Visualized macrophage dynamics and significance of S100A8 in obese fat

Ryohei Sekimoto; Shiro Fukuda; Norikazu Maeda; Yu Tsushima; Keisuke Matsuda; Takuya Mori; Hideaki Nakatsuji; Hitoshi Nishizawa; Ken Kishida; Junichi Kikuta; Yumiko Maijima; Tohru Funahashi; Masaru Ishii; Iichiro Shimomura

doi:10.1073/pnas.1409480112

New Research In

Edited by Jason G. Cyster, University of California, San Francisco, CA, and approved March 16, 2015 (received for review May 26, 2014)

Significance

Infiltration of immune cells into adipose tissue has been observed in obesity, but, until now, these data were based on immunohistological microscopic analysis. Here, to our knowledge for the first time, we applied an intravital multiphoton imaging technique to adipose tissue with lysozyme M-EGFP transgenic (LysM^EGFP) mice whose EGFP was expressed in the myelomonocytic lineage. Mobility of LysM^EGFP-positive macrophages was activated just 5 d after high-fat and high-sucrose (HF/HS) diet before the development of obesity. Furthermore, a significant increase of S100A8, one of the alarmins, was detected in fat tissue just 5 d after HF/HS diet. S100A8 stimulated chemotactic migration, and neutralization of S100A8 suppressed the HF/HS diet-induced activation of LysM^EGFP-positive cells. Time-lapse intravital imaging first identified the very early event exhibiting increased mobility of adipose macrophages.

Abstract

Chronic low-grade inflammation of adipose tissue plays a crucial role in the pathophysiology of obesity. Immunohistological microscopic analysis in obese fat tissue has demonstrated the infiltration of several immune cells such as macrophages, but dynamics of immune cells have not been fully elucidated and clarified. Here, by using intravital multiphoton imaging technique, to our knowledge for the first time, we analyzed and visualized the inflammatory processes in adipose tissue under high-fat and high-sucrose (HF/HS) diet with lysozyme M-EGFP transgenic (LysM^EGFP) mice whose EGFP was specifically expressed in the myelomonocytic lineage. Mobility of LysM^EGFP-positive macrophages was shown to be activated just 5 d after HF/HS diet, when the distinct hypertrophy of adipocytes and the accumulation of macrophages still have not become prominent. Significant increase of S100A8 was detected in mature adipocyte fraction just 5 d after HF/HS diet. Recombinant S100A8 protein stimulated chemotactic migration in vitro and in vivo, as well as induced proinflammatory molecules, both macrophages and adipocytes, such as TNF-α and chemokine (C-C motif) ligand 2. Finally, an antibody against S100A8 efficiently suppressed the HF/HS diet-induced initial inflammatory change, i.e., increased mobilization of adipose LysM^EGFP-positive macrophages, and ameliorated HF/HS diet-induced insulin resistance. In conclusion, time-lapse intravital multiphoton imaging of adipose tissues identified the very early event exhibiting increased mobility of macrophages, which may be triggered by increased expression of adipose S100A8 and results in progression of chronic inflammation in situ.

Obesity, especially visceral fat obesity, is a central player in the development of metabolic syndrome and in its clinical consequences (1 ⇓⇓–4). Chronic low-grade inflammation of adipose tissue has been demonstrated to be critical in the pathogenesis of obesity (5, 6). In terms of innate immune cells, the defining feature of adipose tissue inflammation in obesity is a marked increase of macrophage infiltration into adipose tissue (6, 7). The macrophages in inflamed adipose tissue are predominantly inflammatory M1 macrophages producing proinflammatory cytokines such as TNF-α (8). The other immune cells, such as T/B lymphocytes (9, 10), neutrophils (11), and eosinophils (12), have also been shown to play significant roles in adipose tissue inflammation and consequent metabolic disorders. However, most of these observations are based on immunohistological and flow cytometric analyses. The dynamic nature of immune cells has not been fully elucidated in adipose tissue during the progression of obesity. Adipose tissue is considered to be a key site of interaction between adipocytes and other immune system effectors, which may highlight the necessity to analyze immune cell dynamics in obese adipose tissue in vivo.

Chemokines and other inflammatory mediators have been proposed to be involved in adipose tissue inflammation in obesity. Several potential targets of modulation of inflammatory response have been demonstrated. Among them, there is considerable evidence for the pathophysiological role of the chemokine (C-C motif) ligand 2 [CCL2; monocyte chemoattractant protein-1 (MCP-1)]/chemokine (C-C motif) receptor 2 (CCR2) pathway in monocyte/macrophage infiltration into obese adipose tissues (13, 14). In addition, danger-associated molecular patterns, also called alarmins, which are released from damaged tissues and stressed cells, have caused sterile immune responses such as obesity-induced chronic inflammation. Pattern-recognition receptors such as Toll-like receptors (TLRs) and NOD-like receptors are involved in binding and responding to alarmins, including high-mobility group box chromosomal protein 1 (HMGB1) (15), S100 proteins (16, 17), saturated fatty acid (18, 19), oxidized LDL (20), and degradation products of extracellular matrices (21). Furthermore, adipose-derived saturated fatty acids have been shown to directly activate TLR4 and TLR2, not only of macrophages but also of adipocytes themselves, as endogenous ligands, resulting in the production of proinflammatory cytokines and chemokines and the dysregulation of adipocytokines (18, 19).

Chronic inflammation is a complex phenomenon, and multiple factors have been shown to be intricately involved in the progression of adipose tissue inflammation when it has happened. However, the early events triggering the chronic inflammatory cascades in adipose tissues have not been understood. Here, by means of intravital multiphoton microscopy of adipose tissues, we visualized time-dependent changes of immune cell dynamics during the development of obesity, and succeeded in identifying the early triggering event of chronic inflammation in obese adipose tissue.

Results

Visualization of LysM^EGFP-Positive Macrophages in Adipose Tissues.

To analyze visually the dynamic behaviors of macrophages within adipose tissue of living mice, we used lysozyme M-EGFP transgenic (LysM^EGFP) mice whose EGFP was specifically expressed in the myelomonocytic lineage to visualize immune cell dynamics (22). To analyze immune cell dynamics in adipose tissue in the process of obesity, intravital imaging in epididymal white adipose tissue (WAT) of LysM^EGFP mice was carried out under a normal chow (NC) or a high-fat and high-sucrose (HF/HS) diet (23). To analyze the cellular dynamics in early and late stages of obesity, we obtained images from mice fed with HF/HS diet for 5 d and 8 wk, respectively (Fig. 1 A and B). A few moving LysM^EGFP-positive cells could be detected in adipose tissues in mice fed an NC diet (Fig. 1A, Left, and Movie S1). On the contrary, it was of note that cell migration was apparently enhanced in mice fed with an HF/HS diet for just 5 d (Fig. 1A, Middle, and Movie S2). Because histological changes such as enlargement of adipocyte sizes were not obvious at this time point, this early change in cellular dynamics could be detected by the intravital imaging technique. In 8 wk feeding with HF/HS diet, a number of LysM^EGFP-positive cells were recruited and vigorously migrating, and some were surrounding dead adipocytes, forming crown-like structures (CLSs) (6, 7) (Fig. 1A, Right, and Movie S3), when the obesity-induced histological changes were prominent. Quantitative image data analyses showed that cell tracking velocity was significantly increased, by approximately twofold, after just 5 d of HF/HS diet feeding, and its increase was maintained at 8 wk after HF/HS diet (Fig. 1B). Histological analysis of WAT revealed that adipocyte size was not changed at day 5 under HF/HS diet compared with NC diet, and adipocytes were significantly larger at 8 wk after HF/HS diet (Fig. 1C).

Fig. 1.

Intravital adipose tissue imaging. LysM^EGFP mice were fed NC or HF/HS diet for 5 d (5d) or 8 wk (8w). (A) Representative intravital two-photon images in epididymal adipose tissues of LysM^EGFP mice fed with NC diet (Left; Movie S1), HF/HS diet for 5 d (Middle; Movie S2), and HF/HS diet for 8 wk (Right; Movie S3). For adipose tissue imaging, adipocytes (blue) were stained with BODIPY and blood vessels (red) were stained by i.v. injection with nontargeted Q-dots just before imaging. LysM^EGFP-positive cells appear in green. Colored lines show the associated trajectories of LysM^EGFP-positive cells. (Scale bar: 100 μm.) (B) Mean tracking velocity and tracking length of LysM^EGFP-positive cells. LysM^EGFP-positive cells in sections were tracked, and mobility was quantified from continuous time-lapse imaging for 1 h by using IMARIS software. Data points (n = 354 for NC; n = 424 for HF/HS 5 d; n = 413 for HF/HS 8 wk) and bars represent the values for individual cells compiled from five independent experiments and averages, respectively. (C) Cell size of adipocytes (Left) and frequency of adipocyte size (Right) of epididymal fat pad. To analyze adipocyte cell size, low-power field images were obtained from five mice per each group, and the area of 35 adipocytes in each field were measured. Data are expressed as means ± SEM (***P < 0.001).

Identification of LysM^EGFP-Positive Cell Population in Adipose Tissues.

Flow cytometry analysis was next performed to identify the cell population of LysM^EGFP-positive cells in adipose tissues. Fig. 2A presents the comparison of LysM^EGFP-positive cell populations in WAT and circulating blood. In contrast to the cell population in blood, the frequency of LysM^EGFP-positive neutrophils was very low in WAT, whereas 60–70% of LysM^EGFP-positive adipose cells were identified to be macrophages (Fig. 2A). As shown in Table S1, the proportions of adipose EGFP-positive cells in neutrophils (CD11b⁺ Ly-6G⁺), macrophages (CD11b⁺ F4/80⁺), and inflammatory monocytes (CD11b⁺ Ly-6C^hi) were 95.3%, 91.8%, and 56.6%, respectively, under NC diet.

Fig. 2.

Populations of LysM^EGFP-positive cells in adipose tissue and blood. (A) Percent population of neutrophils, macrophages, and inflammatory monocytes in LysM^EGFP-positive cells in epididymal adipose tissue (Left) and blood (Right) from LysM^EGFP mice fed with NC. (B) Representative flow cytometry dot plots showing CD11b⁺ Ly6G⁺ (neutrophils), CD11b⁺ F4/80⁺ (macrophages), CD11b⁺ F4/80⁺ CD11c⁺ (M1 macrophages), and CD11b⁺Ly6C^hi (inflammatory monocytes) in epididymal adipose tissue of LysM^EGFP mice fed an NC diet or HF/HS diet for 5 d (5d) or 8 wk (8w). (C) Number of LysM^EGFP-positive cells in epididymal adipose tissue. Data are expressed as means ± SEM. Dots represent values from individual mice (n = 6; *P < 0.05 and ***P < 0.001).

Fig. 2B shows representative changes of LysM^EGFP-positive cell population in WAT under HF/HS diet study. The population of neutrophils (CD11b⁺ Ly-6G⁺) was not changed on day 5, but it was rather decreased on 8 wk after HF/HS diet feeding. The total macrophage population (CD11b⁺ F4/80⁺) was not altered during HF/HS diet feeding. However, the population of CD11c⁺ macrophages (CD11b⁺ F4/80⁺ CD11c⁺) was increased after just 5 d of HF/HS diet feeding, and the increase was maintained at 8 wk after HF/HS diet. Population of inflammatory monocytes (CD11b⁺ Ly-6C^hi) was not obviously increased during HF/HS diet. Fig. 2C demonstrates the number of LysM^EGFP-positive cells in WAT. Concordant with the data shown in Fig. 2B, total LysM^EGFP-positive cells were not altered on day 5 after HF/HS diet and were slightly but significantly increased at 8 wk after HF/HS diet. Neutrophils (CD11b⁺ Ly-6G⁺) were not altered on day 5 but tended to be decreased at 8 wk after HF/HS diet. The number of macrophages (CD11b⁺ F4/80⁺) was not changed during HF/HS diet, but CD11c⁺ macrophages (CD11b⁺ F4/80⁺ CD11c⁺) were gradually increased on day 5 after HF/HS diet and were significantly augmented at 8 wk after HF/HS diet. Inflammatory monocytes (CD11b⁺ Ly-6C^hi) were not altered on day 5, but they tended to increase at 8 wk after HF/HS diet. The proportion of adipose EGFP-positive cells in macrophages (CD11b⁺ F4/80⁺) was unchanged during HF/HS diet feeding (Table S1). These results indicate that the highly mobile LysM^EGFP-positive cells in the very early stage of obesity would be macrophages, which would be the initial phenomenon observed during the course of obesity-induced chronic inflammation.

Up-Regulation of Adipocyte S100A8 mRNA Expression in the Very Early Stage of Obesity.

Metabolic parameters (Fig. 3A) and adipose tissue mRNA expression levels (Fig. 3B) were examined to understand the mechanism for the increased mobility of adipose macrophages at day 5 after HF/HS diet. Body weights of the HF/HS diet group were similar to those of the NC diet group at day 5, whereas they were significantly elevated at 8 wk after HF/HS diet. Weights of WAT were slightly but significantly increased in the HF/HS group compared with the NC group at day 5 and markedly increased at 8 wk after HF/HS diet. Plasma glucose concentrations were not changed at day 5 after HF/HS diet, but were elevated at 8 wk after HF/HS diet. There were no significant differences in plasma triglyceride levels between NC and HF/HS groups. Levels of plasma total cholesterol were significantly elevated in the HF/HS group at day 5, and the increase was maintained at 8 wk after HF/HS diet. These results suggest that metabolic status is largely unchanged at day 5 after HF/HS diet, although there are marked differences observed between NC and HF/HS groups at 8 wk.

Fig. 3.

Metabolic characteristics of LysM^EGFP mice after HF/HS diet feeding. (A) Metabolic parameters of LysM^EGFP mice fed with NC or HF/HS diet for 5 d (5d) or 8 wk (8w). (B) Gene expressions in epididymal adipose tissue under NC or HF/HS diets. (C) S100A8 mRNA levels in MAF and SVF under NC or HF/HS diet. (D) S100A8 protein levels in epididymal adipose tissue under NC or HF/HS diets. Data are expressed as means ± SEM; n = 6 per group (*P < 0.05, **P < 0.01, and ***P < 0.001).

We next examined the inflammatory gene expressions in WAT to explore the early response leading to enhanced migration of macrophages and resultant adipose chronic inflammation (Fig. 3B). Among a large number of molecules we tested, only one molecule, S100A8, one of the alarmins, was significantly up-regulated in mRNA level (Fig. 3B) just 5 d after HF/HS diet feeding, whereas the other alarmin molecules such as S100A9, HMGB1, and IL-1α were not altered at day 5 after HF/HS diet. Several chemokines and proinflammatory cytokines, such as CCL2, CCL3, and IL-1α (Fig. 3B), which already have been reported to be critically involved in obesity-induced chronic inflammation of adipose tissues, were all elevated in the later phase of obesity (8 wk) but essentially unaltered in the very early stage (day 5). These results suggest the possibility that S100A8 contributes to the very early pathological event in adipose inflammation. To examine the cell type expressing S100A8, WAT was next fractionated into mature adipocyte fraction (MAF) and stromal vascular fraction (SVF; Fig. 3C). A significant increase of S100A8 mRNA level was observed in the MAF but not in the SVF on day 5 after HF/HS diet. Moreover, similar to the mRNA changes, S100A8 protein level was increased on day 5 after HF/HS diet and was further augmented at 8 wk after HF/HS diet (Fig. 3D), suggesting that adipose S100A8 immediately responded to the HF/HS diet.

Inflammatory Responses to S100A8 in RAW264.7 Macrophages and 3T3-L1 Adipocytes.

To clarify the functional role of S100A8 in obese adipose tissue, chemotaxis assay was next performed by using a ready-made chemotaxis assay chamber, EZ-TAXIScan, which is useful to assess chemotactic activity visually and quantitatively (Fig. 4A and Movies S4–S6). As a result, a macrophage cell line, RAW264.7, exhibited a positive chemotaxis toward recombinant S100A8 proteins in a concentration-dependent manner, as previously reported (24).

Fig. 4.

S100A8-induced inflammatory responses in RAW264.7 cells and 3T3-L1 adipocytes. (A) Recombinant S100A8 protein-induced in vitro dynamic chemotaxis of RAW264.7 cells visualized by EZ-TAXIScan. RAW264.7 cells were loaded onto the lower chamber. The upper chambers were filled with medium (Cont, control; Left; Movie S4) or medium containing 1 μg/mL of S100A8 (Middle; Movie S5) or 10 μg/mL of S100A8 (Right; Movie S6). Mean migration speed (Left) and cell number (Right) are shown (Lower). Data points (n = 180 per group) and bars represent the values for individual cells compiled from six independent experiments and averages, respectively (Left). Data are expressed as means ± SEM and individual values (n = 6; Right). (B) Changes of inflammatory gene expressions in response to stimulation with recombinant S100A8 protein (1 or 10 μg/mL) or LPS (1 or 10 ng/mL) in the absence or presence of the LPS inhibitor polymyxin B (Poly B; 25 μg/mL) in RAW264.7 cells (n = 4 per group). (C) Changes of inflammatory gene expression in response to stimulation with recombinant S100A8 protein (1 or 10 μg/mL) or LPS (1 or 10 ng/mL) in the absence or presence of polymyxin B (Poly B; 25 μg/mL) in 3T3-L1 adipocytes (n = 4 per group). SAA3, serum amyloid A3. Data are expressed as means ± SEM (*P < 0.05, **P < 0.01, and ***P < 0.001).

Treatment with recombinant S100A8 protein increased TNF-α and MCP-1 mRNA levels in RAW264.7 cells (Fig. 4B, lanes 1–3). To rule out the effect of endotoxin contamination in recombinant S100A8 protein, RAW264.7 cells were treated with the LPS inhibitor polymyxin B. S100A8-induced increases of TNF-α and MCP-1 mRNA levels were partly reduced by polymyxin B (Fig. 4B, lanes 4–6), whereas polymyxin B efficiently blocked the LPS-induced increases of TNF-α and MCP-1 mRNA levels (Fig. 4B, lanes 7–12). In addition, also in 3T3-L1 adipocytes, recombinant S100A8 protein induced mRNA expression levels of serum amyloid A3, MCP-1, chemokine (C-X-C motif) ligand 1 (CXCL1), and CXCL5 in the presence of polymyxin B (Fig. 4C, lanes 4–6), whereas such increases were not observed in LPS treatment under polymyxin B (Fig. 4C, lanes 10–12). These results suggest that adipose S100A8 acts not only as a chemoattractant to macrophages but also as a proinflammatory factor on macrophages and adipocytes.

In Vivo Effect of S100A8 on the Mobility of LysM^EGFP-Positive Cells in Adipose Tissues.

We next examined whether S100A8 can also accelerate macrophage mobility in adipose tissues in vivo by using LysM^EGFP mice. Similar to the experiments shown in Fig. 1 A and B, intravital imaging of adipose tissue was conducted by adding recombinant S100A8 protein or an equal amount of LPS directly onto WAT. The mobility of adipose LysM^EGFP-positive cells was gradually increased at 1 h after S100A8 administration (Fig. 5A and Movie S7), whereas such increased mobility was not observed in the LPS administration group (Movies S8 and S9). Tracking velocity was significantly increased by the treatment with recombinant S100A8 protein after 1 h (Fig. 5B), suggesting that S100A8 is indeed functional for mobilizing adipose tissue-resident macrophages in vivo.

Fig. 5.

Effect of S100A8 and S100A8 antibody on the mobility of LysM^EGFP-positive cells and diet-induced insulin resistance. Intravital two-photon imaging (A) and mean tracking velocity (B) in epididymal adipose tissues of LysM^EGFP mice in the absence (Left) or presence (Right) of recombinant S100A8 protein (Movie S7). Colored lines show the associated trajectories of LysM^EGFP-positive cells. (Scale bar: 100 μm.) Data points (n = 469 from four mice for control; n = 565 from five mice for S100A8) and bars represent the values for individual cells and averages, respectively (***P < 0.001). (C) Intravital two-photon images after treatment of IgG isotype control antibody (Left) or S100A8 antibody (Right) in epididymal adipose tissues of LysM^EGFP mice fed with HF/HS diet for 5 d (Movie S10). Colored lines show the associated trajectories of LysM^EGFP-positive cells. (D) Line graph of percent changes in mean tracking velocity of LysM^EGFP-positive cells after antibody administration under HF/HS diet for 5 d. The mobility of LysM^EGFP-positive cells was set in 100% at pretreatment (n = 4 per group). Data are expressed as means ± SEM (**P < 0.01 and ***P < 0.001 vs. IgG control group at the same period). (E) Mean tracking velocity of LysM^EGFP-positive cells after the indicated antibody treatment under HF/HS diet for 5 d. Data points (n = 677 for IgG control Ab; n = 811 for S100A8 Ab) and bars represent the values for individual cells compiled from four independent experiments and averages, respectively (***P < 0.001). (F) Line graph of plasma glucose under insulin tolerance test. C57BL6 mice were treated with IgG isotype control antibody or S100A8 antibody and fed with HF/HS diet from 11 wk of age. For insulin tolerance test, mice received 0.75 U/kg insulin i.p. on day 21 (n = 4 per group). Data are expressed as means ± SEM (*P < 0.05 vs. IgG control group). (G) Area under the curve (AUC) of insulin tolerance test calculated from 0 min to 60 min (n = 4 per group). Data are expressed as means ± SEM (*P < 0.05).

In Vivo Effect of S100A8 Antibody on the Mobility of LysM^EGFP-Positive Cells in Adipose Tissues and Insulin Sensitivity.

To examine whether enhanced macrophage mobility at an early stage of obesity was caused by adipose S100A8, we tested the effect of inhibition of endogenous S100A8 by its antibody on the mobility of adipose LysM^EGFP-positive cells on day 5 after HF/HS diet. The intravital imaging was performed until 150 min after the administration of S100A8 antibody or IgG isotype-matched control antibody (Fig. 5 C–E). As a result, the S100A8 antibody treatment almost completely blocked the HF/HS diet-induced mobility of adipose LysM^EGFP-positive cells (Fig. 5C, Right, and Movie S10), whereas no significant changes on LysM^EGFP-positive cell mobility was observed when IgG isotype control antibody was administered (Fig. 5C, Left, and Movie S10). Fig. 5D shows the time course of mean tracking velocity integrating every 30 min and indicates that S100A8 antibody significantly suppressed the HF/HS diet-induced activation of adipose LysM^EGFP-positive cell mobility at 90–120 min and 120–150 min. Mean tracking velocity of adipose LysM^EGFP-positive cells was significantly suppressed later than 90 min after the treatment of S100A8 antibody (Fig. 5E). The present results clearly indicate that adipose S100A8 in the early stage of obesity did mobilize LysM^EGFP-positive macrophages in adipose tissues.

Finally, mice were administered anti-S100A8 antibody or IgG isotype control antibody under HF/HS diet and were challenged with an insulin tolerance test (Fig. 5 F and G). As a result, anti-S100A8 antibody treatment significantly improved insulin sensitivity compared with the IgG isotype control antibody group without changes in body weight and fat mass, suggesting the pathophysiological significance of S100A8-mediated macrophage control during chronic inflammation in obese adipose tissues.

Discussion

The major findings of the present study of intravital imaging of adipose tissues during development of obesity were as follows. (i) Adipose LysM^EGFP-positive cells mainly consisted of macrophages, and their mobility was activated just 5 d after HF/HS diet, before the increase of adipose LysM^EGFP-positive cells. (ii) S100A8 mRNA and protein levels were increased in adipose tissues just 5 d after HF/HS diet. (iii) S100A8 accelerated migration of RAW264.7 cells and activated inflammatory response in macrophages and adipocytes. (iv) S100A8 activated the mobility of adipose LysM^EGFP-positive cells whereas its antibody effectively suppressed the HF/HS diet-induced activity of adipose LysM^EGFP-positive cell mobility. Finally, (v) Administration of S100A8 antibody improved insulin sensitivity under HF/HS diet.

Obesity is recognized as a state of chronic low-grade inflammation (5, 6). Accumulating evidence demonstrates that the progression of adipose tissue inflammation is strongly associated with overreactive innate and adaptive immune responses (6 ⇓⇓⇓⇓–11). However, it is difficult to predict the balance of immune responses by conventional research using immunohistological and flow cytometric procedures because cellular responses to inflammation are particularly dynamic and complicated. To analyze the dynamic processes underlying adipose inflammation may require intravital imaging of intact tissues in living animals. Here, for the first time to our awareness, we succeeded in visualizing immune cell dynamics from the beginning of the development of obesity. Strikingly, the mobility of LysM^EGFP-positive cells infiltrating in adipose tissues was increased in the very early stage of obesity without the prominent hypertrophy of adipocytes and the increase in the number of macrophage accumulation. However, it remains to be elucidated where such activated LysM^EGFP-positive cells relocate to different positions in adipose tissues. The present study also could not provide evidence whether LysM^EGFP-positive cells interact with other cell types in adipose tissues. There are several technical limitations at present, including that intravital imaging in adipose tissues cannot be performed more than 3 h to obtain a live image. Further technical advances may provide novel insights into the activated macrophages and other immune cells in adipose tissues.

On the contrary, it should be noted that LysM^EGFP-positive cells in adipose tissues contain some undefined fractions (∼30%), although most of the EGFP-positive cells are macrophages (∼60%; Fig. 2A), which limits the interpretation of the present results. LysM is specifically expressed in myelomonocytic cells such as monocytes/macrophages, neutrophils, and granulocytes (22). In other experiments, proportions of eosinophils, basophils, and dendritic cells were less than 1% among LysM^EGFP-positive cells in adipose tissues under NC diet. Generation of monocyte/macrophage-specific fluorescent protein-transgenic mice would be necessary to better understand the visible activity of macrophages in adipose tissue. Eosinophils have been shown to migrate into adipose tissue and maintain metabolic homeostasis in mice (12). Talukdar et al. clearly demonstrated that neutrophils infiltrated into adipose tissue just 3 d after 60% high-fat diet, and deletion of neutrophil elastase decreased adipose tissue inflammation in their diet-induced obesity mice (11). However, the present study shows that the proportion of adipose neutrophils to stromal vascular cells (SVCs) was not increased after 5 d of HF/HS diet feeding (Fig. 2C). The reason for such a difference remains uncertain, although the age of mice starting high-fat diet feeding and the fat composition of high-fat diet in the present study (30% fat) were not similar to those in the previous study (11). It might be possible that immune response may vary with murine age and fat composition of meals.

Increases of chemokines and their related molecules may be involved in the development of adipose tissue inflammation. Several potential modulators of inflammatory and immune responses have been identified in the development of obesity, e.g., CCL2 (MCP-1), CCL3 [macrophage inflammatory protein-1α (MIP-1α)], CCL8 (MCP-2), CCL7 (MCP-3), CCL5 (RANTES), and CXCL14 (13, 25, 26), and accumulating evidence shows the crucial role of the CCL2/CCR2 pathway in adipose inflammation (13, 14), although the initial factors triggering the inflammation in this context had still been unknown. The present study suggested that S100A8 is a critical component regulating the initial step of inflammatory cascade in obese adipose tissue. Importantly, S100A8 was increased in the very early stage of obesity, whereas no changes in CCL2 (MCP-1) and CCL3 (MIP-1α) mRNA expression levels were observed in this stage (Fig. 3 B and C). However, we cannot completely exclude the possibility that MCP-1 or other chemokines would be secreted locally with limited concentrations that were undetectable by quantitative PCR of whole adipocyte fractions. The mobility of adipose LysM^EGFP-positive cells and the migration of macrophages were increased by S100A8 administration (Figs. 4A and 5 A and B), whereas the mobility of adipose LysM^EGFP-positive cells was suppressed by treatment with S100A8 antibody under HF/HS diet (Fig. 5 D and E). However, the present study could not fully show whether S100A8 directly or indirectly induced mobility of adipose LysM^EGFP-positive cells. The possibility could not be excluded that S100A8 induces the local production of MCP-1 and/or other chemoattractant factors that mobilize adipose tissue-resident LysM^EGFP-positive cells. LysM^EGFP mice with macrophage-specific ablation of MCP-1 and/or other chemotactic factors will be needed in the future to determine whether S100A8 possesses a direct effect on mobility of adipose LysM^EGFP-positive cells.

As shown in Fig. 5 C and D, S100A8 antibody-mediated suppressive effect on the HF/HS diet-induced mobility of adipose LysM^EGFP-positive cells was not observed immediately (it took approximately 90 min), which may reflect the time course for penetrating S100A8 antibody into the tissues or that necessary for S100A8 antibody to block the release of other chemokines locally. In vitro examinations also showed the proinflammatory effect of S100A8 on macrophages and adipocytes (Fig. 4 B and C). A series of the present data may provide the possibility that adipose S100A8 increases in the very early stage of obesity, recruits macrophages into adipose tissue, and induces local inflammation by its effect on adipocytes and macrophages. S100A8 antibody treatment also enhanced insulin sensitivity under HF/HS diet (Fig. 5 F and G), suggesting that increased mobility of adipose macrophages via S100A8 impact on the development of insulin resistance and diabetes in obesity. Generation and analysis of adipose-specific S100A8 transgenic and/or KO mice would be required to confirm the significant role of S100A8 in adipose inflammatory response and insulin resistance.

Two members of the S100 protein family, S100A8 and S100A9, have been identified as important endogenous alarmins that are released from the activated phagocytes and are recognized by TLR4 on monocytes (16). Significance of S100A8 has not been clarified in adipose tissue, but our previous cDNA microarray analysis identified, to our knowledge for the first time, the increased expression levels of S100A8 in obese adipose tissue (27), and we recently reported that adipose S100A8 mRNA level was highly expressed in MAF in obesity (28). Moreover, circulating S100A8/A9 heterodimer complex (calprotectin) positively correlated with adiposity in human subjects (28, 29). Interestingly, PPAR-γ agonists significantly reduced adipose S100A8 mRNA level and plasma S100A8/A9 complex (calprotectin) concentration in obese diabetic mice (27) and human subjects (30). Furthermore, the present study demonstrated, to our knowledge for the first time, that blocking against S100A8 ameliorated diet-induced insulin resistance. These results suggest that S100A8 plays an essential role in the inflammatory responses of obese adipose tissue and contributes to the pathogenesis of obesity from the very early stage. Chronic inflammation usually takes a long course, and intricate inflammatory chain reactions will occur when it has happened. In light of the treatment, it would be desirable if we can block the initial trigger and completely halt the very early event of this vicious cycle. From the point of view, S100A8 would become a good target for a new line of therapeutics against obesity-induced chronic inflammation.

Materials and Methods

Animals.

LysM^EGFP mice (22) were maintained with NC diet. At 10 wk of age, LysM^EGFP mice were continued with NC diet or were changed to HF/HS diet (F2HFHSD; Oriental Yeast) (31). LysM^EGFP mice were analyzed at 5 d and 8 wk after NC or HF/HS diet. Mice were anesthetized with an i.p. injection of a mixture of medetomidine (0.3 mg/kg body weight), midazolam (4 mg/kg body weight), and butorphanol tartrate (5 mg/kg body weight). Blood samples were collected from the inferior vena cava after 16 h of fasting, and epididymal WAT was excised for RNA extraction and histological analysis. Plasma glucose, triglyceride, and cholesterol concentrations were measured with glucose CII, triglyceride E, and cholesterol E tests (Wako Pure Chemical Industries), respectively. To analyze adipocyte size, low-power field images were acquired from five animals in each group, after which the area of 35 adipocytes in each field were measured. The histogram shown was constructed from 175 cells from each group. For insulin tolerance test study, C57BL6 mice were injected i.p. with IgG isotype control antibody or S100A8 antibody at 50 μg per mouse three times per week and started feeding HF/HS diet from 11 wk of age. Mice were subjected to insulin tolerance test by i.p. administration of 0.75 U/kg insulin on day 21. Mice were kept in rooms set at 22 °C with a 12:12 h dark/light cycle (lights on from 800 h to 2000 h).

Multiphoton Intravital Adipose Tissue Imaging.

The imaging system was composed of a multiphoton microscope (A1-MP; Nikon) driven by a laser (Chameleon Vision II Ti: Sapphire; Coherent) tuned to 800–840 nm and an inverted microscope equipped with a 20× multiimmersion objective lens (Plan Fluor; N.A., 0.75; Nikon). Fluorescent signals were detected through band-pass emission filters at 500/50 nm (for GFP), at 563/25 nm [for 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY)], and at 601/56 nm (for Qtracker). For adipose tissue imaging, mice were anesthetized with isoflurane, and epididymal WAT was carefully exposed and incubated with BODIPY (Life Technologies). Vessels were visualized by the i.v. injection of Qtracker (Life Technologies) just before imaging. Epididymal WAT was covered with saline solution-moistened sheets of paper, and mice were set in a heating box to maintain body temperature during intravital imaging. Sequential images were taken every 30 s for 1 h. Image stacks were collected at a 5-μm vertical step size. To quantify leukocyte dynamics, raw imaging data were processed as follows: EGFP-positive cells in sections were tracked, and mobility was calculated as the mean tracking velocity and mean tracking length by using IMARIS 3D and 4D real-time image processing and analysis software (Bitplane) (32, 33).

Evaluation of the Effect of S100A8 on Adipose Tissue by Intravital Imaging.

For the recombinant S100A8 protein treatment study, the epididymal WAT of LysM^EGFP mice was exposed, and human recombinant S100A8 protein (10 μg per mouse; GIOTTO) was administered locally onto the epididymal WAT after staining. Sequential images of WAT were taken every 30 s from 1 h to 2 h after treatment with or without S100A8 recombinant protein. To assess the effects of S100A8 antibody on the mobility of macrophages induced by HF/HS feeding, the epididymal WAT of LysM^EGFP mice under HF/HS diet for 5 d was exposed, and S100A8 antibody (50 μg per mouse, clone 335806; R&D Systems) or control rat IgG (50 μg per mouse; clone 141945; R&D Systems) was administered directly onto the epididymal WAT after staining. Sequential images of WAT were taken every 30 s from −30 min to 150 min in the course of S100A8 antibody treatment. The mean tracking velocity of LysM^EGFP-positive cells before antibody treatment was set as 100%. Continuous time-lapse imaging after antibody treatment was divided into five sections of 30-min intervals and quantified.

Isolation of the SVF and Flow Cytometry.

To isolate the SVF, minced adipose tissue in Krebs–Ringer bicarbonate Hepes buffer [120 mmol/L NaCl, 4 mmol/L KH₂PO₄, 1 mmol/L MgSO₄, 1 mmol/L CaCl₂, 10 mmol/L NaHCO₃, 30 mmol/L Hepes, 20 μmol/L adenosine, and 4% (wt/vol) BSA (Calbiochem)] was centrifuged to remove blood cells and then incubated in a collagenase solution (2 mg/mL) at 37 °C for 30 min under continuous shaking. The tissue was then centrifuged, and the resultant pellet containing the SVF was filtered through 100-μm mesh and finally resuspended in PBS solution. Cells were first incubated with anti-mouse CD16/CD32 (BioLegend) for 15 min at 4 °C to block Fc receptors and then incubated with labeled monoclonal or isotype control antibody. Flow cytometric analysis was performed by using a FACSCanto II flow cytometer (BD Biosciences) and FlowJo software (Tree Star). The antibodies used were as follows: anti-CD11b (clone M1/70; BioLegend), anti-Ly-6G (clone 1A8; BioLegend), anti-F4/80 (clone BM8; BioLegend), anti-Ly-6C (clone AL-21 ;BD Pharmingen), and anti-CD11c (clone N418; eBioscience).

Quantitative Real-Time PCR.

Total RNA was isolated by using RNA STAT-60 (Tel-Test) according to the protocol supplied by the manufacturer. The quality and quantity of total RNA were determined by using an ND-1000 Spectrophotometer (Thermo Scientific). First-strand cDNA was synthesized from 1 μg of total RNA using Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics). Real-time quantitative PCR amplification was conducted with a Viia7 real-time PCR system (Life Technologies) using Thunderbird quantitative PCR mix (Toyobo) according to the protocol recommended by the manufacturer. The results for each sample were normalized to the respective 36B4 mRNA levels. The primer sets were as follows: mouse 36B4, 5′-GGCCAATAAGGTGCCAGCT-3′ (forward) and 5′-TGATCAGCCCGAAGGAGAAG-3′ (reverse); mouse S100A8, 5′-GACAATGCAATTAACTTCGAGGAG-3′ (forward) and 5′-TGTGGCTGTCTTTGTGAGATGC-3′(reverse); mouse S100A9, 5′-CACAGTTGGCAACCTTTATG-3′ (forward) and 5′-CAGCTGATTGTCCTGGTTTG-3′ (reverse); mouse HMGB1, 5′-TGGCAAAGGCTGACAAGGCTC-3′ (forward) and 5′-GGATGCTCGCCTTTGATTTTGG-3′ (reverse); mouse CCL2, 5′-CCACTCACCTGCTGCTACTCAT-3′ (forward) and 5′-TGGTGATCCTCTTGTAGCTCTCC-3′ (reverse); mouse CCL3, 5′-CCAAGTCTTCTCAGCGCCAT-3′ (forward) and 5′-GAATCTTCCGGCTGTAGGAGAAG-3′ (reverse); and mouse IL-1α, 5′-CAAACTGATGAAGCTCGTCA-3′ (forward) and 5′-TCTCCTTGAGCGCTCACGAA-3′ (reverse).

Immunoblotting.

Epididymal adipose tissues were frozen in liquid nitrogen immediately and lysed in RIPA buffer (50 mM Tris⋅HCl, pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 1% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitor mixture (Cell Signaling). Lysates were incubated in lysis buffer for 30 min on ice before centrifugation and collection of the supernatant. Total protein concentration was determined by using the BCA-200 Protein Assay Kit (Pierce). For immunoblotting, protein samples were incubated with a reducing sample buffer [2% (wt/vol) SDS, 50 mmol/L Tris⋅HCl, pH 6.8, and 10% (vol/vol) glycerol with 2-mercaptoethanol] and separated by 4–20% gradient SDS/PAGE. Human recombinant S100A8 protein (GIOTTO) was used as a positive control. Immunoblot analysis was performed first with antibody S100A8 (Santa Cruz), followed by incubation with secondary antibody conjugated with horseradish peroxidase. The ECL Prime system (GE Healthcare) was used for detection of the protein signal.

Cell Culture.

RAW264.7 cells were maintained in the regular medium (DMEM supplemented with 10% FCS and 1% penicillin/streptomycin (P/S)]. RAW264.7 cells were plated in 12-well plates (2 × 10⁵ cells per well) and were cultured for 24 h before the experiment. RAW264.7 cells were treated with or without 1 and 10 μg/mL of S100A8 (GIOTTO) or 1 and 10 ng/mL of LPS (Sigma-Aldrich) in the presence or absence of 25 μg/mL of polymyxin B (Sigma-Aldrich), and cells were harvested after the indicated treatment for 24 h. 3T3-L1 cells were maintained and differentiated, as described previously (34). Briefly, cells were grown to confluence and differentiated by induction medium (DMEM supplemented with 10% FCS and 1% P/S containing 0.5 mM 1-methyl-3-isobutylxanthine, 1 μM dexamethasone, and 5 μg/mL insulin). After incubation with the induction medium for 48 h, the medium was changed to maintenance medium (DMEM supplemented with 10% FCS and 1% P/S). On day 7 after differentiation, 3T3-L1 adipocytes were incubated with or without 1 and 10 μg/mL of S100A8 (GIOTTO) or 1 and 10 ng/mL of LPS (Sigma-Aldrich) in the presence of 25 μg/mL of polymyxin B (Sigma-Aldrich), and cells were harvested after the indicated treatment for 24 h.

Chemotaxis Assay.

Chemotaxis experiments were conducted in an EZ-TAXIScan chamber according to the manufacturer’s protocol (GE Healthcare), as described previously (32, 33). Briefly, EZ-TAXIScan is a visually accessible chemotactic chamber in which one compartment contains the ligand (S100A8) and another compartment contains the cells (RAW264.7 cells), and these compartments are connected by a microchannel. A stable concentration gradient of chemoattractant can be reproducibly formed and maintained through the channel without medium flow. Images of RAW264.7 cells undergoing chemotaxis in response to S100A8 were obtained at 1-min time-lapse intervals for 3 h (37 °C). Sequential image data were processed by using the ImageJ program (National Institutes of Health) with an add-on program, MT Track J.

Statistical Analysis.

The statistical significance of differences between groups was determined by two-tailed t tests or Dunnett test for Gaussian-like distributions. The Wilcoxon rank-sum test or the Steel test was used to calculate the P values for highly skewed distributions. P values less than 0.05 were considered significant. All analyses were performed with the JMP Statistical Discovery Software (version 11.0; SAS Institute).

Study Approval.

All animal studies were approved by the Ethics Review Committee for Animal Experimentation of Osaka University School of Medicine and also conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

Acknowledgments

We thank Miyuki Nakamura (Department of Metabolic Medicine, Graduate School of Medicine, Osaka University) for excellent technical assistance, and all members of the III^rd laboratory (Adiposcience Laboratory), Department of Metabolic Medicine, Osaka University, for helpful discussion on the project. This work was supported in part by Grant-in-Aid for Scientific Research (C) no. 22590979 (to N.M.), Grant-in-Aid for Scientific Research (B) no. 24390238 (to I.S.), Grant-in-Aid for Scientific Research (A) no. 25253070 (to M.I.), and Grants-in-Aid for Scientific Research on Innovative Areas no. 22126008 (to T.F.) and no. 22113007 (to M.I.), from the Ministry of Education, Science, Sports and Culture of Japan.

Footnotes

↵¹To whom correspondence may be addressed. Email: norikazu_maeda{at}endmet.med.osaka-u.ac.jp or mishii{at}icb.med.osaka-u.ac.jp.

Author contributions: R.S., N.M., T.F., M.I., and I.S. designed research; R.S., S.F., N.M., Y.T., K.M., T.M., H. Nakatsuji, J.K., and Y.M. performed research; R.S., N.M., H. Nishizawa, K.K., J.K., Y.M., and M.I. analyzed data; and R.S., N.M., T.F., M.I., and I.S. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1409480112/-/DCSupplemental.

View Abstract

References

↵
1. Wellen KE,
2. Hotamisligil GS
(2005) Inflammation, stress, and diabetes. J Clin Invest 115(5):1111–1119
.
OpenUrl CrossRef PubMed
↵
1. Badman MK,
2. Flier JS
(2005) The gut and energy balance: Visceral allies in the obesity wars. Science 307(5717):1909–1914
.
OpenUrl Abstract/FREE Full Text
↵
1. Gesta S,
2. Tseng YH,
3. Kahn CR
(2007) Developmental origin of fat: Tracking obesity to its source. Cell 131(2):242–256
.
OpenUrl CrossRef PubMed
↵
1. Matsuzawa Y,
2. Funahashi T,
3. Kihara S,
4. Shimomura I
(2004) Adiponectin and metabolic syndrome. Arterioscler Thromb Vasc Biol 24(1):29–33
.
OpenUrl Abstract/FREE Full Text
↵
1. Xu H, et al.
(2003) Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 112(12):1821–1830
.
OpenUrl CrossRef PubMed
↵
1. Weisberg SP, et al.
(2003) Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 112(12):1796–1808
.
OpenUrl CrossRef PubMed
↵
1. Murano I, et al.
(2008) Dead adipocytes, detected as crown-like structures, are prevalent in visceral fat depots of genetically obese mice. J Lipid Res 49(7):1562–1568
.
OpenUrl Abstract/FREE Full Text
↵
1. Lumeng CN,
2. Bodzin JL,
3. Saltiel AR
(2007) Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest 117(1):175–184
.
OpenUrl CrossRef PubMed
↵
1. Nishimura S, et al.
(2009) CD8+ effector T cells contribute to macrophage recruitment and adipose tissue inflammation in obesity. Nat Med 15(8):914–920
.
OpenUrl CrossRef PubMed
↵
1. DeFuria J, et al.
(2013) B cells promote inflammation in obesity and type 2 diabetes through regulation of T-cell function and an inflammatory cytokine profile. Proc Natl Acad Sci USA 110(13):5133–5138
.
OpenUrl Abstract/FREE Full Text
↵
1. Talukdar S, et al.
(2012) Neutrophils mediate insulin resistance in mice fed a high-fat diet through secreted elastase. Nat Med 18(9):1407–1412
.
OpenUrl CrossRef PubMed
↵
1. Wu D, et al.
(2011) Eosinophils sustain adipose alternatively activated macrophages associated with glucose homeostasis. Science 332(6026):243–247
.
OpenUrl Abstract/FREE Full Text
↵
1. Kanda H, et al.
(2006) MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest 116(6):1494–1505
.
OpenUrl CrossRef PubMed
↵
1. Weisberg SP, et al.
(2006) CCR2 modulates inflammatory and metabolic effects of high-fat feeding. J Clin Invest 116(1):115–124
.
OpenUrl CrossRef PubMed
↵
1. Park JS, et al.
(2004) Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein. J Biol Chem 279(9):7370–7377
.
OpenUrl Abstract/FREE Full Text
↵
1. Vogl T, et al.
(2007) Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock. Nat Med 13(9):1042–1049
.
OpenUrl CrossRef PubMed
↵
1. Foell D,
2. Wittkowski H,
3. Vogl T,
4. Roth J
(2007) S100 proteins expressed in phagocytes: A novel group of damage-associated molecular pattern molecules. J Leukoc Biol 81(1):28–37
.
OpenUrl Abstract/FREE Full Text
↵
1. Shi H, et al.
(2006) TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 116(11):3015–3025
.
OpenUrl CrossRef PubMed
↵
1. Lee JY, et al.
(2004) Saturated fatty acid activates but polyunsaturated fatty acid inhibits Toll-like receptor 2 dimerized with Toll-like receptor 6 or 1. J Biol Chem 279(17):16971–16979
.
OpenUrl Abstract/FREE Full Text
↵
1. Miller YI, et al.
(2003) Minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD-2, and inhibits phagocytosis of apoptotic cells. J Biol Chem 278(3):1561–1568
.
OpenUrl Abstract/FREE Full Text
↵
1. Schaefer L, et al.
(2005) The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest 115(8):2223–2233
.
OpenUrl CrossRef PubMed
↵
1. Faust N,
2. Varas F,
3. Kelly LM,
4. Heck S,
5. Graf T
(2000) Insertion of enhanced green fluorescent protein into the lysozyme gene creates mice with green fluorescent granulocytes and macrophages. Blood 96(2):719–726
.
OpenUrl Abstract/FREE Full Text
↵
1. Nishimura S, et al.
(2008) In vivo imaging in mice reveals local cell dynamics and inflammation in obese adipose tissue. J Clin Invest 118(2):710–721
.
OpenUrl PubMed
↵
1. Fujiu K,
2. Manabe I,
3. Nagai R
(2011) Renal collecting duct epithelial cells regulate inflammation in tubulointerstitial damage in mice. J Clin Invest 121(9):3425–3441
.
OpenUrl CrossRef PubMed
↵
1. Huber J, et al.
(2008) CC chemokine and CC chemokine receptor profiles in visceral and subcutaneous adipose tissue are altered in human obesity. J Clin Endocrinol Metab 93(8):3215–3221
.
OpenUrl CrossRef PubMed
↵
1. Nara N, et al.
(2007) Disruption of CXC motif chemokine ligand-14 in mice ameliorates obesity-induced insulin resistance. J Biol Chem 282(42):30794–30803
.
OpenUrl Abstract/FREE Full Text
↵
1. Hiuge-Shimizu A, et al.
(2011) Dynamic changes of adiponectin and S100A8 levels by the selective peroxisome proliferator-activated receptor-gamma agonist rivoglitazone. Arterioscler Thromb Vasc Biol 31(4):792–799
.
OpenUrl Abstract/FREE Full Text
↵
1. Sekimoto R, et al.
(2012) High circulating levels of S100A8/A9 complex (calprotectin) in male Japanese with abdominal adiposity and dysregulated expression of S100A8 and S100A9 in adipose tissues of obese mice. Biochem Biophys Res Commun 419(4):782–789
.
OpenUrl CrossRef PubMed
↵
1. Catalán V, et al.
(2011) Increased levels of calprotectin in obesity are related to macrophage content: impact on inflammation and effect of weight loss. Mol Med 17(11-12):1157–1167
.
OpenUrl PubMed
↵
1. Nakatsuji H,
2. Kishida K,
3. Funahashi T,
4. Shimomura I, Senri Study II Group
(2012) Three-month treatment with pioglitazone reduces circulating levels of S100A8/A9 (MRP8/14) complex, a biomarker of inflammation, without changes in body mass index, in type 2 diabetics with abdominal obesity. Diabetes Res Clin Pract 95(3):e58–e60
.
OpenUrl CrossRef PubMed
↵
1. Maeda N, et al.
(2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8(7):731–737
.
OpenUrl CrossRef PubMed
↵
1. Ishii M, et al.
(2009) Sphingosine-1-phosphate mobilizes osteoclast precursors and regulates bone homeostasis. Nature 458(7237):524–528
.
OpenUrl CrossRef PubMed
↵
1. Kikuta J, et al.
(2013) Sphingosine-1-phosphate-mediated osteoclast precursor monocyte migration is a critical point of control in antibone-resorptive action of active vitamin D. Proc Natl Acad Sci USA 110(17):7009–7013
.
OpenUrl Abstract/FREE Full Text
↵
1. Maeda N, et al.
(2001) PPARgamma ligands increase expression and plasma concentrations of adiponectin, an adipose-derived protein. Diabetes 50(9):2094–2099
.
OpenUrl Abstract/FREE Full Text

Article Alerts

Email Article

Citation Tools

Request Permissions

Proceedings of the National Academy of Sciences: 112 (16)

Table of Contents

Submit

Article Classifications

Biological Sciences
Medical Sciences

[1] ↵
Wellen KE,
Hotamisligil GS
(2005) Inflammation, stress, and diabetes. J Clin Invest 115(5):1111–1119
.
OpenUrl CrossRef PubMed

[2] Wellen KE,

[3] Hotamisligil GS

[4] ↵
Badman MK,
Flier JS
(2005) The gut and energy balance: Visceral allies in the obesity wars. Science 307(5717):1909–1914
.
OpenUrl Abstract/FREE Full Text

[5] Badman MK,

[6] Flier JS

[7] ↵
Gesta S,
Tseng YH,
Kahn CR
(2007) Developmental origin of fat: Tracking obesity to its source. Cell 131(2):242–256
.
OpenUrl CrossRef PubMed

[8] Gesta S,

[9] Tseng YH,

[10] Kahn CR

[11] ↵
Matsuzawa Y,
Funahashi T,
Kihara S,
Shimomura I
(2004) Adiponectin and metabolic syndrome. Arterioscler Thromb Vasc Biol 24(1):29–33
.
OpenUrl Abstract/FREE Full Text

[12] Matsuzawa Y,

[13] Funahashi T,

[14] Kihara S,

[15] Shimomura I

[16] ↵
Xu H, et al.
(2003) Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 112(12):1821–1830
.
OpenUrl CrossRef PubMed

[17] Xu H, et al.

[18] ↵
Weisberg SP, et al.
(2003) Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 112(12):1796–1808
.
OpenUrl CrossRef PubMed

[19] Weisberg SP, et al.

[20] ↵
Murano I, et al.
(2008) Dead adipocytes, detected as crown-like structures, are prevalent in visceral fat depots of genetically obese mice. J Lipid Res 49(7):1562–1568
.
OpenUrl Abstract/FREE Full Text

[21] Murano I, et al.

[22] ↵
Lumeng CN,
Bodzin JL,
Saltiel AR
(2007) Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest 117(1):175–184
.
OpenUrl CrossRef PubMed

[23] Lumeng CN,

[24] Bodzin JL,

[25] Saltiel AR

[26] ↵
Nishimura S, et al.
(2009) CD8+ effector T cells contribute to macrophage recruitment and adipose tissue inflammation in obesity. Nat Med 15(8):914–920
.
OpenUrl CrossRef PubMed

[27] Nishimura S, et al.

[28] ↵
DeFuria J, et al.
(2013) B cells promote inflammation in obesity and type 2 diabetes through regulation of T-cell function and an inflammatory cytokine profile. Proc Natl Acad Sci USA 110(13):5133–5138
.
OpenUrl Abstract/FREE Full Text

[29] DeFuria J, et al.

[30] ↵
Talukdar S, et al.
(2012) Neutrophils mediate insulin resistance in mice fed a high-fat diet through secreted elastase. Nat Med 18(9):1407–1412
.
OpenUrl CrossRef PubMed

[31] Talukdar S, et al.

[32] ↵
Wu D, et al.
(2011) Eosinophils sustain adipose alternatively activated macrophages associated with glucose homeostasis. Science 332(6026):243–247
.
OpenUrl Abstract/FREE Full Text

[33] Wu D, et al.

[34] ↵
Kanda H, et al.
(2006) MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest 116(6):1494–1505
.
OpenUrl CrossRef PubMed

[35] Kanda H, et al.

[36] ↵
Weisberg SP, et al.
(2006) CCR2 modulates inflammatory and metabolic effects of high-fat feeding. J Clin Invest 116(1):115–124
.
OpenUrl CrossRef PubMed

[37] Weisberg SP, et al.

[38] ↵
Park JS, et al.
(2004) Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein. J Biol Chem 279(9):7370–7377
.
OpenUrl Abstract/FREE Full Text

[39] Park JS, et al.

[40] ↵
Vogl T, et al.
(2007) Mrp8 and Mrp14 are endogenous activators of Toll-like receptor 4, promoting lethal, endotoxin-induced shock. Nat Med 13(9):1042–1049
.
OpenUrl CrossRef PubMed

[41] Vogl T, et al.

[42] ↵
Foell D,
Wittkowski H,
Vogl T,
Roth J
(2007) S100 proteins expressed in phagocytes: A novel group of damage-associated molecular pattern molecules. J Leukoc Biol 81(1):28–37
.
OpenUrl Abstract/FREE Full Text

[43] Foell D,

[44] Wittkowski H,

[45] Vogl T,

[46] Roth J

[47] ↵
Shi H, et al.
(2006) TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 116(11):3015–3025
.
OpenUrl CrossRef PubMed

[48] Shi H, et al.

[49] ↵
Lee JY, et al.
(2004) Saturated fatty acid activates but polyunsaturated fatty acid inhibits Toll-like receptor 2 dimerized with Toll-like receptor 6 or 1. J Biol Chem 279(17):16971–16979
.
OpenUrl Abstract/FREE Full Text

[50] Lee JY, et al.

[51] ↵
Miller YI, et al.
(2003) Minimally modified LDL binds to CD14, induces macrophage spreading via TLR4/MD-2, and inhibits phagocytosis of apoptotic cells. J Biol Chem 278(3):1561–1568
.
OpenUrl Abstract/FREE Full Text

[52] Miller YI, et al.

[53] ↵
Schaefer L, et al.
(2005) The matrix component biglycan is proinflammatory and signals through Toll-like receptors 4 and 2 in macrophages. J Clin Invest 115(8):2223–2233
.
OpenUrl CrossRef PubMed

[54] Schaefer L, et al.

[55] ↵
Faust N,
Varas F,
Kelly LM,
Heck S,
Graf T
(2000) Insertion of enhanced green fluorescent protein into the lysozyme gene creates mice with green fluorescent granulocytes and macrophages. Blood 96(2):719–726
.
OpenUrl Abstract/FREE Full Text

[56] Faust N,

[57] Varas F,

[58] Kelly LM,

[59] Heck S,

[60] Graf T

[61] ↵
Nishimura S, et al.
(2008) In vivo imaging in mice reveals local cell dynamics and inflammation in obese adipose tissue. J Clin Invest 118(2):710–721
.
OpenUrl PubMed

[62] Nishimura S, et al.

[63] ↵
Fujiu K,
Manabe I,
Nagai R
(2011) Renal collecting duct epithelial cells regulate inflammation in tubulointerstitial damage in mice. J Clin Invest 121(9):3425–3441
.
OpenUrl CrossRef PubMed

[64] Fujiu K,

[65] Manabe I,

[66] Nagai R

[67] ↵
Huber J, et al.
(2008) CC chemokine and CC chemokine receptor profiles in visceral and subcutaneous adipose tissue are altered in human obesity. J Clin Endocrinol Metab 93(8):3215–3221
.
OpenUrl CrossRef PubMed

[68] Huber J, et al.

[69] ↵
Nara N, et al.
(2007) Disruption of CXC motif chemokine ligand-14 in mice ameliorates obesity-induced insulin resistance. J Biol Chem 282(42):30794–30803
.
OpenUrl Abstract/FREE Full Text

[70] Nara N, et al.

[71] ↵
Hiuge-Shimizu A, et al.
(2011) Dynamic changes of adiponectin and S100A8 levels by the selective peroxisome proliferator-activated receptor-gamma agonist rivoglitazone. Arterioscler Thromb Vasc Biol 31(4):792–799
.
OpenUrl Abstract/FREE Full Text

[72] Hiuge-Shimizu A, et al.

[73] ↵
Sekimoto R, et al.
(2012) High circulating levels of S100A8/A9 complex (calprotectin) in male Japanese with abdominal adiposity and dysregulated expression of S100A8 and S100A9 in adipose tissues of obese mice. Biochem Biophys Res Commun 419(4):782–789
.
OpenUrl CrossRef PubMed

[74] Sekimoto R, et al.

[75] ↵
Catalán V, et al.
(2011) Increased levels of calprotectin in obesity are related to macrophage content: impact on inflammation and effect of weight loss. Mol Med 17(11-12):1157–1167
.
OpenUrl PubMed

[76] Catalán V, et al.

[77] ↵
Nakatsuji H,
Kishida K,
Funahashi T,
Shimomura I, Senri Study II Group
(2012) Three-month treatment with pioglitazone reduces circulating levels of S100A8/A9 (MRP8/14) complex, a biomarker of inflammation, without changes in body mass index, in type 2 diabetics with abdominal obesity. Diabetes Res Clin Pract 95(3):e58–e60
.
OpenUrl CrossRef PubMed

[78] Nakatsuji H,

[79] Kishida K,

[80] Funahashi T,

[81] Shimomura I, Senri Study II Group

[82] ↵
Maeda N, et al.
(2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8(7):731–737
.
OpenUrl CrossRef PubMed

[83] Maeda N, et al.

[84] ↵
Ishii M, et al.
(2009) Sphingosine-1-phosphate mobilizes osteoclast precursors and regulates bone homeostasis. Nature 458(7237):524–528
.
OpenUrl CrossRef PubMed

[85] Ishii M, et al.

[86] ↵
Kikuta J, et al.
(2013) Sphingosine-1-phosphate-mediated osteoclast precursor monocyte migration is a critical point of control in antibone-resorptive action of active vitamin D. Proc Natl Acad Sci USA 110(17):7009–7013
.
OpenUrl Abstract/FREE Full Text

[87] Kikuta J, et al.

[88] ↵
Maeda N, et al.
(2001) PPARgamma ligands increase expression and plasma concentrations of adiponectin, an adipose-derived protein. Diabetes 50(9):2094–2099
.
OpenUrl Abstract/FREE Full Text

[89] Maeda N, et al.

Main menu

User menu

Search