A neutralizing antibody against receptor for advanced glycation end products (RAGE) reduces atherosclerosis in uremic mice

Atherosclerosis 201 (2008) 274–280

A neutralizing antibody against receptor for advanced glycation end products (RAGE) reduces atherosclerosis in uremic mice Susanne Bro a,b,∗ , Allan Flyvbjerg d , Christoph J. Binder e , Christian A. Bang b , Larry Denner f , Klaus Olgaard a , Lars B. Nielsen b,c a

e

Department of Nephrology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark b Department of Clinical Biochemistry, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark c Department of Biomedical Sciences, University of Copenhagen, Copenhagen, Denmark d Medical Department M (Diabetes and Endocrinology), The Medical Research Laboratories, Clinical Institute, Aarhus University Hospital, Aarhus, Denmark Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Vienna, Austria f Stark Diabetes Center, University of Texas, Medical Branch, Galveston, TX, USA Received 11 July 2007; received in revised form 18 November 2007; accepted 31 January 2008 Available online 16 February 2008

Abstract Chronic renal failure markedly accelerates atherogenesis in apolipoprotein E-deficient (apoE−/−) mice. To study the putative role of receptor for advanced glycation end products (RAGE) in development of uremic atherosclerosis, apoE−/− mice received intraperitoneal injections thrice weekly of a neutralizing murine RAGE-antibody (RAGE-ab) (n = 21) or an isotype-matched control antibody (placebo-ab) (n = 23). Treatment was started 4 weeks after surgical 5/6 nephrectomy in 16 weeks old mice and continued for 12 weeks. The RAGE-ab did not affect blood pressure, plasma cholesterol or measures of uremia. However, the aortic plaque area fraction was reduced by 59% in RAGE-ab compared with placebo-ab-treated mice (0.016 ± 0.002 versus 0.039 ± 0.005, P < 0.001). In plasma, the RAGE-ab reduced concentrations of oxidized phospholipid neo-epitopes in plasma as detected by the specific monoclonal antibody EO6 (P < 0.05) and titers of IgG antibodies against oxidized low-density lipoprotein (P < 0.001). In the aorta of treated mice, the RAGE-ab did not affect the mRNA expression of eight selected genes associated with inflammation. The results suggest that blockade of RAGE reduces the proatherogenic effects of uremia, possibly through a systemic decrease in oxidative stress. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Antibodies against oxidized low density lipoprotein; Atherosclerosis; EO6; ICAM-1; Oxidized phospholipid neo-epitopes; Receptor for advanced glycation end products; Renal failure; VCAM-1

1. Introduction The risk of cardiovascular disease in patients with renal failure is far greater than in the general population, even when kidney function is only moderately reduced [1]. Although highly prevalent, the increased risk cannot be explained by the classical risk factors alone [2]. Ure∗ Corresponding author at: Department of Nephrology P 2131, Rigshospitalet, University of Copenhagen, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. Tel.: +45 35454157; fax: +45 35452524. E-mail address: [email protected] (S. Bro).

0021-9150/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2008.01.015

mia accelerates atherosclerosis in apolipoprotein E-deficient (apoE−/−) mice [3–5]. The aortic lesions are characterized by intimal accumulation of macrophage-derived foam cells and cholesterol esters [6]. The lesions in uremic mice also display accumulation of nitrotyrosine (a marker of reactive oxygen species (ROS) modification of proteins) [3,4], and increased expression of intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) [6]. Moreover, plasma concentrations of antibodies recognizing epitopes in oxidized low density lipoprotein (OxLDL) and soluble (s) ICAM-1 and VCAM-1 are increased in uremic apoE−/− mice [7]. Both oxidative stress and

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

vascular inflammation promote atherogenesis [8]. Hence, the increased oxidative stress and inflammation may be pivotal in the acceleration of atherosclerosis seen in uremia. In accordance with this idea, inhibition of the renin–angiotensin system reduced plasma concentrations of antibodies against OxLDL, reduced aortic expression of VCAM-1, and attenuated atherogenesis in uremic mice [7]. Renal dysfunction causes increased plasma concentrations of advanced glycation end products (AGEs) [9]. AGEs are formed by non-enzymatic glycation in a series of biochemical reactions between glucose and reactive carbonyl compounds, proteins, lipids or nucleic acids [10,11]. AGEs bind to and activate RAGE, the receptor for AGEs [12]. Other RAGE ligands include S100/calgranulins, amphoterin, and Mac-1 [13,14]. RAGE is expressed in cultured endothelial cells, monocytes/macrophages, and smooth muscle cells [12]; all three cell types participate in atherosclerotic lesion formation. In vitro, ligand interaction with RAGE leads to oxidative changes including increased expression of NAD(P)H oxidase and formation of ROS [12]. RAGE stimulation may also activate NF-␬B and as such increase the expression of VCAM-1 and other proinflammatory molecules [12]. Upregulation of RAGE expression has been demonstrated in blood vessels from diabetic patients [15]. An important role of RAGE in atherogenesis is suggested by suppressed development of atherosclerosis in diabetic apoE−/− mice upon blockade of RAGE with a soluble extra-cellular ligand-binding domain of RAGE [16,17]. RAGE blockade has also been achieved using a neutralizing anti-RAGE-antibody (RAGE-ab) which reduced the adverse effects of diabetes on kidney morphology and function [18,19]. Uremic individuals without diabetes also display upregulation of vascular RAGE expression [20]. We hypothesized that signaling through RAGE might increase oxidative stress and/or inflammation in uremia and as such promote development of uremic atherosclerosis. To test this idea, we examined the effect of RAGE blockade with a neutralizing antibody on formation of aortic atherosclerosis in uremic apoE−/− mice.

2. Materials and methods 2.1. Animals Male apoE−/− mice on a C57BL/6 background (Taconic M&B, Denmark) and male C57BL/6 mice (used to study the tissue expression pattern of RAGE mRNA) were kept on a 12-h light/dark cycle in a temperature-controlled room at 21–23 ◦ C with free access to water and a standard mouse diet (Altromin 1314, Altromin, Lage, Germany). The apoE−/− mice that were made uremic and the control apoE−/− mice were litter mates. The experiments were performed according to the principles stated in the Danish law on ani-

275

mal experiments, and approved by the Animal Experiments Inspectorate, Ministry of Justice, Denmark. 2.2. Surgical procedures Renal failure was induced by a two-step surgical procedure [6]. First, at 10 weeks of age the upper and lower poles of the right kidney were resected. Second, 2 weeks later the entire left kidney was removed. Anesthesia was achieved with a mixture of fentanyl 0.079 mg/ml, fluanisone 2.5 mg/ml, and midazolam 1.25 mg/ml (Hypnorm/Dormicum) (8–10 ␮l/g body weight, subcutaneously). Buprenorphine (0.1 ␮g/g body weight, subcutaneously) was used as analgesia after surgery. 2.3. Experimental protocol To study the involvement of RAGE in development of uremic atherosclerosis, uremic apoE−/− mice were treated with a neutralizing monoclonal murine RAGE-antibody (RAGE-ab) (n = 21) or an isotype-matched, control antibody (placebo-ab) (n = 23). The RAGE-ab and placebo-ab were administered intraperitoneally as an initial bolus of 300 ␮g/mouse followed by injections of 100 ␮g/mouse three times weekly; antibodies were dissolved in 0.154 mol/l NaCl and injected in a volume of 0.5 ml. Treatment was started 4 weeks after the second operation in 16 weeks old mice and continued for 12 weeks. Control mice did not undergo surgery, and received no treatment (n = 4). The preparation and characterization of the monoclonal IgG3␬ RAGE-ab have been previously described [18,19]. A NF-␬B reporter-gene assay using Nε -(carboxymethyl)lysine-modified human serum albumin as a ligand for RAGE on human THP-1 monocytic leukemia cells documented the neutralizing activity of the RAGE-ab [18,19,21]. The placebo-ab was a monoclonal anti-diphtheria IgG3␬ antibody produced by the cell line HYB 123-1 and purified by Protein G affinity chromatography (obtained from Dr. Klaus Koch, University of Southern Denmark). 2.4. Quantification of atherosclerotic lesions After 12 weeks of treatment, each mouse was anesthetized and perfused with 0.9% NaCl (0 ◦ C) through the left ventricle. The aorta from the heart to the iliac arteries was removed, freed of connective tissue under a dissection microscope, opened longitudinally, and placed between a microscope slide and a cover slip. The intimal surface was scanned with an AGFA Snapscan e50 flatbed scanner (AGFA-Gevaert, Glostrup, Denmark) before storage of aortas at −80 ◦ C for subsequent RNA analyses. Aortic total area and lesion area were determined by digital image analysis with the Multi-Analyst/PC Version 1.1 software from BioRad Laboratories (Hercules, CA, USA). The inter-observer and intra-observer variability were 8.9% and 5.3%, respectively.

276

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

2.5. Blood pressure

cent substrate LumiPhos as described before [23]. Thus, this assay determines the total OxPL-EO6 reactivity in plasma.

Systolic blood pressure (BP) was measured after 10 weeks of antibody treatment with a tail-cuff system (BP 2000; Visitech Systems, Apex, NC, USA) as described previously [3,7]. 2.6. Plasma biochemistry Blood from the retro-orbital venous plexus was collected during anesthesia at the end of the study. Whole blood hemoglobin was determined using an OSM3 hemoximeter (Radiometer, Denmark). Plasma urea, creatinine, total calcium, and phosphate were measured with a Modular P Automatic analyzer (Roche A/S, Hvidovre, Denmark). Plasma total cholesterol was assayed with an enzymatic kit [22] and plasma concentrations of sICAM-1 and sVCAM-1 were measured with monoclonal antibody-based sandwich ELISA kits (Catalog Nos. MVC00 and MIC100, R&D Systems Europe, Abingdon, Oxon, UK). Plasma titers of antibodies reacting with malondialdehyde-modified (MDA)–LDL and Cu2+ -oxidized (CuOx)–LDL were determined by chemiluminescent enzyme immunoassays, as previously reported [7]. The presence of oxidized phospholipid (OxPL) neo-epitopes in total plasma was measured using a chemiluminescent ELISA. Plasma samples were diluted 1:100 in PBS containing 0.27 mM EDTA, and the dilutions were coated overnight at 4 ◦ C in microtiter wells. After a washing step, wells were blocked with PBS/EDTA containing 1% BSA for 15 min, followed by another washing step. OxPL epitopes were then detected using a biotinylated mouse antibody EO6, which specifically recognizes the phosphorylcholine moiety of OxPL, followed by alkaline phosphatase conjugated neutravidin and the chemilumines-

2.7. Quantitative real-time PCR Isolation of aortic RNA and real-time PCR quantification of gene transcripts were done with a LightCycler (Roche) as previously described [6]. The primers [product size] were (m-RAGE-55: 5 -cagggtcacagaaaccgg-3 ) and (m-RAGE-35: 5 -attcagctctgcacgttcct-3 ) [214 bp]; (mMCP1-51: 5 -aggtccctgtcatgcttctg-3 ) and (m-MCP1-31: 5 -tctggacccattccttcttg-3 ) [249 bp]; (m-IL1␣-51: 5 -cccgtccttaaagctgtctg-3 ) and (m-IL1␣-31: 5 -aattggaatccaggggaaac-3 ) [161 bp]; (m-IL6-51: 5 -agttgccttcttgggactga-3 ) and (m-IL6-31: 5 -tccacgatttcccagagaac-3 ) [159 bp]; (mTNF␣-51: 5 -agcccccagtctgtatcctt-3 ) and (m-TNF␣-31: 5 -ctccctttgcagaactcagg-3 ) [212 bp]; (m-TGF␤1-51: 5 -ttgcttcagctccacagaga-3 ) and (m-TGF␤1-31: 5 -tggttgtagagggcaaggac-3 ) [183 bp]; (m-RelA-51: 5 -gcgtacacattctggggagt-3 ) and (m-RelA-31: 5 -accgaagcaggagctatcaa-3 ) [179 bp]. The primers for ICAM-1, VCAM-1, and ␤-actin have been reported elsewhere [6]. To evaluate the tissue expression pattern of RAGE mRNA, RNA was isolated from 12 different tissues from 20 weeks old male C57BL/6 mice (n = 10) and used for cDNA synthesis. 2.8. Statistical analyses Data were analysed with Kruskal–Wallis test followed by Mann–Whitney post-tests comparing RAGE-ab and placebo-treated mice (to evaluate the effect of RAGE blockade) and placebo-treated uremic versus control mice (to evaluate the effect of uremia). The post-tests were only performed if Kruskal–Wallis test showed significant differences

Table 1 Effects of uremia and RAGE-ab treatment on body weight, blood pressure, plasma indices of uremia, plasma cholesterol, and soluble adhesion molecules Uremic

n Body weight (g) BP (mm Hg)a Blood haemoglobina Urea Creatinine Phosphate Calcium Ca × P (mmol/l)2 Total cholesterol sICAM-1 (␮g/l) sVCAM-1 (␮g/l)

Controls

Placebo-ab

RAGE-ab

No treatment

23 26.3 ± 0.6 122.8 ± 2.2 7.9 ± 0.3 32.3 ± 2.6 0.029 ± 0.002 2.63 ± 0.22 2.93 ± 0.07 7.49 ± 0.39 15.09 ± 1.63 571.8 ± 62.3 935.3 ± 85.8

21 27.1 ± 0.7 124.6 ± 3.3 7.1 ± 0.4 31.8 ± 1.9 0.031 ± 0.002 2.81 ± 0.17 3.00 ± 0.06 8.34 ± 0.45 14.04 ± 0.88 623.1 ± 64.5 964.3 ± 42.0

4 32.3 ± 1.0b NA NA 8.8 ± 0.6b 0.010 ± 0.001b 2.29 ± 0.15 2.65 ± 0.02 6.07 ± 0.37 9.33 ± 1.26 363.7 ± 107.8 467.6 ± 18.4c

Values are from 16 weeks after 5/6 nephrectomy (uremic) or no surgery (controls). RAGE-ab or placebo-ab was administered from 4 to 16 weeks after 5/6 nephrectomy. Blood and plasma concentrations are in mmol/l unless otherwise indicated. NA: not available; s: soluble. Values are mean ± S.E.M. a BP and blood haemoglobin were only measured in placebo-ab and RAGE-ab-treated mice. b P < 0.01 compared with uremic placebo-ab-treated mice. c P < 0.05 compared with uremic placebo-ab-treated mice.

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

(P < 0.05) between the three groups. Data are presented as mean ± S.E.M., with n indicating the number of mice studied.

3. Results Surgical subtotal 5/6 nephrectomy was used to induce moderate uremia in apoE−/− mice. The uremic mice had reduced body weight, and increased plasma urea and creatinine concentrations (Table 1), as well as markedly increased aortic atherosclerosis (Fig. 1). Previous studies developed a monoclonal RAGE-ab that blocks RAGE and inhibits signaling events elicited by RAGE ligands in vitro [21] and adverse renal effects in diabetic mice in vivo [18,19]. To test the effect of blocking RAGE on development of atherosclerosis in uremic apoE−/− mice, the RAGE-ab or a placebo-ab was injected thrice weekly for 12 weeks. The RAGE-ab did not affect body weight, BP, plasma indices of uremia, or plasma cholesterol (Table 1), but

Fig. 1. Effects of uremia and RAGE-ab treatment on the development of aortic atherosclerosis. Aortic atherosclerosis was measured in uremic apoE−/− mice at 16 weeks after 5/6 nephrectomy. RAGE-ab or placebo-ab was administered during weeks 4–16 after 5/6 nephrectomy. Control apoE−/− mice did not undergo surgery, and received no treatment. (A) En face pictures showing atherosclerotic lesions in aortas from a RAGE-ab and a placeboab-treated uremic mouse. (B) The plaque area fraction was quantitated in each mouse aorta by computerized morphometry. Values are mean + S.E.M.

277

reduced the aortic plaque area fraction from 0.039 ± 0.005 in placebo-treated uremic mice to 0.016 ± 0.002 (P < 0.001) (Fig. 1). The antiatherogenic effect of the RAGE-ab could be mediated by direct effects on the arterial wall or systemic effects due to blockade of RAGE in other organs. Indeed, examination of multiple tissues from wild-type mice showed that RAGE mRNA was widely expressed with the highest expression in the lung, but also clearly detectable in the aorta (Fig. 2A). Importantly, aortic RAGE mRNA expres-

Fig. 2. (A) Tissue expression pattern of RAGE mRNA in non-uremic wildtype mice. RNA was isolated from 12 different tissues from 20-week-old male C57Bl/6 mice (n = 10) and used for cDNA synthesis. For each tissue, cDNA preparations from all animals were pooled before quantification of RAGE mRNA. Messenger RNA expression of RAGE was normalized by ␤-actin. The expression in aorta was set to 1. (B) Effects of uremia and RAGE-ab treatment on aortic expression of RAGE mRNA. Aortic expression of RAGE mRNA was measured in uremic apoE−/− mice at 16 weeks after 5/6 nephrectomy. RAGE-ab or placebo-ab was administered during weeks 4–16 after 5/6 nephrectomy. Messenger RNA expression of RAGE was normalized by ␤-actin. Further, the mRNA expression in uremic mice was normalized by the average of mRNA expression in control apoE−/− mice, which was set at 1. Control mice did not undergo surgery, and received no treatment. Values are mean + S.E.M.

278

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

sion was increased in uremic compared with non-uremic apoE−/− mice (Fig. 2B). Treatment of uremic mice with the RAGE-ab did not affect RAGE mRNA expression (Fig. 2B). Since downstream effects of signaling through RAGE have been reported to involve NF-␬B-dependent upregulation of adhesion molecule expression, aortic VCAM-1 and ICAM1 mRNA expression were compared in RAGE-ab and placebo-ab-treated uremic mice. However, the aortic mRNA expression of VCAM-1 and ICAM-1 were similar in the two groups (Table 2). Although the plasma concentration of sVCAM-1 increased in uremic versus non-uremic apoE−/− mice (Table 1), it was not affected by treatment with the RAGE-ab (Table 1). In addition, the RAGE-ab did not affect the aortic mRNA expression of the inflammatory mediators IL-1␣, IL-6, MCP-1, TGF-␤ or TNF-␣, nor of RelA (P65) (a component of the NF-␬B complex) (Table 2). RAGE signaling has also been reported to increase oxidative stress [12]. We recently reported that uremic apoE−/− mice are characterized by increased oxidative stress as reflected by elevated plasma concentrations of OxPl on circulating apoB and increased plasma titers of antibodies against neo-epitopes in oxidized LDL [7]. In accord with those find-

Table 2 Effects of uremia and RAGE-ab treatment on aortic mRNA expression of genes associated with inflammation Uremic

n ICAM-1 VCAM-1 IL-1␣ IL-6 MCP-1 TGF-␤ TNF-␣ RelA

Controls

Placebo-ab

RAGE-ab

No treatment

10 1.80 ± 0.59 1.92 ± 0.25 1.92 ± 0.10 2.11 ± 1.01 2.84 ± 0.36 2.90 ± 0.96 1.70 ± 0.87 1.53 ± 0.47

12 1.83 ± 0.19 2.03 ± 0.33 2.83 ± 1.00 1.87 ± 0.39 3.70 ± 1.10 3.19 ± 0.88 1.91 ± 0.41 1.81 ± 0.34

4 1.00 ± 0.31 1.00 ± 0.42 1.00 ± 0.72 1.00 ± 0.55 1.00 ± 0.40 1.00 ± 0.09a 1.00 ± 0.77 1.00 ± 0.33

Aortic mRNA expression was measured 16 weeks after 5/6 nephrectomy (uremic) or no surgery (controls). RAGE-ab or placebo-ab was administered from 4 to 16 weeks after 5/6 nephrectomy. Messenger RNA expression was normalized by ␤-actin. Further, the mRNA expression in uremic apoE−/− mice was normalized by the average of mRNA expression in control mice, which was set at 1. Values are mean ± S.E.M. a P < 0.01 compared with uremic placebo-ab-treated mice.

Fig. 3. Effects of uremia and RAGE-ab treatment on formation of antibodies against oxidized LDL and on plasma concentrations of EO6-reactive oxidized phospholipid epitopes. Plasma titers of antibodies reacting with Cu2+ -oxidized (CuOx)–LDL (A), malondialdehyde modified (MDA)–LDL (B), and plasma concentrations of EO6-reactive oxidized phospholipid (OxPL) epitopes (C) were measured in uremic apoE−/− mice at 16 weeks after 5/6 nephrectomy. RAGEab or placebo-ab was administered during weeks 4–16 after 5/6 nephrectomy. Control apoE−/− mice did not undergo surgery, and received no treatment. Antibody titers and EO6/OxPl reactivity were measured in relative light units (RLUs). Antibody titers in uremic mice were normalized to control values. Although not indicated in the figure, all control antibody titers were significantly lower than values in uremic mice. Please note the logarithmic scale. All values are mean + S.E.M.

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

ings, the uremic apoE−/− mice in the present study showed increased plasma titers of IgM and IgG antibodies reacting with MDA–LDL and CuOx–LDL and increased OxPl/EO6 reactivity in plasma, indicating increased oxidative stress (Fig. 3A–C). Treatment with RAGE-ab reduced the titers of IgG antibodies against both MDA–LDL and CuOx–LDL, especially the IgG1 titers (P < 0.0001), but did not affect IgM antibody titers (Fig. 3A and B). Moreover, the OxPl/EO6 reactivity in plasma was also reduced in the RAGE-ab-treated group (P < 0.05) (Fig. 3C).

4. Discussion The present study shows that treatment with a neutralizing RAGE-ab attenuates the development of atherosclerotic lesions in the aortas of uremic mice. Of note, the antiatherosclerotic effect was seen without effects on parameters of uremia, BP, plasma cholesterol, or body weight. Thus, the data suggest that activation of RAGE accelerates the development of uremic atherosclerosis. Uremia is associated with increased plasma markers of oxidative stress both in humans [24] and mice [5,7]. In the current mouse model, uremia caused increased plasma concentrations of EO6-reactive OxPl [7]. Moreover, we previously observed that uremia in apoE−/− mice increases plasma titers of antibodies against neo-epitopes in OxLDL [7] which likely reflect increased OxLDL formation and subsequent activation of adaptive immune responses. Remarkably, treatment with the RAGE-ab, but not with the isotypematched control-ab, reduced both plasma concentrations of EO6-reactive OxPl and the titers of IgG antibodies against OxLDL in uremic apoE−/− mice. Activation of RAGE conveys oxidative stress, i.e., AGE–RAGE interaction stimulates the production of ROS in cultured endothelial cells, monocytes and smooth muscle cells [12]. ROS is presumed to participate in atherogenesis in several ways including the formation of OxLDL [25]. An antioxidative effect of the RAGE-ab could explain the reduced development of atherosclerosis in RAGE-ab-treated uremic apoE−/− mice since OxPl have several proatherogenic effects including endothelial damage and accelerated foam cell formation [26]. It is unknown whether the effect of RAGE blockade reflects a direct effect of RAGE in the arterial wall in addition to a systemic decrease in oxidative stress. Although RAGE mRNA was expressed in the arterial wall and increased in uremic versus non-uremic apoE−/− mice, the tissue expression was >1000-fold lower in the aorta than in the lung. It should be noted, that this number was obtained in non-uremic wildtype mice by analyzing cDNA pools and as such should be interpreted with caution. Nevertheless, there was no effect of treatment with the RAGE-ab on aortic VCAM-1 and ICAM-1 mRNA expression. RAGE signaling activates NF-␬B resulting in increased VCAM-1 expression in cultured cells [12] and the RAGE-ab employed in the present study reduces

279

NF-␬B activation in cultured THP-1 cells [21]. Thus, the present results contrast the findings in both type 1 and type 2 diabetic apoE−/− mice where a soluble extra-cellular ligandbinding domain of RAGE resulted in decreased VCAM-1 protein expression in aorta as judged by immunohistochemical staining [17,27]. The apparent discrepancy might reflect subtle differences in effects of the two modalities used to block RAGE, e.g., that different downstream pathways may be affected depending on the agent used for RAGE blockade. The complexity of the downstream effects of ligand interaction with RAGE is also exemplified by the observations by Valencia et al. [28] that AGE-modified albumin, in contrast to the alternate RAGE ligand S100b, failed to induce an inflammatory response in endothelial cells. Also, it is possible that the effects of RAGE blockade on atherosclerosis in uremic apoE−/− mice depend on other effects than those mediated through NF-␬B. Hence, a previous in vivo study also saw no effect on NF-␬B activation despite effects on renal morphology and function when treating diabetic mice with the RAGE-ab [18]. In addition to the lack of effect on VCAM-1 and ICAM-1 expression, we did not see any effects of RAGE-ab treatment on the mRNA expression of selected inflammatory genes, including the proinflammatory cytokines IL-1␣ and IL-6, MCP-1, TNF-␣, and TGF-␤ that all have been shown to play pivotal roles in progression of atherosclerosis [29]. Thus, the data did not provide support for the idea that the RAGE-ab had major direct antiinflammatory effects in the arterial wall. In conclusion, the data suggest that RAGE blockade reduces the proatherogenic effects of uremia. The mechanism remains elusive, but the current data suggest that the effect mainly may reflect a systemic reduction of oxidative stress rather than direct effects in the arterial wall. Nevertheless, treatments directed toward blocking RAGE or the formation of AGEs are promising candidates to prevent cardiovascular disease in uremia.

Acknowledgments The Danish Medical Research Council, The Danish Heart Foundation, The Danish Kidney Foundation, The Danish Diabetes Association, The Copenhagen Hospital Corporation Research Council, and The Helen and Ejnar Bjoernow Foundation supported this study. We thank Kirsten Bang, Hanne Kjaergaard, Tina Axen, Charlotte Wandel, Karen Rasmussen, Kirsten Hansen and Maria Ozsvar-Kozma for technical assistance at various stages of this project, and Dr. Klaus Koch for producing the placebo antibody.

References [1] Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;351:1296–305.

280

S. Bro et al. / Atherosclerosis 201 (2008) 274–280

[2] Cheung AK, Sarnak MJ, Yan G, et al. Atherosclerotic cardiovascular disease risks in chronic hemodialysis patients. Kidney Int 2000;58:353–62. [3] Bro S, Bentzon JF, Falk E, Andersen CB, Olgaard K, Nielsen LB. Chronic renal failure accelerates atherogenesis in apolipoprotein edeficient mice. J Am Soc Nephrol 2003;14:2466–74. [4] Buzello M, Tornig J, Faulhaber J, Ehmke H, Ritz E, Amann K. The apolipoprotein e knockout mouse: a model documenting accelerated atherogenesis in uremia. J Am Soc Nephrol 2003;14:311–6. [5] Massy ZA, Ivanovski O, Nguyen-Khoa T, et al. Uremia accelerates both atherosclerosis and arterial calcification in apolipoprotein E knockout mice. J Am Soc Nephrol 2005;16:109–16. [6] Bro S, Moeller F, Andersen CB, Olgaard K, Nielsen LB. Increased expression of adhesion molecules in uremic atherosclerosis in apolipoprotein-E-deficient mice. J Am Soc Nephrol 2004;15: 1495–503. [7] Bro S, Binder CJ, Witztum JL, Olgaard K, Nielsen LB. Inhibition of the renin-angiotensin system abolishes the proatherogenic effect of uremia in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2007;27:1080–6. [8] Falk E. Pathogenesis of atherosclerosis. J Am Coll Cardiol 2006; 47:C7–12. [9] Degenhardt TP, Grass L, Reddy S, Thorpe SR, Diamandis EP, Baynes JW. Technical note. The serum concentration of the advanced glycation end-product N epsilon-(carboxymethyl)lysine is increased in uremia. Kidney Int 1997;52:1064–7. [10] Brownlee M, Cerami A, Vlassara H. Advanced glycosylation end products in tissue and the biochemical basis of diabetic complications. N Engl J Med 1988;318:1315–21. [11] Miyata T, van Ypersele de S, Kurokawa K, Baynes JW. Alterations in nonenzymatic biochemistry in uremia: origin and significance of “carbonyl stress” in long-term uremic complications. Kidney Int 1999;55: 389–99. [12] Goldin A, Beckman JA, Schmidt AM, Creager MA. Advanced glycation end products: sparking the development of diabetic vascular injury. Circulation 2006;114:597–605. [13] Schmidt AM, Yan SD, Yan SF, Stern DM. The biology of the receptor for advanced glycation end products and its ligands. Biochim Biophys Acta 2000;1498:99–111. [14] Chavakis T, Bierhaus A, Al Fakhri N, et al. The pattern recognition receptor (RAGE) is a counterreceptor for leukocyte integrins: a novel pathway for inflammatory cell recruitment. J Exp Med 2003;198:1507–15. [15] Ritthaler U, Deng Y, Zhang Y, et al. Expression of receptors for advanced glycation end products in peripheral occlusive vascular disease. Am J Pathol 1995;146:688–94.

[16] Park L, Raman KG, Lee KJ, et al. Suppression of accelerated diabetic atherosclerosis by the soluble receptor for advanced glycation endproducts. Nat Med 1998;4:1025–31. [17] Wendt T, Harja E, Bucciarelli L, et al. RAGE modulates vascular inflammation and atherosclerosis in a murine model of type 2 diabetes. Atherosclerosis 2006;185:70–7. [18] Flyvbjerg A, Denner L, Schrijvers BF, et al. Long-term renal effects of a neutralizing RAGE antibody in obese type 2 diabetic mice. Diabetes 2004;53:166–72. [19] Jensen LJ, Denner L, Schrijvers BF, Tilton RG, Rasch R, Flyvbjerg A. Renal effects of a neutralising RAGE-antibody in long-term streptozotocin-diabetic mice. J Endocrinol 2006;188:493– 501. [20] Greten J, Kreis I, Wiesel K, et al. Receptors for advance glycation end-products (AGE)—expression by endothelial cells in non-diabetic uraemic patients. Nephrol Dial Transplant 1996;11:786–90. [21] Yeh CH, Sturgis L, Haidacher J, et al. Requirement for p38 and p44/p42 mitogen-activated protein kinases in RAGE-mediated nuclear factorkappaB transcriptional activation and cytokine secretion. Diabetes 2001;50:1495–504. [22] Bartels ED, Lauritsen M, Nielsen LB. Hepatic expression of microsomal triglyceride transfer protein and in vivo secretion of triglyceride-rich lipoproteins are increased in obese diabetic mice. Diabetes 2002;51:1233–9. [23] Reardon CA, Miller ER, Blachowicz L, et al. Autoantibodies to OxLDL fail to alter the clearance of injected OxLDL in apolipoprotein Edeficient mice. J Lipid Res 2004;45:1347–54. [24] Himmelfarb J, Stenvinkel P, Ikizler TA, Hakim RM. The elephant in uremia: oxidant stress as a unifying concept of cardiovascular disease in uremia. Kidney Int 2002;62:1524–38. [25] Cathcart MK. Regulation of superoxide anion production by NADPH oxidase in monocytes/macrophages: contributions to atherosclerosis. Arterioscler Thromb Vasc Biol 2004;24:23–8. [26] Nielsen LB. Atherogenecity of lipoprotein(a) and oxidized low density lipoprotein: insight from in vivo studies of arterial wall influx, degradation and efflux. Atherosclerosis 1999;143:229–43. [27] Bucciarelli LG, Wendt T, Qu W, et al. RAGE blockade stabilizes established atherosclerosis in diabetic apolipoprotein E-null mice. Circulation 2002;106:2827–35. [28] Valencia JV, Mone M, Koehne C, Rediske J, Hughes TE. Binding of receptor for advanced glycation end products (RAGE) ligands is not sufficient to induce inflammatory signals: lack of activity of endotoxinfree albumin-derived advanced glycation end products. Diabetologia 2004;47:844–52. [29] Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005;352:1685–95.