Increased Adiponectin Secretion by Highly Purified Eicosapentaenoic Acid in Rodent Models of Obesity and Human Obese Subjects

【摘要】  Objectives— Fish oil rich in n-3 polyunsaturated fatty acids (PUFAs) or n-3 PUFAs have been shown to reduce the incidence of coronary heart disease. Here we investigated the effect of highly purified eicosapentaenoic acid (EPA) on production of adiponectin, the only established antiatherogenic and antiinflammatory adipocytokine, in rodent models of obesity and human obese subjects.

Methods and Results— We demonstrated that EPA increases adiponectin secretion in genetically obese ob/ob mice and high-fat diet–induced obese mice. In the in vitro coculture of adipocytes and macrophages, EPA reversed the coculture-induced decrease in adiponectin secretion at least in part through downregulation of tumor necrosis factor- in macrophages. We also showed significant increase in plasma adiponectin concentrations in human obese subjects after a 3-month treatment with EPA (1.8 g daily). Multivariate regression analysis revealed that EPA treatment is the only independent determinant of plasma adiponectin concentrations.

Conclusion— This study demonstrates that EPA increases adiponectin secretion in rodent models of obesity and human obese subjects, possibly through the improvement of the inflammatory changes in obese adipose tissue. Because EPA has reduced the risk of major coronary events in a large-scale, prospective, randomized clinical trial, this study provides important insight into its therapeutic implication in obesity-related metabolic sequelae.

Here we show that highly purified EPA, the only class of n-3 PUFAs used clinically to treat hyperlipidemia, increases adiponectin secretion in rodent models of obesity and human obese subjects possibly through the improvement of adipose tissue inflammation, thereby providing important insight into its therapeutic implication in obesity-related metabolic sequelae.

【关键词】  adipocytes adiponectin EPA macrophages obesity

Introduction

The adipose tissue has a high capacity to secrete many biologically active substances (or adipocytokines) such as leptin and tumor necrosis factor- (TNF ). 1 Dysregulation of pro- and antiinflammatory adipocytokine production is associated with the metabolic syndrome, suggesting that inflammatory changes within obese adipose tissue may critically contribute to the development of many aspects of the metabolic syndrome and results in diabetes and atherosclerosis. Among numerous adipocytokines, adiponectin is unique in that it is the only established adipocytokine with antiatherogenic and antiinflammatory properties. 2,3 It also increases tissue fat oxidation, leading to reduced levels of fatty acids (FAs) and tissue triglyceride content, thus enhancing insulin sensitivity in the liver and skeletal muscle. 2,4,5 Because plasma adiponectin concentrations are decreased in obese subjects, 1,2 extensive researches have been aimed at the upregulation of adiponectin and its cognate receptors (AdipoR1 and AdipoR2) for the treatment of obesity-related metabolic sequelae. 2

Previous studies showed that the adipose tissue is markedly infiltrated by macrophages in several models of rodent obesities and human obese subjects, 6,7 suggesting that macrophages participate in the inflammatory pathways that are activated in obese adipose tissue. Using an in vitro coculture system composed of adipocytes and macrophages, we have demonstrated that a paracrine loop involving saturated FAs and TNF derived from adipocytes and macrophages, respectively, establishes a vicious cycle that augments the inflammatory changes; ie, marked upregulation of proinflammatory adipocytokines such as monocyte chemoattractant protein-1 (MCP-1) and TNF and downregulation of adiponectin. 8 These findings led us to speculate that macrophages, when infiltrated, induce the release of saturated FAs from adipocytes via lipolysis, which, in turn, may serve as a proinflammatory adipocytokine in the adipose tissue. 9 Interestingly, n-3 polyunsaturated fatty acids (PUFAs) such as eicosapentaenoic acid (EPA) are unable to activate macrophages or even antagonize the proinflammatory effect of saturated FAs, 8,10 suggesting the structural specificity of FAs in the induction of inflammatory changes.

In epidemiological and clinical trials, fish oil rich in n-3 PUFAs or n-3 PUFAs have reduced the incidence of coronary heart disease. 11,12 Given the pleiotropic effect of n-3 PUFAs, 11,13 it is tempting to speculate the beneficial effect of n-3 PUFAs on the dysregulation of adipocytokine production. There are a couple of recent reports showing that fish oil or n-3 PUFAs increase adiponectin mRNA expression and/or secretion in several models of rodent obesities. 14–16 However, it is still unknown whether EPA, the only class of n-3 PUFAs used clinically to treat hyperlipidemia, increases adiponectin production in obesity, and if so, how it does in obese adipose tissue remains to be elucidated. Furthermore, there has been no report showing the direct effect of EPA on adiponectin secretion in human obese subjects.

Here we report that serum adiponectin concentrations are increased by EPA in several models of rodent obesities and in human obese subjects. Using the in vitro coculture of adipocytes and macrophages, 8 we also show that EPA reverses the coculture-induced decrease in adiponectin secretion at least in part through downregulation of TNF in macrophages. Because EPA has reduced the risk of major coronary events in a large-scale, prospective, randomized clinical trial, 12 this study provides important insight into its therapeutic implication of obesity-related metabolic sequelae and thus the metabolic syndrome.

Methods

Materials

Preparation and characterization of highly purified eicosapentaenoic 98%, Mochida Pharmaceutical Co Ltd) used in animal and clinical studies was reported elsewhere. 17 Other materials are described in supplemental Methods (available online at http://atvb.ahajournals.org.).

Animals

Male C57BL/6J ob/ob mice and their wild-type (WT) littermates were purchased from Charles River Japan (Tsukuba, Ibaraki, Japan). The animals were housed in individual cages in a temperature-, humidity-, and light-controlled room (12-hour light and 12-hour dark cycle) and allowed free access to water and fish meal–free diet (fish meal–free F1; 362 kcal/100 g, 4.4% energy as fat; Funabashi Farm, Chiba, Japan). All animal experiments were conducted in accordance to the guidelines of Tokyo Medical and Dental University Committee on Animal Research (No. 0060026).

Administration of EPA in ob/ob Mice

Six-week-old male ob/ob mice and WT littermates had unrestricted access to the fish meal–free diet (control group) or fish meal–free diet supplemented with 5% EPA (wt/wt) (EPA-treated group) for 4 weeks (n=10 to 14). In the short-term administration protocol, 8-week-old male ob/ob mice were treated with EPA for 2 weeks (n=7 to 8). All diets were changed every day and served with nonmetallic feeder to prevent oxidization of fatty acids. At the end of the experiments, mice were euthanized after 5-hour starvation under intraperitoneal pentobarbital anesthesia (30 mg/kg).Blood glucose and serum concentrations of triglyceride (TG) and free fatty acid (FFA) were measured as previously described. 18 Serum EPA concentrations were measured by gas chromatography.

Histological Analysis

The epididymal white adipose tissue (WAT) was fixed with neutral-buffered formalin and embedded in paraffin. Five µm–thick sections were stained with hematoxylin and eosin and studied under x 200 magnification to compare the adipocyte cell size using the software Win Roof (Mitani Co Ltd). More than 200 cells were counted per each section. The presence of F4/80-positive macrophages in WAT was detected immunohistochemically using the rat monoclonal anti-mouse F4/80 antibody described elsewhere. 19 The number of F4/80-positive cells was counted in more than 10 µm 2 area of each section and expressed as the mean number/µm 2.

Coculture of Adipocytes and Macrophages

Details are described in supplemental Methods.

Quantitative Real-Time Polymerase Chain Reaction

Total RNA was extracted from mouse epididymal WAT and cultured cells using TRIzol reagent (Invitrogen) and quantitative real-time polymerase chain reaction was performed with an ABI Prism 7000 Sequence Detection System (Applied Biosystems) as described. 8,9 Primers used in this study were described elsewhere. 8 Levels of mRNA were normalized to those of 36B4 mRNA.

Measurement of Adiponectin Concentrations

Serum adiponectin concentrations in ob/ob mice and diet-induced obese mice were determined by a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Otsuka Pharmaceutical Assaypro). Adiponectin concentrations in culture supernatants were also determined by an ELISA kit (Otsuka Pharmaceutical).

Human Study

Details are described in supplemental Methods.

Statistical Analysis

Data are presented as mean±SE, and P <0.05 was considered statistically significant. In cell culture experiments and animal studies, statistical analysis was performed using analysis of variance followed by Scheffe test.

Results

Administration of EPA in ob/ob Mice

Increased Serum Adiponectin Concentrations in ob/ob Mice Treated With EPA

To elucidate the effect of EPA on adiponectin production in obesity, we performed 4-week administration of EPA in ob/ob mice. Body weight gain did not change in ob/ob and WT mice by EPA treatment throughout the experiment ( Figure 1 A). There was no appreciable difference in the weights of the adipose tissue and liver between EPA-treated and control WT mice, whereas the weights of the epididymal and subcutaneous WATs and liver were significantly reduced in EPA-treated ob/ob mice relative to control ob/ob mice (epididymal WAT, P <0.01; subcutaneous WAT and liver, P <0.05; Figure 1 B). Blood glucose concentrations did not change in ob/ob and WT mice by EPA treatment, whereas serum TG and FFA concentrations were significantly decreased in EPA-treated WT mice ( Figure 1 C). Histological analysis revealed that hepatic steatosis is markedly improved in ob/ob mice by EPA treatment (supplemental Figure IA). Both EPA-treated ob/ob and WT mice exhibited significant increase in serum EPA concentrations (WT, 5.30±0.35 versus 260.78±16.39; ob/ob, 16.23±0.78 versus 422.43±18.11 µg/mL, n =3 to 4, P <0.01). Serum adiponectin concentrations tended to be increased in WT mice and were significantly increased in ob/ob mice by EPA treatment ( P <0.05; Figure 1 D). These observations indicate that administration of EPA increases serum adiponectin concentrations in ob/ob mice in parallel with the reduction of adipose tissue weight.

Figure 1. Body weight, adipose tissue and liver weight, and serum parameters in mice treated with EPA for 4 weeks. A, Growth curve of ob/ob and WT mice on either control fish meal–free diet (EPA (–)) or 5% EPA-supplemented diet (EPA (+)). Open triangle, WT-EPA (–) (n=14); closed triangle, WT-EPA (+) (n=10); open circle, ob/ob -EPA (–) (n=10); closed circle, ob/ob -EPA (+) (n=10). B, Weight of the epididymal (epi), mesenteric (mes), and subcutaneous (sub) WATs and liver in ob/ob and WT mice on either EPA (–) or EPA (+) diet. Ln 1, WT-EPA (–); lane 2, WT-EPA (+); lane 3, ob/ob -EPA (–); lane 4, ob/ob -EPA (+). C, Blood glucose levels and serum concentrations of TG and FFA. D, Serum adiponectin concentrations in ob/ob and WT mice. * P <0.05; ** P <0.01. n=10 to 14.

Effect of EPA Treatment on Adipose Tissue Histology and mRNA Expression

Histological examination of the epididymal WAT revealed that adipocyte cell size is significantly reduced by EPA treatment in both ob/ob and WT mice (mean cell diameters: WT, 44.6±0.4 versus 37.4±0.3 µm; ob/ob, 99.5±1.2 versus 85.5±1.0 µm; P <0.01; Figure 2A and 2 C). We next examined mRNA expression in the epididymal WAT. As reported previously, 20 adiponectin mRNA expression was markedly reduced in ob/ob mice relative to WT mice ( Figure 2 E). On the contrary to the increment of serum adiponectin concentrations, there was no significant change in adiponectin mRNA expression by EPA treatment ( Figure 2 E). Although F4/80-positive macrophage accumulation in the adipose tissue was significantly increased in EPA-treated ob/ob mice relative to control group ( Figure 2B and 2 D, P <0.01), there was no significant difference in mRNA expression of TNF, a proinflammatory cytokine expressed predominantly in macrophages ( Figure 2 E). Plasminogen activator inhibitor (PAI)-1 and heparin-binding epidermal growth factor–like growth factor (HB-EGF) mRNA expression was significantly decreased in EPA-treated ob/ob mice ( P <0.01), although there was no significant change in MCP-1, interleukin (IL)-6, and resistin mRNA expression (supplemental Figure II). These observations indicate that EPA treatment increases serum adiponectin concentrations in ob/ob mice without affecting its mRNA expression, and attenuates inflammatory changes in the adipose tissue.

Figure 2. Histological analysis and mRNA expression in epididymal WAT from ob/ob and WT mice treated with EPA for 4 weeks. A, Hematoxylin and eosin staining of the epididymal WAT from ob/ob and WT mice. B, F4/80 staining of the epididymal WAT from ob/ob mice. Original magnification, x 200. Scale bars, 100 µm. C, Histogram of diameters of adipocytes in the epididymal WAT. D, Cell count of F4/80-positive cells in the epididymal WAT. E, Expression of adiponectin and TNF mRNAs in the epididymal WAT. n.s., not significant. ** P <0.01. n=10 to 14.

Effect of a Short-Term EPA Treatment in ob/ob Mice

To further examine whether the reduction of adipocyte cell size is related to the increased serum adiponectin concentrations by EPA treatment, we examined the metabolic phenotypes of ob/ob mice that received a short-term (or 2-week) treatment with EPA. There were no significant differences in body weight and adipose tissue weight between EPA-treated and control groups at the end of the experiment (supplemental Table I). The liver weight was significantly reduced ( P <0.01) and hepatic steatosis was histologically improved in EPA-treated group relative to control group (supplemental Figure IB). There was no significant difference in adipocyte cell size (mean cell diameters: 97.5±0.9 versus 94.1±0.9 µm) and mRNA expression of F4/80 and TNF in the epididymal WAT between EPA-treated and control groups ( Figure 3A and 3 B). PAI-1, HB-EGF, and resistin mRNA expression was significantly reduced with 2-week EPA administration ( P <0.05; supplemental Figure III). mRNA expression of IL-10 and arginase, which is specific for antiinflammatory M2-polarized macrophages, 21 tended to be increased, and that of inducible nitric oxide synthase (iNOS), which is specific for proinflammatory M1-polarized macrophages, tended to be decreased with EPA treatment (supplemental Figure IV). There was no significant change in mRNA expression of M1-specific CD11c (supplemental Figure IV). In this study, adiponectin mRNA expression was reduced in the epididymal WAT by the 2-week EPA treatment ( Figure 3 C). In this setting, serum adiponectin concentrations were significantly increased ( P <0.01; Figure 3 D), which is comparable to those attained by the 4-week EPA treatment. These observations indicate that EPA treatment increases serum adiponectin concentrations independently of adipocyte cell size.

Figure 3. Effect of a 2-week EPA treatment on adiponectin secretion in ob/ob mice. A, Hematoxylin and eosin staining of the epididymal WAT from ob/ob mice. Original magnification, x 200. Scale bars, 100 µm. B, Expression of F4/80 and TNF mRNAs in the epididymal WAT. mRNA expression levels (C) and serum concentrations (D) of adiponectin. n.s., not significant. * P <0.05, ** P <0.01. n=7 to 8.

Administration of EPA in Mice With High-Fat Diet–Induced Obesity

We next examined the effect of EPA on adiponectin production in mice with diet-induced obesity. Ten-week-old male C57BL/6J mice were fed high-fat diet, with or without EPA by daily oral administration for 8 weeks. The mice on high-fat diet weighed 14% more than those on standard diet at the end of the experiment (supplemental Figure VA). By histological analysis, adipocyte hypertrophy and macrophage accumulation in high-fat diet–induced obese mice were much milder than those in ob/ob mice (data not shown), suggesting that mice with diet-induced obesity and ob/ob mice used in this study represent models of the early and advanced stages of obesity, respectively. There was no significant difference in body weight between the groups fed high-fat diet. Plasma adiponectin concentrations in obese mice with EPA treatment were significantly increased relative to those without EPA treatment as early as 2 weeks after the treatment ( P <0.01; supplemental Figure VB), whereas there was no significant upregulation of adiponectin mRNA levels (supplemental Figure VC). These observations, taken together, indicate that EPA treatment effectively increases plasma adiponectin concentrations without increase in adiponectin mRNA expression at the early stage of obesity as well as at the advanced stage of obesity.

Effect of EPA in the Coculture of Adipocytes and Macrophages

To explore the molecular mechanism for the EPA-induced increase in adiponectin secretion in vivo, we examined the effect of EPA on adiponectin production in the in vitro coculture of 3T3-L1 adipocytes and RAW264 macrophages. As we reported previously, 8,9 adiponectin mRNA expression and secretion to the media were significantly reduced in 3T3-L1 adipocytes cocultured with RAW264 macrophages ( P <0.01; Figure 4 A). Treatment with EPA reversed significantly the coculture-induced decrease in adiponectin secretion in a dose-dependent manner ( P <0.05), whereas adiponectin mRNA expression was unchanged ( Figure 4 A). In this study, EPA did not affect adiponectin secretion and mRNA expression in 3T3-L1 adipocytes alone (or in the control culture; Figure 4 A). We found that treatment with EPA at doses of 100 and 200 µmol/L suppresses significantly the coculture-induced increase in TNF mRNA expression ( P <0.05; supplemental Figure VIA), which occurs predominantly in RAW264 macrophages in the coculture system. 8 Furthermore, blockade of TNF using an anti-TNF neutralizing antibody at a dose that can abolish the effect of recombinant TNF reversed significantly the coculture-induced decrease in adiponectin secretion ( P <0.05; supplemental Figure VIB). We also observed that adiponectin secretion is reduced significantly in 3T3-L1 adipocytes treated with recombinant TNF at a dose of 200 pg/mL ( P <0.05; supplemental Figure VIC), which is roughly comparable to those detected in the coculture system. 8 In this study, adiponectin mRNA expression was unaffected, although both adiponectin mRNA expression and secretion are shown to be suppressed in vitro when treated with TNF at much higher doses. 22

Figure 4. Effect of EPA on adiponectin secretion and mRNA expression in the coculture of adipocytes and macrophages. A, Effect of EPA on adiponectin secretion and mRNA expression in 3T3-L1 adipocytes cocultured with RAW264 macrophages. B, Effect of EPA on the palmitate-induced increase in TNF mRNA expression and NF- B activation in RAW264 macrophages. B-luc, a luciferase reporter vector for NF- B; pGL3, a promoterless luciferase reporter vector; RLA, relative luciferase activity. C, Effect of EPA on the LPS-induced increase in TNF mRNA expression and NF- B activation in RAW264 macrophages. EPA 50 to 200 µmol/L; pal, palmitate 200 µmol/L; LPS 1 ng/mL. n.s., not significant. * P <0.05, ** P <0.01. n=4 to 6.

Because saturated FAs is an important adipocyte-derived paracrine signal to induce TNF mRNA expression and NF- B activation in macrophages, 8,9 we also examined the effect of EPA on saturated FA-induced increase in the inflammatory changes in RAW264 macrophages. The basal NF- B activity in RAW264 macrophages transfected with pGL3 vector as a promoterless control was approximately 1/20 relative to those transfected with the luciferase reporter vector for NF- B activity, and EPA tended to decrease TNF mRNA expression in RAW264 macrophages ( Figure 4B and 4 C). Treatment with palmitate, a major saturated FA released from 3T3-L1 adipocytes, 8 induced significantly TNF mRNA expression in RAW264 macrophages ( P <0.01), which were effectively reversed by EPA ( P <0.01; Figure 4 B). Furthermore, a luciferase reporter assay showed that EPA suppresses the palmitate-induced activation of NF- B in RAW264 macrophages ( Figure 4 B). Because saturated FAs act as a proinflammatory adipocytokine through toll-like receptor 4 (TLR4), 9,10 a bona fide receptor for the recognition of lipopolysaccharide (LPS), we also examined the effect of EPA on the LPS-induced increase in TNF mRNA expression and NF- B activation. In this study, EPA was capable of reversing dose-dependently the LPS-induced increase in TNF mRNA expression and NF- B activation ( Figure 4 C). Moreover, EPA significantly reduced FFA release from 3T3-L1 adipocytes cocultured with RAW264 macrophages ( P <0.05; supplemental Figure VII). These observations, taken together, suggest that EPA reverses the coculture-induced decrease in adiponectin secretion at least in part through the reduction of the inflammatory changes induced by the interaction between adipocytes and macrophages.

Treatment With EPA in Human Obese Subjects

We also examined whether EPA increases adiponectin secretion in human obese subjects. There were no significant differences between EPA-treated and control groups in all measured variables before EPA treatment ( Table 1 ). After the 3-month EPA treatment, plasma concentrations of EPA were significantly increased in EPA-treated group relative to control group ( P <0.01). In EPA-treated group, plasma TG concentrations were significantly decreased ( P <0.05) relative to control group, although there were no changes in BMI, waist circumference, systolic blood pressure (SBP), fasting plasma glucose (FPG), total cholesterol, high density lipoprotein cholesterol (HDL-C), and low density lipoprotein cholesterol (LDL-C) between both groups. Plasma adiponectin concentrations were increased after the treatment with EPA ( P <0.01), but unchanged in control group. To determine the risk factors independently influencing the changes of plasma adiponectin, multivariate regression analysis was performed using the variables shown in supplemental Table II. Only the treatment with EPA was an independent determinant of plasma adiponectin concentrations in obese subjects. (β=0.378, P =0.018). By stepwise multivariate regression analysis (r 2 =0.170), plasma adiponectin concentrations were significantly correlated only with the treatment with EPA (F=9.24, P =0.004).

Clinical Characteristics and Metabolic Parameters Before and After the Follow-Up Period

Discussion

In a large-scale, prospective, randomized clinical trial, highly purified EPA has reduced the risk of major coronary events via cholesterol-independent mechanisms. 12 However, the detailed mechanism is still poorly understood. Here we examined the effect of EPA on adiponectin production in rodent models of obesity and obese human subjects. Using the in vitro coculture system composed of 3T3-L1 adipocytes and RAW264 macrophages, 8 we also examined how EPA increases adiponectin production in the interaction between adipocytes and macrophages in obese adipose tissue.

This study demonstrates that EPA is capable of increasing serum adiponectin concentrations in genetically obese ob/ob mice at the late stage of obesity, when macrophages are markedly infiltrated into the adipose tissue. The effect was observed as early as 2 weeks and persisted up to 4 weeks after the administration. In this study, there was no appreciable increase in adiponectin mRNA expression in the adipose tissue, indicating that EPA increases adiponectin secretion rather than its mRNA expression. Using an in vitro coculture of adipocytes and macrophages, 8 we demonstrated that treatment with EPA reverses the coculture-induced decrease in adiponectin secretion, with no effect on its mRNA expression. These observations indicate that EPA increases adiponectin secretion rather than its biosynthesis in vitro. We also found that EPA reduces the coculture-induced increase in TNF mRNA expression, which is considered to reflect the inflammatory state of macrophages. Importantly, blockade of the TNF action effectively reversed the coculture-induced decrease in adiponectin secretion. These observations, taken together, suggest that EPA reverses the coculture-induced decrease in adiponectin secretion possibly through the attenuation of inflammatory changes in macrophages. We have recently demonstrated that saturated FAs, which are released via the macrophage-induced adipocyte lipolysis, serve as a naturally occurring ligand for TLR4, thereby inducing the inflammatory changes in macrophages through NF- B activation. 9 In this study, EPA suppressed the palmitate- and LPS-induced increase in TNF mRNA expression and NF- B activation in macrophages. It is, therefore, likely that EPA acts directly on macrophages, thereby interrupting the vicious cycle created by the interaction between adipocytes and macrophages in obese adipose tissue.

Histological analysis revealed significant reduction of adipocyte cell size in ob/ob mice treated with EPA relative to control group, which is consistent with previous reports. 23,24 In this study, the adipose tissue weights tended to be reduced by 2-week administration of EPA, and were significantly reduced after 4-week or longer-term administration. The reduction of adipose tissue weights and serum FFA concentrations may be secondary to improved lipid metabolism, as n-3 PUFAs are shown to act as peroxisome proliferator-activated receptor- (PPAR ) ligands, thereby inducing the upregulation of genes involved in fatty acid oxidation in the liver. 25 Although it is largely unknown about the effect of n-3 PUFAs on the regulation of FFA production in the adipose tissue, we have observed that EPA significantly suppresses the coculture-induced increase in FFA release from 3T3-L1 adipocytes. Flachs et al also reported that n-3 PUFAs increase mRNA expression of PPAR coactivator 1- and nuclear respiratory factor-1, which in turn stimulate β-oxidation and mitochondrial biogenesis in the adipose tissue and 3T3-L1 adipocytes. 26 We also observed increased adiponectin secretion in ob/ob mice during the 2-week administration of EPA, with no change in adipocyte cell size. These observations, taken together, suggest that the positive effect of EPA on adiponectin secretion is not related to the adipocyte cell size.

In this study, it is interesting to note that macrophage accumulation is apparently increased in the adipose tissue from ob/ob mice treated with EPA for 4 weeks, although TNF mRNA expression is not increased in parallel with the number of macrophages. Because we used fish meal–free diet (control group) and fish meal–free diet supplemented with EPA, the difference in fat intake between EPA-treated and control groups (4.4 versus 9.4% , respectively) may influence macrophage infiltration. We and others have recently demonstrated that TLR4-deficient or mutant mice rendered obese by high-fat diet feeding show decrease in expression of inflammatory markers and increase in serum adiponectin concentrations relative to wild-type mice, with no changes in the number of macrophages infiltrated. 27,28 Moreover, Lumeng et al recently reported that adipose tissue macrophages isolated from lean and obese mice are in different activation state like M1 or "classically activated" (proinflammatory) and M2 or "alternatively activated" (antiinflammatory) polarization, suggesting the implication of macrophage subpopulation. 21 These findings suggest that not only the number but also the activation state of macrophages in the adipose tissue affects adipocytokine production. Indeed, n-3 PUFAs are known to suppress expression of inflammatory markers such as TNF and IL-1β in macrophages in vitro. 29 Collectively, we speculate that macrophages infiltrated into the adipose tissue of ob/ob mice treated with EPA are not activated to induce adipose tissue inflammation. Further studies are needed to evaluate the activation state of macrophages in vivo.

In this study, we confirmed that EPA increases serum adiponectin concentrations in mice rendered mildly obese by a short-term high-fat diet at the early stage of obesity, when macrophages are not apparently infiltrated. These observations suggest that EPA acts directly on adipocytes in the adipose tissue, where it increases adiponectin secretion. Interestingly, there was no significant increase in adiponectin mRNA expression in these animals. In this regard, Nishizawa et al demonstrated that testosterone inhibits adiponectin secretion but not adiponectin mRNA expression in 3T3-L1 adipocytes. 30 There may be unknown mechanisms involved in the regulation of adiponectin secretion in adipocytes, whose activity is modulated by EPA. Indeed, Xie et al recently reported that intracellular trafficking and secretion of adiponectin are uniquely mediated by specific coated vesicles formed at the trans-Golgi network of adipocytes, whereas those of another adipocytokine, leptin are not. 31 EPA is known to be taken into cell membrane phospholipids, 32 and it may affect adiponectin secretion through modulation of fatty acid composition of adipocyte cell membrane. On the other hand, Neschen et al reported that fish oil increases both adiponectin mRNA expression and secretion in mice on relatively a short-term (or 2-week) high-fat diet. 15 This may be because of the potential difference in antiinflammatory effect between EPA used in this study and fish oil or the doses of administration. In addition, we and others observed that n-3 PUFAs do not affect adiponectin mRNA expression and secretion in 3T3-L1 adipocytes, 15 suggesting the difficulty to examine the direct effect of n-3 PUFAs on adiponectin production in adipocytes in vitro. Further studies are needed to elucidate how EPA regulates adiponectin mRNA expression or secretion in mildly obese animals in vivo.

The limitation to this study includes the difference between the in vivo and in vitro data as to the mechanism by which EPA increases adiponectin secretion. To get insight into the effect of EPA on the inflammatory changes in obese adipose tissue, we used the unique in vitro coculture system composed of adipocytes and macrophages, by which we have reproduced the proinflammatory gene expression profile found in obese adipose tissue. 8,9 In this study, we found that EPA reverses the coculture-induced reduction of adiponectin secretion possibly through the attenuation of macrophage activation. However, although EPA reduces the coculture-induced increase in TNF production in vitro, TNF mRNA expression is not significantly reduced in the epididymal WAT from EPA-treated ob/ob mice. The difference between the in vivo and in vitro settings may be because the adipose tissue shows dramatic changes in cellular population during the development of obesity. 6 Otherwise, 3T3-L1 adipocytes may not respond to EPA in vitro. Nonetheless, we found that treatment with EPA for 2 and 4 weeks reduces mRNA expression of several proinflammatory adipocytokines including PAI-1 and HB-EGF, whereas there was no apparent change in that of MCP-1, IL-6, and resistin in this study. These observations suggest that EPA reduces the inflammatory changes in obese adipose tissue in vivo and those induced by the interaction between adipocytes and macrophages in vitro, thereby leading to increased adiponectin secretion.

We also demonstrated increased plasma adiponectin concentrations in obese human subjects after the treatment with EPA. Multivariate regression analysis revealed that EPA administration is only the independent determinant of plasma adiponectin concentrations, suggesting the direct involvement of EPA in adiponectin secretion. It was reported that weight loss improves inflammation in the adipose tissue and that n-3 PUFAs has additive effect on plasma adiponectin concentrations in combination with weight loss. 33,34 In this study, there was no significant change in body weight of our study subjects, suggesting that EPA itself increases adiponectin secretion without body weight change. Although we did not obtain the adipose tissue samples used for histological examination, it has been reported that there exists macrophage infiltration in the adipose tissue from obese subjects with BMI of approximately 30. 35 We speculate that EPA increases adiponectin secretion at least partly by interrupting the vicious cycle created by adipocytes and macrophages in human obese subjects as in the in vitro coculture experiments.

In conclusion, this study demonstrates that EPA increases adiponectin secretion in rodent models of obesity and human obese subjects. Given that hypoadiponectinemia has been shown to increase the risk of coronary heart disease, 36,37 the beneficial effect of EPA may be attributable at least in part to the modulation of the inflammatory changes in obese adipose tissue and increased adiponectin secretion. Because EPA is the only class of n-3 PUFAs, which has proved to reduce the risk of major coronary events, 12 the data of this study provide important insight into its therapeutic implication in obesity-related metabolic sequelae.

Acknowledgments

We thank M. Yashima for secretarial assistance and the members of the Ogawa laboratory for helpful discussions.

Sources of Funding

This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and Ministry of Health, Labour, and Welfare of Japan, and research grants from The Naito Foundation, The Novo Nordisk Insulin Study Award, the Danone Institute of Japan, Nestlé Nutrition Council, Japan, Research Fund of Mitsukoshi Health and Welfare Foundation, The SKYLARK Food Science Institute, Suzuken Memorial Foundation, Ono Medical Research Foundation, Japan Heart Foundation/Novartis Grant for Research Award on Molecular and Cellular Cardiology, and Takeda Medical Research Foundation. M. Tanaka is supported by Mishima Kaiun Memorial Foundation.

Disclosures

None.

【参考文献】
  Matsuzawa Y, Funahashi T, Nakamura T. Molecular mechanism of metabolic syndrome X: contribution of adipocytokines adipocyte-derived bioactive substances. Ann N Y Acad Sci. 1999; 892: 146–154.

Kadowaki T, Yamauchi T, Kubota N, Hara K, Ueki K, Tobe K. Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest. 2006; 116: 1784–1792.

Berg AH, Scherer PE. Adipose tissue, inflammation, and cardiovascular disease. Circ Res. 2005; 96: 939–949.

Civitarese AE, Ukropcova B, Carling S, Hulver M, DeFronzo RA, Mandarino L, Ravussin E, Smith SR. Role of adiponectin in human skeletal muscle bioenergetics. Cell Metab. 2006; 4: 75–87.

Maeda N, Shimomura I, Kishida K, Nishizawa H, Matsuda M, Nagaretani H, Furuyama N, Kondo H, Takahashi M, Arita Y, Komuro R, Ouchi N, Kihara S, Tochino Y, Okutomi K, Horie M, Takeda S, Aoyama T, Funahashi T, Matsuzawa Y. Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med. 2002; 8: 731–737.

Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003; 112: 1796–1808.

Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A, Ross JS, Tartaglia LA, Chen H. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest. 2003; 112: 1821–1830.

Suganami T, Nishida J, Ogawa Y. A paracrine loop between adipocytes and macrophages aggravates inflammatory changes: role of free fatty acids and tumor necrosis factor. Arterioscler Thromb Vasc Biol. 2005; 25: 2062–2068.

Suganami T, Tanimoto-Koyama K, Nishida J, Itoh M, Yuan X, Mizuarai S, Kotani H, Yamaoka S, Miyake K, Aoe S, Kamei Y, Ogawa Y. Role of the toll-like receptor 4/NF- B pathway in saturated fatty acid-induced inflammatory changes in the interaction between adipocytes and macrophages. Arterioscler Thromb Vasc Biol. 2007; 27: 84–91.

Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. J Biol Chem. 2001; 276: 16683–16689.

Kris-Etherton PM, Harris WS, Appel LJ. Fish consumption, fish oil, omega-3 fatty acids, and cardiovascular disease. Circulation. 2002; 106: 2747–2757.

Yokoyama M, Origasa H, Matsuzaki M, Matsuzawa Y, Saito Y, Ishikawa Y, Oikawa S, Sasaki J, Hishida H, Itakura H, Kita T, Kitabatake A, Nakaya N, Sakata T, Shimada K, Shirato K. Effects of eicosapentaenoic acid on major coronary events in hypercholesterolaemic patients (JELIS): a randomised open-label, blinded endpoint analysis. Lancet. 2007; 369: 1090–1098.

Calder PC. Polyunsaturated fatty acids, inflammation, and immunity. Lipids. 2001; 36: 1007–1024.

Flachs P, Mohamed-Ali V, Horakova O, Rossmeisl M, Hosseinzadeh-Attar MJ, Hensler M, Ruzickova J, Kopecky J. Polyunsaturated fatty acids of marine origin induce adiponectin in mice fed a high-fat diet. Diabetologia. 2006; 49: 394–397.

Neschen S, Morino K, Rossbacher JC, Pongratz RL, Cline GW, Sono S, Gillum M, Shulman GI. Fish oil regulates adiponectin secretion by a peroxisome proliferator-activated receptor- -dependent mechanism in mice. Diabetes. 2006; 55: 924–928.

Todoric J, Loffler M, Huber J, Bilban M, Reimers M, Kadl A, Zeyda M, Waldhausl W, Stulnig TM. Adipose tissue inflammation induced by high-fat diet in obese diabetic mice is prevented by n-3 polyunsaturated fatty acids. Diabetologia. 2006; 49: 2109–2119.

Mita T, Watada H, Ogihara T, Nomiyama T, Ogawa O, Kinoshita J, Shimizu T, Hirose T, Tanaka Y, Kawamori R. Eicosapentaenoic acid reduces the progression of carotid intima-media thickness in patients with type 2 diabetes. Atherosclerosis. 2006; 191: 162–167.

Kouyama R, Suganami T, Nishida J, Tanaka M, Toyoda T, Kiso M, Chiwata T, Miyamoto Y, Yoshimasa Y, Fukamizu A, Horiuchi M, Hirata Y, Ogawa Y. Attenuation of diet-induced weight gain and adiposity through increased energy expenditure in mice lacking angiotensin II type 1a receptor. Endocrinology. 2005; 146: 3481–3489.

Kitagawa K, Wada T, Furuichi K, Hashimoto H, Ishiwata Y, Asano M, Takeya M, Kuziel WA, Matsushima K, Mukaida N, Yokoyama H. Blockade of CCR2 ameliorates progressive fibrosis in kidney. Am J Pathol. 2004; 165: 237–246.

Delporte ML, El Mkadem SA, Quisquater M, Brichard SM. Leptin treatment markedly increased plasma adiponectin but barely decreased plasma resistin of ob/ob mice. Am J Physiol Endocrinol Metab. 2004; 287: E446–E453.

Lumeng CN, Bodzin JL, Saltiel AR. Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest. 2007; 117: 175–184.

Maeda N, Takahashi M, Funahashi T, Kihara S, Nishizawa H, Kishida K, Nagaretani H, Matsuda M, Komuro R, Ouchi N, Kuriyama H, Hotta K, Nakamura T, Shimomura I, Matsuzawa Y. PPAR ligands increase expression and plasma concentrations of adiponectin, an adipose-derived protein. Diabetes. 2001; 50: 2094–2099.

Rossi AS, Lombardo YB, Lacorte JM, Chicco AG, Rouault C, Slama G, Rizkalla SW. Dietary fish oil positively regulates plasma leptin and adiponectin levels in sucrose-fed, insulin-resistant rats. Am J Physiol Regul Integr Comp Physiol. 2005; 289: R486–R494.

Raclot T, Groscolas R, Langin D, Ferre P. Site-specific regulation of gene expression by n-3 polyunsaturated fatty acids in rat white adipose tissues. J Lipid Res. 1997; 38: 1963–1972.

Jump DB. The biochemistry of n-3 polyunsaturated fatty acids. J Biol Chem. 2002; 277: 8755–8758.

Flachs P, Horakova O, Brauner P, Rossmeisl M, Pecina P, Franssen-van Hal N, Ruzickova J, Sponarova J, Drahota Z, Vlcek C, Keijer J, Houstek J, Kopecky J. Polyunsaturated fatty acids of marine origin upregulate mitochondrial biogenesis and induce β-oxidation in white fat. Diabetologia. 2005; 48: 2365–2375.

Suganami T, Mieda T, Itoh M, Shimoda Y, Kamei Y, Ogawa Y. Attenuation of obesity-induced adipose tissue inflammation in C3H/HeJ mice carrying a Toll-like receptor 4 mutation. Biochem Biophys Res Commun. 2007; 354: 45–49.

Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS. TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest. 2006; 116: 3015–3025.

Yaqoob P, Calder P. Effects of dietary lipid manipulation upon inflammatory mediator production by murine macrophages. Cell Immunol. 1995; 163: 120–128.

Nishizawa H, Shimomura I, Kishida K, Maeda N, Kuriyama H, Nagaretani H, Matsuda M, Kondo H, Furuyama N, Kihara S, Nakamura T, Tochino Y, Funahashi T, Matsuzawa Y. Androgens decrease plasma adiponectin, an insulin-sensitizing adipocyte-derived protein. Diabetes. 2002; 51: 2734–2741.

Xie L, Boyle D, Sanford D, Scherer PE, Pessin JE, Mora S. Intracellular trafficking and secretion of adiponectin is dependent on GGA-coated vesicles. J Biol Chem. 2006; 281: 7253–7259.

Kamada T, Yamashita T, Baba Y, Kai M, Setoyama S, Chuman Y, Otsuji S. Dietary sardine oil increases erythrocyte membrane fluidity in diabetic patients. Diabetes. 1986; 35: 604–611.

Krebs JD, Browning LM, McLean NK, Rothwell JL, Mishra GD, Moore CS, Jebb SA. Additive benefits of long-chain n-3 polyunsaturated fatty acids and weight-loss in the management of cardiovascular disease risk in overweight hyperinsulinaemic women. Int J Obes (Lond). 2006; 30: 1533–1544.

Clement K, Viguerie N, Poitou C, Carette C, Pelloux V, Curat CA, Sicard A, Rome S, Benis A, Zucker JD, Vidal H, Laville M, Barsh GS, Basdevant A, Stich V, Cancello R, Langin D. Weight loss regulates inflammation-related genes in white adipose tissue of obese subjects. FASEB J. 2004; 18: 1657–1669.

Curat CA, Wegner V, Sengenes C, Miranville A, Tonus C, Busse R, Bouloumie A. Macrophages in human visceral adipose tissue: increased accumulation in obesity and a source of resistin and visfatin. Diabetologia. 2006; 49: 744–747.

Kumada M, Kihara S, Sumitsuji S, Kawamoto T, Matsumoto S, Ouchi N, Arita Y, Okamoto Y, Shimomura I, Hiraoka H, Nakamura T, Funahashi T, Matsuzawa Y. Association of hypoadiponectinemia with coronary artery disease in men. Arterioscler Thromb Vasc Biol. 2003; 23: 85–89.

Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB. Plasma adiponectin levels and risk of myocardial infarction in men. J Am Med Assoc. 2004; 291: 1730–1737.

作者单位：Department of Molecular Medicine and Metabolism (M.I., T.S., K.T.-K., X.Y., M.T., Y.K., Y.O.) and the Center of Excellence Program for Frontier Research on Molecular Destruction and Reconstitution of Tooth and Bone (M.T., Y.O.), Medical Research Institute, Tokyo Medical and Dental University; the Cl