**4.3 RAGE and obesity**

### **A. RAGE, adiposity and atherosclerosis in mouse model**

Recent reports suggest that RAGE could be involved in progression of obesity. Recent study in humans shows RAGE mRNA expression in subcutaneous adipose tissues 88. Although this study does not delineate which cells in adipose tissue express RAGE, our current animal study shows RAGE expression in adipocyte as wells as endothelial cells in adipose tissues 16. We have shown by using apo E/RAGE double knockout mice that progression of atherosclerosis is closely associated with RAGE-regulated adiposity in non-diabetic conditions 16. As shown in Figure 3, apoE-/-RAGE-/- mice fed either with standard or atherogenic diet exhibited significantly decreased atherosclerotic plaque area in aorta as compared with apoE-/-RAGE+/+ mice. Importantly, apoE-/-RAGE-/- mice also exhibited significantly less body weight, epididymal fat weight and epididymal adipocyte size than apoE-/-RAGE+/+ mice at 20-weeks of age (Figure 4). Decreased body weight, epididymal fat weight, and adipocyte size are associated with higher plasma adiponectin levels and decreased atherosclerosis progression. RAGE is involved in adiposity even in apo E+/+ genetic background. At 20-weeks of age, epididymal adipocyte size of RAGE-/- mice was significantly smaller than that of RAGE+/+ mice (data not shown).

under the conditions of oxidative stress, and this may be important for the ability of ROS to mediate insulin resistance. RAGE has a short cytosolic portion that contains 43 amino acids 72. So far, adaptors and/or scaffold proteins that interact with the cytosolic tail of RAGE has barely been identified. The RAGE mutant lacking the 43-residue C-terminal tail fails to activate NF-κB, and expression of the mutant receptor results in a dominant negative effect against RAGE-mediated production of proinflammatory cytokines from macrophages 56, 57.

Fig. 2. RAGE and insulin signaling. RAGE is known to activate JNK pathway, which could phosphorylate serine-residue of insulin receptor substrate (IRS) and inhibit its activity. RAGE mediated generation of reactive oxygen spices (ROS) may alternatively influence

Recent reports suggest that RAGE could be involved in progression of obesity. Recent study in humans shows RAGE mRNA expression in subcutaneous adipose tissues 88. Although this study does not delineate which cells in adipose tissue express RAGE, our current animal study shows RAGE expression in adipocyte as wells as endothelial cells in adipose tissues 16. We have shown by using apo E/RAGE double knockout mice that progression of atherosclerosis is closely associated with RAGE-regulated adiposity in non-diabetic conditions 16. As shown in Figure 3, apoE-/-RAGE-/- mice fed either with standard or atherogenic diet exhibited significantly decreased atherosclerotic plaque area in aorta as compared with apoE-/-RAGE+/+ mice. Importantly, apoE-/-RAGE-/- mice also exhibited significantly less body weight, epididymal fat weight and epididymal adipocyte size than apoE-/-RAGE+/+ mice at 20-weeks of age (Figure 4). Decreased body weight, epididymal fat weight, and adipocyte size are associated with higher plasma adiponectin levels and decreased atherosclerosis progression. RAGE is involved in adiposity even in apo E+/+ genetic background. At 20-weeks of age, epididymal adipocyte size of RAGE-/- mice was

insulin signaing.

**4.3 RAGE and obesity** 

**A. RAGE, adiposity and atherosclerosis in mouse model** 

significantly smaller than that of RAGE+/+ mice (data not shown).

Fig. 3. RAGE deficiency suppresses atherosclerotic progression in apoE deficient mice. Representative aortas from apoE-/-RAGE+/+ and apoE-/-RAGE-/- mice (20-weeks old) fed with atherogenic diet were shown in left panel. Right panel summarizes the quantitative analyses. Plaque area was represented as percentages of the total plaque area. Columns represent mean ± standard deviation. Black columns represent apoE-/-RAGE+/+ mice, and grey columns, apoE-/-RAGE-/- mice. P values were analyzed by Student's unpaired t-test. Reproduced from ref 16.

### **B. Roles of inflammatory cells?**

RAGE is also known to play fundamental role in functions of inflammatory cells 61, 89, 90, raising an intriguing possibility that RAGE's function on adiposity may be mediated through its function in inflammatory cells infiltrated in adipose tissues. In our study in apoE-/- genetic background fed with atherogenic diet, numbers of Mac-3-positive inflammatory cells infiltrated in the epididymal adipose tissues of RAGE+/+apoE-/- mice and RAGE-/-apoE-/- did not show significant differences, and crown-like structure were barely detected in epididymal adipose tissue in both groups even at 20-week of age. In standard diet-fed mice, even though the adiposity was significantly different between RAGE+/+apoE- /- and RAGE-/-apoE-/- mice, crown-like structure were not detected in epididymal adipose tissues in both groups even at 20-week of age. Further in apoE+/+ genetic background at 10 week of age when significantly different pattern of gene expression was observed between WT and RAGE-/- mice, no marked differences in expressions of macrophage markers were observed as analyzed by gene microarray. At that age, macrophage infiltration in adipose tissues is also reported to be scant 91. Thus, it appears infeasible to RAGE acting primarily at inflammatory cells at least in early phase of adiposity, while RAGE expressed in endothelial cells or adipocyte might play fundamental roles.

### **C. RAGE-regulated genes in adipose tissue: gene chip analysis**

To explore potential mechanisms underlying RAGE-regulation of adiposity, mRNA expression profile in epididymal adipose tissue was compared between RAGE+/+ and RAGE-/- mice using Affymetrix GeneChip Mouse Genome 430 2.0. We isolated total RNA from epididymal adipose tissue at 10-weeks of age, at which phenotypic change in adipocyte size was not observed. Using 3 μg of total RNA, 59.8% and 61.4% of 45,037 genes were revealed to be present in RAGE+/+ and RAGE-/- adipose tissue, respectively. Comparison analysis of the genes (RAGE+/+ adipose tissue as base line) revealed that 10.3% of the total genes were decreased, while 11.7% increased in RAGE-/- adipose tissue. As compared with RAGE+/+ adipose tissue, 623 genes were downregulated to less than a half, and 2,470 genes upregulated more than 2 fold in RAGE-/- adipose tissue.

Fig. 4. RAGE deficiency is associated with decreased body weight, epididymal fat weight and adipocyte size in apolipoprotein E (apoE)-deficient genetic background. (A) Comparisons of body weight between apoE-/-RAGE+/+ and apoE-/-RAGE-/- mice fed with standard or atherogenic diet. (B) Comparisons of epididymal fat weight between apoE-/- RAGE+/+ and apoE-/-RAGE-/- mice fed with standard or atherogenic diet. (C) Comparisons of adipocyte size in epididymal adipose tissues. Columns represent mean ± standard deviation. P values were analyzed by Student's t-test. Modified from ref 16.

### **D. RAGE-regulated genes in adipose tissue: ontology analysis**

To mine specific group of genes involved in adiposity regulated by RAGE, gene ontology analyses were performed. Downregulated genes in RAGE-/- adipose tissue were significantly accumulated in the ontology terms of metabolic process including acetyl-CoA biosynthetic process, neutral lipid biosynthetic process, pyruvate metabolic process, gluconeogenesis, glycogen biosynthetic process, and NADPH regeneration. Interestingly, genes involved in fat cell differentiation were also identified to be accumulated as down-regulated in RAGE-/ adipose tissue. Ontology terms of glucose transport and neutral amino acid transport were also significantly extracted as downregulated in RAGE-/- adipose tissue. Insulin receptor signaling pathway was a highly significant ontology term downregulated in RAGE-/- adipose tissue. On the contrary, many of the genes upregulated in RAGE-/- adipose tissue were

Fig. 4. RAGE deficiency is associated with decreased body weight, epididymal fat weight

Comparisons of body weight between apoE-/-RAGE+/+ and apoE-/-RAGE-/- mice fed with standard or atherogenic diet. (B) Comparisons of epididymal fat weight between apoE-/- RAGE+/+ and apoE-/-RAGE-/- mice fed with standard or atherogenic diet. (C) Comparisons of adipocyte size in epididymal adipose tissues. Columns represent mean ± standard

To mine specific group of genes involved in adiposity regulated by RAGE, gene ontology analyses were performed. Downregulated genes in RAGE-/- adipose tissue were significantly accumulated in the ontology terms of metabolic process including acetyl-CoA biosynthetic process, neutral lipid biosynthetic process, pyruvate metabolic process, gluconeogenesis, glycogen biosynthetic process, and NADPH regeneration. Interestingly, genes involved in fat cell differentiation were also identified to be accumulated as down-regulated in RAGE-/ adipose tissue. Ontology terms of glucose transport and neutral amino acid transport were also significantly extracted as downregulated in RAGE-/- adipose tissue. Insulin receptor signaling pathway was a highly significant ontology term downregulated in RAGE-/- adipose tissue. On the contrary, many of the genes upregulated in RAGE-/- adipose tissue were

and adipocyte size in apolipoprotein E (apoE)-deficient genetic background. (A)

deviation. P values were analyzed by Student's t-test. Modified from ref 16.

**D. RAGE-regulated genes in adipose tissue: ontology analysis** 

accumulated in ontology terms including cell adhesion, endocytosis, T cell activation, prostaglandin biosynthesis, protein binding, protein folding, processing and glycoprotein biosynthetic process, many of which are known be associated with cellular mechanisms for inflammation and defensive process. Nitrogen compound metabolic process, including amino acid metabolic process, was also identified to be a significant ontology term upregulated in RAGE-/-. Interestingly, upregulated genes in RAGE-/- tissue were also significantly accumulated in ontology term for cell redox homeostasis process.

## **E. RAGE-regulated genes in adipose tissue: pathway analysis**

To further identify potential pathways involved in RAGE-regulation of adiposity, KEGG pathway analyses were performed (Table 1). In accordance with the ontology analyses, insulin signaling pathway, pyruvate metabolism, fatty acid biosynthesis and gluconeogenesis were identified to be downregulated pathways in RAGE-/- adipose tissue. PPAR signaling and adipocytokine signaling were also identified to be downregulated in RAGE-/- adipose tissue. Similar to gene ontology analyses, inflammatory pathways including cell adhesion molecules and leukocyte transendothelial migration were the significant pathways upregulated in RAGE-/- mice. Pathways including amino acid metabolic pathways, nitrogen metabolism, glycan biosynthesis, structure and degradation were the pathways significantly upregulated in RAGE-/- adipose tissues.


Table 1. Pathway analyses of the genes differentially expressed in WT vs. RAGE-/ epididymal adipose tissue.


Table 1. Pathway analyses of the genes differentially expressed in WT vs. RAGE-/ epididymal adipose tissue (continuation).

### **F. RAGE-regulated genes in adipose tissue: real time RT-PCR confirmation**

Adipogenesis related genes including, lipin 1, peroxisome proliferator-activated receptor (PPAR)-γ, adipose differentiation related protein, were shown to be downregulated in RAGE-/- mice. Fatty acid binding protein 5, 1-acylglycerol-3-phosphate O-acyltransferase 2, diacylglycerol O-acyltransferase 2, monoacylglycerol O-acyltransferase 1, acetoacetyl-CoA synthetase, acetyl-coenzyme A carboxylase α were downregulated in RAGE-/- adipose tissue, which could be an essential mechanisms for decreased adiposity in RAGE-/- mice. In insulin signaling, phosphatidylinositol 3-kinase (p85α), adaptor protein with pleckstrin homology and src (APS), sorbin and SH3 domain containing 1 (CAP), insulin receptor substrate (IRS) 1 and 3, thymoma viral proto-oncogene 2 / similar to serine/threonine kinase (Akt), Protein phosphatase 1 regulatory (inhibitor) subunit 3C, facilitated glucose

Arginine and proline metabolism 6/33 0.0008 Glycine, serine and threonine metabolism 7/47 0.0010 Glycerolipid metabolism 5/41 0.0128 N-Glycan degradation 3/15 0.0136 Cyanoamino acid metabolism 2/6 0.0163 One carbon pool by folate 3/16 0.0164 Glycosphingolipid biosynthesis - ganglioseries 3/16 0.0164 Polyunsaturated fatty acid biosynthesis 3/17 0.0194 Ether lipid metabolism 4/32 0.0234 Cell adhesion molecules (CAMs) 10/147 0.0301 Prostate cancer 7/88 0.0317 Nitrogen metabolism 3/21 0.0343

Glycan structures - biosynthesis 1 8/114 0.0428

Arginine and proline metabolism 4/33 0.0053 Polyunsaturated fatty acid biosynthesis 3/17 0.0054 Glycerolipid metabolism 4/41 0.0114 Leukocyte transendothelial migration 7/115 0.0118 Glutathione metabolism 4/42 0.0124 Cell adhesion molecules (CAMs) 8/147 0.0137 O-Glycan biosynthesis 3/27 0.0199 Thyroid cancer 3/28 0.0219 Glyoxylate and dicarboxylate metabolism 2/14 0.0358 Glycan structures - biosynthesis 1 6/114 0.0364 One carbon pool by folate 2/16 0.0459 Pantothenate and CoA biosynthesis 2/16 0.0459

Table 1. Pathway analyses of the genes differentially expressed in WT vs. RAGE-/-

Adipogenesis related genes including, lipin 1, peroxisome proliferator-activated receptor (PPAR)-γ, adipose differentiation related protein, were shown to be downregulated in RAGE-/- mice. Fatty acid binding protein 5, 1-acylglycerol-3-phosphate O-acyltransferase 2, diacylglycerol O-acyltransferase 2, monoacylglycerol O-acyltransferase 1, acetoacetyl-CoA synthetase, acetyl-coenzyme A carboxylase α were downregulated in RAGE-/- adipose tissue, which could be an essential mechanisms for decreased adiposity in RAGE-/- mice. In insulin signaling, phosphatidylinositol 3-kinase (p85α), adaptor protein with pleckstrin homology and src (APS), sorbin and SH3 domain containing 1 (CAP), insulin receptor substrate (IRS) 1 and 3, thymoma viral proto-oncogene 2 / similar to serine/threonine kinase (Akt), Protein phosphatase 1 regulatory (inhibitor) subunit 3C, facilitated glucose

**F. RAGE-regulated genes in adipose tissue: real time RT-PCR confirmation** 

3/21 0.0343

WT<RAGE-/- (>= 2 fold)

biosynthesis

WT<RAGE-/- (>= 3 fold)

epididymal adipose tissue (continuation).

Glycosylphosphatidylinositol(GPI)-anchor

Fig. 5. Gene microarray, Ontology and KEGG pathway analyses suggest that insulin signaling and adipocyte differentiation are the potential pathways regulated by RAGE. (A) Figure summarizes the results of Ontology and KEGG pathway analyses. Genes suppressed in RAGE-/- adipose tissue were described in black circles. (B) Changes in mRNA expression obtained by gene microarray analyses were confirmed by real-time quantitative RT-PCR analyses. All changes in gene expression were statistically significance (p<0.05, Student's ttest). IRS-1: insulin receptor substrate 1, PI3K: phosphatidylinositol 3-kinase (p85α), AKT: thymoma viral proto-oncogene 2 / similar to serine/threonine kinase, APS: adaptor protein with pleckstrin homology and src, CAP: sorbin and SH3 domain containing 1, Glut4: facilitated glucose transporter member 4, FABP4: fatty acid binding protein 4, PPAR-γ: peroxisome proliferator-activated receptor, Steap4: six-transmembrane epithelial antigen of prostate 4, ACC: acetyl-coenzyme A carboxylase α, Gldx: glutaredoxin.

transporter member 4 (Glut 4) were identified to be downregulated in RAGE-/- adipose tissue. Figure 5A shows genes specifically suppressed in RAGE-/- adipose tissue (closed circles) in insulin signaling and adipocyte differentiation pathways. Real-time quantitative RT-PCR analyses confirmed the genes in the pathways were indeed down-regulated in RAGE-/- adipose tissue (Figure 5B). These results altogether suggest direct role of RAGE in adiposity. Although in which cell types RAGE is principally working, insulin signaling and adipocyte signaling pathway in adipose tissue appear to play important part in RAGE regulation of adiposity.
