Advanced glycation end products (AGEs) are proteins or lipids that become glycated after exposure to diverse reducing sugars. Accumulation of AGEs induces diabetes complications. Microinflammation is a common major mechanism in the pathogenesis of diabetic vascular complications. Activation of monocytes/macrophages and T cells plays roles in the pathogenesis of atherosclerosis. The activation of T cells requires the enhanced expression of adhesion molecules on monocytes. AGEs activate monocytes by engaging the receptor for AGE (RAGE); however, little is known about the profile of agonist activity of diverse AGE moieties on monocytes. We investigated the effect of four distinct AGE subtypes (AGE-modified bovine serum albumin; AGE-2, AGE-3, AGE-4, and AGE-5) at concentrations ranging from 0.1 to 100 μg/ml on the expression of intercellular adhesion molecule-1, B7.1, B7.2, and CD40 on monocytes and its impact on the production of interferon-γ and tumor necrosis factor-α in human peripheral blood mononuclear cells. Among the AGEs examined, AGE-2 and AGE-3 selectively induced adhesion molecule expression and cytokine production. Antagonism experiments using antibodies against adhesion molecules demonstrated that cell-to-cell interaction between monocytes and T/natural killer cells was involved in AGE-2- and AGE-3-induced cytokine production. AGE-2 and AGE-3 up-regulated the expression of RAGE on monocytes. The effects of AGE-2 and AGE-3 were inhibited by nuclear factor-κB and p38 mitogen-activated protein kinase inhibitors. These results indicated that AGE-2 and AGE-3 activated monocytes via RAGE, leading to the up-regulation of adhesion molecule expression and cytokine production.
Advanced glycation end products (AGEs) are a heterogeneous class of compounds and modifications of proteins or lipids that become nonenzymatically glycated and oxidized after contact with aldose sugars (Schmidt et al., 1994). Accumulation of AGEs in disorders such as diabetes, renal failure, Alzheimer's disease, and natural aging has suggested their potential contribution to the pathogenesis of complications that typify these conditions (Brownlee, 1995; Takedo et al., 1996). Direct immunochemical evidence for the existence of six distinct AGE structures, including AGE-1, AGE-2, AGE-3, AGE-4, AGE-5, and AGE-6, was provided from the analysis of AGEs within modified proteins and peptides (Takeuchi and Yamagishi, 2004). Recently, it was demonstrated that toxic AGE structures, glyceraldehyde-derived AGE (AGE-2) and glycolaldehyde-derived AGE (AGE-3), have diverse biological activities on vascular endothelial cells, vascular smooth muscle cells, mesangial cells, Schwann cells, malignant melanoma cells, and cortical neurons (Okamoto et al., 2002). AGEs have also been suggested to have profound effects on inflammatory and immune cells (Figarola et al., 2007); however, it is still not clear which AGE subtypes play a role in the modulation of immune response.
Receptor for AGE (RAGE), a member of the immunoglobulin superfamily, was first described as a cell surface receptor for AGEs (Neeper et al., 1992). It is reported that AGEs ligate cell surface RAGE on the vascular endothelium, mononuclear phagocytes, vascular smooth muscle, and neurons to activate cell signaling pathways such as p44/p42 mitogen-activated protein kinase (MAPK) and nuclear factor-κB (NF-κB) (Yan et al., 1994), leading to redirect cellular function in a manner linked to the expression of inflammatory and prothrombotic genes important in the pathogenesis of chronic disorders such as diabetic microvascular disease and amyloidosis (Schmidt et al., 1994; Miyata et al., 1996; Park et al., 1998). Interestingly, AGEs up-regulate RAGE expression in various tissues, facilitating the AGEs-RAGE response by forming a positive feedback loop (Yamagishi and Imaizumi, 2005). Conversely, it is suggested that interruption of the interaction of AGEs with RAGE in vivo, by the administration of soluble form of RAGE (sRAGE), an extracellular ligand-binding domain of RAGE, reversed vascular hyperpermeability and suppressed accelerated atherosclerotic lesion development in diabetic rodents (Miyata et al., 1996; Park et al., 1998). Thus, ligation of AGEs with RAGE might play an important role in the development of various diabetic complications, including atherosclerosis.
Microinflammation is a common major mechanism in the pathogenesis of diabetic vascular complications. It is reported that diabetes has greater macrophage and T-cell infiltration in atherosclerotic plaques (Burke et al., 2004). Macrophages are recruited by abnormal endothelium over developing atherosclerotic plaques (Boyle, 2005). In addition, recruitment of monocytes/macrophages and T cells in diabetic glomeruli during the early stage of diabetes is considered to be involved in the progression of diabetic nephropathy (Sugimoto et al., 1997). Monocyte/macrophage activation, adhesion, and migration are key events in the pathogenesis of atherosclerosis (Figarola et al., 2007). Activated T cells induce the progression of inflammatory atherosclerotic plaques (Stoll and Bendszus, 2006). AGEs are implicated in the pathogenesis of atherosclerotic vascular disease of diabetic etiology (Stitt et al., 1997). The stimulation of RAGE is reported to induce plaque rupture in diabetic patients (Cuccurullo et al., 2006).
It is known that the enhanced expression of adhesion molecules, including intercellular adhesion molecule (ICAM)-1, B7.1, B7.2, and CD40, on monocytes results in the activation of T cells (Durie et al., 1994; Ranger et al., 1996; Camacho et al., 2001). We also found that cell-to-cell interactions mediated by the engagement between ICAM-1, B7.1, B7.2, and CD40, respectively, on monocytes and their ligands, lymphocyte function-associated antigen-1, CD28, and CD40 ligand, on T cells were involved in T-cell activation, leading to induce the production of interferon (IFN)-γ and tumor necrosis factor (TNF)-α in human peripheral blood mononuclear cells (PBMC) (Takahashi et al., 2003). However, little is known about the effect of AGE subtypes on adhesion molecule expression as well as cytokine production. Therefore, we examined the effect of AGE-2, AGE-3, AGE-4, and AGE-5 on the expressions of ICAM-1, B7.1, B7.2, and CD40 on monocytes and the production of IFN-γ and TNF-α in PBMC to better understand the profile of each AGE in monocyte activation. Moreover, we analyzed changes in the expression of RAGE on monocytes induced by AGEs, the involvement of adhesion molecules in the cytokine-producing action of AGEs, and the relevant signal pathways triggered by AGEs.
Materials and Methods
Reagents and Drugs. Recombinant human IFN-γ, TNF-α, anti-IFN-γ Ab and anti-TNF-α Ab were purchased from BD Biosciences (San Jose, CA). SN50 and SB203580 were purchased from Calbiochem (San Diego, CA). AGE-modified bovine serum albumin (BSA) (Sigma-Aldrich, St. Louis, MO) was prepared as described previously (Takeuchi et al., 2000). In brief, each protein was incubated under sterile conditions with glyceraldehyde-3-phosphate (AGE-2) (Sigma-Aldrich), glycolaldehyde (AGE-3) (Sigma-Aldrich), methylglyoxal (AGE-4) (Sigma-Aldrich), or glyoxal (AGE-5) (Tokyo Kasei, Tokyo, Japan) in 0.2 M phosphate buffer, pH 7.4, at 37°C for 7 days. BSA was incubated under the same conditions. AGE-BSA and BSA were dialyzed for 2 days at 4°C. The endotoxin concentration of AGEs at 100 μg/ml described above was measured at SRL (Okayama, Japan), and it was found to be 1.2 pg/ml. AGE-specific fluorescence was measured at 450 nm after excitation at 390 nm with a fluorescence spectrophotometer (Hitachi, Tokyo, Japan). sRAGE was prepared as described previously (Goren et al., 2007). In brief, sRAGE was cloned into the Escherichia coli expression vector pASK-IBA32; some modifications to the pASK-IBA32 polylinker region were made for cloning purposes. sRAGE, from amino acids 23 to 340 (Neeper et al., 1992), was amplified by polymerase chain reaction using the following oligonucleotides (MWG Biotech, High Point, NC); the underlined bases designate restriction sites: 5′-CTGACCTATG CGGCCGCTGC TCAAAACATCACAGC-3′ and 5′-GACTGAATTC ATCAGTGATGATGGTGATGG TGAGTTCCCA GCCCTGATCC-3′. Anti-human RAGE rabbit polyclonal Ab was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Isolation of PBMC, Monocytes, and T Cells. Normal human PBMC were obtained from ten healthy volunteers after acquiring Institutional Review Board approval (Okayama University Institutional Review Board no. 106). Samples of 20 to 50 ml of peripheral blood were withdrawn from a forearm vein, after which the PBMC were prepared, and monocytes isolated from PBMC were separated by counterflow centrifugal elutriation as described previously (Takahashi et al., 2003). T cells were then enriched from PBMC by passing them through a nylon wool column to a purity of 85% T cells as determined by flow cytometry with FITC-conjugated anti-CD3 antibody. PBMC, monocytes, and T cells were then suspended at a final concentration of 1 × 106 cells/ml in the medium as described previously (Takahashi et al., 2003).
Measuring AGEs-RAGE Binding Using Microtiter Plate in Vitro. Ninety six-well flat-bottomed microtiter plates were coated with BSA-AGEs at increasing concentrations from 0.3125 to 20 μg/ml and were incubated at 4°C for 16 h with gentle shaking on an orbital microplate shaker. After three times washings with washing buffer (10 mM Tris-buffered saline containing 0.05% Tween 20, pH 7.5), plates were blocked with 10% BSA at 4°C for 16 h. The plates were then incubated with His-tagged sRAGE at 0.83 μg/ml in 10 mM Tris-buffered saline at 4°C for 16 h. After three times washings, nickel-nitrilotriacetic acid-horseradish peroxidase conjugate (QIAGEN, Osaka, Japan) was added to the well, and the incubation was performed at room temperature for 1 h. The reaction was developed by the addition of 0.15% H2O2 and 2.5 mM 2,2-azinobis (3-ethylbenzothiazoline-6-sulfonic acid ammonium salt) (Tokyo Kasei Kogyo Co., Ltd., Tokyo Japan) in 0.2 M citrate buffer, pH 4.0.
Flow Cytometric Analysis for Adhesion Molecule Expression. For flow cytometric analysis, FITC-conjugated mouse IgG1 monoclonal (m) Ab against ICAM-1/CD54 and phycoerythrin-conjugated anti-CD14 mAb were purchased from Dako Denmark A/S (Glostrup, Denmark). FITC-conjugated mouse IgG1 mAb against B7.1 was purchased from Immunotech (Marseille, France), FITC-conjugated mouse IgG1 mAb against B7.2 and CD40 was from BD Biosciences Pharmingen (San Diego, CA), and FITC-conjugated IgG1 isotype-matched control was obtained from Sigma-Aldrich. Changes in the expression of human leukocyte antigens, ICAM-1, B7.1, B7.2, and CD40, on monocytes were examined by multicolor flow cytometry using a combination of anti-CD14 Ab with anti-ICAM-1, anti-B7.1, anti-B7.2, or anti-CD40 Ab. In addition, to analyze changes in the expression of human RAGE on monocytes by multicolor flow cytometry using a combination of phycoerythrin-conjugated anti-human CD14 mouse IgG (Dako Denmark A/S) and anti-human RAGE rabbit IgG (Santa Cruz Biotechnology, Inc.) followed by FITC-conjugated anti-rabbit IgG goat IgG (Dako Denmark A/S) was performed. PBMC at 1 × 106 cells/ml were incubated for 24 h. Cultured cells at 5 × 105 cells/ml were prepared for flow cytometric analysis as described previously (Takahashi et al., 2003). The cells were analyzed with a FACSCalibur (BD Biosciences). Data were processed using the Cell Quest program (BD Biosciences).
Cytokine Production. PBMC at 1 × 106 cells/ml were used for analyzing IFN-γ and TNF-α production. After culturing for 24 h at 37°C in a 5% CO2, air mixture, the cell-free supernatant was assayed for IFN-γ and TNF-α protein by enzyme-linked immunosorbent assay (ELISA) using the multiple Abs sandwich principle (R&D Systems, Minneapolis, MN). The detection limit of ELISA for IFN-γ and TNF-α was 10 pg/ml.
Immunofluorescence Staining. For double immunofluorescence staining, PBMC at 1 × 106 cells/ml were incubated for 24 h in the presence or absence of AGE-2 AGE-3 and BSA at 100 μg/ml. The cultured cells were fixed in 1% paraformaldehyde and stained with anti-human CD14 mouse IgG (Dako Denmark A/S) followed by Alexa 555-conjugated anti-mouse IgG rabbit IgG (Millipore Bioscience Research Reagents, Temecula, CA) and anti-human RAGE rabbit IgG (Santa Cruz Biotechnology, Inc.) followed by Alexa Fluor 488-conjugated anti-rabbit IgG goat IgG (Millipore Bioscience Research Reagents), respectively. The stained cells were mounted for viewing by fluorescent confocal microscopy (Biozero BZ8000; Keyence, Osaka, Japan).
Western Blot Analysis. The effect of AGE-2 and AGE-3 on the expression of RAGE was determined by Western blot analysis. Anti-human RAGE rabbit IgG was obtained by immunization of rabbit with recombinant human sRAGE. The lysates of human monocytes or A549 cells, a human pulmonary epithelial cell line, were electrophoresed on SDS-polyacrylamide gel electrophoresis gel. The fractionated proteins were transferred to nitrocellulose membrane. The blotting of β-actin was used as a loading control.
Statistical Analysis. Statistical significance was evaluated using analysis of variance followed by Dunnett's test. A probability value of less than 0.05 was considered to indicate statistical significance. The results are expressed as the means ± S.E.M. of triplicate findings from five donors.
AGE Subtype Binding to sRAGE. To evaluate the binding of AGE subtypes to RAGE, we established the in vitro assay by using the immobilized AGE subspecies and the His-tagged sRAGE protein (Fig. 1). AGE-2 and AGE-3 showed relatively high-affinity binding for sRAGE, whereas AGE-4 and AGE-5 showed a moderate affinity (8 times less potent than AGE-2 and AGE-3) for sRAGE. We confirmed that the immobilization efficiency of each AGE was the same when detected by anti-BSA Ab (data not shown).
Time Course Effects of AGEs on the Expression of ICAM-1, B7.1, B7.2, and CD40 on Monocytes and the Production of IFN-γ and TNF-α in PBMC. To determine the proper incubation time, we examined the kinetics as shown in Figs. 2 and 3. In the absence of AGEs and BSA, the expression of ICAM-1, B7.1, B7.2, and CD40 moderately increased at 16 h and thereafter up to 24 and 48 h (Fig. 2A). AGE-2 and AGE-3 at 100 μg/ml significantly enhanced the expression of ICAM-1, B7.1, B7.2, and CD40 at 16 h and thereafter up to 24 and 48 h (Fig. 2, B, C, and G), whereas AGE-4, AGE-5, and BSA at 100 μg/ml had no effect at all (Fig. 2, D–F). In the absence of AGEs and BSA, the production of IFN-γ and TNF-α was under the detection limit during the incubation period (Fig. 3A). AGE-2 and AGE-3 at 100 μg/ml also time-dependently induced the production of IFN-γ and TNF-α in PBMC (Fig. 3, B and C). Neither AGE-4, AGE-5, nor BSA induced cytokine production (Fig. 3, D–F).
Dose-Response Relationship of AGEs on the Expression of ICAM-1, B7.1, B7.2, and CD40 on Monocytes, and the Production of IFN-γ and TNF-α in PBMC. The level of glyceraldehyde-derived AGE (AGE-2) is reported to be 17 μg/ml in the serum of patient with diabetes (Enomoto et al., 2006; Nakamura et al., 2007). It is reported that AGEs at the concentrations ranging from 50 to 200 μg/ml remarkably induce human monocyte adhesion to bovine retinal endothelial cells (Mamputu et al., 2004). AGEs at 200 μg/ml induce the expression of CD40, CD80, and CD86 and the production of IFN-γ in dendritic cells (Ge et al., 2005). Therefore, the effects of AGE-2, AGE-3, AGE-4, AGE-5, and BSA at concentrations ranging from 100 ng/ml to 100 μg/ml on the expression of ICAM-1, B7.1, B7.2, and CD40 and the production of IFN-γ and TNF-α were determined (Figs. 4 and 5). AGE-2 and AGE-3 concentration-dependently induced adhesion molecule expression with similar potency (Fig. 4, A and B), whereas AGE-4, AGE-5, and BSA showed no effect (Fig. 4, C–E). Consistent with the up-regulation of adhesion molecules, AGE-2 and AGE-3 concentration-dependently induced the production of IFN-γ and TNF-α (Fig. 5, A and B). The levels of IFN-γ and TNF-α production induced by AGE-2 at 100 μg/ml were 820 and 800 pg/ml, and those by AGE-3 at 100 μg/ml were 805 and 810 pg/ml, respectively. When we assumed that the effects of AGE-2 and AGE-3 were maximal at the concentration of 100 μg/ml, the ED50 values of AGE-2 for the induction of ICAM-1 expression and TNF-α production were calculated to be 3 and 5 μg/ml, and those of AGE-3 were 10 and 5 μg/ml, respectively. However, AGE-4, AGE-5, and BSA had no effect on cytokine production (Fig. 5, C–E).
Involvement of IFN-γ and TNF-α in the Effect of AGE-2 and AGE-3 on Adhesion Molecule Expression. We examined the involvement of IFN-γ and TNF-α in the AGE-2 and AGE-3-induced expression of ICAM-1, B7.1, B7.2, and CD40 on monocytes (Fig. 6). IFN-γ and TNF-α at 10 ng/ml remarkably increased the expression of all adhesion molecules examined, and the expression reached a maximum level at 100 ng/ml. Moreover, the effects of cytokines and AGEs were additive. IFN-γ and TNF-α at 10 and 100 ng/ml enhanced the actions of AGE-2 and AGE-3, but those at 1 ng/ml had no effect (Fig. 6, A and B). Alternatively, anti-IFN-γ and anti-TNF-α Abs up to 100 ng/ml, which blocked the effect of exogenous IFN-γ and TNF-α at 100 ng/ml on adhesion molecule expression (data not shown), had no effect on the actions of AGE-2 and AGE-3 at 100 μg/ml (Fig. 6, C and D).
Involvement of Adhesion Molecule in the Effect of AGE-2 and AGE-3 on Cytokine Production. We examined the involvement of ICAM-1, B7.1, B7.2, and CD40 in the AGE-2- and AGE-3-induced production of IFN-γ and TNF-α in PBMC. As shown in Fig. 7, A and C, AGE-2 and AGE-3 concentration-dependently induced the production of IFN-γ and TNF-α in monocytes isolated from PBMC, exhibiting 20% of the amount obtained in PBMC, as shown in Fig. 5, B and C. AGE-2 and AGE-3 had no effect on the production of IFN-γ and TNF-α in T cells isolated from PBMC (Fig. 7, A and C). As shown in Fig. 7, B and D, anti-ICAM-1, anti-B7.1, anti-B7.2, and anti-CD40 Abs concentration-dependently inhibited the AGE-2- and AGE-3-induced production of IFN-γ and TNF-α in PBMC. The maximal inhibitory effect obtained by each Ab at 10 μg/ml was varied, and the rank order for the inhibition was as follows: ICAM-1 > CD40 > B7.1 > B7.2.
Expression of RAGE on Monocytes. The expression of RAGE on monocytes was determined by immunocytochemical staining (Fig. 8, A–D) and flow cytometry (Fig. 9A) using anti-RAGE Ab. AGE-2 and AGE-3 remarkably enhanced the expression of RAGE 24 h after stimulation, but AGE-4 and AGE-5 had no effect (data not shown). To confirm the binding specificity of anti-RAGE Ab to RAGE, we used a combination of anti-RAGE Ab with sRAGE for flow cytometry and immunocytochemical staining in the presence or absence of AGE-2 and AGE-3 (Figs. 8, B and D, and 9A). The detection of RAGE in the presence or absence of AGE-2 and AGE-3 was inhibited completely by the addition of sRAGE, indicating that anti-RAGE Ab recognized the extracellular domain of RAGE. As shown in Fig. 9B, Western blot analysis showed that RAGE is highly expressed on A549 cells, a human pulmonary epithelial cell line, as reported previously (Nakano et al., 2006). We confirmed that AGE-2 and AGE-3 at 100 μg/ml induced the up-regulation of protein on human monocytes 24 h after the start of stimulation (Fig. 9B).
Involvement of NF-κB and MAPK in the Actions of AGE-2 and AGE-3. We examined the involvement of NF-κB and p38 MAPK activation in AGE-2- and AGE-3-induced expression of ICAM-1, B7.1, B7.2, and CD40 on monocytes and production of TNF-α and IFN-γ in PBMC (Fig. 10). A NF-κB inhibitor, SN50, and a p38 MAPK inhibitor, SB203580, reduced the AGE-2- and AGE-3-enhanced adhesion molecule expression as well as cytokine production. Even in the absence of AGE-2 and AGE-3, the SN50 and SB203580 inhibitors concentration-dependently inhibited adhesion expression, but neither had any effect on basal cytokine production.
The diabetic state produces micro- and macrovascular lesions via various metabolic derangements, leading to crucial complications, including acquired blindness, end-stage renal failure, and neuropathies. In diabetic patients, the infiltration of macrophages and T cells into atherosclerotic plaques seems to be facilitated (Burke et al., 2004). Monocytes or macrophages, major components of atherosclerotic lesions and in contact with circulating plasma, play important roles in the hypercoagulable state and the progression of vascular injury, which develop during the progression of diabetes mellitus (Radoff et al., 1990). Macrophages are activated by atherosclerotic risk factors, including oxidized low-density lipoprotein and AGEs, leading to death of cells and degradation of the extracellular matrix (Figarola et al., 2007). AGEs are involved in the cause of foam cell formation via the increased expression of oxidized low-density lipoprotein receptors in accelerated atherosclerotic lesions of patients with diabetes (Iwashima et al., 2000). NO by macrophages up-regulates the expression of Fas on vascular smooth muscle cells, priming them for apoptosis. The phenotypes of advanced diabetic nephropathy were prevented by administering an AGE inhibitor, (±)-2-isopropylidenehydrazono-4-oxothiazolidin-5-ylacetanilide, in RAGE-transgenic mice crossbred with transgenic mice carrying human cDNA for inducible NO synthase under the control of the insulin promoter (Yamamoto et al., 2001). We therefore suggested that the activation of T cells by the enhancement of adhesion molecule expression on monocytes might result in the development of diabetic microangiopathy.
AGEs develop diabetic microangiopathy through the stimulation of RAGE (Stern et al., 2002). AGEs and RAGE are also detected in atherosclerotic plaque of diabetic patients (Cuccurullo et al., 2006). Takeuchi and Makita (2001) provided direct immunological evidence for the existence of four distinct AGE classes (AGE-2–5) among the AGE-modified proteins and peptides found in the serum of diabetic patients on hemodialysis. Antibodies, which cross-reacted with BSA modified by AGE-2, AGE-3, AGE-4, and AGE-5, recognized non–N-carboxymethyllysine AGE-2, AGE-3, AGE-4, and AGE-5 in serum samples obtained from type 2 diabetic patients, respectively. Among various subtypes of AGE, it has been shown that AGE-2 and AGE-3 are the main structures of AGEs that are detectable in the serum of diabetic patients (Takeuchi and Yamagishi, 2004). It is reported that AGE-2 induces diabetic microangiopathy (Takeuchi et al., 2000). Therefore, it has been suggested that blockade of the engagement of AGEs and RAGE is a target for the treatment of diabetic microangiopathy. However, little is known about the effect of AGE subtypes on the engagement of monocytes and T cells.
Several reports have shown a positive correlation between serum levels of TNF-α and circulating AGE levels. TNF-α promotes macrophage-induced apoptosis of human vascular smooth muscle cells (Boyle et al., 2003). Goren et al. (2007) reported that anti-TNF-α therapy, widely used in chronic inflammatory diseases in humans, improves healing of diabetes skin ulcers by targeting activated TNF-α-expressing macrophage subsets. Although it is reported that AGEs induce the production of TNF-α in monocytes (Boyle et al., 2001), little is known about the mechanism of enhancement of TNF-α production in monocytes. In the present study, we examined, for the first time, the affinity of AGE-2, AGE-3, AGE-4, and AGE-5 for RAGE by using in vitro assay system (Fig. 1). AGE-2 and AGE-3 have higher affinity for RAGE compared with AGE-4 and AGE-5. As shown in Figs. 2, 3, 4, 5, we examined the effect of AGE subtypes on the expression of ICAM-1, B7.1, B7.2, and CD40 on monocytes, and the production of IFN-γ and TNF-α in PBMC. AGE-2 and AGE-3, but not AGE-4 and AGE-5, increased the expression of ICAM-1, B7.1, B7.2, and CD40 and the production of IFN-γ and TNF-α. Consistent with high-affinity binding of AGE-2 and AGE-3 to RAGE in vitro, our results concerning the adhesion molecule expression and cytokine production may support the idea that AGE-2 and AGE-3 function as toxic AGEs and play a central role in the pathophysiological processes associated with AGE formation. As shown in Fig. 6, A and B, the actions of AGE-2 and AGE-3 might be independent of the endogenous IFN-γ and TNF-α. IL-18 and IL-12, as well as IFN-γ and TNF-α, are associated with accelerated atherosclerosis (Gerdes et al., 2002; Wen et al., 2006). Recently, we reported that IL-18 induced the production of IFN-γ and TNF-α in PBMC (Takahashi et al., 2003). We found that the levels of IL-18 and IL-12, monocyte-derived cytokines, were under the detection limit (10 pg/ml) in the presence of AGE-2 and AGE-3 (data not shown). Anti-IL-18 and anti-IL-12 Abs had no effect on the AGE-2- and AGE-3-induced expression of ICAM-1, B7.1, B7.2, and CD40 and production of IFN-γ and TNF-α (data not shown), indicating that the actions of AGE-2 and AGE-3 might be also independent of IL-18 and IL-12. As shown in Fig. 7, the enhanced production of IFN-γ and TNF-α by AGE-2 and AGE-3 may require the enhancement of cell-to-cell interaction between monocytes and T cells through the induction of plural adhesion molecule expression on monocytes.
As shown in Figs. 8 and 9A, we found, for the first time, the up-regulation of RAGE expression on monocytes by AGE-2 and AGE-3 using flow cytometry and immunocytochemistry. Moreover, we confirmed the up-regulation of RAGE protein induced by AGE-2 and AGE-3 using Western blot analysis (Fig. 9B). It is reported that the interaction of AGE-2 with RAGE alters intracellular signaling, gene expression, and the release of proinflammatory molecules in monocytes and macrophages (Sato et al., 2006). AGEs are reported to induce the activation of NF-κB in monocytes via RAGE (Hofmann et al., 1999). In endothelial cells, the activation of NF-κB induces the expression of cytokines, including IL-1 and transforming growth factor-β, and adhesion molecules, including ICAM-1 and vascular cell adhesion molecule-1 (Miyata et al., 1996). The interaction of AGEs and RAGE is reported to result in the up-regulation of RAGE through the activation of NF-κB in monocytes (Li and Schmidt, 1997). Consistent with this finding, pharmacological experiments using the inhibitors of NF-κB and p38 MAPK suggested that the activation of NF-κB and p38 MAPK might be in part involved in the effects of AGE-2 and AGE-3 on adhesion molecule expression and cytokine production (Fig. 10), although it is not excluded that other pathways may be involved in the actions of AGEs. Further studies on the signal pathway leading to the up-regulation of RAGE and adhesion molecule expression should be performed.
In conclusion, among AGEs, AGE-2 and AGE-3 selectively induced the expression of ICAM-1, B7.1, B7.2, and CD40, and the production of IFN-γ and TNF-α. Although adhesion molecule expression on monocytes was independent of endogenous cytokine production in the presence of AGE-2 and AGE-3, the induction of IFN-γ and TNF-α depended on the engagement of monocytes and T cells through the up-regulation of ICAM-1, B7.1, B7.2, and CD40. The actions of AGE-2 and AGE-3 might be through the stimulation and up-regulation of RAGE and the activation of NF-κB and p38 MAPK in monocytes. Together with these results, toxic AGE-dependent responses, including the enhancement of adhesion molecule expression on monocytes, may partially contribute to facilitation of the development of atherosclerotic plaques in diabetes.
We thank Miyuki Shiotani for technical assistance.
This work was supported in part by the Japan Society for the Promotion of Science [Grants 18590509, 20590539, 17659159, 19659061]; and by the Takeda Science Foundation.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: AGE, advanced glycation end product; RAGE, Receptor for advanced glycation end product; MAPK, mitogen-activated protein kinase; NF-κB, nuclear factor-κB; sRAGE, soluble form of receptor for advanced glycation end product; ICAM, intercellular adhesion molecule; IFN, interferon; TNF, tumor necrosis factor; PBMC, peripheral blood mononuclear cells; Ab, antibody; SN50, H2N-Ala-Ala-Val-Ala-Leu-Leu-Pro-Ala-Val-Leu-Leu-Ala-Leu-Leu-Ala-Pro-Val-Gln-Arg-Lys-Arg-Gln-Lys-Leu-Met-Pro-OH; SB203580, 4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyridyl)1H-imidazole; BSA, bovine serum albumin; FITC, fluorescein isothiocyanate; mAb, monoclonal antibody; ELISA, enzyme-linked immunosorbent assay; IL, interleukin.
- Received January 7, 2009.
- Accepted April 17, 2009.
- The American Society for Pharmacology and Experimental Therapeutics