5, incubation with GS (1000 mg/mL); lane 6: control. Data are representative of results in three experiments.

Figure 7.

 

Immunofluorescence assays for the p65 subunit of NF-κB. (A, C, E, G) on sections stained with antibody to p65. (B, D, F, H) Sections stained with propidium iodide (2.0 μg/mL) for the localization of nuclei. (A) Cytoplasmic immunostaining was seen in control cells. (C) Incubation with GS (1000 mg/L) alone also produced cytoplasmic staining. (E) When ARPE-19 cells were stimulated with TNF-α (20 ng/mL) for 1 hour, there was intense nuclear immunostaining. (G) When ARPE-19 cells were preincubated with GS (1000 mg/mL) and then stimulated with TNF-α (20 ng/mL) for 1 hour, there was a significant decrease in p65 nuclear immunostaining compared with TNF-α-stimulated cells. Data are representative of results in three experiments. Magnification, ×200.

Figure 8.

 

Effects of GS on the inhibitory protein IκBα in TNF-α-induced NF-κB activity. ARPE-19 cells were preincubated with GS for 1 hour and then stimulated with TNF-α or IFN- ã for 45 minutes. Protein levels of IκBα were quantified by Western blot and are expressed as ratios to the mean of the control group. Data are shown as the mean ± SEM; n = 4; *P < 0.05 compared with the control. #P < 0.05 compared with the TNF-α-treated cells.

Figure 9.

 

Effect of GS on STAT1 phosphorylation induced by IFN-γ. ARPE-19 cells were preincubated with GS for 1 hour and then stimulated with TNF-α or IFN-γ for 45 minutes. Protein levels of phosphorylated STAT1 in cytosolic and nuclear fraction were quantified by Western blot and are expressed as ratios to the average of the control group. Data are shown as the means ± SEM; n = 4; *P < 0.05 compared with the control. #P < 0.05 compared with IFN-γ-treated cells.